CONCEPTUALIZATIONS OF EFFECTIVE CHEMI CAL DBMONSTRAT ION TEACHING AMONG EXPERIENCED AND NOVICE CH~t1! CA:~, DEMONS TRATORS AND THE INFLUENCE OF INTENSIVC INSERVICING by Christian P . Clermont Di ssertation submitted to the Faculty of the Graduate School of the University of Maryland in partial fulfillment of the requirements for t he degree of Doctor of Phi l osophy 1989 !) i ; Advisory Committee : 1\/ I Dr. Joseph S. Krajcik - Dissertation advisor Dr . Jon M. Bel lama .- Dr . Hilda Barko .I Dr. Neil A. Davidson l~ Dr . William G. Hol l iday /, ! I ' J I 0 Copyright by Christian Peter Clermont 1989 APPROVAL SHEET Title of Thesis: CONCEPTUALIZATIONS OF EFFECTIVE CHEMICAL DEMONSTRATION TEACHING AMONG EXPERIENCED AND NOVICE CHEMICAL DEMONSTRATORS AND THE INFLUENCE OF INTENSIVE INSERVICING Name of candidate: Christian P. Clermont Doctor of Education, 1989 Ph.D. Thesis and Abstract Approved: r. Joseph S. Krajci Assistant Professor Department of Curriculum and Instruction University of Maryland Date Approved: I I ABSTRACT Title of Dissertation: CONCEPTUALIZATIONS OF EFFECTIVE CHEMICAL DEMONSTRATION TEACHING AMONG EXPERIENCED AND NOV I CE CHEMICAL DEMONSTRATORS AND THE INFLUENCE OF INTENSIVE INSERVICING Chr istian Peter Clermont , Doctor of Philosophy , 1989 Dissertation directed by : Dr. Joseph S. Krajcik Assistant Professor Department of Curriculum and Instruct ion This study examined the pedagogical content knowl edge (PCK) of two groups of physical sci e nce teachers identified as experienced and novice chemical demonstrators. Science teachers, as do other teacher s , use their pedagog ical content knowledge to make decisions on how to teach very specific subject matter topics to students of various ages and abilities . Given the importance of thi s knowl edge system for effective sc i ence teaching , this study also examined the influence of an intensive chemical demonstration workshop (a NSF summer institute program) on fostering pedagogical content knowledge growth among a group of eight novice chemical demonstrators . A clinical interview, consisting of a critical-stop task and a semi - structured interview , served to probe the exper i enced and novice chemical demonstrators ' PCK . The critical-stop task required subjects to view videotaped chemical demonstrations and stop and di scuE~ the tape at segments they perceived contributed to effective o,- ineffective chemical demonstrating . The clinical in~-~1~, r .' 'J :f ,_,? :t1.sed on teachers' pedagogical knowledge of demonstrating der : ~:..::,r.: ;.~ ~ r pressure . Domain , taxonomic , componential, and theme analyses and several quantitative content analyses were conducted on the verbal data . The data indicated that experienced chemical demonstrators possess quantitatively greater and qua l itatively richer ped3gogical content knowledge for demonstrating basic chemical concept.s than novice chemical demonstrators. Experienced demonstrators , unlike nov i ces , possess a large body of knowledge about chemical demonstrations; chemical demonstration variations , and accompanying inquiry strat~yies for presenting subject-matter topics to middl e school students . The study further showed that the two-week chemical demons tration workshop produced an increase in the number of chemical demonstrati on s and demonstration variations novice chemical demonstrators could discuss on the targeted concepts , density and air pressure. The novice demonstrators also became more cognizant of the complexity of several chemical demonstrations , how these complexities could interfere with learning , and how simpl ified variations of these chemical demonstrations could promote science concept learning. After the workshop , novices' verbalizations also conta i ned fewer references to pedagogically unsound chemical demonstrations on the targeted concepts . Many of these changes brought novices cloaer t o th0 characteristics of experienced chemical demonstrato~s? PCK . ii DEDICATION To Brenda , Mom, and Dad. iii ACKNOWLEDGEMENTS A doctoral dissertation study is an undertaking that is completed only through the supportive help of a numbe r of caring people. I would like to acknowledge several of the key players in making this study possible. First, I would like to thank Dr. Joseph Krajcik, my Lesearch advisor, for hi s tremendous encouragement and support in helping me begin , conduct, and complete this study. Without his energetic enthusiam and insightful guidance, this study would not have been possible. Dr. Krajcik has become a well - re spected friend and professional colleague during the dissertation phase of my graduate program at the University of Maryland. I would also like to extend my gratitude to Dr. Hilda Bod.a f'):r her helpful critiques during various stages of this study. He r expertise in qualitative research helped me shape and r e f i ne the j ~t a gathering strategy and analysis methods. To Dr. Jon M. Bellama for his leadership in the ICE progran , support of my involvement in this program, and dedication tu t he professional development of science teachers. I would further like to express my thanks to Dr . Neil OY : :- ') Experienced and Novice Chemical Demonstrat0rs .?. :i "- 14 Average Number of Pedagogical Enhancements Mad~ by Experienced and Novice Chemical Demonstrat ors According to Pedagogical Knowledge Cate~ory t53 xiii 15 Frequency of AIR PRESSURE Demonstrations Cited by Experienced and Novice Chemical Demonstrators 157 16 Frequency of DENSITY Demonstrations Cited by Experienced and Novice Chemical Demonstrators 158 17 Theme Analysis of Critical-Stop Discourses According to Pedagogical Knowledge System 185 18 Frequency of AIR PRESSURE Demonstrations Cited by Pre- and Post-Workshop Novice Chemical Demonstrators 212 19 Frequency of DENSITY Demonstrations Cited by Pre- and Post-Workshop Novice Chemical Demonstrators 213 20 Frequency of AIR PRESSURE AND DENSITY Demonstrations Cited by Experienced and Novice Chemical Demonstrators 217 xiv LIST OF FIGURES Figure# Title Page No. 1 Model of Pedagogical Reasoning 28 1 CHAPTER 1 INTRODUCTION "How should I teach it?" This simple question f i ll s the mind of many beginning science teachers on almost a daily basis. Many reflective science teachers also ask themselves this question in an effort to improve the quality of their science t eaching. This question emerges whenever science teachers plan l essons on science concepts that students find difficult to learn. Thinking through this planning question provides a major challenge to both beginning and experienced science teache rs whose profess ional responsibility includes the goal of taking their comprehension of a science discipline and making it comprehensible to learners. According to Shulman (1 986 , 1987), one of the characteristics of teachers that distinguishes them from content specialists is their knowledge of teaching, called pedagogical content knowledge . Thi s body of knowledge represents an integra tion of content and pedagogy that provides teachers with an understanding of how particular s ubj ect matter topics, problems , and i ss ues are organized, represented, and adapted to the diverse interests and abilities of l earners , and then presented for instruction. This body of knowledge plays a vit a l role in teacher planning of specific t opics in a di scipline . Most research on teachers' pedagogical content ;~;-,o,-,ledge ha s been conducted in the context of how beginning teachers l2Jrn to transform their knowledge of subject matter into a form s uite'.-. }_; for tr~aching (Shulman 1986, 1987; Wilson & Gudmundsdottir, 1986; h:i.lson, Shulman , & Richert, 1987). These studies have focused on how classroom t eaching experiences and the process of pedagogical reasoning promote 2 pedagogical content knowledge growth. Current findings seem to indicate that pedagogical content knowledge growth is s l ow and largely dependent on the motivation , creativity , and reflective thinking skills of the beginning teacher. This study examined pedagogical content knowledge growth resulting from an intensive two-week i nservi ce intervention designed to help teachers develop their chemical demonstration teaching skills. The findings of this study may help determine whether knowledge growth in teaching can be accelerated through highly-focused, intensive inservice efforts . To help establ ish a frame of reference , the pedagogical content knowledge growth of novice chemical demonstrators was contrasted with the pedagogical content knowledge of experienced chemical demonstrators. The inservice program of interest to the present study is the Institute for Chemical Education (ICE) Workshop B: Chemistry Supplements for Pre-High School Classes . This NSF-sponsored program provided two weeks of intensive training in chemical demonstrating. The central goal of the inservice workshop was to increase elementary, middle , and high school science teachers ' use of chemical demonstrations (Bell , 1987) . Two cognitive research methods were used to probe demonstrator s ' pedagogical content knowledge (PCK) and general pedagogica l knowledge (GPK) of effective chemical demonstration teaching. The s8 methods included a think-aloud task and a semi-structured iri':.e ;:ni.ew . Such methods have been effectively used in other cognitL;?e studies probing teachers' thoughts and knowledge structures (Berli~2r , 1987a; Borke & Livingston , 1988 ; Elbaz, 1981; Peterson & Comeaux , 190"/ ; Yinger & Clark , 1982 ). 3 Purpose This study had a two-fold purpose. First, this study explored the nature of science teachers' pedagogical content knowledge and general pedagogical knowledge (Shulman , 1986 , 1987) as it relates to effective demonstration teaching. It specifically compared experienced and novice chemical demonstrators' knowledge of how to demonstrate basic chemical concepts at the middle school level. The chemical concepts examined were density and air pressure. Second, this study investigated the effects of a two-week intensive chemical demonstration workshop on novice demonstrators' pedagogical knowledge growth. This growth was gauged by comparing novice demonstrators' pre- and post - workshop clinical interview responses. Changes in their response, together with a comparison of experienced demonstrators' clinical interview responses, helped determine whether pedagogical content knowledge and general pedagogical knowledge growth can be acce l erated among elementary and secondary school science teachers participating in an intens ive, skills-oriented inservice workshop. Research Questions Three major research questions are addressed i;1 L1?.is study: 1. What are the domains of knowledge that chara~t~,:i.ze experienced and novice chemical demonstrat ,/ -: i' tJ? '.rlagogi cal discourses on effective chemical demonstrat :~0 11 tecching? 2 . What are the commonalities and distinguishing characteristics of experienced and novice chemical demonst rator s ' discourses 4 on effective chemical demonstration teaching in the identified domains? 3. How does an intensive (two-week) , skills-oriented (chemical demonstration) teacher workshop alter novice chemical demonstrators' evaluative judgments and knowledge of effective chemical demonstration teaching of targeted chemical concepts? Rationale for the Study A Theory-Based Rationale for Examining Teachers' Pedagogical Knowledge Why examine teachers' pedagogical knowledge (i.e., their PCK and GPK)? Over the past decade, researchers have given greater attention to teachers ' thought processes and knowledge structures as part of a broader effort to understand the process of teaching. This research has been guided by the premise that what teachers do is influenced by what they know and how they think about teaching. Several reviews on teacher effectiveness , for example , have stressed the importance of linking teachers ' actions and teachers ' cognition (Clark & Peterson, 1986; Gleissman & Pugh , 1981; MacLeod, 1977 ; Sacerdoti, 1977; Shulman, 1986). Such an integrated view of teaching coincides with schema theory in cognitive psychology , which suggests that teachers' mental processes and associated knowledge systems strongly iu f l1 ~ cce behavior in successfully performing complex tasks. Research c0111.(uct2d by Leinhardt and Greeno (1986) and Shulman (1987 ) sugge2~s t t1 ~t effective teaching is a complex cognitive skill involving nurr.r.:: ,:c.::1. '.' i:;1ought processes that interact with various knowledge systtm~ - Teachers frequently rely on a form of knowledge ( 5~ led pedagogical content knowledge . According to Shulman (1986 , 1987) 5 pedagogical content knowledge refers to a unique body of knowledge that teachers possess that allows them to make subject matter content comprehensible to students of various ages and abilities. This knowledge system grows gradually as a result of classroom teaching experiences and the process of pedagogical reasoning. It plays a vital role in teachers ' pre-active and interactive teaching. Although some research exists on the pedagogical content knowledge of beginning teachers in English and social studies, a paucity of research currently exists regarding the nature and growth of this knowledge system among science teachers. This study examines science teachers ' pedagogica l content knowledge and the influence of an intensive , inservice chemical demonstration workshop on fostering PCK growth among novice chemical demonstrators . Rationale for Examining Experienced and Novice Science Teachers Studies that attempt to understand the cognitive processes and relevant knowledge systems associated with performance on complex tasks frequently examine expert-novice differences . Comparisons of this nature have been effective in identifying the particular thought processes and knowledge structures associated with competent performance in a variety of disciplines ; e.g. , math , science , engineering , and chess (Bloom, 1986 ; Chi , Glaser , & Rees , 1981; Gick , 1986; Larkin , McDermott, Simon , & Simon , 1980) . Studi es that have probed the cognitive aspects of ~~~~jing have also examined expert/novice teacher differences , pa;:::i. ?1:_ :)_.:-1 ,.<'.::1 with respect to teachers' thinking about students, lesson s t =ucture, and classroom events (Berliner , 1986; Calderhead, 1983; H0~sner & Griffey , 1985; Huberman , 1985; Leinhardt & Greeno , 1986). These studies have 6 provided detailed accounts of the routine behaviors and knowledge systems characteristic of expert teachers. The rich descriptions generated by these expert/novice teacher studies make it possible to infer the knowledge structures, or schemata , that are unique to highly ski lled and master teachers (Berliner , 1986, 1987a; Leinhardt & Greeno, 1986). In addition, these studies generate hypotheses about the mental processes and knowledge systems that provide a basis for effective teaching (Borko & Shavelson, in press; Clark & Peterson , 1986; Fogarty et al., 1982; Peterson & Comeaux , 1987). By probing experienced and novice teachers ' thinking and the various components of their professional knowledge base, it is possible to identify a set of cognitive characteristics unique to experienced and novice teachers. In keeping with this goal, this study sought to examine the nature of experienced and novice chemical demonstrators' pedagogical content knowledge and general pedagogical knowledge (Shulman, 1986, 1987; Wilson, Shulman , & Richert, 1987) needed to demonstrate basic chemical concepts . The differences that emerge could provide a set of "neutral data" or standards that Yarger and Galluzzo (1983) and Joyce and Showers (19 80 ) indicate are needed to evaluate the level of impact of teacher education programs. The data obtai ned may also help identify some of the substantive needs among practicing science teachers, which together with the assessment of teachers' preceived needs, provide two important strategies in the development of effective inservice programs (Evans , 1987; Jones & Hayes, 1980 ; Yarger & Galluzzo, 1983). Finally , if Shulman ' s (1 987 ) suggestion for the professional testing of teachers incl ude less emphasis on the generic aspects of 7 teaching and greater emphasis on the teaching of specific subject matter topics , then a thorough analysis of what experienced and novice teachers actually know about teaching specific science concepts is warranted. Rationale for Examining Science Inservice Workshops The 1980's has seen a remarkable interest in science teacher inservice education. This interest comes from leaders in education as well as those in political circles seeking to improve science education across America (Evans , 1987). Thi s interest has been tied to several factors, including reported declines in student achievement and interest in science (Rakow , Welch, & Hueftle, 1984; Yager & Penick, 1984) and a growing shortage of qualified science teachers in this country (Bethel , 1985; National Science Board, 1983; Yankwich , 1984). Educators generally agree that students' science achievement scores and attitudes towards science will not improve much if the quality of teaching does not improve substantially (DeRose , Lockard, & Paldy, 1978; Medley, 1982), and that science teaching will not improve much without dramatic improvements in teacher education (Lanier , 1986). At the present time, federal support exists for the inse rvice training of science teachers (Evans , 1987) as evidenced by the revitalization of NSF s ummer institutes (Bell , 1987; Lippincott, 1985; O'Brien, 1987). These programs reflect the convictions of leaders in both educat ion and government that the quality of science t2aching can be rapidly improved by concentrating on inservice teacher training. Amid the numerous benefits cited for having federally-f unded inservice programs, research continues to fall short in providing 8 conclusive evidence for the effectiveness of these programs on improving teacher performance (Joyce & Showers, 1980; Wade, 1984 - 1985; Yarger & Galluzzo, 1983). This situation exists, in part, because of the difficulty in conducting sound experimental studies in inservice teacher education and also because of the difficulty in finding acceptable standards and procedures for comparing the dive r s ity of inservice studies found in the literature (Yarger & Galluzzo, 1983). Joyce and Showers (1980) have noted that many inservice research studies do not meet acceptable standards of evaluation, such as the collection of pretraining observations of teaching to determine entry level skills. A majority of the evaluations use client perceptions of a variety of inservice activities for evaluation purposes. Few studies provide concrete evidence of workshop effects on teachers' pedagogical knowledge gains. Most inservice evaluations simply represent statements of participant satisfaction, which the evaluators use to determine the success of the program (Wade, 1984-1985). Such studies do little in the way of directly assessing knowledge growth in teaching (Shulman, 1986) or improvements in the use of complex teaching skills (Yarger & Galluzzo , 1983). This study specifically examined science teachers' pedagogical content knowledge growth resulting from participation in an intensive chemical demonstration workshop. Growth was gauged using cognitive resea rch methods which, in turn, became the means of evaluating the inserivce wo~kshop. Thi s study may therefore suggest the usefulness of cognitive r.esearch methods for a more objective assessment of workshop succ1:,~;.s . Examining the influence of an intensive (two-week ), ski ll s- oriented (chemical demonstration) workshop may also provide educators 9 with evidence that inservice summer institutes with certain design characteristics can stimulate pedagogical content knowledge growth among practicing science teachers. Shulman ' s model of learning to teach , which addresses pedagogical content knowledge growth , was born out of numerous case studies of beginning teachers learning to teach specific subject-matter content to their students (Shulman, 1986, 1987 ; Wilson, Shulman , & Richert, 1987 ) . The study conducted by this researcher may show that Shulman ' s model of learning to teach may also be appl icable in the context of science teacher inservice education. Rationale for Examini ng Science Demonstration Teaching National studies of science teaching at the precollege level have documented the central role of the textbook and the lecture-discussion teaching strategy as predominant features of science teaching in U. S. schools , with hands-on inquiry-oriented labs and activities receiving less attention (Harms & Yager, 1981 ) . An overreliance on teacher- centered approaches, however , keeps students from going beyond the mastery of a series of facts (Yager & Lunetta , 1984 ) and may contribute to students ' less than positive attitudes toward science (Hurd , 1982 ; Yager & Penick , 1984 ) . Research suggests that skillfully presented science demonstrations might be as effective as laboratory approaches in promoting student cognitive growth because of its potential to be delivered as a guided discovery teaching strategy that increases pupil attention and task involvement (Beasley , 1982; Garrett & Roberts , 1982 ). It is , therefore , important to provide science teachers with opportunities to acquire these skills and to increase their use in the science classroom . 10 Rationale for Examining the Target Concepts, Density and Air Pressure Teachers' pedagogical knowledge was examined with respect to the demonstration teaching of two basic chemical concepts, density and air pressure. These concepts were selected because of their frequent inclusion in the elementary/secondary school science curriculum, their association with other fundamental concepts in chemistry, their abstract nature (Hewson & Hewson, 1983; Swamy, 1986), and the variety of chemical demonstrations available for illustrat ing these concepts. Hewson and Hewson (1983) have shown that middle and secondary school s tudents often hold alternative conceptions of density that are at variance with the scientifically acceptable conceptions, equating density and relative density with terms such as mas s , weight, or denseness (crowdedness ). St udents also possess several mi sconceptions regarding the behavior of gases (Swamy, 1986). These alternative conceptions may be confronted through conceptual change strategies that include the use of chemical demonstrations. Significance of the Study Thi s study was designed to contribute to an understanding of science teachers' pedagogical content knowledge and factors that contribute to its growth. By documenting subs t antive differences between expe rienced and novice science t eachers ' pedagogical content knowledge and the influence of intensive inservicing on pedagogical content knowledge growth, this study may be useful in several ways. First , this study may s how the applicability of Shulman's (19 86 , 1987) model of t eaching to sc i ence education. It may provide evidence that science teachers possess a body of professional knowledge, called 11 pedagogical content knowledge, and that the breadth and depth of this knowledge system is related to their exper i ence in teaching science and to their knowledge of science. By highlighting the similarities and differences between experienced and novice chemical demonstrators' knowledge of chemical demonstration teaching, this study may contribute to the process of validating Shulman's model of teaching (e . g ., the components of teachers ' pedagogical content knowledge system and the factors that contribute to its growth ). Such findings would contribute to a better understanding of the professional knowledge base of experienced science teachers (Valli & Tom, 1988). Second , such a study may also contribute to the body of literature examining cognitive differences between experts and novices in various disciplines (Bloom, 1985; Gick , 1986; Chi , Glaser , & Rees , 1981; Larkin, McDermott, Simon , & Simon , 1980 ), including education (Berliner , 1986, 1988; Borko & Livingston, 1988; Leinhardt , 1983) . Documenting cognitive differences between experienced and novice chemical demonstrators , for example , may lend s upport to Leinhardt and Greeno ' s (1 986 ) assertion that effective teaching involves complex cognitive skills and knowledge structures that are ill formed in novice t eachers and well developed in expert teachers. Third , this s tudy may identify some of the cognitive outcomes of an intensive , skills-oriented inservice workshop for sc i ence teachers. In the context of chemistry teaching, this study may help determine the nature of pedagogical content knowledge growth as it occurs during two weeks of chemical demonstration training. Knowledge growth would be reflected in novice demonstrators ' think-aloud discourses and interview responses to probes of effective chemical demonstration 12 teaching. Fourth , this study may provide support for the usefulness of cognitive approaches (e.g., think-aloud tasks and semi-structured interviews) , for evaluating skills-oriented workshops. In the past, these approaches have been used as tools in educational research, not inservice program evaluation. Therefore , this study may suggest the utility of cognitive research methods for evaluating skills-oriented workshops. This information would be useful to inservice program implementors seeking to use more direct methods for evaluating teachers and the success of their programs . La s t, by documenting pedagogical content knowledge growth in an inservice context , this study may help provide funding agencies and decision makers with data on the importance of supporting discipline-specific (e . g., biology , chemistry , physics) inservice workshops that focus on teachers' instructional skills. Beyond these theoretical , methodological, and policy-making contributions , the most important beneficiaries of science teachers ' pedagogical content knowledge growth are the students themselves . As science teachers l ea rn how to better teach complex science concepts , students ultimately benefit by having a greater understanding of science and the world in which they live. Assumptions Thi s st udy makes several theoretical and methodological ass umpti ons : 1. Shulman's (1986 , 1987) model of teaching (Wilson , Shulman , & Richert , 1987) provides an adequate mode l for describing the 13 various components of teachers' pedagogical content knowledge system (seep. 28 ), and knowledge acquired in at least one of the components of the model reflects meaningful PCK growth. 2. The nature of experienced and novice chemical demonstrators' pedagogical content knowledge can be reasonably inferred from the verbal statements they elicited during the clinical interview. 3 . Teachers evaluated the videotaped chemical demonstrations with candor and professional integrity; therefore , the think-aloud discourses represent a reasonably accurate reflection of teachers' knowledge and judgments of effective chemical demonstration teaching. 4. The semi-structured interview contains questions that are sufficiently non-directive and do not distort teachers' reporting of their pedagogical knowledge , judgments , and science concept understanding . 5 . The experienced and novice chemical demonstrators selected for thi s study are representative of the two populations under investigation . Furthermore , the verbalizations obtained from the experienced and novice demonstrators in this study are assumed to be comparable to those that would be obtained from other chemical demonstrators possessing similar levels of demonstration experienced and content knowledge . Scope and Li mit ati ons The number of chemical demonstrators available to this researcher and the time available to conduct the clinical interviews delimited the scope of this study . A few conceptual and methodological 14 shortcomings also existed that placed restraints on what could be known about the population under investigation and about the impact of the inservice workshop. These limitations could not be overcome during the course of this investigation. The limitations and delimitations of this study included the following : 1 . Experienced and novice chemical demonstrators ' analysis of the videotaped chemical demonstrations and their response to the follow-up interview may not fully tap teachers ' pedagogical knowledge (i.e. , their PCK and GPK ) about how to demonstrate the targeted chemical concepts: density and air pressure. 2 . Descriptions of experienced and novice chemical demonstrators' pedagogical knowledge of how to demonstrate the concepts , density and air pressure , to students at the middle school level do not necessarily reflect the nature of their pedagogical knowledge in other chemistry content areas. 3. The generalizability of the findings is somewhat limited by the small number of experienced and novice demonstrators examined and the possible nonrepresentativeness of the voluntary populations. 4. The experienced and novice chemical demonstrators examined in this study differed not only in their experience in conducting chemical demonstrations (i . e. , confidence and weekly use in conducting chemical demonstrations) , but also in their chemistry content knowledge. This difference in content knowledge undoubtedly contributed to some of the observed differences in experienced and novice chemical demonstrators ' clinical interview verbalizations. 15 Definition of Terms The following theoretical and operational definitions apply to this study . Chemical Demonstration - "a type of interactive teaching s trategy where the teacher presents and manipulates a real chemical system or a model thereof to introduce , illustrate in concrete form, and/or challenge students ' understanding of a particular principle or concept by engaging them in observing , questioning , and reasoning. As such , the demonstration includes both the advanced preparation and follow - up activities. " (O' Brien, 1987) Critical Feature - a teaching behavior that is discussed by a chemical demonstrator during the critical-stop task. This behavior , as displayed by a videotaped teacher performing a chemical demonstration, is judged by the viewer as contributing to or hindering effective chemical demonstrating. Critical Incident - the occurrence of a critical feature. The emphasis is on the time and place a critical feature or t eaching behavior was manifested. Occasionally , the terms "critical incident " and "critical feature " are used interchangeably to describe the videotaped teachers ' demonstration skills. Experienced Chemical Demons trator, (E) - an expe rienced chemi s try or physical science teacher who regularly incorporates chemical demonstrations into his/her teaching and who has conducted teache r training workshops and outreach programs using the chemical demon stration teaching strategy . General Pedagogical Knowledge , (GPK) - knowledge of gene ric t eaching skill s . Thi s knowledge includes principles or me thods of cl ass r oom 16 organization and management as well as knowledge of the learner and their backgrounds (i.e., individual differences among students) (Shulman , 1986). ICE Workshop B - an intensive two-week chemical demonstration workshop sponsored by the Institute for Chemical Education and held at the University of Maryland, College Park , during the summer of 1987. In this workshop , participating teachers observed chemical demonstrations being modeled and discussed by the workshop staff. As part of the workshop, teachers also practiced and presented chemical demonstrations before peers , middle school students, and ICE workshop staff. Institute for Chemical Education (ICE ) - '' an institute centered at the University of Wisconsin-Madison whose objective is to promote scientific literacy and chemical education at all levels, from elementary through advanced university studies. ICE activities include sponsoring a variety of teacher workshops and programs at various sites around the country. " (O ' Brien, 1987) Novice Chemical Demonstrator , (N) - an elementary , middle school, or high school science teacher who has had at least one year of chemistry or physical science teaching experience , who possesses minimal competency in demonstration teaching as suggested by his/her low self-reported confidence in chemical demonstration teaching , and who incorporates few chemical demonstrations into his/her own teaching (frequency of one or less per week). Pedagogical Content Knowledge , (PCK) - "that special amalgam of content and pedagogy that is uniquely the province of teachers, their own special form of professional understanding " of how to 17 teach specific sub j ect-matter topics (e.g ., density ) to students of various ages and abilit i es. "It represents the blending of content and pedagogy into an understanding of how particul ar topics, probl ems , or issues are (i ) organized , (ii ) represented , and (iii) adapted to the diverse interests and abilities of learners , and (iv) presented for instruction." (Shulman , 1987, p. 8). It includes (i ) how to present the idea , (ii) what materials to use, (i i i ) how to sequence those materials , and (iv ) what misconceptions the students have about the topics of interest (Wil son & Gudmundsdottir , 1986). In this study , these definitions of PCK apply exc lus i vely to science teachers and to their knowledge of how to perform effective chemical demonstrations on t he concepts density and air pressure at the middle school level. Pedagogical Knowledge - the body of knowledge that includes pedagogical content knowledge and general pedagogical knowledge (knowledge of ge ner ic teaching skills ) . Teacher ' s Evaluative Judgments - one form of mental process ing of classroom teaching information performed by teacher s . In this s tudy , this cognitive activi ty permits teachers to provide critical analysis of an observed chemi cal demonstration teaching episode as being ineffective , effective, or neutral. It r eflect s both subjective and objective aspects of teachers ' pedagogical knowledge of effective chemica l demonstration teaching . Chapter Summary This chapter provided a brief overview of the study. It presented a case for examining sc i e nce t eachers ' pedagogical content 18 knowledge and its growth during intensive inservicing , assumptions and limitations in conducting such a study , and the potential contribution this study makes to the field of science education. Chapter 2, which fo l lows, presents a review of the literature with respect to the theoretical and methodological underpinnings of this study. 19 CHAPTER 2 REVIEW OF THE LITERATURE The fol l owi ng rev i ew of the l iter ature examines the professional development of practicing science teachers. The review begins by examining research on teachers ' thinking , with particular emphasis on teachers ' pedagogica l knowledge . This is followed by a discussion of qua l itative methods used to explore teachers ' knowledge of teaching and the techniques used to analyze verbal data derived from these methods. The review culminates with an examination of the impact of i nservice programs on fostering teachers ' professi onal development with special consideration given to the impact of past National Science Foundation inservice programs on promoting science teachers ' teachi ng skills . Research on Teachers' Thinking Teaching is a complex process. To explore this complexity , behavioral studies on teaching are increasingly being complemented by studies exami ning teaching from a cogni t ive perspective . Because of this trend , research on teachers ' cognition has grown into a highly visible body of research ca l led research on teachers ' thinking. Thi s body of research has a common concern with the ways in which knowledge is actively acquired and used by teachers and the circums t ances that affect its acquisition and employment (Calderhead , 1987 ). Research on teachers ' thinking consists of t hree broad areas of i nvestigation . One area probes the source of teache rs ' unde rstandings , contrasting knowledge gained from class r oom experi ence 20 with knowledge acquired through more formal modes of teacher preparation (Pigge & Reed , 1985; Queen & Gretes , 1982). The second area of re search deals with teachers' decision-making and problem-solving activity during pre-active and interactive teaching , (Clark & Peterson , 1986; Shavelson & Stern , 1981). And the third area of research involves studying the content and structure of teachers ' professional knowledge base (Shulman , 1986, 1987; Berliner, 1987a ). Because this study is concerned with science teachers' pedagogical knowledge of demonstrating basic chemical concepts , the literature review that follows will focus primarily on this third area of teachers ' thinking . The review will give particular emphasis to the nature of teachers ' pedagogical knowledge , the influences on its formation , how it is applied to the analysis of teaching situations , and how it becomes embedded in teachers ' actions. Teachers ' Professional Knowl edge Base Researchers have begun to make concerted efforts at understanding the professional knowledge base of teaching. Much of this research has been prompted by the recognition that teachers possess extensive and speciali zed knowl edge about children , curricula , classroom organi zation , and methods of instruction . This knowledge helps them relate to childr en , manage classrooms , decide how to teach a particular topic, maintain children's interest , and instruct them . Teachers use their spec ialized knowledge to guide their actions and cope with a constant ba rrage of complex s i t uat i ons while teaching. This knowledge a lso influences the development of teachers ' classroom routines and r esponses to classroom events (Calderhead , 1987 ). -- ------- 21 A Theoretical Model A theoretical model has recently been proposed by Wilson, Shulman , and Richert (1987 ) to describe the structural components of the professional knowledge base of teaching. This model is referred to as a "logical model " derived l argely from the general educational literature but also from a series of studies conducted at Stanford University seeking to understand knowledge growth in teaching. The Stanford studies indicate that beginning and experienced t eachers draw upon many different knowledge systems as they make decisions about the content of the courses they teach. The model proposed by Wilson , Shulman, and Richert (1987) that describes the professional knowledge base of teaching includes the following components: (1 ) knowledge of subject matter , (2) knowledge of other content , (3) knowledge of learners , (4) knowledge of curriculum, (5) knowledge of educational aims , (6) knowledge of educational contexts , (7) general pedagogical knowledge (GPK) , and (8) pedagogical content knowledge (PCK). As teachers make pedagogical decisions they must draw upon their content knowledge, i .e. , their knowledge of subj ect matter. Within this particular knowledge system teachers must understand nume rous facts and concepts in the discipline they teach. They mus t also have knowledge of the substantive structure of the discipline (Schwab , 1964) , that is , how the fundamental principles of the di scipline are organized and related to one another. In addition, teache r s must possess knowledge of the syntactic structure of the discipline whi ch provides the rules that guide inquiry in the di scipline and he lp establish evidence and proof. 22 Teachers also draw upon their knowledge of other content that i s not within the scope of the discipline they are teaching. Examples of this knowledge include knowledge derived from other disciplines, current events , popular television programs, or even personal experiences that could be introduced into a lesson and linked to a specific discipline topic. Teachers also have knowledge of l earner s that includes a knowledge of student characteristics and cognitions as well as knowledge of motivational and developmental aspects of how students learn (Wilson , Shulman, & Richert , 1987). Teachers frequently use their curricular knowledge to make pedagogical dec isions. This body of knowledge consists of an understanding of the programs and materials available for teaching particular topics and subjects at a given level. In addition , teachers have knowledge of educational aims, goals, and purposes that enter into their teaching and planning decisions . Shulman (1987) also discusses the importance of yet another category of teachers ' professional knowledge base, namely, knowledge of educational contexts. It refers to knowledge ranging from the workings of student groups and classrooms, to the governance and financing of school districts , to the character of communities and cultures. Each of these factors enter into t eachers ' pedagogical decisions which , in turn, influences the kind of instruction that occurs in the classroom. Teachers regularly make use of their general pedagogical knowledge during pre -active and interactive teaching. Thi s knowledge consists of pedagogical principles and techniques that are not bound by topic or subject matter , techniques that have been the focus of 23 much research on teaching (Shulman, 1986). It includes knowledge of (i) verbal and nonverbal communication skills, (ii) techniques for presenting information such as using advanced organizers and determining prior knowledge, and (iii) questioning skill s such as using different levels of difficulty in questioning, different types of questions, wait time, and providing feedback (Arends , 1988 ). 1 Teachers' general pedagogical knowledge also includes an unders tanding of generic principles or methods of classroom organization and management such as (i) maintaining order, (ii) maintaining equity in the distribution of teacher time among students , (iii) dete rmining the number of students who can work together on a given task, (iv) regulating the amount of student-student interaction , and (v) gauging timeliness in teacher-student inte ractions (Cohen , Intili, & Robbins, 1979). It also includes knowledge of procedures for (i ) controlling student movement and student talk, (ii) ensuring smoothness and momentum during instruction, (iii) cuing and s ignaling, (iv) managing inappropriate and disruptive behavior, (v) using rewards and punishment, and (vi) establishing rapport with students (Arends , 1988) . The preliminary findings of Shulman and hi s colleagues suggest that teachers also possess another type of subject matter knowl edge that is enriched and enhanced by teachers' various knowledge s ystems and is called pedagogica l content knowledge. It is a form of content knowledge that: ... embodies the aspects of content most germane to it s teachability. Within the category of pedagogical content knowledge I include , for the most regularly taught topic s 1 See Skill Locator Guide on p.xv -xvi of Arends (1 988 ). 24 in one's s ubject area , the most useful forms of r ep r esen- tation of those ideas, the most powerful analogies , illustrations, examples , explanations , and demonstrations - in a word, the ways of representing and formulating the subject that make it comprehensible to others . .. [It] also includes an understanding of what makes the l earning of specific topics easy or difficult: the conceptions and preconcept i ons that students of different ages and backgrounds bring with them to the learning. (Shulman , 1986, p. 9) [It is] that special amalgam of content and pedagogy that is uniquely the province of teachers, their own special form of professional understanding .... It represent s the blending of content and pedagogy into an understanding of how particular topic s , problems, or issues are organized, represented , and adapted to the diverse interests and abilities of learners , and presented for instruction. (Shulman, 1987 , p. 8) . Pedagogical content knowledge refers to a knowledge of how to teach the s ubject matter. It includes how to present the idea, what materials to use , how to sequence those materials, and what misconceptions the students have about the topics of interest (Shulman 1986 , 1987 ). This form of content knowledge is discussed below in greater detail because of its central importance to this r esearcher ' s study . Case Studies of Teachers' Pedagogical Content Knowledge The research literature provides several examples of the various facets of the term pedagogical content knowledge. These examples are taken from a growing body of casework based on experienced and novice secondary school teachers planning and teaching specific topics in English and history. Most of the reported examples are taken from studies conducted by researchers at Stanford who are giving special attention to knowledge growth in teaching. Two case studies are briefly examined to illustrate the construct ' pedagogical content knowledge .' Wil son , Shulman , and Richert (1 987 ) 25 have conducted numerous interviews on teachers preparing to teach a specific topic. One case study reported on an English teacher , who was required to teach Shakespeare ' s Julius Caesar to a class of ninth graders towards the end of the school year. In an attempt to introduce the play in a meaningful way, one that would take into account their interests and their level of intellectual sophistication, the teacher presents his students with a popular television program (Star Trek) and a scenario of how the students would confront the central character (Captain Kirk) whose adventurous campaigns are needlessly risking the lives of others (the crew). The analogy presented to the students illustrates one teachers' favored representation of the idea of moral conflict. It further represents this teachers ' transformation of the subject matter for teaching that is l ikely to initiate understanding on the parts of students. Wilson , Shulman, and Richert (1987) maintain that as beginning teachers experience pedagogical content knowledge growth they develop multiple representations of the subject matter they teach , and thus build upon their schema for teaching a given concept. Work based on the Stanford studies has fostered the idea that beginning teachers should develop a ' representational repertoire ' for the subject matter they teach because a multiplicity of connections renders understanding more durable and rich. The possession of variations of representations also permits instruction to be tailored and adapted to the various types of learners teachers confront. Another case study (Wilson , Shulman , & Richert , 1987) exemplifies a teachers ' efforts at transforming subject-matter for the purposes of teaching and provides information on the nature of 26 pedagogical content knowledge growth in teaching. The case is taken from another secondary school English teacher who was going to teach the concept of ' theme ' for the first time. When this teacher was asked during a pre-instruction interview how he would introduce this concept to his class of tenth grade boys , this teacher had several ideas, one of which included tracing a theme for his students after they have read a short story. In addition to exemplifying this term for students , the teacher a l so planned to use a metaphor he created (a baseball game) to help him communicate his tacit knowledge of theme and themat i c analysis to students. After teaching the concept of theme to his students , he was asked during a post-instruction interview what he knew about the concept of theme after having the experience of teaching it. It became apparent to him that his students had difficulty understanding how the concept of theme was related to patterns observed in a baseball game. In addition , they still did not understand how to find a theme in a text , the teachers ' major lesson objective . Wilson, Shulman , and Richert ' s (1987 ) case study furthe r documented how this beginning English teacher , though frustrated, searched for a better metaphor to help his students trace and understand theme. The teacher transformed the content again, this time focusing on the trail of a wounded animal . Through the process of planning , teaching , adapting the instruction, and reflecting on the instructional experience , this English teacher slowly added a new representation to his teaching repertoire , making his pedagogica l content knowledge richer and fuller. For this English teacher , the representation is a more refined understanding of the use of analogies I . 27 for instruction. The new knowledge acquired also included an understanding of some of the difficulties students have with understanding the concept of theme. 1 As novice teachers begin to transform the i r content knowledge for instruction suitable for a particular level of student , they engage in and develop their pedagogical content knowledge. Wilson, Shulman, and Richert (1987) indicate , however, that pedagogical content knowledge is not simply a repertoire of multiple representations of the subject matter . It is also characterized by a way of thinking called pedagogical reasoning. This type of reasoning facilitates the generation of subject matter transformations into teachable forms. It plays a pivotal role in fostering pedagogical content knowledge growth. In this study , the representations that experienced and novice chemical demonstrators have for the demonstration teaching of dens ity and air pressure provi de a perspective of these science teachers ' pedagogical content knowledge . A Mode l of Pedagogical Reasoning Based on numerous interviews with novice teachers Wilson , Shulman, and Richert (1987 ) developed a model for the process of pedagogical reasoning , a model which attempts to account for growth in teachers ' pedagogical content knowledge base. This mode l is depicted in Figure 1. 1 Precisely how this English teacher came up with the hunter metaphor is not stated in Wilson , Shulman , and Richert ' s (1987 ) article. One is left with the impression it was an act of creative thinking associated with reflection and l esson planning . 28 Instruction New comprehension Evaluation Reflection Figure 1 Model of Pedagogical Reasoning The process of pedagogical reasoning consists of six aspects related to the act of teaching: comprehension, transformation, instruction , evaluation , reflection, and new comprehension. The pedagogical reasoning process begins with comprehension. This requires a teacher to critically understand the substantive and syntactic structure of the content to be taught. In addition , it refers to an understanding of how the content is related to ideas in other domains. 29 The transformation process , the centra l component of the pedagogical reasoning model, occurs next . It describes the thought processes that enter into the lesson planning task . It involves the following five cognitive subprocesses : critical interpretation , representat i on , instructional selection , adaptation , and tailoring. These five subprocesses contribute vitally to the growth of teachers' pedagogical content knowledge system. The first of these subprocesses, critical interpretation , involves reviewing instructional materia l s in light of one ' s own understanding of the subject matter, including the identification of errors in a text as well as changes in understanding of a particular construct since the text ' s publication date . Once the instructional materials have been critically examined , teachers often consider alternative ways of representing the subject matter to students , the second step in the transformation process (see Figure 1) . Thi s may inc l ude the development of analogies and metaphors , or numerous othe r forms of representation to help convey to students particular subject-matter concepts or problem-solving strategies. Ideally , teachers should possess a representational repertoire that consists of metaphors , analogies , illustrations, activities , assignment s , examples , and other forms of representation that t eachers could use t o transform the content for instruction (Wilson , Shulman, & Richert , 1987 ) . A third aspect of the transformation process , in s tructi onal selection , refer s to the embodiment of representations in an instructional form or me thod (Shulman , 1987 ) . This process r equires teachers to draw upon a repertoire of instructional strategies of 30 teaching to present the chosen representation. This r eportoire include s a wide variety of teaching strategies such as l ecture , recitation, seatwork , cooperative learning, reciprocal teaching, inquiry methods, student projects , and several other instructional approaches. Adaptation , the fourth aspect of the transformation process , r efers to the process of fitting the subject matter representation to the characteristics of students in general. If students possess particular misconceptions about light that would interfere with learning, a teacher might present or introduce an activity about light in such a way as to clear up the misconception or in a way that would avoid reinforcing the misconception. Tailoring, the last process , involves adapting the materials and representations to specific students in one class rather than to the student population in general. Together, the four mental processes of interpretation , representation, adaptation , and tailoring produce a plan, a set of strategies for teaching a lesson, a unit, or a course. Instruction refers to the observable performance of the teacher during teaching. It includes all the teacher behaviors related to effective instruction in areas such as management, grouping, pacing, coordination of learning activities , explanation, questioning, and discussion. Evaluation is the teacher' s efforts at checking for understanding in their students during and after instruction. Teachers use two modes of evaluation : formal modes and informal modes. Formal modes of evaluation include the use of unit tests and quizzes; informal mode s include strategies such as questioning during interactive teaching. 31 Checking for student understanding and misunderstanding is usually tied to specific school subjects and to individual topics within a subject. Shulman (1987) , therefore , indicates that evaluation represents an important way in which pedagogical content knowledge is used. Teachers also engage in an evaluation of their own teaching through a process of reflection . Reflection refers to the process of learning from experience. It involves a mental recollection of the teaching and learning that has occurred over a given span of time, highlighting the accomplishments and shortcomings that have occurred along the way. The final phase of the pedagogical reasoning cycle i s a new comprehension of the purposes of instruction , the sub ject matter , the students , and the instructional strategies. Findings from the Stanford studies have shown that new comprehension, the last phase in the pedagogical reasoning model , is not always discernible after each unit of teaching (Wilson, Shulman, & Richert , 1987 ) . In some instances, enriched understanding in teaching of a specific topic grows slowly by accretion, wherea s in other instances a single experience promotes a significant leap in new comprehension. In many cases , teachers showed no changes for long periods of time . An understanding of the factors influencing these various growth patterns awaits further research. The two models presented above (the components of the professional knowledge base of teaching and the model of pedagogical reasoning , Figure 1) provide a theoretical base for r esearch on knowledge growth in teaching. Although these models have empirical support and are l ogica lly based, it is also clear that more de tail ed 32 conceptua l ization and testing of these models are still needed (Wi l son , Shulman , & Richert , 1987 ) . Such efforts would help contribute to a clearer understanding of how teachers perform the crucial task of transforming their content knowledge for the purposes of teaching and the factors that contribute to pedagogical content know l edge growth . Studies on Teachers ' Professiona l Knowledge Base This section reviews several studies that have examined the nature of teachers ' professional knowledge and influences upon it. The review will focus primarily on four knowledge sy st ems compri s ing teachers' professional knowledge base , namely , their subj ec t -matte r knowledge , pedagogical content knowledge , general pedagogical knowledge , and knowledge of students. Attention is first give n to recent studies on teachers' pedagogical content knowledge and how it is used during planning and interactive teaching . This is followed by a review of expert/novice teacher studies examining the othe r three know l edge systems that teachers frequently use in their t eaching. Factors Influencing Teachers ' Pedagogical Content Knowledge One question of particular interest is the role that t eache r s ' sub ject-matter knowledge plays in instruction. It can be s t ated as follows : "How do teachers use their knowledge of content for teaching? " A study by Hashweh (1987 ) examined this question by investigating the influence of teachers' subject-matte r knowledge on their transformation of the content into a form suitable for instruction. The study compared two groups of scie nce t eacher s wh o 33 differed in subject-matter knowledge as they prepared to teach identical material. The two groups of teachers were asked to teach two lessons, one involving content that they knew intimately, and one involving content they had never encountered. This situation was created by having a group of three biology teachers and three physics teachers each plan a biology and a physics lesson using selected chapters out of a popular biology and physics textbook during think-aloud interview sessions. Hashweh (19 87) observed that prior subject-matter knowledge affected the transformation of chapter organization. In situations where science teachers were presented with a chapte r in which the theme was meaningless or unsound from a disciplinary pe r spective , only the knowledgeable teachers rejected the chapte r structure and established their own organizing scheme. Unknowledgeable teache r s closely followed the chapter structure . Knowledgeable t eacher s were al so more prone to make additions that were consistent with their organizing theme for the chapter and deletions that were not essenti al for developing the theme. Hashweh attributed these planning behavi or s to the rich knowledge schema the knowledgeable teachers po s ses sed with respect to the targeted science concepts (subject-matte r knowledge ) and related science activities (pedagogical content knowledge )? Hi s findings suggest that the organizational and adaptation aspects of science t eacher s ' pedagogical content knowledge depended grea tly on teachers' subject-matte r knowledge . Teachers posses s ing a good unde rstanding of the s ub stantive and syntactic s tructures of a particular sub-di scipline of soc i a l stud i e s showed similar patterns in their cognitive abilities t o t ra nsform t he 34 subject matter into a teachable form (Wilson & Gudmundsdottir, 1986). Among beginning social studies teachers , subject-matter restructuring was limited to the content they knew, with very little emphasis on the restructuring of the content to include an interdisciplinary perspective (Gudmundsdottir, 1987). Although these beginning teachers possessed a novitiate form of pedagogical content knowledge, their experiences during the first year of teaching reportedly provided these teachers with some measure of pedagogical content knowledge growth. The most striking shifts in content presentations reflected a change in instructional selection. The lecture method gave way to story-telling encounters that included methods and activities for getting students engaged in learning by having them take on the interviewing role of cultural anthropologists. Most studies reported in the literature have examined how teachers' pedagogical content knowledge grows as a result of classroom experience. Little is known about how this knowledge system differs betwee n experienced and novice science teachers and how it is effected by intensive inservicing; hence , a need for studies in these areas exists. This review continues by examining what is known about experienced and novice teachers ' thinking in those areas of teachers' professional knowledge base that have received recent research attention. Exploring Teachers' Cognitions Through Expert - Novice Teacher Studies Researchers have recently begun to use the theories and methods from cognitive psychology to study the nature of pedagogical expertise . These studies tend to focus on the thought processes and knowledge structures of experienced or expert teachers and those of 35 minimally experienced or novice teachers. Studies involving experienced teachers typically examine individuals who have comple t ed four or more years of elementary or secondary school teaching (Hou s ne r & Griffey , 1985; Yinger & Clark, 1982). Studies involving expert teachers typically involve those who have had several years of teaching experience and have a reputation of teaching e xce llence that has been recognized by colleagues or school administrator s . Identifying expert teachers may also include the use of s tudent s tandardized test scores and validated observational measures used by independent observers (Berliner , 1986). Each of the se measures , however, has its limitations as an indicator of pedagogical expe rti se . Consequently, the distinction between 'experienced' and ' expert' teachers is not always straightforward , and as a r esult , the two t erms are often used interchangeably (Berline r, 1986; Pe t e rson & Comeaux, 1987). In the review that follows, an attempt is made to use the terms following the criteria discussed above. In a study of experienced and inexperienced (novi ce ) social s tudy t eache rs , Peterson and Comeaux (1987) examined t eache r s ' thinking during classroom ins truction in terms of their recall, r epresent ati on, and analysis of problem event s during interactive teaching. The th r ee cognitive abilities were probed by presenting teache rs with videot aped recordings of problem events associated with handing back t es t s , a class discussion, and cheating. Structured inte rvi ews that fo ll owed the videotaped viewings revealed that experienced t eache r s s howed greater r ecall of class room events, r elied more on procedural knowledge and principles in analyzing the problem epis odes , and gave more jus tifications for the ir deci s i ons and comment s . Novi ces t ended 36 to recall more of the surface or literal characteristics of classroom events and their solutions to classroom problems tended to be more simplistic and specific , conveying fewer under l ying pedagogical principles. From these findings , it was suggested that experi enced teachers have better developed knowledge structures or "schemata" for phenomena related to classroom teaching and learning than novice teachers . In a related study , Carter , Cushing , Sabers , Stein , and Berliner (1 988) noticed dist i nct qua l itative di fferences between expert and novice math and science teachers in terms of their ability to process visual information of math and science classrooms. When these teachers were presented with a series of slides taken from an actual math or chemistry class , experts appeared better able to (1) pe rceive and interpret the importance of one piece of visual information against another , (2) form connections among pieces of infor mation , (3) make sound inferences of classroom events , and (4) represent management and instructional situations in terms of meaningful problem units . The findings indicated that expert teachers pos ses s a rich store of classroom knowledge about students and events , and that they use that knowledge to understand and explain classroom phenomenon . These findings suggest that expert teachers posses s comparative ly richer schemata than novices for ascribing meaning to visual classroom information . Studies of experienced and novice teachers engaged in actual, rather than simulated, lesson planning and interactive teaching ta s ks showed other interesting cognitive differences. Housne r and Griffey (1985 ), for example, conducted an analysis of expertise as it 37 pertained to the planning and interactive decision making of experienced and inexperienced (or novice) physical education teachers. An analysis of teachers ' think- aloud verbalizations showed that the experienced teachers were better able to anticipate possible teaching situations and generate appropriate contingency plans to meet the demands of these situations than novice teachers , e.g. , what to do if students showed high ability or if there was extra time. The experienced teachers , thus, had a larger number of alternative instructional strategies and adaptations stored in clinical memory than did inexperienced teachers. The experienced teachers also made over twice as many instructional strategy decisions than novice s during their lesson planning task. In this regard , they gave considerably more attention to the establishment of rule s and routines for managing student activities and to giving students feedback that would facilitate skill acquisition. During interactive instruction , the experienced physical education teachers ' attended more to cues related to individual student performance , while novices attended more to cues related to student interest and overall class behavior. These differences in expe rienced and inexperienced teachers' general pedagogical knowledge , their thinking about instruction, and their processing of classroom information further reflects the enhanced cognitive abilities that experienced teachers have over their novi ce counterparts. Leinhardt and Greeno (1986) examined teachers' cognitions by looking at the activity structures by which ce rtain t eaching pl ans (e.g. , opening homework r eview , guided practice, transitions ) were carried out among expert and novice mathematic s teache r s . The study 38 focused on routines embedded in relatively common teaching activities. A micro-analysis of the opening homework review activity, for example , showed that the expert teachers performed this activity in one-third less time than novices , and in the process were much more thorough in picking up information about student attendance, homework completion, and who would need help later in the lesson. For experts , these routines are virtually automated pieces of action that allow student s and teachers to devote their attention to other, perhaps more important matters in the lesson. For novices, these activity structures were largely absent or they showed day-to-day change s in the way an activity, such as opening homework review , were performed. These patterns reflected the hypothesized presence of ill - formed action schemata characteristic of less expert teachers. Borko and Livingston (1988) examined the planning , t eaching, and post-lesson reflections of three expert and three novice math teachers . Their study showed that novices showed more time-cons uming, less efficient planning behaviors for presenting lesson content. They also made less selective use of information during planning, showed limited ability to see relationships across the curriculum, and revealed less ability to anticipate student problems. During interactive teaching , novices made less use of instructional and management routines , and appeared to be more distracted by s tudent responses that diverged from scripted lesson plans. Novices a l so reported more varied, less selective post-lesson reflections than experts . Their post - lesson reflections showed a lack of cons i s t ent focus , with thoughts ranging from concerns about their e ff ect ive ness as teachers, to student unde rstanding concerns, to conce rns r egard ing 39 the overal l lesson structure . These researchers' findings suggested fu ndamental differences in expert and novice teachers' cognit i ve schemata. In contrast with experts , novices appeared to possess l ess-devel oped propositional structures for pedagogical content knowledge and l imited inter-connections among their cognitive structures for content knowledge , pedagogical content knowledge , and knowledge of students. Other studies invo l ving experienced and novice teachers' cognition include those examining teachers ' knowledge of s tudent s . Ca l derhead ' s (1983 ) study , for example, showed that expe rienced teachers have amassed a large quantity of information about students and , in a sense , "they seemed to ' know ' their students before they met them (p.5) . " Novices did not show as well-developed student schemat a . In a related study , Carter, Sabers , Cushing , Pinnegar , and Berliner (1 987 ) examined differences between expert and novi ce math and science teachers with respect to the way they processed and used information about students in simulated teaching tasks . These researchers identified nine propositions that distinguished expert and novice teachers in their thinking about students. These diffe rences included the importance teachers ' attributed to available student information before taking over a math or science class, their i nc l ination to accept as valid the information provided by the previous teacher , the way they talk and think about individual students , the kinds and quality of solution strategies they proposed for classroom problems , the strategies for taking over a new cl ass , routines for getting to know student s and for assess ing what t he students have l earned, the types and amount of information they 40 remembered about students , the attention given to student test and homework information , and the amount of time allocated for planning instruction based on available student information. These researchers concluded that , "expert teachers , like other experts , appear to bring rich schemata to the interpretation of phenomena , and these schemata appear to provide them with a framework for meaningfully interpreting information (p. 156 )." Most of the studies reviewed above compared experienced and novice teachers in terms of their skill in planning instruction , processing information derived from student performance , and the adherence to interactive teaching routines. Berliner (1988 ) recently proposed a model suggesting that the development of expertise in pedagogy consists of 5 stages of skill development and that most expert-novice teacher studies only examine the extremes in teacher performance and cognition. Berliner suggests that there are severa l stages teachers move through in the development of pedagogical expertise . He identifies these stages as follows : Let us view the development of expertise in pedagogy as consisting of 5 stages of ski l l development , following the general model presented by two Berkley professors, the philosopher Hubert Dreyfus and his brother , computer scientist Stuart Dreyfus (1986) . we begin with novices , who , with experience , develop into advanced beginners . Most of these individuals then go on to become competent teachers . It should be the goal of teacher education colleges to he lp prepare the novice and assist the advanced beginner to become a competent teacher. Competence , I believe, should be our goal. Perhaps in the fifth year [of teaching), a modest number of teachers may move into a further stage of development , that of proficient. Some of these proficient teachers will reach the highest stage , achieved by very few members of a field , that of expert. (pp. 2-3) The current study makes no attempt to identify teachers ' stage of chemical demonstration expertise . It does, however, attempt to 41 describe the characteristics of science teachers at relatively early and advanced stages of development in their pedagogical content knowledge of chemist ry and demonstration teaching. Our attention next turns to the influence of inservice workshops on promoting science teachers' professional development. Science Teacher Inservice Education The 1980's has seen an unprecedented interest in science teacher inservice education by the science education community and people concerned with improving the teaching and learning of science (Evans , 1987). Several factors and events that took place during the 1970's and early 1980 ' s help explain the current interest in science t eacher inservice education. These factors have included (1) the steady growth in the number of stable, tenured science teachers whose training was in need of updating, (2) the alarming findings of several federally funded studies and reports that documented a serious erosion in the quality and quantity of science taught in the U.S. at the K- 12 l evels (National Commission on Excellence in Education, 1983; National Sc ience Board , 198 3; National Academy of Science, 1982), and (3) recent educationa l reforms and initiatives attempting to se t higher standards in science teaching and science teacher education (Bell, 1987; Holmes Group, 1986; National commission for Excellence in Teacher Education, 1985; Joyce & Clifts , 1984; Yankwich , 1984) ? One l ogical outcome of these developments has been the increased interest in science teacher inservice education . 42 The National Science Foundation and Inservice Prog r ams The federal government, through the agency of the National Science Foundation, has had a long and varied history in the inservice training of precollege science and mathematics teachers . Since the mid-1950's, NSF has has spent over $750 million in support of inservice programs, the most familiar being the NSF summer institutes (Dyche , 1974; Willson & Lawrenz, 1980). Summer institutes offered prior to 1970 had a central goal of increasing the effectiveness of teachers by broadening and updating their scientific backgrounds. During the 1970's there appeared to be a shift in NSF summer inst itute goals in s upport of curriculum implementation institutes. Efforts at addressing science teaching methodology were purposely omitted . Only recently have NSF inserv i ce programs been implemented that have stressed teacher acquisition of specific science teaching skills , such as chemical demonstration teaching , as exemplified by the summer workshops offered by the Institute for Chemical Education (Bell , 1987; Chelimsky , 1984 ; Lippincott, 1985). Evaluating the Effectiveness of Inservice Programs Evaluation of Past NSF Inservice Programs Much of the literature on NSF inservice programs indicates that these federally funded programs have received wide acceptance by universities and participants alike (Highwood & Mertens , 1972; Willson & Lawrenz, 1980). Most of these studies focus ed on participant perceptions of various aspects of NSF summer institute programs . Some r epresented follow-up studies on participants ' self-assessed use of 43 workshop skills, curriculum, and supporting materials (While, 1984). Comparatively few studies exist on the influence NSF summer institutes have on variables such as student achievement and attitudes towards science. One study conducted by Willson and Geribaldi (1976) showed a consistent trend in the direction of better student performance with increased NSF participation. In another study, Willson and Lawrenz (1980) reported a small, but significant, relationship between institute attendance and student attitudes toward science. St udies examining the influence of NSF summer institutes on the deve lopment of teachers' pedagogical knowlege base seem to be lacking, a situation which this researcher seeks to address. General Problems in Evaluating Inservice Programs Reviews of the literature pertaining to the evaluation of sc i ence inservice education programs and inservice programs in genera l often concur with Hounshell and Liggett ' s (1976) assessment that "Research in the area of inservice education is meager and often poorly planned, organized, and executed.'' Such conditions have lead to nume rous contradictory and confusing findings regarding whether implement ed programs were effective in bringing about the desired changes in teachers and their students (Wade, 1984-1985). One of the problems frequently encountered in the research and cited in reviews involves the lack of pretraining observations of teaching so that entry level skills of teachers prior to inservicing are documented (Joyce & Showers, 1980). In addi tion , many studies on inservice education lack sufficient descriptions to indicate what actually occurred during the training of teachers. Most inservice eva luation studies are typically non- or quasi -experimental studies . 44 Such designs prevent the random assignment of subjects into experimental and control groups. Therefore, thorough descriptions of teacher characteristics and inservice treatment should be col l ected when conducting research on teacher education programs (Strother , 1983 ) . These descriptions are necessary to minimize threats to validity resulting from confounding variables and to maximize reliability and generalizability of the findings (Boulanger , 1981; Druva & Anderson, 1983 ) Some educators have criticized inservice teacher education evaluation efforts in terms of the overreliance placed on self-reports to assess workshop success (Boschee & Hein , 1980). It is charged that these self-report measures have little relationship to t eaching behavior or student achievement. Furthermore , self-reports of teachers ' attitudes and satisfaction with various workshop components, although easily and economically obtained, may be subject to biases (e . g ., leniency , lack of discrimination , lack of internal consistency ) . The measurement instruments used to assess teacher attitudes and perceptions of the workshop have also been quest ioned in terms of t heir validity (Jones & Hayes , 1980 ) . Research on Teacher Education The literature on inservice education discusses severa l program elements that appear to contribute to changes in teacher behaviors . Gliessman ' s (1981 ) review of the literature on teacher education (preservice and inservice) indicated that simple and complex t eaching skills can be learned through a number of well -defined processes. These processes include (1) learning through observation , (2) concept 45 learning , (3) learning through practice, and (4) learning from feedback. According to Gliessman (1981 ), learning through observation (i . e. , modeling or imitation ) is a particularly effective strategy for helping beginning teachers acquire questioning skills as well as the emotive aspects of teaching , such as enthusiasm and warmth. Observation strategies have al so been effective with preservice teachers in terms of influencing behaviors that give greater use to a student-centered teaching style . Beginning and experienced teachers also learn to teach through the acquisition of concepts . In this regard , Gliessman (1981) stated the following , ''Viewing a teaching skill as a concept to be acquired implies that the immediate goal of instruction is conceptual : To be learned are the essential characteristics of a skill, its specific uses in teaching , and how it is distinguishable from other skill s ." Conceptual learning has been particularly useful in changing teacher performance from teacher - centered to student-centered styles and in increasing teacher use of probing and informing skills during interactive teaching (Wagner , 1973 ) . Learning through practice refers to the act of teaching unde r controlled conditions with the intention of improving one ' s performance . Microteaching in a laboratory setting is probably the most familiar example of learning through practice (Allan , 1966; Brown , 1975 ). Gleissman (1981 ) indicated that, among preservice teachers , most microteaching efforts appear effective in a laboratory setting but often fail t o produce transfer t o classroom settings , 46 unless the acquired practices are phi l osophical l y accepted by trainees and modeled by supervising personnel. Learning from feedback also contributes to the process of learning to teach. Variou s feedback media (for example , videotape replay , audiotape replay , and verbal report) can stimulate change in teacher performance . Changes that have been reported as a res ult of focused and immedia te feedback include increased skill in varying the level of questions directed to students , greater interactions with lower ability students , and improved styles of body pos tures and teaching mannerism (reviewed by Gleissman , 1981). Evans (1987 ) examined several research studies and r eviews on inservice sci ence teacher education in order to generat e guide l ines for conducting effective inservice programs. He suggests that science inservice education programs aimed at the enhancement of teaching skills should include modeling , practice , and feedback. Inservice education programs aimed at the acquisition of teaching s kills should include theory , modeling , practice , and feedback . Although the research literature suggests that inservice programs designed around these three or four training components would contribute to the acquisition of basic teaching skills , these st udi es fail to show how these training interventions influence knowledge growth in teaching of specific subject-matter topics . In this regard , the question of how science inservice workshops can influe nce the development of teachers' pedagogica l reasoning skills and fo ster pedagogica l content knowledge growth is of particular inte r est in the quest to understand the factors that influence science teachers ' professional development and how this development impacts st udent 47 l earning. Assessing Teachers' Knowledge Research in cognitive science has brought significant developments in methods for assess ing knowledge, cognitive structures , and mental processes used by individuals engaged in complex , discipline-specific tasks. Most of these methods yield verbal data rather than numbers for their raw data . This section r evi ews some of the data gather ing methods used to probe teachers' knowledge and thinking about pedagogy. A later section gives attention to the techniques used to analyze the verbal reports generated by these methods. Data Gathering Methods Educators have used several different qualitative methods to assess the various components of teachers' professional knowledge base. Three methods frequently used to probe teachers' knowledge and thinking have included: (1) clinical interviews using the think-aloud technique , (2) critical-incident methods, and (3) structured and semi-structured interviews. The se strategies are particularly useful for assessing the nature of teachers ' knowledge and thinking about students , classroom events , and pedagogy. Although each of these methods have distinctive characteristics , they are often used collectively or s imultaneous l y in cognitive research studies . The major features that di st inguish these methods are briefly discussed because they reflect characteristics of the methods used in this study 48 to probe teachers' specialized knowledge of chemical demonstration t eaching. Clinical Interviews Using the Think- Aloud Techni que. In the field of cognitive science, the interview is one of the most direct and widely used methods to gather information about a subject ' s internal cognitive states (Ericsson & Simon , 1980; Garner, 1987). It consists of a face-to-face meeting where the interviewer seeks information from an individual in the form of verbalized thoughts. These verbalizations provide data on the interviewed subject ' s knowledge or thought processes within specific content domains. With interviews, known as clinical interviews, an individual is specifically asked to provide descr i ptions , predictions, and explanations of events relevant to the content domain of interest (Finley, 1986) . They typically consist of one or more standardized think-aloud tasks along with interviewer questions to test the genuineness and consistency of the interviewee ' s response (Posner & Gertzog, 1982). Clinical interviews can be designed to examine both the declarative and procedural knowledge held by an individual. Declarative knowledge represents the large body of facts and concept s , and their inter-relationships, tha t an individual stores in memory . Procedural knowledge refers to t he mental processes an individual uses to manipulate these facts and concepts to so l ve problems or make decisions. Gitomer and Pellegrino (1985) describe declarative knowledge as "knowing what" or "knowing that" and procedural knowl edge as "knowing how. " This researcher's study focuses primarily on the retrieval of declarative information from science t eache r s (i. e , t he i r 49 pedagogical content knowledge). Posner and Gertzog (1982 ) describe the knowledge- probing function of clinical interviewing as follows : Its chief goa l is to ascertain the nature and extent of an individual ' s knowledge about a particular domain by specifying the relevant conceptions he or she holds and the perceived relationship among those conceptions . Once obtained , this information could be represented in a suitable format, such as a semantic network , which would be equivalent to a partial representation of the individua l ' s cognitive s tructure . The most prevalent use of the cl inical interview method has been in conjunction with studi es probing students ' cognitive s t ructure (Finley , 1986 ; Posner & Gertzog , 1982 ) . Science educators have found this interview s trategy to be particularly effective in examining the process of conceptual development among students and inservice teachers learning abs tract concepts in science , such as heat , temperature, equilibrium, pressure-related gas behavior , and magnetism through a variety of instructional strategies (Benbow , 1987 ; Crosby , 1987 ; Finle y , 1986; Swamy , 1986; Layman & Krajcik, 1988). The term ' clinical inte rview ' is encountered in the lite r at ure on teachers' thinking in the context of examining teachers ' implicit theories of teaching and l earning , and in the context of examining t eacher s ' interactive thoughts and decisions (reviewed by Clark & Peterson, 1986 ) . The c l inica l intervi ew format can vary from highly f l exib l e (with both tasks and questions varying from subject to s ubject ) to highly standardized (with carefully specified tasks and questioning patte rn s ) (Novak & Gowin , 1984) . Interviews designed to obtai n a de tailed picture of t eachers ' pedagogical knowledge often use think-a l oud ta s ks that s imulate various pre-active and interactive teaching be havior s . 50 Structured or semi-structured interview questions often accompany these simulated teaching tasks (Clark & Peterson , 1986; Medley , 1984). The think-aloud method has numerous research applications . It has been widely used in clinical interview studies where individual s are engaged in problem-solving or decision-making tasks. In studies examining teachers' thinking , these tasks have resembled simulated teaching tasks such as (1) lesson planning (Peterson , Marx , & Clark , 1978) or (2) making judgments about curriculum materials and student s (Berliner , 1987a; Yinger & Clark, 1982) . The teacher ' s verbalizations are recorded on audiotape and later transcribed to create protocol s . The protocols are then analyzed according to various coding schemes t o produce descriptions of the content of teacher thinking and of the sequence of cognitive processes that teachers fo ll ow while planning, making decisions , and teaching. The think-aloud method has been used in several r ecent s tudies t o examine expertise and the role of experience in teaching (Berline r, 1987a; Carter et al ., 1988; Cushing et al. , 1986 ; Leinhardt , 1983 ) ? In these studies experienced and novice classroom teachers wer e provided with tasks that required teachers to process class room information and to discuss their thoughts about common pedagogi ca l issues. One task required teachers to look very briefly at s lides of mathematics and science classes , and describe what they saw (Carter et al. , 1988; Cushing et al., 1986). Another task required teache r s to look at multiple-choice items from a standardized test and to es timate the percentage of students at a specif i ed-age and exposure to a specified curriculum, who would get such items correct and to expl a in why certain di s tractors would be frequently or infrequently chosen 51 (Leinhardt, 1983). Another task required the viewing of three television screens simultaneously, each one showing different part s of a classroom during a lesson. The teachers had to provide think-aloud comments on what they were seeing and hearing, and then answer questions about their viewing when the videotaped lesson was over (Sabers et al., in progress ; cited in Berliner , 1987a) . These think-aloud tasks suggested that the experienced teachers possessed richer schemas about students than less experienced teachers , e .g., richer schemata for 'typical' or 'normal' students and for other sense-making categories for thinking about students. The finding s obtained from these think-aloud tasks also suggested that experienced teachers possess many more cognitive skills of teaching (e.g ., information processing skills , interpret ive skills , problem-solving skills for addressing classroom events) than less experienced teachers. The think-aloud task developed for this st udy ha s sci ence teachers observing videotaped models of chemical demonstration teaching and , on second viewing , critiquing the performance at variou s teacher-selected intervals . In this think-aloud ta s k, the primary stimuli provided to the teache rs comes from the videotaped chemical demonstration presentations . Researcher-initiated clarificati on questions were also interjected during the think-a loud ta s k as a secondary source of response stimuli. This think-aloud task , called the critical-stop task , was designed to help probe the content of experienced and novice chemical demons trators' pedagogical knowl edge base associated with the demonstration t eaching of bas ic chemi cal concepts. 52 Critical-Incident Methods A review of the literature reveals three types of critical-incident methods used in studying teachers ' cognitions and the process of teaching. They include (1) the critical-incident technique , (2) critical-incident analysis , and (3) obtaining a critical incident record. While the names of these methods sound very much alike , they represent techniques obtained from different lines of qualitative research. The first two of these techniques show some s imilarities to the critical-stop task used in this s tudy to probe teachers' knowledge of demonstration pedagogy. They are discussed in some detail , below. The third method, obtaining a critica l incident record, refers to a strategy for making direct, systematic observations in an educat ional setting (Evertson & Green , 1986) ? It has little to do with making direct probes of teachers' knowledge in an interview setting and is therefore not part of thi s r eview. Critical-incident technique. The critical-incident t echnique, as typically employed , uses interviews to obtain descriptions from a group of informed individuals (such as principals) who have knowledge about the performance of a group of target individual s (such as teachers) (Borg & Gall , 1983, p. 509 ). During the inte rvi ew the informants desc ribe "critical incidents" about the performance of t he target group . Each critical incident provides a descript ion of a specific behavior pattern that is considered critical to the skil l or trait being studied. This technique offers a simple yet ef f ect ive alternative to training observers and having them carry out l engthy observations on a group of individuals in a variet y of educational settings . The technique , as it probes the knowledge of particular 53 individuals, is particularly useful in understanding the performance of a group of target individual s . The critical-incident technique is well-suited for providing detailed case de sc riptions of behaviors and characteristics associated with a complex theoretical construct that is not well understood or defined, e .g., professionalism in education (Le l es , 1968 ). A difficulty frequently encountered with the t echnique , however, is that many of the recorded incidents appear to be global eva luations about a sub j ect ' s performance rather than specific incidents involving the subject , e .g., see Flanagan (19 54 ). The data often needs to be screened so that only specific incidents are used in the data ana lysis . Perhaps the most serious problem associated with the critical-incident technique is ascertaining whether identified incidents obtained from the interviewed informants are truly critical to the behavior or skil l being studied , i.e. , the incidents can truly differentiate between s uccessful and unsuccessful behavior. The critical-incident technique, even with some of its limitations , shows considerabl e utility in exploratory and theory- generating studi es . The usefulness of thi s technique may be enhanced if the "interviewed " sub jects rely less on their long term memory (LTM) and more on shor t term memory (STM) to discuss critical incidents . This may be possible by having interviewed informants observe and analyze videotaped segments of classroom teaching during a clinical interview (e . g. , as described in chapter 3). Thi s approach would be consistent with Ericsson and Simon ' s (1 980 ) claim that verbal reports t end to be more reliable and valid as data when a pe r son 54 reports on the contents of short-term memory, that is, that which he or she is currently attending to. Critical-incident analysis. Another critical-incident me thod used in qualitative research involves the use of informants analyzing researcher-identified critical incidents associated with t eaching. The analyses are conducted in a think-aloud fashion. This technique , which I shall call critical - incident analysis, has been effectively used to assess teacher ' s knowledge of students and student misconceptions. Hashweh (1 986 ) used this technique when he presented science teachers with several critical teaching incidents involving hypothetical student answers to textbook questions. The fabric a t ed student answers contained misconceptions that revealed l ea rning difficulties commonly encountered by students solving work and ene r gy problems in physics and photosynthesis problems in biology. The technique was des i gned in order to assess how topic-knowledgeable and unknowledgeable science teachers would respond to commonly encount ered sc i ence teaching problems. Analysis of teachers' r e sponses to these critical incidents revealed major differences between teache r s ' sub j ect - matter knowledge in dealing with these general class difficulties. Calderhead (1 981a ) also used this technique in a study on beginning and experienced teachers when he orally presented t eachers with descriptions of common critical incidents related t o classroom discipline and management. He then as ked them how they might respo nd to these incidents during interactive t eaching. The study s howed a marked difference in the nature and sophis tication of r esponse bet ween 55 experienced and novice teachers in their interpretation and ana lys i s of classroom events. The crit ical-stop task used in the present study asked teacher s to both identify as well as analyze chemical demonstration t eaching incidents observed on videotape. The data derived from such a critical-incident task helped explore science teachers' pedagogical knowledge of chemical demonstration teaching. Semi-Structured Interviews Semi-structured interviews have significant application in probing teachers' thinking (Calderhead , 1981a). A semi- structured interview is one where the interviewer asks a series of s tructured questions and then probes more deep l y, using open-ended ques ti ons , t o obtain a more thorough understanding of the respondent's answe r s and the reasons behind them. Semi-structured interview questions are often integrated into a larger clinica l interview design. Novak and Gowin (1984 ) provide severa l practical suggesti ons f or conducting semi - structured interviews that probe conceptions and propositions that individuals hold in their cognitive structures. To conduct effective interviews , interviewers must be thoroughly f amili ar with the material to be covered , carefully listen to an in f ormant ' s response, create a calm and relaxed interview atmospher e , use probes or rephrase questions when insufficient or "don ' t know" r esponses are provided (particularly with semi-s tructured interviews ), use the informant ' s own language, avoid irre levant discussion, and e nd the interview on a positive note . In a s tudy conducted by Pet e r s on and Comeaux (1 987 ), a se t of structured inte rview questi ons wer e directed to t eache r s afte r 56 viewing videotaped teaching episodes of a high school history class. The interview questions asked novice and experienced teachers to recall as many classroom events as possible and to discuss alternative interactive decisions the videotaped teacher could have made . The interview helped these researchers distinguish between experienced and novice teachers in terms of their schemata for classroom event s and their analysis of classroom management problems arising during interactive teaching. (See discussion of findings on p. 35 ) . Carter , Sabers , Cushing, Pinnegar, and Berliner (19 87 ) employed interviewing techniques to examine teachers' knowledge of s tudents . In their study , teachers were given 40 minutes to prepare a two-day lesson plan in mathematics or science after receiving extens ive information about the class they were to take over . The semi - st ructured interview which followed this task asked t eachers to recall, generalize , and explain issues related to their l esson plan and the st udents . The protocols obtained from the ques t ions yi e l ded basic data on the information processing abilities of experienced and novice teacher s . In this study, a semi-structured interview followed the critical-stop task . Its chief goal wa s to further det e rmine t he nature and extent of an individual's pedagogical content knowl edge with respect to the demons tration teaching of two bas ic chemical concepts by soliciting and probing for information not volunteered during the think-aloud task. Reliability and Validity in Qualitative Research In as much as reliability is concerned with the replicability of scientific finding s , validity i s concerned with the accuracy of 57 scientific findings. Questions about reliability and validity can be raised at two levels in qualitative studies . One is at the data gathering (measurement ) leve l, t he second is at the analysis (interpretation) level. Reliability at the measurement level. Discussions about the reliability of measurement typical ly center around the i s sue of whether repeated measures with the same instrument on a given sample would yield similar results. In qualitative studies , the reliability of the verbal report data themselves is seldomly discussed. Thi s i s because verbal protocols based on a subject ' s performance on two equivalent interview tasks are not expected to be identical . Words , sentences , and paragraphs will differ with each administrati on of a task . In qualitative research the issue of reliability of verba l report data is some times resolved by determining its validity (e .g., predicting behavior on similar tasks) (Shavelson, Webb, & Burs t e in, 1986) . Adequate reliability at the measurement level is inferred i f predictive validity can be demonstrated. Reliability at the interpretation level. Clearly , verbal protocols obtained from subjects engaged in equivalent tasks would not be identical. Nevertheless , one would expect that for simil ar t as ks and leve ls of performance , similar decision-making or problem- solv i ng processes would be evident . In this context , reliability es timat es refer to the consistency with which a particular coding scheme can be applied - one that records , for example, the kinds and numbe r of statements about subjects ' interactive verbal behaviors (Shave l son, Webb, & Burste in, 1986). This type of reliability i s ofte n r e f e rred to as interceder reliability. It corresponds to a measure of 58 reliability at the data analysis level (not the data gathe ring leve l) of qua l itative research. Goetz and Lecompte (1984, p. 210) similarly define (internal) reliability in qualitative research as follows: "Internal reliability refers to the degree to which other r esearche rs , given a set of previously generated constructs, would match them with data in the same way as did the original researcher." Reliability in coding verbal di scourses can be estimat ed in several ways. The simplest estimate requires having two code r s categorize subjects' statements into one of several pre - de termined categories and then calculating percent coder agreement. The coding categories can be derived fr om theory or previous re search, or t hey may emerge from the data it self through an ite rative process . A combination of these two sources i s also feasible. Borg a nd Gal l (1983, p. 479) indicate that in observational re sea r ch, intercoder agreement levels above 70 % are usually obtained among t rained observe r s who mu s t make infe r ences or evaluati ons about a give n behavior or set of behaviors. In coding interview protocol s and journal entries , researchers have reported intercede r agreement l evels of about 75 % or above (Borko, Lalik & Tomchin, 1987; J ohans s on, Marton, & Svensson, 1985). If the number of codi ng cat egories is small, e . g. l es s th an 4, chance contributes measurably t o intercede r agreement val ues . I n s uch instances , Kang (1987) s ugges t s using Scott' s Pi to correc t for chance coding agreement. This correction provides a l ess biased va lue for intercoder r eliability. Validity at the measurement l evel. Several va lidity i s s ues ari se when interview methods are used t o gather in fo rmat ion on an 59 individual's internal cognitive states . Schuster (1983) discusses some of these issues in terms of confounding variables that dese rve attention when a researcher presents himself as the primary research instrument, i.e., as interviewer. These variabl es include cueing, prompting, and suggestibility by the interviewer. Other confounding variabl es include concept modification and learning by the interviewed sub j ect during the interview session. Some of these potential threat s to validity can be minimized through the acquisition of basic clinica l interviewing skills. These ski ll s , include keeping informant s motivated and at ease , and avoiding the above-mentioned pitfall s discussed by Schuster . Such ski ll s can be developed through prac t ice , feedback , and conscious efforts (Pines , 1978; Posner & Gertzog , 1982). Field testing of interview questions in terms of clarity and proper sequencing can further contribute to the generation of accurat e da t a . Ericsson and Simon (1 980 ) have theorized about the probabl e validity of verbal reports generated by several process-tracing techniques (e . g ., think aloud , r e trospective interview, s timulated recall ) as a source of legitimate data on cognitive processes . On t he basis of their theory of human cognition and from studies of the characteristics of the verbal reports produced by vari ous process-tracing techniques , they reported that the think-al oud met hod , when applied to verbal tasks, theoretically produce s the mos t val i d data of an individual's cognitive proces ses. Shavelson, Webb, a nd Burstein (1 986 ) indicate that the think- aloud me thod t ypically produces verbal protocols as complete as possible , ha s a negligible effect on process ing time , and does not distort the s tructure and course of cognitive processes . Whe n think - aloud me thods a r e appl ied 60 to nonverba l (e.g. , visual) tasks , it appears to increase interviewees ' processing time but does not distort the data. Retrospective methods , on the other hand, are susceptible to some decrements in completeness and some distortion when subjects are asked to give an account of earlier thinking processes. Validi ty at the interpretation level. Dur i ng the process of interpreting verbal data, threats to internal and external validity need to be given careful consideration (Krathwohl, 1985; Goet z & Lecompte , 1984 ; Drew & Hardman , 1985 ) . In experimental studies, internal validity is characterized by successfully controlling (or accounting for ) all systematic influences between two groups except the one under study (Drew & Hardman , 1985). Internal validity in qua l itative research more typically refers to the extent to which scientific observations and measures are authentic representations of a particular reality (Goetz & Lecompte , 1984). Threats to internal va l idity common to both research paradigms include history and maturation , observer effects (instrumentation), selection and regression, mortality , and interview/test practice . External validity refers to the generalizability of re sult s to other relatively similar situations. For qualitative research, Goe tz and Lecompte (1984 , p . 210 ) provide the following definition: "External validity refers to the degree to which representations [of some reality ] can be compared legitimately across groups ." Threats to external validity arise from population - sample differences, artificial research arrangements, and multiple treatment interference. 61 Analyzing Qualitative Data on Teachers ' Knowledge Analysis of verbal data can take on two general forms: one focuses on the meaning of the content present in a written transcript, the other on the frequency with which specific forms of writte n communication appear in print. The first approach , with its emphasis on meaning, is characterized as Spradley ' s approach, and consi s ts of a collection of techniques for analyzing qualitative data. The second approach is a content analysis (Berelson, 1952 ; Williamson , Karp , Dalphin , & Gray, 1982) . Both of these methods of analysis are discussed in the review which follows and represent techniques used in the present study. Spradley's Approach Spradley (1980) recommends several useful methods for analyzing qualitative data . His description of these methods is given from an ethnographic perspective, but the methods are equally applicable to the analysis of qualitative data associated with science educati on research (Crosby , 1987; Swamy , 1986). His methods were written to provide ethnographic researchers with a systematic approach for analyzing field notes obtained from direct observation of a cultural setting and from interviews held with informants. Spradley's techniques are particularly useful in generating grounded hypotheses and understanding the meaning of terms and ideas held by subj ect s in a given cultural setting. His approach consists of four analytic strategies: domain analysis , taxonomic analysis , componential analysis , and theme analysis. These four strategies we r e appli ed t o the data gathered in this study. 62 Domain analysis. Domain analysis involves a systematic search for conceptual categories by grouping verbal izations that po ssess simi l ar linguistic patterns or semantic relationships (Spradley, 1979 , 1980). The first step in identifying semantic relationships in the verbal data is to identify meaningful statements derived from t he written protocol s . These statements , somet imes referred to as proposit i ons (a statement with a subject and predicate) , provide the raw data for a domain analysis. Domains emerge from the data by the nature of the semantic relationships that occurs between the subj ect and predicate in given propo sitions. These r e lationships may be characterized as descriptive , cause and effect , evaluative , r at i onal e , attribution , function , sequence , or some other lingui s tic fo rm . The purpose of a domain analysis is to provide a broad ove rview of the data (Spradley , 1980). Once such an analysis ha s been conducted , a taxonomic analysis can be pe rformed on the propositions that have been coded into domains. Taxonomic ana l ysis. This analysis cons ist s of a sear ch for categories and subcategories within a recognized domain . Spradl ey (1 980 ) provide s an i llustrative example of an ethnographer who seeks to understand the written forms of communication in a parti cul ar culture . From the researche r ' s field notes he identifies several domains of printed materials , including book s , journal s , magazi nes , notes and l etters . A taxonomic analys is of the magaz ine domain yielded seve ral categories including literary magazines , practical magazines , comics , and news magazine s . Furthe r taxonomic analy s i s of the news magazines category yielded subcategories cons i s ting of specific examples such as Time , Newswee k, and U.S. News & Worl d 63 Report. Although this example involves an analysis of a cultural setting of interes t to ethnographers , it brings out the pur pose of a taxonomic analysis , namely , to search for simi l arities and dif fe r ences within a specified domain. Collective ly, a domain analys i s and taxonomic analysis serve to organize and repre se nt l arge quantit ies of verbal data in t e rms of meaningful cat egories . Componential analysi s . A componential analys is i s a s ys t ema t ic search for contras ts within a domain or taxonomic category. Con trast may be found be tween terms (unit s of meaning ) or be tween cont ras ting groups of subj ect s . With the available data , the researcher decides which doma i ns and taxonomic cat egories r ece ive a componential analysi s . Thi s analysis may extend to include all domains and categories identi f ied by the re searche r or it may include onl y those domains and cat egories most central t o the s tudy . Theme analysi s . A theme analy s i s involves a sea r ch for t he relationships among domains and f or how they ar e linked t o a particular se tting , e . g ., middle school sc i ence t eachi ng . A theme reflect s a principle evident across a numbe r of domains . Its presence is supported by implicit and explicit s tatement s made by i nformants . In this s tudy , the goal of the theme analy s i s was not t o ident ify new themes or components of teache r s ' profess i onal knowledge base , as much as it wa s to he lp ve rify the ex t ent that the l it erature-ident i f i ed themes , pedagogica l content knowledge and gene r al pedagogi cal knowledge , were actually evident in the data . In the current s tudy , the dat a analys i s t echniques recommended by Spradley (1 979, 198 0) a r e used t o (1) examine the na t ure of chemical demons trator s ' pedagogical discourses , (2) cont ras t chemica l 64 demonstrators ' pedagogical discourses , and (3) provide categories for a quantitative content ana l ysis of the verbal data. Content Analysis Content analysis represents a broad and complex type of research technique for analyzing qualitative data. Berelson (1952) describes i t as an "objective , systematic and quantitative description of the manifest content of communication. " It is most often used to detail the freque ncy with which symbols , concepts , or themes appear in a written document (Williamson , Karp , Dalphin , & Gray , 1982) . This technique has been successfully employed in several qualitative studies in education. They include studies that have examined the nature and emphasis of supervising teachers ' written evaluations of student teachers (Cici relli , 1969 ) and of college students ' conceptions of chemical processes (Basili , 1988; Crosby , 1987; Swamy , 1986). It has also been used to study the relationship of s cience teachers' subject - matter knowledge to their thinking about preactive and interactive teaching (Hashweh , 1987 ) . In each of these studies , the investigator coded and enumerated the frequency of verbal statements within pre-defined categories or categories that emerged from the data. Content analysis places emphasis on a quantitative descrip t i on of communications. It allows a researcher to characterize a large volume of materials rather efficiently with one or a small number of frequency tables. These frequency tables essentially summariz e the number of times each content category or subcategory is present in a written document or a verbal protocol. ----~-- 65 Chapter summary This chapter examined the components of teachers' professional knowledge base with particular emphasis on teachers' pedagogical content knowledge and its development through pedagogical reasoning. It also examined the influence of inservice education on science teachers ' professional development. ? The chapter culminated with a review of several cognitive methods used to probe teachers ' thinking . This body of information served as a theoretical and empirical base for the present study. I ,1,I .:, ' II 'I ' I I I I I . 66 CHAPTER 3 RESEARCH DESIGN AND PROCEDURES Th is study probes the pedagogical content knowledge and general pedagogical knowledge systems (Shulman, 1986, 1987) of experienced and novice chemi ca l demonstrators. It also investigates the nature of nov i ce chemical demonstrators' pedagogical knowledge growth re s ulting from participation in a two-week chemical demonstration workshop . This chapter describes the characteristics of the research s ubj ect s and inservice workshop, the methods used to assess t ea che r s ' pedagogical knowledge of chemical demonstrating, the r esearch design , and the data analysis procedures . Characteristics of Research Subjects The research participants involved in this study cons i sted of eight science teachers who were novices at chemical demons trating and five who se rved as experienced demonst rator s . The novices wer e se l ected from a pool of workshop participants possessing lower l eve l s of chemical demonstration t eaching exper ience . The experienced demonstrators were ins tructors in a two-week chemical demonst r ation workshop (Workshop B: Chemistry Supplements for Pre-High School Cla sses ). The workshop was offered by the Institute for Chemical Education (I CE ) at the University of Maryland (UMCP ) during the summer of 1987. Enrollment in ICE Works hop B (Session II) consisted of twenty-three science teachers who were fully or provi s ionally certified at the e l ementary, middle, and high school l evel. Al l t he ----- - - - . .. -~.. -?-- .. - -?- ----- 67 participants had teaching assignments that included physical science. ICE staff carefully selected the participants from a larger pool of applicants based on their potential to transfer the knowledge and skills acquired at the workshop to other science teache rs in the i r school district . This potential was based on prior experience in conducting inservice science programs in their local school districts and on leadership skil l s reflected in their I CE workshop application (Appendix A, the second short answer question regarding applicants ' prior experiences in present ing outreach programs and developing educationa l material s ). Before the inservice workshop began , pertinent background , I ' I: II information on each of the summer institute participants was collected using the Participant Information Form (PIF) 1 and participants' application to the program . The Participant Information Form was mailed to the participant s three weeks before the workshop started . Twenty -two participants completed this questionnaire and returned it to the researcher prior to the s tart of the workshop . One person, living overseas , did not receive the questionnaire in time. The workshop instructors al s o completed the participant questionnaire. This researcher rank-ordered all Workshop B participants on the basis of their self-reported confidence and weekly use of chemical demonstrations as indicated on the PIF. Appendix C shows the grouping of the ICE teachers according to demonstration confidence and experience . Tho se in the lower 50th percentile along both indices 1 The PIF is described in a later section under Data Collecti on Procedures (p. 80 ). (See Appendix B for the PIF ). 68 were considered novices and candidates for the study . A few sub ject s that ranked low on one index and not the other were also considered. One subject , (N6 ), reported novice-intermediate credentials on both scales , however , his limited teaching experience and limited chemi s try background qua l ified him as a novice candidate . Two ICE workshop instructors later added support to his classification as a novice chemical demonstrator by providing this researcher with a general assessment of his public demonstrations. The PIF identified twelve less experienced (novice) chemi cal demonstrators among the workshop participants. This identi f ication 1; ,,,, was found to be consistent with background information provided on the ,, 1: workshop application form. ;; Each of the identified novice demonstrators was mailed a l e tt e r seven to ten days prior to the start of the workshop r eques ting voluntary participation in this research study (see Appendi x D). Follow-up phone calls confirmed the voluntary participation of s i x novices . Two others confirmed their interest in the study on the day of their arrival. These eight teachers served as a group of novice chemical demonstrators and the workshop instructors served as experie nced chemical demonstrators . The four novices wh o did not participate in the study either arrived at the workshop afte r the eight interview slots were filled or could not s chedule time for a pre-workshop interview . Table 1 describes the self - assessed confidence and weekly use in conducting chemical demons trations reported by the five experienced chemical demonstrators and e ight novice chemical demonstrators who volunteered for this s tudy. The - ? __ _.._., - ____ ..,. _.;_,_ _ _ 69 Table 1 Reported Confidence and Use of Science Demonstrations Group (Mean ) Characteristic Novice Experienced n = 8 n = 5 1. Confidence in using chemical demonstrations 2.4 4.8 2 . Confidence i n using other demonstrations 3.8 4 . 8 3. Number of chemical demonstrations per week 0. 5 a 4.4 4. Number of other demonstrations per week 2. 2 a 4.7 Notes . ~fidence Scale: 1/Low ---> 5/High a PIF response of< 1 was considered "zero " for the purpose of calculating a group mean. confidence scale consisted of a five-point rating scale with categories ranging from 1 (Low ) to 5 (High ) . The sampling strategy of using novices and experienced chemical demonstrators in this study served the purpose of potentially discriminating between subjects having high and low levels of specialized knowledge of chemical demonstration teaching. Novice demonstrators were also of interest to this study because they had the pot ential for showing the greatest cognitive gains resulting from workshop participation. Furthermore , the selection of novice subject s mitiga t ed against possible ceiling effects that could have been encountered with more experienced workshop participants. Description of the Experienced and Novice Chemical Demonstrators The experienced chemical demonstrators were the instructors for ICE Workshop B. Their exper ience in conducting chemical demonstrations and i n providing numerous inservice workshops on 70 chemistry teaching strategies provide justification for their classification as experienced chemical demonstrators. Because the research literature continues to debate the validity of the criteria used to identify expert pedagogues (Berliner, 1986; Sloan & Capie, 1987) , the term "experienced chemical demonstrators" will be used instead. The term "proficient" (Berliner, 1988) , which probably represents a conservative description of the modal skill l evel of the five workshop instructors, has not yet received wide adoption in the literature. Verification of even such skill levels , as with expert skill levels, remains a problem; therefore , the term "experienced" becomes the preferred descriptor for the workshop instructors examined in this study. The five experienced demonstrators included the four instructors of ICE Workshop Band one special guest lecture-demonstrator who made two presentations at the workshop. The eight novice demons trat or s represented about a third of the participants who enrolled in the 1987 ICE Workshop. The novice demonstrators included one element ary school teacher, five middle school teachers , and two high school t eache r s . Tables 2 and 3 provide additional background information on the experienced and novice chemical demonstrators, in parti cul ar, thei r science teaching experience , college chemistry training, and experience in conducting chemical demonstrations works hops . The information suggests that the experienced demonstrators' spec i ali zed knowledge and skill in conducting chemical demonstrations i s der ived from their extensive (1) science/chemistry teaching experience , (2) uppe r-leve l college chemistry knowledge , and (3) experience (wee kly 71 Table 2 Teaching Experience and Chemistry Training of the Novice and Experienced Chemical Demonstrators Group (Mean or Range ) Teaching Experience and Chemistry Training Novice Experienced 1. Number of years teaching experience 7 . 0 13.0 in the sciences (bio , chem, phys sci ) 2 . Number of college chemistry courses 3 . 4 12.4 3. Grade l evels most experienced Up.Elem- HS Up.Elem-Coll. use ) in conducting chemical demonstrations in science class rooms . These demonstrators have also had (4) considerable experience in conducting chemical demonstration workshops (Table 3). Novices ' pri or knowledge and skill in conducting chemical demonstrations wa s de rived from their (1) physical science teaching experience , (2) introductory college chemistry coursework , and (3) occasional experience (weekly use ) in conducting chemi cal/science demonstrations (see Tables 1 and 2). These teachers had essentially no experience in conducting chemical demons t ration workshops (Table 3) . The pre-workshop grouping of the participants and instruc tors (Appendix C) reveals the range of demonstration experie nce and chemistry training found among the two groups of chemical demonstrators participating in this study . The background informati on obtained from the experienced s ubjects (workshop instructors) s howed two dis t inct subgroups in terms of their chemistry training and teaching grade level (Appendix C, Table C-2 ) . One subgroup consi s t ed 72 Table 3 Confidence and Experience with Chemical Demonstration Outreach Programs Group (Mean) Outreach-Rel ated Criteria Novice Experienced 1. Confidence in chemical demonstration 2 . 6 4.8 outreach programs 2. Number of chemical demonstration outreach 0 . 1 a 4.3 programs per year to students 3. Number of chemical demonstration outreach 0. 2 a 4. 3 programs per year to other teachers Notes . ~fidence Scale: 1/Low - --> 5/High a PIF response of< 1 was cons i dered "zero " for the purpose of calculating a group mean. of an upper elementary and a middle school science teacher, (i. e . , experienced demonstrators El and E2 ), who had completed two and seven courses in college-level chemistry , respectively . The other subgroup consists of two high school chemistry teachers and one community college chemistry teacher , all having completed over 12 college-level chemistry courses. Novices showed a smaller range of chemical demonstration experience and chemistry training (Appendix C, Table C-1) . This l imited range is partially attributed to the consistently low levels of chemical demonstration experience and basic college chemistry training the novice group possessed . Experienced and novice subjects showed a similar range in teaching grade level . Throughout this document , the term "demonstrator " will refe r to both the experienced and novice chemical demonstrators . When 73 referring to a specific group, the terms "exper i enced chemical demonstrator " or "novice chemi ca l demonstrator" will be used . Characteristics of the Inservice Workshop This sect ion describes the objectives and basic components of ICE workshop B. It discusses the time allocated for t he various workshop components , the content of the staff presentations , and the training model used to foster professional development among the participating teachers. Workshop Goa l s and Objectives The ICE workshops offered during the summer of 1987 we r e wide ly publi cized through the mailing of several thousands of brochures to classroom teachers around the country. I CE Workshop B (Chemistry Supp l ements for Pre-High Schoo l Classes ) was one of the workshops f or which teachers could apply. Workshop B was offered at four s ites around the country, i ncluding the University of Maryland (College Park). The 1987 ICE brochure and application form (Appendix A) describes the dual objectives of Workshop B: To help teachers (1) "lea r n and practice effective and safe demonstrations, experiment s , and activit i es appropriate for younger students and (2) l ea rn interactive teaching methods ." These ob jectives are cons i s t ent with the goa l of all ICE workshops : "to provide first - hand expe rience wi th descriptive teaching methods, and to help participants s tre ngthen their background in chemistry so that they can encourage more questioning and exploration by students . . . [and] to he lp parti c ipant s _-_: ?_:.? _. ~? _-_ -:,. __ __ ,. -- -~.. ..... .... .... . .. _.. __ __ 74 prepare to present in-service workshops for teachers at other schools in their communities " . Workshop Components and Schedul e ICE Workshop B was designed around five inservice workshop components (Bell, 1987 ; O' Brien , 1987 ). These components included: (1 ) staff content presentations and demonstrations - model presentations and teaching tips on how to utilize demon s trations in a classroom setting and how to design outreach programs ; (2) participant "library " research - an opportunity to examine various sourcebooks of chemical demonstrations to select suitable demonstrat i ons; (3) individual participant "lab time " to practice self-selected demonstrations from 50 boxed demonstrations (listed in Appendix E); (4) individual presentations of demonstrations before peers (microteaching ); and (5) work with the laboratory-based Summer Chem Camp for sixth through eighth graders . A schedule of the workshop (Appendix F) shows how time was allocated for the major workshop components . Table 4 summarizes the time distr i bution . This table shows that Workshop B provided partic i pants with over 70 hours of chemical demonstration education across a two-week period . Approximately one - third of this time wa s devoted to staff and guest presentations . Participant preparation, presentation , and f eedback of chemical demonstrations performed before peers and middle school students accounted for most of the remaining two-thirds of the workshop . Each staff presentation was videotaped or audiotaped in order to capture the content of the presentations and to document the formal 75 Tab l e 4 Core Tra i ning Components for Al l Participants Component Allocated Time % Total (Hours ) Theory and Modeling: Introduction/Logistics 4 . 5 6 Staff Presentations 15 . 5 21 Guest Presentations 4 . 5 6 Subtotal : 24 . 5 34 Practice and Feedback : Participant l aboratory time 18 . 5 25 Participant publ ic demos w. 26 . 0 35 feedback Videotape Feedback 1. 25 2 (Individually scheduled ) Subtotal : 45 . 75 62 Social events 3 4 Eva l uat i on 1 1 Total 74 . 25 100 instruction t he participants received . Appendix G give s a brief description of the formal staff presentations. The first staff presentation addressed the expectations for the workshop participant s . A workshop handout summarizes these expectations (Appendix H). The Microteaching Model Three of the five workshop components (1, 3, and 4 listed above ) approximate an "observe , practice , and critique" mode l of microteaching (McIntyre , MacLeod, & Griffiths, 1977 ) . This mode l i s frequently incorporated into preservice and inservice programs that s tress the acquisition of teaching skill s (Gliessman , 1981) . During 76 the observation phase of the microteaching model , participants would observe workshop instructors perform several chemical demonstra t ions . This would be fo l lowed by brief discussions of the characteristics of effective demonstration teaching as modeled by the workshop instructors. This training strategy assumed that participants would transfer some of the knowledge and skills acquired by observational learning (Hunter , 1984) to their own chemical demonstration performances. The workshop participants indicated that they practiced about 15 different boxed chemical demonstrations during the course of the workshop (Appendix E) . These boxed chemical demonstrations were different from those modeled by the workshop instructors. The workshop participants publicly performed at least three of the demonstrations they practiced. Participants were assigned their first chemical demonstration for presentation, whereas they could se l ect demonstrations for subsequent presentations. With the exception of the Chem Camp presentations, each public presentation repre sented a different chemical demonstration. Furthermore , the public presentations given by novice participants r epresented chemical demonstrations other than the two observed on videotape during the pre- and post-workshop interviews. After the workshop participants practiced and presented a chemical demonstration , the workshop instructors and the works hop participants provided brief verbal feedback to the presenter. This researcher , or one of the ICE workshop instructors , would also provide feedback to the participants in a private session using a videotape of their presentation. Providing workshop participants with group and 77 private feedback on their public performance was an important design characteri stic of the workshop. During the videotape playback sessions, novice demonstrators would observe their demonstration performances and critique them using the crit ical-s top method described later in the methods section of th is chapter (p. 86 ). Following this self-evaluation task, an ICE instructor , or thi s researcher, would ask the demonstrator to think about and discuss any additional strengths and weaknesses that would come to mind. The workshop staff (instructors and this researcher) would usually provide a few additional comments at the end of the self-evaluation session in order to reinforce issues di scussed during (1) the formal staff presentations and (2) the feedback sessions that followed the public presentations. The primary emphasis of these private feedback sessions , however, was on having the workshop participants observe , identify, and discuss salient features of their public presentations as seen on videotape . 1 Novice participants involved in this study received feedback on two (in some cases three) of their public demon s trations by critically viewing their videotaped presentations in private session with this researcher . The remaining workshop participants observed their videotaped performances on only one occasion (a half - hour private 1 It is assumed , here, that the closing comments provided by this researcher , who attended all staff presentations , were s imilar in nature to the comments provided by other workshop instructors . Th is researcher's closing comments were based on each novice demonstrators ' self-evaluations as well as on select issues discussed during earlier staff presentations . The workshop instructors did not provide t hi s researcher with formal instructions as to what to say during these private sessions , although on one or two occasions they did suggest bringing up a particular point that may not have been adequately addressed in a given public feedback session. 78 The teach-reteach cycl e frequent l y assoc i ated with microteaching was implemented during the later part of the workshop as teachers presented some of their publica l ly performed demonstrations a second time to Chem Camp Kids . The workshop training model differed from microteaching model in that instead of focusing on and modeling one teaching skill (e . g ., questioni ng for feedback , clarity of explanation , use of examples , higher order questioning and probing), the workshop focused on the integration of several demonstration teaching skills simultaneously. Re s earch Met hods ,,,' This research examined science teachers ' pedagogical content knowledge and general pedagogical knowledge (Shulman, 1986 , 1987) by eliciting teachers' think-aloud critiques of selected chemical demonstration videotapes and by conducting semi -structured inte rview s . These two techniques helped generate qualitative data on the influence of a short-term intensive inservice workshop on novice chemical demonstrators ' pedagogical knowledge growth. The two techniques also served as a probe of experienced chemical demonstrators ' pedagogical knowledge so that their knowledge of chemical demonstrating could be compared to that of pre- and post-workshop novices . Videotape Selection Experienced and novice chemical demonstrators viewed two videotaped chemical demonstrations: the Collapsing Aluminum Can demonstration (Demo A) and the Density Column demonstration (Demo B) . These two demonstrations address the concepts of air pressure and - =- ---,-;:: -r i ,..-, .,....... _- _- _.-..,., ...--.-...._..-- v ? ? ? ? ? 79 dens ity, respectively , and represent two concepts that receive repeated emphasis in pre-col lege physical science classes and in the ICE workshop . A favorab l e and unfavorab l e videotaped vers i on of Demo A and Demo B were selected from a col l ection of videotapes obtained during the 1986 and 198 7 (Session I ) ICE workshops . Demonstration performance qual i ty was determined by a panel of experienced science teachers (i . e ., three science education graduate students and five certified high school sc i ence teachers ) during the pi l ot phase of the study . Performance qua l ity was gauged in two ways : (1) the ratio of strengths to weaknesses identified in each videotape using the critical-stop method (described in detai l bel ow ) and (2) the overall performance r ating of the videotaped demonstration on a Likert - type scale (See Clinical Intervi ew Guide , Question I . 6, Appendix I) . These two criteria were used to help select the videotapes and to assign a general performance rating to the videotaped presentation as either favorable (+) or unfavorable (- ). Ten videotapes were evaluated during the pilot pha se of the st udy . Four of t he tapes showed the desired characteristics usi ng the critical -stop task and Likert-type rating . These four videotapes featured chemical demonstrations performed by four different chemistry/physical science teachers. The targeted chemical concept and the general performance rating reflected by the teachers in the fo ur vi deotapes are summarized in Table 5. The playing times of the videotapes ranged from 3 . 5 to 9.5 minutes. using parallel versions of Demonstrations A and B prevented teachers from having to evaluate the same videotapes at the beginning 80 Tabl e 5 Descriptive Overview of the Four Videotapes Used in the Critical-Stop Task Demo. ID Videotape# , Chemical Concept , Duration of Code Title of General Rating , Videotape (A. /8. ) Demonstration +/ - (min) A. #1 Collapsing Aluminum Can Air Pressure , + 7.0 #2 Collapsing Aluminum Can Air Pressure , - 3 .5 B. #3 Density Column Density , + 7.5 #4 Density Column Density , - 9.5 Note . + Favorable Model Unfavorable Model and end of the workshop . This strategy was designed to minimize problems associated with "test practice" that may influence sub seque nt test response and threaten the internal validity of the inves t iga t ion (Drew & Hardman , 1985 ; Isaac & Michael , 1983). In thi s st udy , a parallel set of tapes refers to two videotapes of a speci f i c chemi ca l demonstration that differ notably in overall pe rformance quality. The average Likert-type performance rating and the number of strengths to weaknesses observed in the four videotapes se l ec t ed f or the study are given in Table 6 . The se data r epresent the ge ne r a l findings of a panel of experienced science t eacher s and sc i ence education graduate students who critiqued the videotapes during the pilot pha se of the study . The data provided prelimina ry evide nce t hat Tapes #1 and #2 (The Collapsing Aluminum Can Demo) and Tape s #3 and #4 (The Density Column Demo) r epresented parallel set s of video t apes . 81 Table 6 Critical -Stop Pi l oting of Videotapes #1 - 4 Concept : Average Number Videotape# , Type of Model , Performance No . Raters (n) Str Wk Rating A. Air Pressure : #1 Favorable Model , (3) 9 5 4.0 #2 Unfavorab l e Model , (4) 4 7 3 . 0 B. Density : Jr3 Favorable Model , (4) 6 3 3.6 #4 Unfavorable Model a , (2) 11 12 2 . 9 Notes. (1)Str = Strengths, Wk= Weaknesses (2) Rating Scale : 1/Low ---> 5/High a Because the average number of strengths and weaknesses identified for t his videotape was similar , the "Unfavorable " designation was determined by the Likert-type rating. Analyses of these videotapes by the expe rienced chemical demonst rators in this study later verified these ratings (See the data discussed in Chapter 4, p. 123 . For a list of the major strengths and weaknesses identified by the experienced group , see also Appendix Q) . The fir st tape i n each set (#1 and #3 ) represents a favorable model of chemical demonstration teaching . The second tape in each set (#2 and 14) represents an unfavorable model. These ratings were gauged by the re l ative number of strengths and weaknes ses identified for each videotape and the Likert-type overall performance rating . Videotapes #2 and #3 were obtained from the 1986 ICE Work s hop B offered at the University of Maryland. Videotape #1, which parall e l ed videotape #2 , wa s obtained during the first s ummer session of Workshop 82 B (1987). Because no equivalent videotape could be found during 4 summer session I (1987) that would parallel Videotape #3, Videotape # was staged. The staged performance involved a certified science teacher performing the Density Column demonstration in front of a small group of science education graduate students. The volunteer teacher received the ICE workshop sourcebook several days before being videotaped so she could familiarize herself with the demonstration. She also received the same demonstration materials the workshop participants used to help her practice and perform the demonstration. 1 11 Appendix J provides the scripts of the four videotaped : :, I, presentations. A description of the Collapsing Aluminum Can demonstration and the Density Column demonstration is given in Appendix K. The descriptions come from the ICE demonstration sourcebook (Sarquis & Sarquis, 1987). The four videotapes provided close-up shots of a chemical demonstrator performing the selected air pressure or density demonstration. The "student audience " for three of the videotapes consisted of summer Session I workshop participants asked to play the role of middle school students. The staged videotape (Videotape #4) used science education graduate students to play the role of middle school students. These "students " were not seen on videotape unless one was asked t o be a volunteer to assist with the demonstration. The sound track on the videotape, however , revealed considerable teacher-student interactions. A content analysis of the scripts showed that about 23 % of the statements across 4 the four videotapes consisted of teacher questions and about s % consisted of declarative teacher statements . Student talk comprised about 24 % of the statements. Most of the questions asked by the -- -- - - --- _.,._,. __ ; ___ , ,_ .. _, __ _ 83 videotaped teachers were di rect questions aimed at the "student" audience ; a smaller number (about 6% ) were r hetorical questions. A content analysis of the verbal i nteract i ons recorded on each videotape is provided at the end of Appendix J (Videotape Scripts , Table J . 1) . Data Col l ection Procedures Participant quest i onnaires and qua litative research methods we re used to help answer the four research questions presented in chapter 1 regarding teachers ' pedagogi ca l content knowledge . The qualitative met hods consisted of a think-aloud task and a semi-structured interview which together constituted the clinical interview. Participant Questionnaires Participant Information Form, PIF. The Participant Information Form (Appendix B) provided this researcher with background information on the workshop participants and instructors . Its purpose was to obtain a pre-workshop assessment of the genera l level of experience each research subject possessed with respect to chemical demons tration teaching. The data helped identify the less experienced/novice chemica l demonstrators participating in the workshop based on their self-reported confidence and weekly use in conducting chemical demonstrations . The PIF used in this study was a modification of a PIF origina l l y developed by O' Brien (1987 ) . The original questionnaire was short ened for this study by deleting two sections that were not pertinent t o this researcher or to the workshop designers . The modified form al so simplified the r e sponse to one item and added a question about teachers' confidence in conducting outreach programs involving 84 chemica l demonstrations. The form was reviewed for clarity and completeness by a s econd science educator before it was mailed to the participants . The PIF contained questions about the research subjects' s cience teaching experience , number of college chemistry courses taken, confidence in conducting chemical demonstrations , and weekl y fr equency in conducting chemical and other science demonstrations. It al so asked about the research subjects ' awareness and use of vari ous chemical demons tration sourcebooks. The remaining three quest i ons on the questionnaire contained items that were of interes t to the workshop designers. The confidence scale used in the ques ti onnaire consist ed of a five-point weighted judgment rating scal e (i. e ., 1 = low, 5 =high). Demons tration Log. This questionnaire /form provided in formation on the chemical demonstrations that the novice demonstrators had an opportunity to practice . The Log (Appendix L) accompa nied each of the 50 boxed demons trations. Works hop participants we r e as ked to s i gn t he de signated Log after successfully practicing a boxed demons trat i on. The information obt ained from thi s Log aided t hi s r esearcher i n keeping track of the kinds and number of chemica l demonstrat i ons practiced by the participant s during the works hop. Den s ity and Air Pressure Demonstration Checkli s t. This for m helped verify the information provided by the Demonst r ation Log . It a l so provide in formation on the demonstrations the works hop participant s had seen pe r f ormed by an ICE wo r kshop instructor or a workshop peer . Thi s form (Appe ndix M) cont ain s the titles of all the boxed demons trat i ons t ha t r e l a t ed to the concepts of de ns ity and a ir 85 pressure. The titles were taken from the ICE sourcebook (Sarquis & Sarquis , 1987) . At the end of the workshop , novice demonstrators were asked to indicate on the checklist which of the listed demonstrations they had an opportunity to observe or perform during the workshop . Responses obtained from this checklist were cross-examined with responses obtained from the Demonstration Log as well as a posted workshop schedule of publicly performed demonstrations. The data obtained from the Checklist provided useful information pertaining to the nature of the workshop intervention experienced by the inservice science teachers , e.g. , the number of boxed demonstrations novices actually practiced with respect to the targeted concepts. These numbers also made it possible to compare the frequency of novices ' post - workshop responses to interview question I.9 (Appendix I) regarding alternative chemical demonstrations on the target ed concepts , to the number of demonstrations they actually pe rf ormed (o r observed) on these concepts. Clinica l Interview A clinical interview was used to as sess subjects ' pedagogical content knowledge. The clinical interview cons i s t ed of two part s , a think-aloud critical-stop task and a semi-structured interview. Novice demonstrators participated in the clinical inte rviews a t the beginning and the end of the two-week workshop. Experi e nced demonstrators were also interviewed twice as their schedules permitted, i . e ., either before, during , and/or after the workshop. Ericsson and Simon (1984) point out that the type of probes used during clinical interviews (e . g. , questions , tasks , and vi s ual stimuli) specifies the type of information r eported by individual s . ----- - - . ; -- -- ---- --~-- --- ---- - -?- 86 Consequently, this study used two research methods (a think-aloud task and a semi - structured interview), two chemical concepts (air pressure and density ) , and four videotapes to help ensure validity of assessment. The strategy served to increase the chance of obtaining a full picture of teachers ' pedagogica l knowledge of demonstrating fundamental chemical concepts. Research subjects were randomly assigned to one of two clini ca l interview testing groups . Half of the experienced and novice subj ect s were randomly assigned to view Tapes #1 and #4 during the pre- workshop interviews while the remaining subjects were assigned to view a parallel set of these Tapes (#2 and #3). This viewing arrangement was reversed in the post-workshop interviews. The technique of counterbalancing the order of administration of the two set s of videotapes was designed to eliminate the possible confounding of order I l of testing effects with treatment effects. Each viewing sessi on ' pre sent ed the more highly rated demonstration fir s t so tha t the viewing sessions could begin in a positive manne r and with an atmosphere of trust and candor between the r esearcher and the s ubj ect . Critical-stop task . During the first part of the clinical interview , subj ects (i. e . , the experienced and novice chemica l demonstrators) were engaged in a critical-stop task. This t as k required the subjects to first view the videotaped demonst r at i on in its entirety , then view it again , stopping the videotape pl aye r whenever the subject judged a critical incident to have occurred on the tape . Thus , critical stops were initiated by the s ubj ec t s . Critical incidents (or critical points) we r e de fined f or the s ub ject s as : stre ngths or wea knesses in the presentation of the demons t rat i on. ----- ---..--- - _____ __, __ . - ? . 87 With each critical stop , the subjects thought - aloud and discussed features they perceived hindered or promoted effective chemical demonstration teaching. The task served as a probe of demonstrator s ' pedagogical knowledge of chemical demonstration teaching of two chemical concepts at a middle school level. Appendix I contains specific instructions that the researcher gave to the demonstrators for the critical - stop task. Before beginning the task , the demonstrators were introduced to the purpose of the interview and the study. They were then provided with instructions on how to conduct the critical-stop task. The demonstrators were asked to think-aloud and discuss fea tures they perceived hindered or promoted effective chemical demons tration teaching during a critical stop. Subjects were told to as s ume that the demonstrations were presented at the middle school level. As the demonstrators described the critical f eatures , the inte rviewer did not provide specific r e inforcements but indicated that the subj ect s we r e providing acceptable re sponses by saying, "Uh huh," "I see ," or "Please tell me more about that." (Novak & Gowin, 1984, p. 130). Semi-structured interview . The critical-s top tas k wa s fo llowed by a semi-structured inter view. During this interview, the researcher asked his subjects seven structured questions (Ques tions I.4 - I.10) and the n probed more deeply using open- ended ques tions in orde r to obtain more comple te data . Borg and Gall (1983) describe thi s type of inte rview as a semi- s tructured inte rview because of the combined use of structured ques tions and open-ended questions that va ry wi t h the nature of the r esponse provided by the interviewee . 88 The interview guide fo und in Appendix I begins with a few warm-up questions (Questions I. l - I. 3), that include questions such as , "Have you ever examined your own or someone e l se's teaching on videotape? What setting was that in? " and "Have you ever seen this demo before? Have you ever performed thi s demo before? " asked before the critical-stop task. Next , the researcher asked several questions designed to give the demonstrators an opportunity to identify additional critical features in the videotaped presentation they did not comment on during the cr i tical - stop task (Question I . 4) and to summarize the ma j or critica l feat ures they observed on videotape (Quest i on I . 5) . Questions of this nature included , "Are the r e any other spec i fic strong or weak points in the presentation you haven ' t already mentioned? What are they? Anything else? " "How would you summarize the ma jor strengths (weaknesses ) of the presentation? " The semi-structured interview was also designed to elicit subjects ' overall reaction to the demonstration performance (Ques ti on I . 6), understanding of the lesson objective (Question I.7) , knowledge of alternative demonstrations that could meet the same l esson objective (Questions I.8 and I . 9), and understanding of the scientif ic pr i nciples illustrated in the videotaped demonstration (Ques ti on 1 .1 0). Pertinent questions inc l uded , "Do you know of any variation , tw i st , or extension of this particular demonstration that could be performed? What would it be? Any others? " "Do you know of any other demonstration that shows the effects of air pressure (shows the concept of density ) ? What would it be? What other demonstrations do you know that shows the effects of air pressure (shows the concept of density ) ?" 89 Question I.7 (the intended lesson objective) and Question I.10 (concept explanations ) were considered optional items and asked by the interviewer only if there was sufficient time to obtain responses . The demonstrators were given 50 - 60 minutes for the critical - stop task and fol l ow - up interview. The interviewer used non-directive questions and responses to help minimize distortion of teachers' evaluative judgments of the videotaped teaching episodes (Shavelson, Webb, & Burstein , 1986). After each response the interviewer prompted the teacher with addit i onal questions such as "Anything else? " or "Can you think of other items? " or "Can you clarify that for me, please?" After the subjects described as many critical features (demonstrations or variations) as he or she was able, the interviewer proceeded to the next item in the interview guide, using the same questioning pattern . The interview questions served as probes of teache r s ' thoughts stored in Long Term Memory (LTM) (Shavelson, Webb, & Burst e in , 1986) . If experienced chemical demonstrators possess richer and mor e complete pedagogical content knowledge of demonstration teaching of chemical concepts , it would be expected that they would also be able to provide more thorough and detailed responses to the interview quest i ons than the novices. Given that the novice chemical demons trators had numerous opportunities to observe , practice, and/or perform chemica l demonstrations during the inservice workshop , it would also be expected that they could provide more complete answers to the interview questions after the workshop intervention . Pilot Study The clinical interview method that was used to probe teachers ' pedagogical knowledge was piloted prior to and during Summer Session I 90 of ICE Workshop B. During this time a col lection of ten videotapes that addressed the targeted concepts was reduced to four. The early pilot studies used an alternate set of air pressure and density demonstration video tapes because parallel sets were not available during this phase of the research. The pilot s tudy al so provided opportunities to clarify t he interview instructions and reduce the ambi guity of interview questions (Appendi x I ) . While piloting the interview guide , the volunteer sc i e nce teache rs we re asked at the end of the interview about the ways the instructions and ques tions could be improved in light of the research goals. These ins ights were incorporated into the revised inte rvi ew guide and again tested on se l ected ICE Works hop part i c ipants during Summer Sess i on I. This strategy he lped increase the face validity and construct validity of the interview guide in asse s sing t eachers ' pedagogical content knowledge . The feedback obtained during the pilot phase included corrm1e nt s that s uggested the need for this researcher to alter or de l e t e s ome of the interview questions in order to r educe r edundancy in respon ses . Several sub j ects also suggested that the term "variation " , and othe r phrases used by the intervi ewer , be explained to the i nt ervi ewee . The r evisions made prompted this researcher to collapse similar ques ti ons in to mor e inclus ive intervi ew questions (I.4 - I.5) and define se l ected terms and phrases more explicitly for the sub jects (I. 8 - I . 10) as shown in the final In terview Guide (Appendix I ). Piloting a l so made it possible to examine how long it would t ake to complete the interview and assess t esting fatigue . Fina lly , the pilot phase of the study helped this r esearche r develop confide nce 91 that the clinical interview research strategies developed for this s tudy could effectively discriminate subjects based on their prior chemistry demonstration teaching experience. Research Design to Study Pedagogical Knowledge Growth The assessment of pedagogical knowledge growth among novice chemical demonstrators was addressed by a one-group quasi-experiment al design (I saac & Michael, 1983 ) . Thi s design is appropriate when a researcher attempts to study a human behavior , attitude , or cognitive characteristic that is fairly stable without intervention e fforts (Borg & Gall , 1983 , p.659). In this study , the dependent variables (e . g ., the gene r al characteristics of teachers ' critical-stop comments and the number of times they elicited responses in a given taxonomic category ) were , I ,l gathered before and after teachers received the inservice works hop ; :J ,l :J intervention (the independent variable). l Several null hypotheses (where u 1 = pretest and u2 = posttest ) were tested to assess the impact of the workshop on novice demonstrators' pedagogical knowledge growth (e .g., changes in discourse frequency within specified taxonomic categories ). for the mean number of criti cal incidents critiqued, (e.g ., Str, Wk) for the mean number of Pedagogical Enhancements , Variations , and Demonstrations elicited during the clinical interview (free-recall). Similar null hypotheses (wher e u 1 = experienced demons trator s and u2 = pre- workshop novice demonstrators or post-workshop novice demonstrators) were tested to assess quantitative differences in experienced and novice chemical demonstrators ' pedagogical knowl edge - ------------------------ 92 base (e.g., differences in discourse frequency within specified taxonomic categories ). The data used to r un these tests were taken from a content analysis of the protocols. Null hypotheses involving frequency of response to several semi -structured interview questions (I.6, I .8, & I .9, Appendix I) were tested with parametric stati stical tests. Hypotheses involving critical-stop tas k critique frequency required the use of non-parametric stat ist ical tes ts. Rejection of the null hypothesis would support the notion that a two-week , skills-oriented workshop can foster pedagogical knowledge growth among novice chemica l demonstrators in the realm of demonstrating fundamental chemical concepts to students at the pre-college level . It wou l d also indicate experienced-novice demonstrator differences with r espect to chemical demonstration teaching knowledge. Moreover , qualitative ana lyses of demonstrators ' clinical interview di scourses were conducted to examine experienced and novice chemical demonstrator differences and pedagogical knowledge growth from a qualitative perspective . The methods used to conduct the qualitative analyses of the verbal reports are discussed in detail in the following section. Qualitative Analysis of the Clinical Interview Protocols This section describes the various types of analyses conducted on the protocols containing the record of participants ' responses during the critical-stop task and fol l ow-up interview. It specifies how the protocol s generated information about t eachers' pedagogical knowledge of chemical demonstration teaching. All of the following analy ses -~- ??-- - . . . - ?------ ------------ -- 93 were conducted by this researcher , many of which were verified by independent coders. Analysis of the Critical-Stop Protocols The research methods of Spradley (1 979, 1980 ) guided the analy s i s of the verbal data obtained from the think-aloud critical-stop task . These methods included a domain analysis , a taxonomic analysis, a componential analysis , and a theme analysis. This section discusses these four qualitative methods as well as a content analysis technique that permits a quantitative examination of the verbal data (Borg & Gall, 1983) . Domain Analysis A domain analysis (Spradley , 1979 , 1980) was performe d on the data by identifying in the written protocols meaningful propositions related to demonstration teaching pedagogy. This was followed by a systematic search for conceptual categories that would group propositions possessing a similar semantic relationship be twee n the subject and predicate. Spradley (1980) provides a list and description of nine domains a researcher could begin with in a na ly zing verbal reports. These nine domains were included in the domain analysis conducted in this study . Additional domains were gene rated whenever propositions did not satisfactorily reflect Spradley' s suggested nine. The domain analysis of the teachers' critical - stop discourses showed that most propositions reflected one of four bas i c domains or semantic relationships . The four domains included one s uggested by Spradley and three identified by this researcher. Although a fe w othe r semantic relations hips we re occasionally evident in the verbat i m 94 transcripts, this s tudy did not attempt to come up with the most comprehensive coding scheme, but one that was sound, robust, and in accord with the research goals. Other semantic relationships ! suggested by Spradley were evident but occurred infrequently. several I of these "infrequent " discourses were shown to adequately fit under I one of the four domains identified by this researcher. Chapter 4 shows that these four domains provide a suitable framework for organizing the critical-stop data and for comparing experienced and novice chemical demonstrators' pedagogical knowledge discourses. Examples of the four semantic forms are provided in Chapt er 4 using data (verbal discourses) obtained from the verbatim transcripts. The domain analysis helped identify the major domains describing ".) '? teachers' pedagogical knowledge discourses (Research Question 1). ,1 Inte r-coder reliability in encoding individual propositions from the verbatim transcripts into the hypothesized domains was det e rmined. Coding instructions for this task are provided in Appendix N. Taxonomic analysis After identifying the major domains and classifying propos itions according to these domains , a taxonomic analysis was conducted. This analysis consisted of a search for categories and subcategories within each of the domains by searching for similarities and contra s t s within the verbal data (Spradley , 1979, 1980). The taxonomic analysis began by inducing categories of pedagogical issues within an identified domain. If , after gene r ating several categories , a given proposition did not code into an ex i s ting category, a new category was generated. After identifying the ma j or 95 domains, categories, and subcategories , inferences were made about the content of teachers' pedagogical knowledge discourses. This study identified the most preval ent categories within each domain. It did not attempt to systematically identify the optimal clustering of subcategories and categories to describe teachers ' pedagogical knowledge discourses because several legitimate ways of grouping the data emerged. In any case , once the categories and subcategories within a domain were identified, a taxonomy of t eache r s ' pedagogical knowledge discourses was formulated in the form of an outline. The domains , categories , and subcategories that emerged from t he critical - stop data became useful in coding the semi-structured interview protocol s and for providing a framework for di scuss ing pedagogical content knowledge differences between experienced and novice chemical demonstrators. The categories generated by a ta xonomic analysis within the four domains provided the coding s chemes needed to conduct content analyses of the verbal data (see di scussion below ) . Componential Analysis Spradley (1979 ) defines componential analy s is as a search f or formal and logical differences among members of the contrast group. In this study , the dimensions of contras t included dif fe r ences between experienced and novice chemical demonstrators' discour se along identified taxonomic categories. Similar contrasts wer e sought between pre - and post-workshop novi ce demonstrat ors r egarding thei r discourses on the attributes associated with effective chemical 96 demonstrat ion teaching of targeted chemi cal concepts and their knowledge of alternative demonstrations on these concepts. Content Analysis Content analysis i s commonly used to detail the frequency wit h which symbols or themes appear in a written document (Williamson, Karp , Dalphin , & Gray , 1982 ) . This technique was used to provide a quantit at ive description of the verbal data gathered in this study. The domains , categories , and themes identified through Spradl ey ' s techniques provided the coding categories for the content analys es . Given that the clinical interview generated over 400 pages of transcripts , the content analyses permitted this researcher to characterize a large volume of material rather efficient ly us ing a few frequency table s . These frequency tables summarized the number of times sub jec ts made comments within each of the pre-defi ned content ,,:. categories or subcategories in their verbal discourses. They provided I a general assessment of the breadth (or extent) of teachers' pedagogical knowledge of eff ect ive chemical demonstrati on teaching. All critical-stop task and f ol low- up interview protocol s were subj ected to content analysis us ing several coding schemes that included: frequency counts of (1) the number of critical stops and the number of perceived strengths, weaknesses , and acceptable- but - could - be-better ratings identified during the think-aloud task , (2) the content foc us of teachers' pedagogical critiques along nine content categories , (3) the number of s uggest i ons for pedagogically enhancing the videotaped demon strations , (4) t he number of variations on the observed chemi ca l demonstration, (5) the number of other demonstrations mentioned that address t he targeted chemical concept , 97 (6) the number of extraneous suggestions, and (7) the distribution of critical-stop and interview comments according to general pedagogical knowledge and pedagogical content knowledge (Wil son , Shulman , & Richert, 1987). The instructions for encoding the protocols according to these schemes are provided in Appendices N, O, and P. Trained coders used these instructions as a guide for coding a set of logically-sampled protocols (usually a sample of 4-6 critical- stop protocols and 4-6 semi-structured interview protocols). Appendix N shows the coding instructions used to code individual propositions according to the four pedagogical knowledge domains . Appendices o and P provide the instructions for coding the protocols according to schemes (1) - (7) listed above. Borg and Gall (1983) discuss the counting procedures used to perform the content analyses . Reliability in coding the data was obtained by determining percent agreement between this researcher and a second coder in coding propositions from a set of logically-sampled protocols into pre-defined coding categories. This researcher trained three coders. Two of the coders have Ph.D.' S in science education and a third code r had a Ph.D . nearing completion (ABD status ). All had at least six years of science teachi ng experience. Each coder received training on a different content analysis coding scheme , i.e ., a coding scheme f or a domain, taxonomic , and theme analysis . The tallies obtained from the three coders were compared to this researchers' coding and percent agreement computed for each scheme. Because this agreement inde x does not consider the extent of inter-coder agreement which may result from chance, Scott ' s pi was computed. This index of reliability provides a suitabl e correction for chance agreement when coding verbal da t a 98 (Kang , 1987). The inter-coder agreements (reliabilities ) are reported and discussed together with the quantitative findings presented in Chapter 4. A non - paramet ric Wilcoxon matched-pairs ranked-signs test (Hul l & Nie, 1981) tested for total frequency differences between the experienced and novice demonstrators conducting the critical - stop ta sk . The two groups of demonstrators each critiqued a total of four videotapes , providing four matched pairs of observations for the ranked-signs test (e .g., experienced and novice group mean frequency scores were paired on each videotape). Wilcoxon matched-pairs ranked-s igns te s ts were also performed on each of the nine taxonomi c categories that emerged from the taxonomic analysis, e.g ., the categories of Inquiry , Ques tioning Strategy, New Terms, Mechanic s of Demonstrat ion. Summed frequency tallies across t he taxonomic categories yielded an additional non -parametric t es t . The unit of analysis used to run the s tatistical tests associated with the critical - stop data r epresented the group frequency means cal cula ted for each videotape. This yielded four sets of paired values . Differences between experienced and novice demonstrators in t e rms of the ave rage number of demonstrations, demonstration variat i ons , pedagogical enhancements , etc ., elicited during the semi -s truc tured interview (see coding schemes 3-7, listed above on pp. 96 - 97 ) we re assessed us ing an independent t-test and a more conservative non-parametric Mann-Whitney U test. The U-test wa s conduc t ed because its t ol erance for using a smal l sampl e size and for working with dat a that deviate from normality . 99 Frequently-Cited Critical Features Content analysis of the critical-stop data provided a means for assessing the level of agreement among experienced and novice demonstrators in identifying videotaped features critical to e ffect i ve chemical demonstrating. The criteria used to identify a "frequently-cited" critical feature was 50 % within-group agreement . This meant that at least half the subjects in the experienced or novice group would need to identify and discuss the same critical feature displayed by the videotaped teachers. The r emaining featur es were labe led "infreqntly-cited" critical features . This convention simplified the protocol analysis task and the search for a set of critical incidents frequently discussed by experienced and/or novi ce demonstrators. Assessing Pedagogical Knowledge Growth in Novice Demonstrators Changes in novice demonstrators' performance on the critical-stop task and semi-structured interview were examined both quantitat ively and qualitatively. A Wilcoxon matched-pairs ranked- signs test (Hull & Nie, 1981) tested for experienced and novice chemical demonstrator differences (Research Question 2) and for quantitative changes in novice performance on the critical-stop tas k resulting from the workshop intervention (Research Ques tion 3). This non- parametric t es t wa s conducted whenever quantitative differ ences in critical-stop task performance was assessed. Thi s test wa s selected because it permitt ed mean critica l -incident frequency scor es for t he comparison groups (e.g ., pre- and post -workshop novices , or pre - workshop novices and experienced demonstrator s ) to be paired with respec t to the f our 100 videot apes. Pair-wise grouping of scores was necessary because the four videotapes used in this study had very different playing t i mes (3 . 5 - 9.5 minutes ), a variabl e that correl ated strongly with the number of critical incident s discus sed by participants. It was al s o nece s s ary because the counterbalanced design used in thi s s tudy had novices randoml y ass i gned to one of two videotape viewing groups , wi th each s ubgroup vi ewing a different set of two videotapes . At t emp t s at using s tatistical t e sts that ignored the pairing of videot ape scores yie lded large variances in scores and no statis tical diffe r ences i n group performance on the think- aloud ta s k, (p > . 30) . Tot al scores across t he four videotapes could not be used as a unit of analy s i s because each novi ce demons t r ator only analyz ed two of t he four videot apes during the pre-workshop and post-workshop clinical i ntervi ews . These condit i on s made the Wilcoxon matched-pair s ranked-s ign t es t the mos t suitable s t ati s tical t es t for analyzi ng the quantitative aspect s of the critical-s top data. Analyzing quantitative changes in novices ' r esponses duri ng the semi -s t r uctured interview helped gauge their knowledge growth i n demons trating targe t ed chemi cal concepts . One-ta i l ed dependent t-tes t s and non-par ametric Mann-Whitney U t ests tes t ed for st ati s t ically s i gnifica nt changes in t he numbe r of (1) alte rna tive chemical demons trati ons , (2) demonstrati on variations , and (3) ex t ra neous exampl es eli cited during the i nterv i ew. These t es t s al so se r ved t o compare the semi-structured intervi ew r esponses of experienced and novi ce chemica l demons trator s . Al l t - tests we r e performed us ing a micr ocomputer and a s t at i st i cal softwar e package 101 (Buhyoff et al ., 1985). Non-parametric statistics were performed using the SPSS stat i stical package (Hul l & Nie , 1981 ) . After novice demonstrators ' pre - and post-workshop transcripts were read , coded , summarized, and contrasted, qualitative changes in novices' pedagogical knowledge resulting from workshop participation were captured through descriptive summaries and illustrated with quotes taken from the verbatim t r anscripts . Quotes from experienced chemical demonstrators were also provided for comparison . These summaries/quotes included the kinds of representative comment s made , examples provided, demonstrations suggested , and illustrations cited by the two groups of demonstrators during the think-aloud task and semi-structured interview. These pre- and post - workshop quotes , together with the descriptive summaries , taxonomic outlines , and the statistical tests run on the frequency data , provide a body of evidence for generating and testing hypotheses regarding (1 ) pedagogical content knowledge differences between experienced and novice chemical demonstrators and (2) pedagogical knowledge growth among novice chemical demonstrators . Chapter Summary This chapter described the research subjects , instrumentation, inservice treatment , and data analysis procedures used to address Research Questions 1 , 2 , and 3. A clinical interview, consisting of a critical-stop task and a semi-structured interview , served to probe various aspects of experienced chemical demonstrators ' PCK and novice chemical demonstrators ' PCK prior to and after participating in an NSF-supported summer workshop. Domain , taxonomic , componential , and 102 theme analyses and several content analyses guided the qualitative analysis of the interview protocols. These research methods pr ovi ded the means for investigating the nature of chemical demonstrators' pedagogical content knowledge and general pedagogical knowledge of demonstrating fundamental chemical concepts. They also provide the means for examining the nature of pedagogical knowledge growth in an inservice context . 103 CHAPTER 4 FINDINGS OF THE STUDY Introduction This chapter presents the findings related to the three major research questions stated in Chapter 1. Research Questions 1 and 2 focus on the commonalities and differences between experienced and novice chemical demonstrators' pedagogical knowledge (i.e ., their pedagogical content knowledge, PCK, and general pedagogical knowledge, GPK ) with respect to the demonstration teaching of two chemical concepts. Research Question 3 addresses the issue of how inte ns ive inservicing can influence science teachers' pedagogical knowledge growth . Chapter 4 is organized around these three research ques ti ons . This chapter specifically identifies and discusses the patterns observed in the verbal reports obtained from participating teachers . '. ' t The identification of these patterns are based on Spradley's (1 980 ) methods for analyzing qualitative data, and include a domain, taxonomic, componentia l, theme, and content analysis. The findings from each of these analyses are presented below and provide insight s into answering the three research questions posed by thi s s tudy. The Identification of the Domains Characterizing Experienced and Novice Chemical Demonstrators' Think-Aloud Discourses The critical-stop protocols obtained from five experienced chemical demonstrators and eight novice chemical demons trators were used to address Research Question 1. This question as ks about t he domains of knowl edge that characterize experienced and novi ce chemical 104 demonstrators ' pedagogical comments on effective chemical demonstration teaching . Genera l Findings A domain analysis (Spradley , 1980 ) of demonstrators' think - al oud, critical-stop discourses of Videotapes #1- 4 identified four maj or domains of verbalization. The analysis showed that greate r than 95 % of the s ubjects' verbal proposit i ons reflected one of the foll owing four knowledge doma i ns pertaining to effective demonstrati on teaching : a . Evaluative Judgments b. Descriptive Knowledge c . Knowledge of Alternatives d. Rationales During each critical-stop of the videotape, the research participant would offer at least one , us ually several ve rba l '; propositions about the quality of the critical f eatures di sp l ayed by ' I the videotaped model . Most of the verbal propositions in t he ve rbatim transcripts (> 98 %) reflected meaningful , comprehensible s t a t ement s about chemical demonstration teaching. The res t of the propos i t i on s in the transcripts (< 2%) represented exclamatory comments , incompl e t e comments, and comments that were undecipherable during the aud iotape transcription. According to two independent coders , l ess t han 5% t he comprehensible comments present in the data did not code in to one of the four domains. These comments were placed into a fif t h domain labeled "Other Knowledge" (See Instructions for Coding Pr opos it i on s into Domains, Appendix N). The "fifth domain" con s i s t s of a cl ust e r of eight other domains di s cussed by Spradley (1980) but rare l y used in teachers' discourses . Although some of these domains we r e 105 infrequently used, coders were still instructed to conside r all 12 domains in the process of coding the verbal data. Intercoder reliability in coding teachers ' critical-stop discourses into the domains was 73 .5%. Scott's Pi coe f fici ent was calculated to be 71.1 % when coding into the 12 domains. Coder differences are accounted for , in part, by the fact that s ome propositions were complex and could be assigned to more than one domain. The nature of the four major domains stated above are now illustrated with think-aloud discourses obtained from two experienced and two novice chemical demonstrators. These four discourses give an exampl e of the robustness of the domains across experienced and novi ce subjects and across favorably and unfavorably-rated critical incidents. The se discourses also provide evidence for the f ace validity of the four domains used to organize the clinica l in terv i ew data . The first critica l incident critiqued represents a "pe r ce ived strength" discussed by experienced demonstrator E2 and novice demonstrator N7 individually observing Videotape #1 (A fav or ab l e mode l of the Collapsing Aluminum Can Demonstration which deal s with t he concept of air pressure). The critical incident relates to the se l ection and use of a student volunteer during the beginning of t he chemical demonstration. Analysis of Critical Incident A E2 ' s Discourse on Critical Incident A: That's a really good thing to do, to get participation from everybody like that , [i.e., polling the class a l ong a given criterion to se lect a s tudent voluntee r to hand-crush an aluminum can (see Video Script , Appendi x -- - - - - ----~=~ ~ - ----- 106 I)]. It makes it , not cutesy, but if he's working with middle school kids, then that particular technique is excel l ent ' cause they want as much involvement as they can and you can see that when we work with the little kids , as soon as you ask for a volunteer, everyones ' hand goes up . So everyone wants to participate. (Videotape #1 , S:4) . Experienced demonstrator E2 stopped the videotape at point S: 4 (Videotape Statement 4 out of 97 ) and discussed this critical incident involving the videotaped t eacher ' s selection and use of a s tudent volunteer . A domai n analysis of the propositions in the above discourse reveals the presence of three of the four coding domains , namely, a , b, and d. a. Eva luative - That ' s a really good thing to do , ... b. Descr i ptive - to get participation from everybody like that , [i. e. , polling the class along a given criteri a t o select a student volunteer to hand-crush an aluminum can]. d. Rationale - I t makes i t , not cutesy , but if he ' s wor king with middle school kids , ... a. Evaluative - then that particul ar technique i s excellent d . Rationale - ' cause they want as much involvement as they can and you can see that when we work with the little kids , as soon as you ask for a volunteer , eve ryones ' hand goes up. So everyone wants to participate . The above clustering of propositions illus trates th ree of the four domains. The opening phrase, "That ' s a really good th i ng to do," reveals that a positive eval uative judgment was r ende r ed by E2 on t h obse rved critical incident. This judgment is evidenced by t he fact that the proposition contains the value terms "r eally good." The semantic re l ationship reflected in thi s opening s t at ement ca n be r epresented by : X (a r eally good thing ) i s an evaluati on of Y (polling 107 students to se l ect a volunteer). Another statement in E2's discourse also reflects this semantic relationship, namely the statement, "then that particular technique is excellent." It too was coded into the evaluative judgment domain. The presence of "positive" value terms, such as "really good" and "excellent", in a think-aloud discourse causes the entire discourse and corresponding critical incident to be classified as a pe r ceived "strength." This coding procedure is important to the next l eve l of data analysis, i.e., a taxonomic analysis of chemical demons trators ' eva luative judgments . (See Research Question 2, p. 114). The remaining portion of the first statement in the discourse above, "... [for him] to get participation from everybody like that ... " represents a brief description of the critical incident observed on Videotape #1 as perceived by experienced demonstrator E2. The description re fers to the method used by the videotaped t eacher for selecting a student to participate in the demonstration. The semantic relationship among the terms in this proposition is descriptive [Xis a behavior of Y]. The proposition provides a basic description of the observed teaching behavior that E2 associated with effective chemical demonstrating. rt reflects descriptive knowledge of chemical demonstration pedagogy, a second domain in the scheme. The remaining comments in the above discourse r epresent a rationale for why chemical demonstrator E2 rated the critical incident as "really good." This demonstrator reasoned that the technique of calling on student volunteers can be an important part of effective chemical demonstrat ion teaching at the middle school level because students at this age have a strong t endency to en joy and desire 108 participation in classroom demonstrations , [" . . . ' cause they want as much involvement as they can .. . "]. The semantic relationship of the terms in this proposition reflects yet another pattern , [Xis a reason for Y]. such propositions represent knowledge of why specific teaching strategies are able to motivate and interest students to learn science. They are indicative of the domain , "Rationales ." Analysis of the above discourse shows that experienced demonstrator E2 discussed the student volunteer incident evoking three domains of pedagogical knowledge: descriptive knowledge of demonstration teaching, evaluative judgments of effective demonstrating, and a rationale for why the observed critical incident was judged favorably. A domain analysis of a novice demonstrator ' s critique of virtually the same critical incident is examined next. N7 ' s Discourse on Critical Incident A: And then what he is going to do right now is in-class participation of the students [i.e . , selecting a s tudent volunteer to hand crush the can]. Again, this gets their attention and I think that is really important. He ' s just not the only one doing the demonstration. In addition, he is getting the class involved. (Videotape #1, S:7). A domain analysis reveals the following: b . Descriptive - And then what he is going to do right now is in-class participation of the students [student volunteer to hand crush the can]. d. Rationale - Again , this gets their attention and a . Evaluative - I think that is really important. d. Rationale - He ' s just not the only one doing the demonstration. In addition, he is getting the class involved. (Videotape #1, S: 7) . 109 From the above domain analysis it is clear that novice chemical demonstrator N7 discusses the student volunteer incident invoking the same three domains of knowledge as experienced demonstrator E2. The se three domains held up equally well in novices' pre- and post - workshop think-aloud discourses . The presence of these three domains r ef l ect s one of the strongest commonalities observed in the discourses of experienced and novice demonstrators ' critique of favorably ra t ed critical incidents. The coding scheme is also applicable when teachers critique critical incidents that are viewed unfavorably. The nex t two discourses illustrate the scheme using unfavorable incident s. These two discourses differ from the ones discussed above in three r espec t s ; namely, they come from different chemical demonstrators comment ing on a different videotape illustrating a different chemica l concept . The first of these discourses comes from subj ect ES critiquing Videotape #3 showing a teacher demonstrating relative densities with a dens ity column . Analysis of Critical Incident ~ ES's Discourse on Critical Incident B: one of the weaknesses is that he seems to be continually walking from the front to the back of the [demons t ra t ion ] bench while he was asking the [class) questions . And he knew he was going to write [the answers) down. He probably should stay at the back or if he wanted t o have one student act as a secretary, it would probably be better to get that arrangement out of the way at the very beginning before the demonstration s tarts because it' s distracting to arrange for that in the middle of t he demonstration. (VT #3 , S:7) A domain analysis of this discourse reveals the foll owing domains , a . Evaluative - one of the wea knesses i s that ... 110 b. Descriptive - he seems to be continuall y walking from the front to the back of the [demonstration ] bench while he was asking the [class ] questions. And he knew he was going to write [the answers ] down. c . Knowledge of Alternatives - He probably should stay at the back or if he wanted to have one student act as a secretary, it would probably be better to get that arrangement out of the way at the very beginning be fore the demonstration starts ... d . Rational e - because it ' s distracting to arrange for that in the middle of the demonstration . The critical incident E5 chose to discuss involves the presence of a large demonstration table which hindered the videotaped t eache r' s access to the blackboard. It was E5 ' s perception that the vi deotaped teachers ' frequent movement around the table to get t o the blac kboard hindered the effectiveness of the chemical demonstrati on presentat i on. The value term "weaknesses " in the opening proposition explicitly indicates that the experienced demonstrator made a negative , or unfavorable , judgment about the incident. The semanti c re l at i onshi p of the terms in this proposition is of the form: X (one of the weaknesses ) is a judgment of Y (continually walking fr om the f r ont to the back .. . ). The discourse above also provides a brief description of the critical incident being evaluated, namely that of the videot aped demonstrator walking back and forth around the demonst ra ti on table to get to the blackboard . The semantic relationship refl ected in th i s proposition can be represented by : X (walking from the front to t he back ) is a behavior of Y (videotaped teacher). The proposition therefore illustrates descriptive knowledge of chemical demons t rat i on t eaching, domain b. _ ,- _? ., ____ ., ____- .;-~ 111 Experienced demonstrator ES also provides a suggestion for dealing with the dilemma of confronting a large demonstration t abl e that makes accessing the blackboard difficult. Subject ES indicates that simply staying back behind the demonstration table would facilitate access to the blackboard. He also indicates that the teacher's eventual use of a student secretary at the blackboard could be further enhanced by arranging for the selection of the student recorder at the beginning rather than during the middle of the demonstration . Both comments represent solutions to the "blackboa rd problem" and the task of coordinating the various components of effective chemical demonstration teaching. The semantic r elationship of the terms in these two statements [Xis another way t o do Y] provides evidence for the domain - Knowledge of Alternatives , domai n c. Experienced demonstrator ES also provides a justification for one of the improvements he advanced. The remark , " .. . because i t ' s I distracting to arrange for that in the middle of the demonstrat ion" represents a practical justification for why the observed t eaching behavior (asking for a student recorder~ several students provided answers to an open-ended question) is perceived to hinder effective demonstration teaching. The terms in this pr oposition are characteristic of a rationale statement , and its presence provides support for the rationale domain in the coding scheme . An examinat i on of many other critical-stop discourses by this resear cher showed that rationale statements were found in conjunction with demons trat ors ' evaluative judgments as well as their suggestions for al te rnative behaviors. - 112 The comments of a novice chemical demonstrator critiquing the s ame general incident mentioned above is now examined for the four domains . N2's Discourse on Critical Incident B: I think it ' s difficult to have to walk around the table as much as he does . Maybe he can walk around the other end because it's shorter . But you still want to be with the kids and not separate it [the presentation? ] from your audience. (VT #3 , S:3) A domain analysis of this discourse reveals the following domains: a. Evaluative - I think it's difficult ... b. Descriptive - to have to walk around the table as much as he did . c. Knowledge of Alternatives - Maybe he can walk around the other end d . Rationale - because it ' s shorter. But you still want to be with the kids and not separate it [the presentation?] from your audience . In this discourse novice demonstrator N2 deals with the critica l incident of the videotaped teacher having to walk around the demonstration table to get to the blackboard. A domain analysis illustrates that novice demonstrator N2 evoked the same knowledge domains as experienced demonstrator ES in her critique of this particular incident. Both judged the incident as contributing t o ineffective chemical demonstration teaching. The evaluative judgment rendered by experienced and novice demonstrators in the four discourses discussed above were stated rather explicitly. Occasionally, demonstrators render ed judgment s in an implicit manner. When this occurred, the direction of the evaluative judgments , strength or weakness, needed to be inferred f rom 113 the avai l able text. An example of an incident where the evaluative judgment was implicitly rather than explicitly stated is given in Appendix O (Coder Instructions: Step 14., Example #3). Most critical-stop discourses coded in this study (> 94 %, based on two independent coders ) contained evaluative judgments that were either explicitly stated or could be implicitly derived from the text. The domain analysi s conducted on the four discourses above illustrates a general pattern observed in coding the remaining critical-stop discourses. Discourses of favorably rated critica l incidents usually contained three major domains (a,b,d), whe r eas di scourses of unfavorably rated incidents contained four (a - d) . Summary of the Domains Characterizing Experienced and Novice Demonstrators' Pedagogical Di scourses Research Question 1 asks about the identity of the ma j or ? ) knowledge domains that characterize experienced and novice chemi ca l ; i demonstrators ' comments pertaining to effective chemical demons t ra t ion teaching. A domain analysis of the cri tica l -stop di scourses s ugges t ed that the domains (i) descriptive knowledge, (ii ) evaluative judgme nt s , (iii) knowledge of alternatives , and (iv) rationales character i ze these teachers ' pedagogical comments . These four domains high l i ght a major commonality between the two groups of chemical demonst r a t ors i n t e rms of their t hink - al oud di scourses on chemical demons tra tion teaching. The se four domains also serve as a useful or gani zing scheme t o present t he fi ndi ngs as sociated with the second r esea r ch ques t ion . The identified domains appear to be logi ca l ly consi s t ent wi t h t he ki nds of in fo r mation one would expect to obt ain from a clini ca l 114 interview designed to probe science teachers' knowledge of chemical demonstrating. Having defined the domains that reflect the nature of demonstrators' pedagogical discourses, a more deta i led taxonomi c analysis of the verbal data can now be conducted. This taxonomic analysis, together with a quantitative content analysis, will help highlight some major differences between the two groups of demonstrators in each of the four knowledge domains discussed above . Pedagogical Knowledge Differences Between Experienced and Novice Chemical Demonstrators An analysis of the ve rbal data gathered from expe rienced and novice demonstrators during the critical-stop task and semi-structured interview addresses Research Question 2. Thi s question explores more deeply the commonalities and differences between experienced and novice chemical demonstrators' knowledge of effective chemical demonstration teaching of specific sub ject-matter topics. Two different methods of data analysis were used to help answer the second research question. One was a taxonomic analysis of the categories and subcategories within the four pedagogical knowl edge domains discussed above (Spradley , 1980). The second method was a quantitative content analysis showing how exper i enced and novi ce demonstrators ' think- aloud comments were distributed acros s the taxonomic categories and subcategories (Eri csson & Simon , 1984) . Because of the close relationship between taxonomic analysi s and content analysis, the findings based on these two methods are presented concurrently. 115 The findings related to Research Ques tion 2 are presented in four parts. Each part focuses on one of the domains that describe the general nature of demonstrators' pedagogical discourses. Data pertaining to teache rs' evaluat i ve judgment is pr esented first followed by an analysis of thei r descriptive comments. Separate attention is then given to each of the domains: knowledge of alternatives and rationales. In this section , all references to novice chemical demonstrators are in regards to novices prior to the ir invo l vement in the chemical demonstration workshop. Chemical Demonstrators ' Evaluative Judgments The Nature of the Evaluative Judgments The evaluative judgments rendered by experienced and novice chemical demonstrators during the critical-stop task were subj ected t o a taxonomic analysis. This analysis revealed that the two groups of demonstrators classified most critical features (> 90 %) as eithe r a "Strength" or "Weakness." In pa r t, this evaluation scheme was imposed on the subjects by the researcher when the subjects were provided wi t h the critical-stop instructions (Appendix I). It should be noted he r e , that in this dissertation , the terms critical features and critica l incidents may be discussed interchangeably. The difference s are subtle , but worth noting. Critica l features r e f e r to speci f ic chemical demonstration teaching behaviors. Critical incidents r efer to the occurance of those behaviors in an instructional setting. (See Chapter 1 , Definition of Terms) . Some of the subjects devised an "acceptable- but - could-be-better " judgment to evaluate a few critical features . This judgment was 11 6 unprompted by the re seache r and was more evi dent among t he experienced demo nst r at ors . These judgment s we r e r e f l ected in s tatements such as : "I would have liked t o have hea rd it earl i er , but I still think it ' s good . . . " (E2) ; "Or be tter yet . . . " (E3 ); " It woul d pr obably be even mor e effective if she would have . . . " (E4 ); and "It might be infrequently cited , but it e ff e cted me when he . . . " (El) . Such s tat ement s s ugges t that experienced demonstrator s we r e abl e t o articul at e judgments that indicated that t hey recognized good and acceptab l e demonst r at i on t eachi ng behav i or s tha t could be even further enhanced. Subject E3 s ummed up thi s f or m of eva l uat i ve j udgmen ts with the fol l owing st at ement: E3 : Mos t of the st r engths and weaknesses , its a matter of degree . It ' s not so much , i n the sense of a wea kness , t hat he di d somet hing that was totally wrong . Some of his strengths he could have done bet t e r on in t erms of vi s ibility and i n terms of drawing out mor e examples from the s tudent s ' l i ves. So , mos t of the char acterist i cs of a good demo seem to be exemplif i ed in this demonstra t ion, in s ome form. It's more a matter , he could impr ove what i s al r eady a pretty good demonst ra t ion . (VT #3; I. 5) Al though this t hird category of evaluative judgments wa s not used as fre quently as t he s trength-weakness cat egories , di fferences between the expe rienced and novice groups of demons t r ators on the use of thi s evaluat i ve judgment category are appa r ent (see Tabl e 7 , page 118 ) . A small per centage (6% ) of t ea che r s ' j udgment s were neut r a or nebulous i n natu re . Th is bri ngs to a total of f i ve the categories used by teachers t o render evaluative j udgments during the critical - stop task , i . e ., strength , weakness , acceptabl e-but-could- be -bette r , neutra l , and nebulous . This researcher used this taxonomy of evaluative ca t egories to perform a quanti tative content ana l ysi s of teachers ' crit i ca l - s t op di s courses (see page 118) . 117 Inter-Coder Reliabi l ities Teachers ' evaluative j udgments of critical features were encoded from the protocol s with 84.2 % coder agreement . This agreement value indicates that two trained coders agreed 84.2 % of the time in the three-fold task of identifying evaluative propositions in the verbat im transcripts , associating these evaluative propositions with the description of the incident , and then categorizing the incident into an evaluative category . (Scott ' s Pi was ca l culated to be 80 . 2%) . Most of the disagreements between coders involved differences in coder percept ion of the number of critical incidents discussed by the demonstrators , i.e ., present in the protocols . One coder , for example , might read two successive verbal propositions in a given discourse and consider the first to be an elaboration of the second and therefore code the discourse as representing only one perceived strength or critical incident. A second coder might consider the two propositions as r epresenting two different pedagogical issues , and thus code the discourse as representing two perceived strengths or critical incidents. Most coder disagreements were of this nature . The remaining cases of coding disagreement involved critical incidents coded "neutral" or "nebulous " by one coder and as a "strength " or "wea kness" by another . This error was infrequent , as suggested by the fact that whenever a particular critical incident discourse was identified by both coders, it was encoded with 97 % coder agreeme nt i nto the same evaluative category . Assessing Critical-Incident Frequency Experienced and novice demonstrators showed considerable differences in the number of critical incidents discussed during the 118 critical-stop task. This difference in performance is shown in Table 7. From this table we see that experienced demonstrators, on the average, make more critical stops of the videotapes (p = .0625) and discuss more critical incidents (strengths and weaknesses; p = . 125 and .0625, respectively) than do their novice counterparts. They basically identified more than one-and-a-half as many critical Table 7 Number of Critical Features Identified and Evaluated by Experienced and Novice Chemical Demonstrators (Analysis of Videotapes #'s 1- 4) a Average Number and Percent Change (or% Diffe rence) in Critical Comment s Made by Chemical Demonstra to r s b Evaluative -------------- - - - - --- ------ Judgment Post-Workshop Experienced Pre -Workshop Novices c, Demonstrators c , Novices, Ave. % Change Ave . % Difference Ave. 21. 2 +49 26.8 * +83 Strength 14.2 30.0 -12 54 . 5 ** +59 Weakness 34.2 (2 . 2 ) (4. 2 ) * Acceptabled -----(-0-. -5-) --------- -- - ------ ----- 51. 2 +6 81.3 ** +68 Total 48 .4 39 . 2 -6 63 . 2 ** +51 Critical Stopse 41.8 Nont e=s. 4-5 subjects per experienced group; n = 8 per novice group. a Total uninterrupted playing time= 27.5 minutes. b All comparison are with respect to pre-workshop novices. c All probablities in comparison to pre-workshop novice d Vaslcuoesr esin. parenthesis represent a subset of those in the weakness column. They were obtained by researcher after obtaining coder reliabilities. e Critical stops do not include neutral or nebulous di scourses . ** p = .0625 * p = .1 25 -- -- --- - - --- - - - ? 119 features that hinder or promote effective chemical demonstrating as novice demonstrators did . The probabilities reported above represent meaningful group differences in performance on the critical-stop task as determined by a non-parametric Wilcoxon matched-pairs ranked-signs test. In this st udy , probabilities of p ~ . 187 were considered to reflect meaningful group differences using this non-parametric test. This probability level was considered meaningful because of the small sample size (four pair-wise values used to run the test). In using a rank-order s tatistical test with a very small sample size, probabilities became greatly inflated whenever the novice group out-performed the experienced demonstrators in one of the four pair-wise comparisons, i.e. , on one of the four videotapes. Further support for establishing P < .187 as a meaningful difference comes from the fact that experienced demonstrators' performance at this level was consistently 25% greater than that of the novices, as suggested by the data presented throughout this chapter . Thus, the data in Table 7 indicate that experienced demonstrators made meaningfully (p = .0625 ) more comments than novices during the critical-stop task regarding effective and ineffective chemical demonstration teaching of specific chemical concepts. Table 7, as well as several other tables in this chapter , include performance data on novice demonstrators after they received the workshop intervention . This information will be discussed separately in a section devoted to post-workshop novice performance (see Influence of Intensive rnservicing, P? 198) ? 120 Group performance on the critical - s top task, when defined in t e rms of the number of critica l feat ures discussed, shows a rea s onable amount of variance. This variance i s illustrated in Table 8 by the range of critical-stop frequency scores obtained for Videotape #3. It is apparent from this table that experienced demonstrators do not always ident i fy the same numbe r of critical features relat ed to effective demonstrating nor do the most experienced chemical demonstrators, s uch as E3 and ES , always make the most frequent cr i tiques. This range of performance among experienced demonstrators was a l so observed in their critiques of Videotapes #1 , #2 , and #4 (Appendix Q). Novices showed simi lar variabili ty in performance. r?' Table 8 Number of Critical Features Identified by Experienced/Novice Chemical Demonstrators for Videotape #3 (Density, + ) Number of Critical Comments Eva luative Made by Chemical Demonstrator s Judgme nt Pre-Workshop Experienced Novices a Demonstrators Nl N2 N3 N4 Ave El E2 E3 E4 ES Ave Strength 7 7 4 2 5.0 11 12 9 9 4 9 . 0 Weakness S 2 10 4 5.2 8 17 7 10 8 10 . 0 Acceptableb ( 0 0 1 0 0. 2) (1 1 3 2 0 1.4) Tota l 12 9 14 6 10.2 19 29 16 19 12 19.0 Critical Stops 9 9 12 S 8.8 18 22 14 18 9 16.2 Notes. ~ 1-N4 represent t he novices that saw videotape #3 before the workshop intervention. b Values in parenthesis represent a subset of those in the weakness column . 121 The range of critical-stop frequency scores for the experienced and novice groups was large enough to produce some overlap in group performance. The data in Table 8 and Appendix Q indicate that a novice would occasionally identify and discuss just as many critical features as an experienced demonstrator. Table 8, for example, shows the performance of novice demonstrator N3 matched (see number of "Strengths " identified) and even exceeded that of experienced demonstrator ES (see the remaining categories ). This frequency comparison, however, ignores qua l itative dist i nctions in the comments made by the two demonstrators. These distinctions are important and are examined later. In terms of overall group performance on the critical-stop ta sk (as defined by average frequency scores; Table 7) , the experienced group of chemical demonstrators out-performed the novice group in every respect. A non-parametric Wilcoxon Matched-Pairs Ranked-S igns test showed that there was a meaningful difference (p ~ .125 ) in the performance of the two groups along each critical-s top parameter , i . e ., the number of critical stops and the number of critical incidents (strengths , weaknesses , total ) identified. Levels of Agreement in Identifying Critical Incidents The level of agreement among experienced chemical demonstrators and among novice chemical demonstrators in identifying the same critical incidents can be determined from the data in Table 8. Again, the data obtained on Videotape #3 (Density Column demo , favorable model ) illustrate these agreement levels . Table 8 shows that E2 identified as many as 25 critical incidents whereas ES only identified twelve . Percent agreement based on these two subjects alone shows 122 that group agreement levels can not exceed 48% (12/25). This maximum value assumes that the twelve features identified by ES were al s o identified by E2 (and by the remaining three experienced demonstrators ) . A content analysis that uses each unique critical incident as a coding category provides the most accurate estimate of group agreement levels . Such an analysis records the nature of each critical incident discussed by a group of demonstrators and then confirms it s occurrence in the demonstration videotape. A tally is kept of critical incident s discussed by more than one subject . Finally , the total number of common incidents critiqued are compared to the total numbe r of incidents discussed . Such an anlysis was conducted on the discourses for Videotape #3 , with the findings shown in Table 9 . The first row of numbers in this table shows that there isn't a single critical incident that all five experienced demonstrators identified and discussed. The third row in the table shows that most of the demonstrators (~ 50 %) did identify and discuss nine of the s ame critical incidents . This number , when compared to the 61 incident s collectively discussed by the group for Videotape #3 , indicates that the nine "frequently cited" incidents represent only 15% of the total identified . A content analysis of novice chemical demons t r ators ' di scour ses of Videotape #3 also showed low levels of group agreement in identifying and discussing the same critical incidents . A t ot al of 31 critical incidents were discussed by the novice group , however , only six were identified as frequently cited , i.e. , identified by at l east half the novice subjects . These frequently cited incident s r epresent 123 Tab l e 9 Percent Agreement Among Five Experier.ced Chemical Demons trators ' Performance on the Critical-Stop Task for Videotape #3 Agreement Criteria No. Common Critical Incident s Percent of Total Number of Exp. Percent I ncidents , Demonstrators of Group Absol ute Cumulative Common 5 100 0 0 0 4 80 5 5 8 3 60 4 9 15 2 40 13 22 36 1 20 39 (Unique) 61 (Total) 100 about 19 % of the total incident s this group identified. '' I The agreement data presented so fa r indicates that experienced '! and novice chemical demonstrat ors both show low levels of within group agreement in identifying identical critical incidents in Videotape i/ 3 . These agreement findings are now extended to a quantitative cont ent ana l ysis of frequently cited i ncidents identified by exper i enced and novice demons trators across the four videotapes . Identifying Frequently Cited Featur e~ Frequently cited critical features are those demonstration teaching behaviors that at least half of the subjects in the experienced or novice group i de ntified during the critical-s top ta sk . These frequently cited feature s represent a set of think - aloud discourses that permitted be tween-group comparisons and within -group 124 generalizations to be made regarding experienced and novice demonstrators' conceptualizations of effective chemical demonstrating derived from commonly discussed critical incidents. "Infrequently ci ted" features , on the other hand, represent those f eatures that l ess than half of the subjects in the experienced or novice group identified during the critical-stop task. Although these features do not provide a constant context for meaningful group comparisons, they do provide valuable information on demonstrators' discourse focus during the critical-stop task. Table 10 shows the total number of frequently cited features identified by the two groups of chemical demonstrators critiquing the four videotapes. Table 10 shows that the experienced demonstrators identified a total of 30 frequently cited features while the novice Table 10 Total Number of "Frequently Cited" Critical Features Identified and Evaluated by Experienced and Novice Chemical Demonstrators Critiquing Four Videotapes Total Number of Critical Comments Made by Evaluative Chemical Demonstrators Judgment - - - -- - ------ - - - - - --- - - - - ---- - - -- - - -- --------------- Pre-Workshop Post-Workshop Experienced Novices , Novices, Demonstrators, Frequently Cited Strengths 9 11 17 Frequently Cited Weaknesses 10 14 13 Frequently Cited Str + Wk 19 25 30 Note. This table shows group totals, not group means, therefore no statistical difference tests were conducted between groups . 125 demonstrators identified 19 frequently cited features. This indicates that the experienced demonstrators rendered judgments on about one-and-a half times as many frequently cited critical features as the novices. The data reflects a pattern similar to one observed in Table 7 which shows the average number of critical features identified per group (76 and 48, respectively ). In both instances , the experienced demonstrators critiqued about one-and-a-half times as many critical features as the novices. An examination of the frequently cited features identified by experienced and novice demonstrators across the four videotapes reveals a total of 43 different frequently cited features . one-sentence summary statements of these 43 features are given in Appendix R. This list of statements shows that the two groups of _. .. , demonstrators generally did not critique the same set of frequently cited features. Of the 43 frequently cited features discussed by the two groups of demonstrators , twenty-three (54% ) were discussed by only the experienced demonstrators, thirteen (30 %) by only the novi ce group , and seven (1 6% ) by both the experienced and novice demonstrators. A total of 274 different critical features (both fr equent ly cited and infrequently cited features ) were identified by the experienced group of demonstrators in their critique of the four videotapes. Wh en this number is compared to the total number of frequently cited features identified by the experienced group , i.e., 30 , it is evident that frequently cited features represent a very small percentage of the total identified (10.9 %). This finding indicates that the critical-stop task shows a very low level of within group agreement 126 among experienced demonstrators in identifying and evaluating identical critical features observed on videotape. Within group agreement among novice chemical demonstrators in identifying and evaluating the same critical features was al so low (11.3 % using a > 50 % group agreement cr i teria) . Experienced and novice chemical demonstrators clearly differed in their performance on the critical-stop task. These differences were apparent in the number of critical stops made and number of evaluative judgments rendered . In the section that follow s , the content of the critica l incidents discussed by the experienced and novice chemical demonstrators is closely examined. The di scuss ion now shifts from the eva luative judgment domain (domain a) to the descriptive knowledge domain (domain b). Analysis of Demonstrators ' Descriptive Knowledge Domain A taxonomic analysis of the 274 critical features identified and evaluated by the experienced demonstrators and the 177 features identified and evaluated by the novices shows that they address nine r ecurring pedagogical issues. These nine issues , or categories, were derived entirely from the verbal data. The names given to the categories were either taken from the literature or created by the re searcher. The nine categories that emerged presumably contribut e to effective chemical demonstration teaching. They include an investigative/inquiry approach to chemical demonstrati ng , questioning s trategies , addressing new terms, quality of explanations, interactive/participatory style , mechanics of the demonstration, use of blackboards and visual aids , overall organization , and presentation s tyle . 127 Table 11 shows the frequency with which these nine pedagogical i s sues we r e di scus sed during the cr i tical - stop task by both experienced and novice chemical demonstrators. Overall group differences along each of the nine taxonomic categories were as sessed using the Wilcoxon Matched-Pairs Ranked-Signs test. With this tes t, significance levels of p = .0625 were obtained whenever the Table 11 Average Number of Critical Comments Made by Experienced and Novice Chemical Demonstrators According to Pedagogical Knowledge Category Average Number of Critical Comme nts Made by Chemical Demons trator s --------- - - - - - -- ---- ---- - - - -- Pedagogical Novice Experi e nced a Categories Pre- Post- Workshop Workshop a 1. Investigative, 6.2 6.2 10. 2 * Inquiry Approach 2. Questioning Strategies 6.8 9 . 0 13. 2 ** 6.0 5.5 7. 8 ** 3. New Terms 8.2 11. 0 * 11. 8 ** 4. Quality of Explanations 5. Interactive, Participatory 2.2 2.5 4. 2 *** Style 4.2 4.8 11. 0 *** 6. Mechanics of Demonstration 7. of Blackboard/Visual Aids 3.5 2.0 4.0 Use Organization 7.8 6.5 8 . 5 8. Overall 1. 8 2.8 6. 8 *** 9. Presentation Style - - - ---- - --- - - - - --- - ------ - -------- 77. 5 *** Ave. Individual Totals: 46.7 50.3 ~Note s.1 1 probabilities in comparison to pre- workshop novice scores . *** p = .0625 ** p = .125 * p = .187 128 experienced demonstrators made more frequent critical comments than novices on all four videotapes. This probability leve l (i .e. , p = . 0625) also represents the lowest achievable probability when working wi t h fou r pair- wise comparisons and the Wilcoxon pair-wise test, (p < .0625 was never obtained with a sample size of four and this non-parametric test). Significance levels of P = .125 were observed whenever an equivalent number of critical comments were made by the experienced and novice group of demonstrators on one of the four videotapes . Probabilities of .187 were observed whenever there was a greater number of critical comments cited by novice demonstrators on one of the four videotapes. Probabilities exceeding .1 87 were observed whenever there was a greater number of critical comment s cited by novice demonstrators on two or more of the four videotapes. For the purpose of this study, non-parametric probabilities showing p < .1 87 were considered a reasonable indicator of meaningful group This level was selected for reasons cited earlie r (see differences . p . 119). Table 11 highlights with asterisks the categories showing the most meaningful group differences between experienced and pre -works hop novice demonstrators . The numerical entries in the table r e fl ec t the mean frequencies with which the nine pedagogical issues were cited and discussed by the two groups of demonstrators critiquing the four videotapes. Reliabi lity of the Taxonomic Analysis The reliability of coding critical-stop discourses into the nine pedagogical categori es showed 74. 3% coder agreement . (Scott ' s Pi was 72.0 %). some of the differences in coding are accounted for by s imple 129 coder oversight (forgetting or not observing specific coding rules ) and cases where one coder placed too much emphasis on one term in a given discourse. Most differences in coding are accounted for by the somewhat subjective, interpretive nature of applying coding rules (Appendix P) to complex written text. For example, some critical incident discourses coded into more than one category, depending upon which statement(s) in the discourse a coder perceived was being emphasized. Coding difficulties were also encountered whenever critical-stop discourses addressed demonstration teaching behavior s that served more than one function, e.g., behaviors that concurrently promoted an "Investigative , Inquiry Approach (Category 1) and an Interactive, Participatory Style (Category 5)" or any other combination of categories . Although the coding instructions provided rules to help guide the coding of complex discourses and verbal propositions , interpretive differences in understanding the content of the protocols and applying coding rules to these complex proposit i ons prevented the attainment of high levels of coder agreement (> 80 %). In spite of these difficulties, acceptable level s of coder agreement were obtained . Content Analysis The information presented in Table 11 shows that the exper i enced demonstrators made significantly more critical comments t han pre-workshop novice subjects in seven out of nine pedagogica l categories . These seven categories include Investigative/Inquiry Approach , Questioning strategies , New Terms, Quality of Explanation, Interactive Participatory Style, Mechanics of Demon s tration, and Presentation Style . once again, probabilities computed with p = .1 87 130 were considered to reflect meaningful group differences using the Wilcoxon matched- pairs ranked- s ign test. Thus , Table 11, indicates that experienced demonstrators made meaningfully (p = .1 87 ) more comments than novices related to inquiry. 1 Table 11 shows that experienced demonstrators made more fr equent mention of two teaching strategies , investigative/inquiry approa ch (Category 1) and interactive, participatory style (Category 5), that can be performed in conjunction with a chemical demonstration. Furthermore , they made more frequent critiques on pedagogical i ssues closely related to chemical content , s uch as questions strategies and the clarity of presenting new terms and concepts (Categories 2 - 4). The experienced demonstrators also made considerably more critical comments about the mechanics of the demonstration (Category 6) than the novice group. This fi nding would be expected, given their experience , i.e. , weekly use in conducting chemical demons trati ons (Table 1 ) . Critical-stop comments related to visual aids and organiza ti onal issues (Categories 7 and 8) were treated similarly by the e xpe r ienced and novice demonstrators. These two issues are not necessarily ti ed to knowledge of chemistry and were therefore confidently discussed by the novice group. Finally , experienced demonstrators we r e mor e attuned to elements of presentation style in conducting chemi ca l 1 This level of significance corresponded to a probability of p = .0 68 when individual scores rather than group scores were also used t o r un the Wilcoxon Matched-pairs test on the "Inquiry" category. The difficulty with this latter approach , however , is that the probabili ty level is unstabl e , ranging from p = .0 28 - .086 : changing with each attempt to randomly pair scores of individual novices with indiv i dua l experienced demonstrators. 131 demonstrations than novices (p = .0625). Table 12 recasts the data presented in Table 11. It shows the similarities and differences between experienced and novice chemica l demonstrators in terms of the relative focus of their crit ical-stop critiques . It contrasts the two groups of demonstators in a manner that shows novices' giving greater relative focus to certain pedagogical issues than is apparent from the frequency data given in Table 12 Focus of Experienced and Novice Chemical Demonstrators' Evaluative Judgments on the Critical-Stop Task Relative Frequencies of Critical Comment s Made by Chemical Demon s trator s Pedagogical Categories Novice Experi enced Pre- Post - Workshop Workshop 1. Investigative, Inquiry Approach .13 .12 .1 3 2. Questioning Strategies . 15 * . 18 * .17 * 3. New Terms .13 .11 .10 4. Quality of Explanations .18 * .22 * .1 5 * 5. Interactive Participatory Style .05 .05 .0 5 6. Mechanics of Demonstration .09 .10 . 14 * 7. Use of Blackboard/Visual Aids . 08 .04 .0 5 8. Overall Organization .17 * .13 * .11 9. Presentation Style . 04 .06 .0 9 ------- - ------- - - - --------------- Total: 1.02 1.01 0. 99 Notes. 1ITTotals that do not sum to 1.00 represent rounding error s . (2) The asterisk (*) r epresents the top three fo cus categori es for each group . 132 Table 11. A within-group comparison of the nine pedagogical categories addressed by the two groups of demonstrators shows that two of the top three issues addressed by experienced and novice demonstrators were questioning strategies (Category 2) and the quality of explanations (Category 4). The third major concern of experienced demonstrators involved issues related to the basic mechanics of the demonstration and how they can contribute to effective demonstrating (Category 6). For the novices, the third major concern involved organizational issues (Category 8) . Experienced and novice demonstrators showed considerable difference in their relative focus on presentation s tyl e (Category 9) , an issue which was of greater relative conce rn to experienced demonstrators than to novices. There was some variati on in demonstrators' content focus depending on the videotape being critiqued, namely , a~ . 03 average variation in relative di scourse frequency per category, (mean frequencies are given in Tabl e 12) . A between-group comparison of the data in Table 12 shows that novice chemical demonstrators gave greater relative emphas i s t o f our pedagogical issues . These issues include attention to new t erms (Category 2), quality of explanations (Category 4) , use of blackboard/visual aids (Category 7) and overall organization (Ca t egor y 8). Two pedagogical issues received equal relative emphasi s by the two groups of demonstrators: investigative/inquiry approach (Ca t egory 1) and interactive participatory style (Category 5). Although novi ces showed equal or greater relative focus on several pedagogical i ssues , the experienced demonstrators still critiqued these is s ues as 133 frequently or more frequently than novice demonstrators in absolute t e rms (see Table 11). Qualitative Differences in Demonstrators ' Critical-Stop Comments in the Nine Pedagogical Categories The next level of anal ysi s , a componential analysis , focuses on the qualitative differences between experienced and novice demonstrators ' critical - stop discourses. These differences are examined in terms of the kinds of comments made in each of the nine taxonomic categories identified above (Tables 11 and 12) . All nine categories deal with pedagogical issues germane to effective chemical demonstration teaching . Differences between experienced and novice chemical demonstrators ' thinking about effective chemical demonstrating are ,,, ?;. ?? highlighted with quotes. These quotes are taken from demonstrator s ' think-aloud critiques of major as well as minor critical incidents . Space only permits a small sampling of comments from the two groups of I subjects. To supplement the quotes supporting the componential analysis , the reader may refer to Appendix S, which contains additional quotes , and to Appendix R, which contains a comprehensive list of all major critical incidents identified and discus sed by the two groups of demonstrators. Appendix Y provides a comprehensive li st of all major and minor critical incidents identified by the subj ect s of this study examining Videotape #3. (Again, space prevents the inclusion of comprehensive critical incident lists obtained for Videotapes #1 , #2 , and #4). In each of the aforementioned appendices , critical incident s are organized taxonomically according to the nine 111111r rnmm -- ---------- _,,- 134 categories discussed above. Samples of complete interview discourses are provided in Appendix T. Inquiry, Investigative Approach. One distinctive difference between experienced and novice chemical demonst r ators in their discourses on effective chemical demonstrating is that the former group tended to make more frequent reference to the term "inquiry" when describing how to go about demonstrating chemical concepts to middle school students . The following quotes illustrate this global perspective of experienced demonstrators. (See also Appendix S.) Note that in these quotes the letter "E" refers to an experienced chemica l demonstrator. El: Didn't use probably as much inquiry as he should have used either . That's a good demonstration. You can really pull things out of people. And there could have been more questions. [VT #3, Q.4] E2: My main [concern] with her, I think, is the same thing as it was with the first demonstration, is the ques tioning. It seems as if these people are not trained to use inquiry. And not that I think inquiry is "it" but as far as demonstrations go, it is "it." I mean, you don't want to tell them the whole answer and then say, "Here it i s !" [VT #4, Q. 4 J Novice demonstrators essentially made no direct refere nce to the term "inquiry" or an inquiry approach to chemical demonstrating in the critiques of the four videotapes . This global conceptuali zation of effective chemical demonstrating at the middle school l evel was a distinct characteristic of the experienced demonstrator s . As well as making direct references to inquiry, the experienced subjects identified and discussed several inquiry issues , including the importance of encouraging student observations, predictions , hypotheses , support ive evidence , and conclusions , as we ll as the 135 testing and discussion of student ideas. Novices discussed several , but not all, of these inquiry behavi ors. The most notable dif fere nces between the two groups of subjects regarding inquiry iss ue s are illustra ted by the quotes and discussion below. Experienced demonstrators frequently discussed the value of drawing students into the demonstrations by encouraging them to verbalize their observations. These demonstrators would also indicate v c h ro ? cal demonst r at or model s good ob s e~va t i on sk i l l s a.ml ~?m@ r~Jl.y qoes not provi de student s wit h basic site observations m k h ms vs . The f o l l owi ng quo s , nct Appendi x !; , ill ?t t~ t his perspeotive . The t e rm "M " r e f ers to he be?? n c it ? qued . E1 : That ? ti? othe good point ? hat he rnacte when he held ?.t up an he s ? d "I want you to make car eful observab .ons . 11 And he 's observing t oo . He ' s act i v l y ob rv? n wh ' ch i a l so good . I t ' s not as i f he is passively l ooking around t he audi e nce. (VT #3 , S:17) E3; Here he ' s invol ving ? ~uden s . Rat h tha n h 'm j us t t al king and showin, h ' s al ady asked s t ude nts for t heir input [i . e ., t he i r observat i ons on t he density eo1umn l . (VT # 3 , S : 2 4) E4: The other good thing he has been say i ng , "Gi ve me more obse rvations on t hi s . " He has been asking for observations at this point. (VT # 3 , S :4 5 ) Only one novice , N7, made a ny ment ion of the value of verbalizing observations during a demon t tion, howevex; , t he emphasi s was on t tie va lue of t eacher - cent e ed obs v on- wh n c ndu ?ng hern ' ca - demons t rat i ons . N7: That ' s good because it ' s pos sibl t hat t he st udent s i n the back might not be abl e to see what is going on. , . [Videotaped teacher said , " I can feel (t he wate r ) . bo i ling und perhaps you can hear it boiling"]. And he's telling, them. what is going on even though they probably know whats going ---- ---p- - - ---1- .,.- _...,,,,., ATT'? 136 on. So , so far he ' s given a pretty detailed account of what he ' s doing. [VT #1 , S:52 ) A componential analysis of demonstrator discourses on inquiry demonstrating showed that only the experienced demonstrators discussed the strategy of having students regul arly verbalize their observations when a teacher performs a chemical demonst r ation. Experienced and novice chemical demonstrators both discussed critical incidents dealing with students making predictions of demonstrat i on outcomes , e.g. , expected properties of a chemical sys t em to be demonstrated , or anticipated outcomes of a set of phy s ical measurements. Both groups considered this a relevant is sue associated with effective chemica l demonstration teaching . The two groups of demonstrators also made frequent mention of the importance of effectively fie l ding multip l e student responses generated by divergent , inquiry-oriented questions , e .g ., observation and prediction questions. Novice demonstrators placed somewhat greater importance on havi ng students try to explain the observed phenomenon by using ques tions such as , "Why did the can collapse? " and "Why is the minera l s it t ing on top?" , questions that go beyond the demonstration . The f oll owing quotes provide illustrations of this thinking . (See Appendi x S f or additional quotes.) N6: So , now he is bringing in some thought ques tions [e .g . , "What forces are responsible for crushing that can? "), for them to ponder and to use their thinking process to go ahea d and say why this is happening. Then he said, "Are there any other options available to you? " So , instead of jus t saying , "That ' s right ," and going on with that , he is trying out mor e thinking process which is probably more toward the objec tive of it . (VT #1 , S:21 ) --- ------- - ?- - - - _ __ .., _,_ ,.,,11""7 l 'T'" 137 N7 : In this section he asks [the student ) to explain why he was able to crush the can. And t he student explained well. And it was good the way he kept probing him, you know , "Is there more? Is there more? ". (VT #1 , S:24 ) Nl : Again , good questioni ng ["How di d this occur?" ] . He ' s stil l keepi ng it going. He ' s trying t o get it out of them trying to make them actually fo l low through. (VT #2 , S:34) Experienced demonstrators , however, went beyond the i ssue of having students provide hypotheses to account for a phenomenon , to the issue of having students provi de supportive evidence to substantiate their hypotheses and to the testing of some of the students ' ideas . The focus was more on the demonstration than on the "answers " to the demonstration . The fol l owing quotes taken from various critical incidents illustrate this difference. E2 : OK , I like that question , [" Is there any evidence to support that? ") . That's one of my favorite questions to use . In fact , no matter what level I teach , elementary all the way through col l ege , that if they are going to gi ve me some kind of response I want some kind of evidence for it. (VT #1 , S: 46) E4 : Now someone (a student ) did come back with the wat er (vapor ) pressure pressing [the air ] up [out of the can ) . And he did ask , "Is there any evidence? ". That was a good point. "What are you bas i ng this upon? " Again , I didn ' t get the answer al l that well . But [a student ) would have an expectation based on their experience that this would probably occur. And that ' s fine to bring in previous experience . That ' s a good strong point . The question did come up, but if it didn ' t , that ' s one of the things you should try to get the group to explain . (VT #1 , S: 50 ) E3 : Here again he is being interactive with the students asking them to make hypotheses [about the identity of the liquids in the column) and then following up with , "How would we know if that wer e true?" In this particular case , he asked them a very broad question , "What liquids might be in the container? " which they probably don ' t have a whole lot of experiences for , but [the students provide a) couple guesses and then he follows it up with , "Well , how can we determine it , if those guesses are in fact valid?" . (VT #3 , S:70) 138 Novices paid very little attention to the value of having students provide evidence for their predictions or hypotheses in response to teacher's questions. One novice that did address the issue of supportive evidence commented on the importance of the teacher providing more evidence for the students so they would be better able to answer the teacher's questions. NB: Most st udents know that a cork board will float on th e water, but since she hasn't given the value for the density of the cork. She hasn't given them enough information to be able to, for them, to deduce accurately [the answer to: "And what do you think the cork will do?"]. (VT #4, S :11 6) Experienced demonstrators were more prone to point out that an effective demonstrator will not spend an excessive amount of time talking or lecturing to the students during a demonstration but will draw students into the process of having them think about possible explanations for the demonstrated chemical phenomenon as well as help them draw conclusions from the resulting class discussions. The final issue raised by demonstrators in terms of an inqui r y approach to chemical demonstrating was that of "forecasting ": the videotaped demonstrator telling the audience what to expect . Experienced demonstrators viewed the issue of "forecasting " as a legitimate threat to an effective chemical demonstration. Thi s i ssue was raised frequently and in several different context s by the experienced group. Novices also recognized this i s sue in seve r a l different contexts but it was discussed in slightly f ewer critical incidents. The two groups of demonstrators viewed foreca s ting from three basic perspectives: forecasting procedures to be performed, forecasting observations to be noticed, and forecasting explanat ions 139 of an observed chemical phenomenon. These three levels of forecasting are exemplified by the critical-stop discourses of experienced and novice demonstrators. Forecasting observations: E3: He sort of forecasts there that, "In a minute or two we are going to see some vapor," and I seem to recall from the first run through where he said something about, "He could hear it boiling." Those are both things he could have asked the s tudents to make observations on rather than t o state. So that it would have been better to let the student make the observation rather than to forecast it. (VT #1, S:53) Forecasting procedures: Nl : I'm not so sure I would have told them what it was I was going to do . I think it's two different ways of looking at it. But maybe if he had just done it, he could have opened up more ques tions as to why, instead of telling them exactly what he ' s going to do. (VT #2 , S:22 ) Forecasting explanations of phenomenon: N2: I get the feeling he's telling them what's going to happen instead of letting them find out. I'm not s ure I like that. Interviewer: How's he tel ling them what' s going to happen? N2: Well , he's explaining that ~e have all thi s air pushin on us, and he could do the experiment and let the kids t ell 2 him that there's air pushing. Let them investigate a little more . (VT #2 , S:3 ) The critical-stop discourses of experienced and novice chemical demonstrators reveal that the inquiry approach to chemical demonstrating plays a very important role in their thinking about demonstrating abstract chemical concepts to middle school st udents. The fact that the experienced demonstrators discussed these i ss ues more frequently suggests a greater knowledge of integrating inquiry methods and demonstration teaching. Questioning Strategies. The next taxonomic category di scussed by experienced and novi ce demonstrators during the critical-stop task 140 involves the effect ive use of teacher questioning strategies during a demonstrat ion. In general, the experienced demonstrator discussed more critical incidents pertaining to the is sue of questioning than novices discussed (see Table 11). A content analysis of these discourses on teacher questioning strategies showed several commonalities and differences between the two groups of demons trat ors . In terms of differences, the experienced demonstrators showed greater proficiency in evaluating the many questions presented by the videotaped teachers. Novices also showed proficiency, but s imply evaluated fewer of the questions presented in the videotapes. (Appendix J, Table J.1 shows that 86 questions were asked by the videotaped teachers. More than half were critiqued by the two group s of chemical demonstrators during the critical-stop task). A componential analysis showed that the experienced and novice demonstrators gave simi lar attention to questioning i ssues such as the timing of questions, fielding and probing of responses to questions , and providing of appropriate feedback and wait time . Many of these issues pertain to generic teaching skills applicable to chemical demonstration teaching. Experienced demonstrators also recognized and di scussed more critical incidents pertaining to the use of leading- quest i ons. They consistently mentioned the ineffectiveness of such quest ions in checking for student understanding. E2: But the next question I thought was leading, [ "Do you think that there was possibly a force r es isting your hand?" (S:17)). I mean he was giving them essentially the answer and just waiting for them to comment on it. He does that throughout his demonstration. (VT #1, S:17) 141 E4: I think it's good that he's trying to indicate that these are solids, "Is this a solid or a liquid?", but I think. if he had worked on the observation a little bit more he probably would have gotten that out of the group rathe r than have to lead them in. Although I'm going to write that off as a function of time factor in this case. (VT #3, S:62). N6: She's trying to ask very leading questions, very helpful questions, but she's hand feeding the people there and if she doesn't change that type of pattern, then kids will make her do all the work. She's doing all the work. The audience is not doing the work that is involved. (VT #4, S:94 ) Experienced chemical demonstrators were also more inclined to discuss crit ical incidents involving teacher questions that probed students' prior knowledge of science concepts. Most novices t ended to i gnore these incidents in their evaluation of the videotaped teachers . Those novices and experienced demonstrators that did address this i ss ue of probing prior knowledge indicated that the probing process he lps the demonstrator fine tune the presentation to the leve l of the st udent. This type of discourse is illustrated with critica l incidents taken from the Dens ity Column Demonstration videotapes. E2: And here there are opportunities to have the kids come up and feel the s ubstance, to talk about it a little bit , "Have you ever used oil?", "Have you ever seen - have you ever poured oil on water? "What happens?" I mean al l t hese diffe rent things could happen. None of thi s happened . She just dumps it (the liquids) in there . (VT #4, S:72) ES: I think one of the strengths of the lesson i s to start with what the s tudents already know and build from t here because , I guess l earning, seems to me , attaches very easi ly, or more eas ily, to things that they already know. (VT #3, S:3) Nl: The fact that he is asking them about their pas t experience, what it is they know already, so he has a cleare r ide a of where to start with them, ins tead of assuming that they all know it or assuming that they don' t kn ow it, he ' s r ea lly getting a hold on, at l eas t, where a few people in the cla ss are. (VT #3, S:3) .? -~-.. ?_:_? . ?: _ .. - ?: ??: . . . . : .. . .... . =. . .. . _? . ? ? ? : . ? :. :: .?: . ; ; ?? ? .: ???. . ?~. ?: . . : : :. ?. . ? ? ,: ? . . 14 2 Experienced demonstrator s evaluated the quali ty of the videot aped teachers ' questions more often and al ong more cri t eria t han novice demonstrators . The experienced demonst rators were more critical of the clarity of questions asked , the appropriateness of the ques ti on f or middle school st udents (i.e ., above or bel ow level ), and the intri nsic va l ue of higher- order ques t i ons to generate st udent thinking . The fo l l owing quotes illust r a t e experienced and novice demons trators ' thought s on t he first of these i ssues , question clarity . (See Appendix S for additional quotes. ) A s imilar pattern was observed for the other two questioni ng issues. ,,'' E3 (Clarity): That seems like she asked the ques tion, 1 expec ting an answer , and then shifted it suddenly be for e t he ( s tudent s had a chance to think about that. "What does it I mean [that i s has a larger density ] ?" is an incredibly vague 'I I question. I f she asks , "Given these numbers , practically , I ~ , / I what does tha t mean? " or "What does that t ran slate to? " But -? > :; i j us t to ask that quest i on [that way i s vague ]. (VT #4 , S: 30) ,/ ~ I '/ E4 (Phra s ing ): She had a little phrasing [problem t here '?' .i",, with ], " .. . more or l es s dense?" She probabl y would have ''11, been be tter off saying, "How would you compa re t he mine r al :j ?,i~:/ oi l to the water? " or "How would you compare the densi ty of mineral oil to t he density of water? " , something like t hat . ~ (VT #4 , S : 77 ) NS (Phrasing ): I t would be ni ce to have a more dramatic openi ng , you know , some way of getting them excited beca use that i s a f lat start , to start with a ques tion like that ["Can anybody tell me what density means? Do you remember hearing that term before? "]; and regardless of how i t is said or who i s saying it , is j us t a f l at way to s tart. (VT #4 , S: 3) N7 (Phrasing ): Her wording even, it ' s j us t not prope r . I t jus t doe sn ' t sound right. [" So alcohol has a grea t e r or l es s density t han wate r ?" ] . (VT #4 , S: 32 ) These quotes , those i n Appendix S, Table 11 (Category 2), and a componential analys i s of experienced and novice demons t rato r s ' di s courses on "Ques t i oning " illus t r ate that the experi enced 143 demonstrators not only made more comments than the novices regarding how to appropriately ask questions, they provided more penetrating analyses of the unsound questions they heard during the critical - stop ta s k by discussing ways of improving poorly phrased questions (domain c; discussed further on p. 150). Experienced and novice demonstrators made frequent New Terms . references to the importance of addressing new scientifi c terms while conducting a chemical demonstration at the middle school level. Table 11 shows that the experienced group discussed s l ightly more critical incidents than novices on this issue. Experienced demonstrators indicated that effective chemical demonstrating at the middle school level requires careful attention to the use of terms such as mass , unit volume, interface, phenolphthalein, solubility , cubic centimeter I barometer, and altimeter when conducting chemical demonstrati ons on density and air pressure. Their critiques generally indicated that scientific terms need to be defined, given greater attention, or deleted in favor of simpler, more appropriate terms. Novices addressed most, though not all , of the terms experienced demonstrators addressed. The nature of their critiques were comparable to those of the experienced subjects in cases involving favorably rated critical incidents (see Appendix S). The experienced subjects were more thorough in their discussion of unfavorably rated critical incidents involving teachers' use of new terms (see domain c , Knowledge of Alternatives , pp. 152-155). Both groups of demonstrators Quality of Explanation~. frequently discussed the quality of the explanations provided by the 144 videotaped teachers. The explanations that were critiqued included explanations of the demonstration observations, procedures , materials , and chemical concepts. These explanations were evaluated in terms of their accuracy, clarity, completeness , and usefulness to middle school learners. The greatest difference between experienced and novice chemica l demonstrators within the category of Quality of Explanations is in their critique of the accuracy of the chemistry information presented in the four videotapes. The experienced demonstrators were much mor e skilled in identifying and discussing informational inaccuracies than were the novices. For example, in Videotape #3 , three of the s i xty - one critical features identified by the experienced demonstrato r s dealt with inaccurate chemistry concept statements made by the videotaped teachers. The following quotes illustrate the three inaccuracies identified by the experienced group. E4: He doesn't understand that [phenolphthalein i s ] mo re soluble in alcohol than it is in water. He has it backwards . It just so happens he added something to the water to give i t color. I know that phenolphthalein is more soluble in alcohol. So he actually had that wrong. In alcohol , the OH group is slightly acidic , so it tends to keep phenolphthal e in colorless. It would tend to end up even more in the al cohol than the water layer. (VT #3; S:88) ES: The generalization , ... "Solids are more dense than liquids? " ... I don ' t think that is nece ssarily true , or I don't think you want to teach that. There are couple of things that float. (VT #3; S:93) E3: Well , basically, he sort of generalized and said that solids are generally more dense than liquids , and with gases it ' s difficult to see them, and so that's why he (the videotaped teacher) [said he] was working with liquids . But in fact there are dense gases that are colored. (VT #3; S:97) 145 Each of these comments represents the i dentification of a chemica l knowledge or generalization inaccuracy made by the videotaped teacher demonstrating relative densities. These inaccuracies were identified e xclusively by the experienced chemical demonstrator s . Novices never discussed these : naccuracies dur i ng their critique of Videotape #3. Simi l ar pattern emerged in novices ' critique of the other three videotapes . Knowledge inaccuracies were rarely discussed by novices . Thr ee points can be made about the two groups of demonstrators with respect to these critical-stop comments . First , prior to the workshop , novices did not identify any of the three critical features discussed above . Thi s finding agrees with the participant background information given in Table 2 wtich shows that novices only completed foundational courses in college chemistry . Second , subjects E3-E5, all of whom possess college chemistry degrees, did not identify all three inaccurate chemical propcsitions. They generally only identified one each . This finding further illustrates the low within - group agreement levels found in experienced and novice demonstrators' critical -stop discourses. Third , some of the expe rienced demonstrators (El and E2 ) did not identify any of the three knowledge inaccuracies in Videotape #3 , although they did identify a few in the other vi~otapes. These demonstrator s , a lthough experienced at conducting chemical demonstrations at the middl e school level , did not appear to possess as much chemistry content knowl edge as did E3-E5 . This finding i s also in agreement with the background information obtained on the five experienced chemical demonstrators concerning their college chemistry training (Table 2). 146 A content analysis of several other critical-s top discour ses on Videotapes #1-4 shows that there is a considerable difference in performance between experienced and novice demonstrators in the s ubcategory Identifying Knowledge Inaccuracies tha t forms a part of Cat egory 4 (Qua lity of Explanations) . This difference is attributed to these demonstrators' college chemistry training. Although these critical features represent a very small percentage of the total number of critical features identified (aprox, 6%), thi s data shows t hat the critical-stop technique i s able to discriminat e s ubjects on the basis of their s ubj ect matter knowledge and how this knowledge influences effect ive chemical demonstration teaching. Interact ive , Par ticipatory Style. In several critical-stop discourses , experienced and novice demonstrators discussed the value of invol ving student participants when conducti ng a chemical demonstration. Thi s would include the use of student volunt eers to assist with the mechanics of the demonstration, to make close-up observa tions , and to r ecord s tudent observations on the blackboard. The demonstrators further indicated that a good interactive , participatory style includes promoting continuous st udent-teacher and s tudent - student dialogs , and cal l ing on st udents by name . Experie nced demonstrators addressed these issues more frequently than novice s in their critical-stop discourses. Qualitative differences betwee n expe ri enced and novice demonst rators in their discourse on these issues were not apparent. Mechanics of Demonstration. Pedagogical issues relat ed to the mechani cs of the demonstration , as discussed during the critical-stop 147 task , included the issues of visibility , handling of equipment, safety , verification , and controls. Experienced demonstrators made more numerous comments than novices regarding the need for e nhanced visibility of demonstration materials. This was particularly apparent in their discussions of the Density Column Demonstration where the size of the graduated cylinder and the use of food coloring notably affects the vi s ibility of the density column. The experienced demonstrator also appeared to be more famil i ar with the demonstration chemicals as suggested by their more frequent discourse on safety and chemica l demonstrating. Experienced demonstrators appeared to address the iss ue of verification more frequently than novices. Verification involves t he l i ' act of demonstrating to students some of the prior preparation work that was conducted by the science teacher . In the case of the Collapsing Aluminum Can demonstration , the teachers in both videota pes added water to an a l uminum can before the demonstration s t a rted and had it sitting over a Bunsen Burner ready to be heated . The f ollowi ng quotes represent the reactions of two experienced chemical demonstrators to this cri tical incident . (See Appendix S f or additional discourses . ) El: Here he says, "There is a little bit of wate r in thi s can, " and I'm wondering what is a little bit of water. And I want to see you put the water in the can. I want t o know what ' s in there. Don't just tell me it ' s in ther e , show me. (VT #2 , S:16 ) ES : I think it's much better to actually pour the wat e r into the can . You rea l ly want the kids to see this. (VT #2 , S:17) 148 Only one novice addressed this critical incident , and from his per spect i ve , the incident was considered a minor issue in the over all scheme of things. NS: At that point it might have been ni ce to go ahead and pour the water in j us t so t he ki ds could see that it was just water and maybe even take i t right out of the tap. But that is j ust really minor details. (VT #1 , S:33) According to the experienced demonstrators , verification of procedural steps taken prior to the demonstration is an important pedagogical principle when demonstrating chemical processes to students at the middle school level. Use of Visua l Aids . Tabl e 11 shows that experienced and novi ce 1 demonstrators identified and discussed a similar number of critical I. . ' incidents re l ated to the effective use of the blackboard and vi sual ; 1 i1 ~ aids during a chemical demonstration . This category covered a broad I. !I .' '' I range of topics that incl ude discussions of what should be vi sualiz ed "..I : ~ I ' on the board , such as the physical system being demonstrated, new terms , numer i ca l va l ues , and important questions. Their di s courses also included critiques on the quality of these blackboard visual izations, in terms of accuracy , completeness , organization, neatness , and visibility. The experienced chemical demonstrators voiced concern about the importance of a blackboard drawing of the physical system bei ng demonstrated while novices foc used more on using the blackboa r d to record student observat i ons. The following quotes illustrate thi s difference. E2: I think it ' s much better to actually pour the wat e r into the [aluminum] can . You really want the kids t o see this . And , when I do this with students, I originally draw 149 on the board a , cat n and f I . have . it f. illed with steam , an d t h en ~aw anot he r pie ure o it.being inverted. It's easy to d[ Just ] talk about these things in a fuzzy way [the h did] . (VT *2 , S:17) ' way e N4: Taking down the [student] .observations and putting it w.h ehr e everyone could see them i. s real important for J' un?i or ig school because they are visually oriented. (VT #3 h S: 49) ' Experienced and novice chemical Overall Organization. demonstrators discussed organizational issues with about the same absolute frequency during the critical-stop task. These di s courses are characterized by discussions of motivational strategies , pacing , sequencing , closure , transitions , relevancy, and complexity of student The activites that followed some of the videotaped demonstrati ons . experienced demonstrators tended to identify a few more problems associated with closure and the sequencing of ideas and demonstration tasks than novice chemical demonstrators. The following quotes, and those in Appendix s , illustrate the thoughts of the experienced demonstrators on the sequencing of events and ideas. El (Ideas ) : Probably the first thing he did incorrectly was try to introduce two concepts at one time. Maybe he should have started with one and later into the lesson , if he had the time, gone to the second one, but that'~ very con f us ing for kids to introduce two concepts at one time . (VT #3, s : 2) E4 (Tasks) : I personally like to have things like that [can ] heat while I'm talking and develop~ng the subject. [With this] variation [I] put the water in the can, set it up on the burner , then go back to the [first] ques tion, and pose the second question while it's heating, [ "Can the air crush an aluminum can? "] . That makes a little bit more effective use of time. one of the problems I have here is - of course we have limited time in class, and you have to be effective - try to be effective on time . It a~so ~akes it easier since you don ' t have to stand there holding it. (VT #1, S:31 ) 150 Novices only discussed the first of these sequence issues , namely the critical incident involving the introduction of two concepts at once at the beginning of the demonstration. N3 (Ideas ): O.K. The first thing I suggest that needs t 0 be mentioned: he "has two ideas that we are going to be concerned with ," and he mentions 'density' and starts walk' off immediately [towards , the black, board]. And I would th i'ni knIg per aps, as a good overview, mention what the two [ideas] h are. (VT #3, S:3) Presentat i on Style. Experienced demonstrators discussed several issues pertaining to an effective chemical demonstration presentation style, including appropriate non-verbal gestures , courte sy/rapport , speech characteristics, humor, enthusiasm, and efficient movement around the classroom. The novices discussed the first three of these issues to a limited extent but made no mention of the lat t er three in their analysis of the videotapes . This section highlighted several experienced and novice chemi ca l demonstrator differences and similarities in their descriptive knowledge discourses (domain b). The section that follows examines demonstrators ' knowledge of alternatives (domain c ) . Experienced and Novice Demonstrators' Knowledge of Alternatives This section examines the various suggestions made by exper ienced and novice chemical demonstrators for improving critical f ea tures t hat received a "weakness " evaluation during the critical-s t op t as k. Thi s section also examines demonstrators ' knowledge of alternative chemica l demonstrations related to the concepts of density and air pressure . lect the third domain of science ref Co l lectively, these responses 151 teachers' discourse on chemical demonstration teaching called Knowledge of Alternatives (Domain c , p.104). A taxonomic analysis of demonstrators' discourses identified as "Knowledge of Alternatives" yielded the following four taxonomic categories: Pedagogical Enhancements , Demonstration Variations, Alternative Demonstrations , and Extraneous Examples. The data obtained from both the critical - stop task and semi-structured interview supported this taxonomy for the "Knowledge of Alternatives " domain . Pedagogical Enhancements Pedagogical enhancements (P) refer to any suggestion for improving the videotaped demonstrations. Most of these suggestions stem from discourses made during the critical-stop task , although some ,, ' ri were also obtained during the semi-structured interview. Subjects :: .,c ' . frequently suggested pedagogical enhancements in conjunction with ,: ,iJ, critical incidents considered to hinder effective chemical " ;1 ?' demonstrating , i.e. , incidents given a "weakness " evaluation (Table 7). Pedagogical enhancements do not include suggestions that r efer to variations on the observed demonstration (V) or to alternative demonstrations (D) on the targeted concepts. Table 13 provides a summary of the number of pedagogical enhancements cited by experienced and novice chemical demonstrators during the clinical interview. Inter-rater reliability in identifying these pedagogical enhancements from the verbatim transcripts wa s 83.0% . The data in Table 13 shows that experienced demonstrators made considerably more suggestions than novices for improving the chemical demonstration teaching observed on videotape (p = .0625, usi ng the 152 Table 13 Frequency of Pedagogical Enhancements Cited b y Experienced and Novice Chemical Demonstrators Average No. P's Cited N, Post Exp Concept Videotape N, Pre 5.5 4. 5 9.2 Air Pressure 1 3.5 2.5 6.5 2 4.3 7.8 7.0 Density 3 4 12.0 8.2 14.8 25.3 23.0 37 . 5 * Totals 1-4 *Nopte. = .0625, when comparing the number of P' s cited by experi enced demonstrators with the numbers cited by pre- or post - workshop ,, novices across the four videotapes. ,.,. ~ ' ,; Wilcoxon's matched-pairs ranked-sign test. Table 13 shows thi s l eve l ; !~ : ~ I of significance to be meaningful , given that the exper i enced subjects ~ provided more suggestions for pedagogical enhancement than novices on all four videotapes . A taxonomic analysis of the pedagogical enhancements mentioned by the experienced and novice demonstrators showed that these statements classify into the same nine pedagogical categories identified for domain b, descriptive knowledge of demonstration teaching. A content analysis of thes e statements indicated that the experienced demonstrators made considerably more suggestions than novices in four of the nine categories (Table 14). They made more suggestions on improving inquiry, attention to new terms , mechanics of the 153 Table 14 Average Number of Pedagogical Enhancements Made by Experienced and Novice Chemical Demons trators According to Pedagogical Knowledge Category Average Number of Pedagogical Enhancement Made by Chemical Demonstrators s Pedagogical ------- --- - --- - - - - - - -- - - - - Categories Novice ------------- --Experienced a Pre- Post- Workshop Workshop a 1. Investigative , Inquiry Approach 3.0 2.2 4. 2 * 2. Questioning Strategies 3.5 3. 8 7.0 3 . New Te rms 2.2 2.2 3 . 8 ** 4. Quality of Explanations 2.5 3.0 1. 8 5 . Interactive, Part icipatory 2.5 1. 8 * 4. 0 Style 6. Mechanics of Demonstration 1. 8 3. 2 5.2 ** 7. Use of Blackboard/Visual Aids 4.8 2 .8 4.2 8 . Overall Organization 4.8 3. 5 6.8 * 9 . Presentation Style 0.2 0.5 0.5 ------------- - ------------------- --- - Ave . Individual Totals : 25 . 3 23.0 37.5 ** Notes . - ?All probabilities in comparison to pre -workshop novi ce scores . ** p = .0 625 * p = .125 demonstration , and over all organization of the density and air pre ssure demonstration . The Wilcoxon matched-pairs ranked-sign test was used to de t ermine group differences along the nine pedagogical categories . Again , for the purpose of this study , non-parametric probabilities showing 154 p ~ . 185 were considered reasonable indicators of meaningful group differences. A few quotes taken from Ca t egory 3, the use of new science terms , illustrate some of the qualitative differences in pedagogical enhancements discussed by experienced and novice chemical demonstrators. The discourses of experienced chemical demon strator s reveals that there are more and less appropriate choices of chemica l terms to explain a demonstration and related concepts to middle school students . El: Right there he says , "What would happen if we eliminated the pressure inside the can?" I think it would have been mor e effective had he said , "Suppose we took the air out of the can?" But by saying , " . . . pressure" it ' s like saying pressure is an object that he's going to take out as oppo sed to air . (VT #2, S: 15 ) E2 : I thought that was a strange comment. I mean , if you ' re , not going to worry about it (phenolphthalein ) , why bring it ' I up? Why don ' t you say it ' s ' coloring ' . (VT #3 , S:83) E3: Also , that definition ' mass per ' [unit volume ' ) - middl e school , even high school peopl e , if you say 'per ' they don't necessarily equate that with the division sign. So , eve n though that ' s the classic definition, I don ' t think it does much for most people (students ) . This right her e , i t would be far , far more effective to either sketch a cubi c centimeter on the board or even better to have people ske tch a cubic centimeter and/or to give them a little sugar cube which is essentia l ly a cubic centimeter , so they get some sense of what that is . Just doing it visually (us ing fingers) . .. is not effective . (VT IM , S:10 , 15) The experienced demonstrators identified several critical incidents where science terms were either improperly used or ineffectively explained to students. In these instances , the experienced demonstrators often provided suggestions (pedagogi cal enhancements) for more appropriate science t e rms and explanations t o use with middle school students. 155 Novices also identified several potentially troublesome science words and phrases for middle school students , but they did not offer as many ideas of how to pedagogically enhance the videotaped teacher's explanations to promote student understanding. In cases where they did provide solutions , the nature of their pedagogical enhancements were often quite general (" He should bring in a few examples ." "He should t ell the students that he will explain that term later." "He should maybe put something on the board. ") rather than content- specific solutions, as illustrated in the quotes above. The following discourses show novices identifying a difficulty associated with the use of a new science term or phrase and then volunteering a rather general so lution , or pedagogical enhancement (P) , to the problem. N4: Here he goes right into the experiment and what I would rathe r see, especially at the middle school , is more of a discussion. ' Mass per unit volume ' is so vague tha t maybe one or two kids might understand it but certainly not the whol e class. If he brought in a few examples or had the kids think of examples of that , or things they were famili ar with that would explain it (density) , rather than a textbook definition . (VT #4 , S:14 ) N3: Again , with middle school students , "We won't worry about what that (phenolphthalein) is used for ... ," but immediately kids would want to know and would really want to have a bit more information . Perhaps later he might say , "If you are interested, etcetera , I'll be happy to explain more about it ." (VT B , S: 84) NS: In terms of the res t of the presentation, I thought it went really well. Right at the end, I don't feel he really cleared up what the concept was, with 'air pressure'. Ther e were a lot of confused people in the room still. They were definitely on the right track and get ting real hot but I don't think he really finished it off. He could have maybe put something on the board or had something e l se to really hammer it (the concept) home. (VT #1 , S:5 ) In general , the experienced chemical demonstrator s provided quantitative ly more and qualitatively different pedagogical 156 enhancements than novice demonstrators for critical incidents judged to be weak. The experienced subjects vo l unteered numerous content-specific solutions/suggestions to help middle school student s better understand abstract chemical terms and concepts mentioned during a demonstration. Novi ces tended to volunteer more generic teaching sol utions to the same problems . Variations , Alternative Demonstrations , and Extraneous Examples "Variations " (V) represent any alternative approach to performing the Collapsing Aluminum Can demonstration or Density Column demonstration observed on videotape. Variations typically pert ain t o alternative demonstration materials or alternative sequences of event s that can enhance a demonstration performance. "Alternative Demonstrations" (D) represent chemical demonstrations that differ markedly from the two demon strations observed on videotape, yet they still address the concepts of air pressure or density. "Demonstration Variations " (DV) repre sent twi s t s or extensions on suggested alternative demonstrations (D). "Extraneous Examples " (XTOT ) refer to (1 ) non-demonstration teaching strategies that address the targeted chemical concepts , e.g., labs , discuss ions , and activities , (coded XPOS ), as well as (2) inappropriate chemical demonstrations for teaching the targeted concepts (coded XNEG). Tables 15 and 16 show that experienced chemical demonstrator s made meaningfully more suggestions than pre-workshop novices in t e rms of discussing variations and alternative chemical demonstrations on air pressure and density. A list of the variations and alternative demonstrations discussed by the demonstrators are given in Appendi ces 157 Tab l e 15 Frequency of AIR PRESSURE Demonstrations Cited by Experienced and Novice Chemical Demonstrators Novices Experienced Type of (Pre-workshop ) Demonstrators Demonstration --- - - - - ----- ---- ------------ - -- t ratio Citation Mean SD Mean SD Variations (V) 0 . 5 0.8 2 . 4 1.1 3.30 * Demonstrations (D) 0 . 8 1. 2 4.8 1. 9 4.25 * Demonstration 0.0 0.0 2.0 0.7 6.32 ** Variations (DV ) Extraneous Examples , 0.6 0 . 7 0.2 0.4 1. 29 Total (XTOT ) Extraneous Examples, 0.4 0.5 0.2 0.4 0.6 5 Negative (XNEG) Extraneous Examples, 0.1 0 . 4 0 . 0 0.0 1.00 Positive (XPOS ) Notes. - (1- ) XTOT Sum of XPOS and XNEG. (2) DV Variations on alternative air pressure demon s trations . *p < . 01. **p < . 001. ;:i p ,, , ,, V and W. Attention should also be give n to the f inding that t he experienced demonstrators cited fewer extraneous examples t han novices in their discourse of how to demonstrate the targeted concepts. Th i s was particularly true for the concept, density . Knowledge of Variations. Five out of eight novi ce chemica l demonstrator s were aware of at least one variation on the Collaps ing Aluminum Can demonstration . This variation involved the popular us e of a larger , ditto-fluid size can inste ad of a soda can to demons tra t e the influence of air pressure . The following two discourses illustrate their knowledge of variations on the Collaps ing Aluminum 158 Table 16 Frequ~ncy of DENSIT~ Demonstrations Cited b Experienced and Novice Chemical Demonst rato~s Novices Experienced (Pre-workshop ) Demonstrators Type of ---------------- t-ratio Demonstrat i on Mean SD Mean SD Citation 0.7 2.8 0.4 7. 34 *** Variations (V) 0.4 5.2 3.1. 2 .84 * 1.1 1.0 Demonstrations (D) o.o o.o 1.0 1.7 1. 29 * Demonstration Variations (DV) 0.2 0.4. 3 . 39 ** 1.8 1. 2 Ext raneous Examples, Total (XTOT ) 0.0 0.0 2 . 97 ** 0.9 0.8 I' Ex traneous ?11 Negative (XNEG ) ,, , 0.4 1. 87 ,, . 1. 0 1.1 0.2 ;: , Extraneous Examples ' ;: ~ 1 ,,., . i, Posit ive (XPOS ) t ;I';. i:.' : :: :1 ***p < .001. < .05 . **p < .01. ' : ' ,,lf,l..i, *p Can demonstration . (Appendix U contains additional quotes ) . N4: The way I do it with [a ) ditto can is a little different. The r e ' s a very small hole in the top just to let the steam escape , and I have it on a hot plate with a smal l amount of water. The students don't know the water is in there. When I feel the water has evaporated , I put a plug in the top and it crumples right there on the hot plate . You don't have to use the ice. (VT #2 , I. 8) N2 : You can use the gasoline type can , with the water heated, and then cap it off, well, cut the heat and cap it off. And watch it shrink on its own. And if you want to increase it, pour ice water over it. It'd shrink much more demonstrably - much more dramatically. (VT, 12, I. 81 The experi enced demons trators also di scussed this variation, but in greater scientific detail and sugges ting a greater variety of 159 procedures and demonstration materials. The fo l lowing two quotes i llust ra t e thi s point. (See Appendix U for addi tional quot es) . E3 : One can also do [this ] on a l arger sca l e with the classic demonstration using duplicating fluid cans , and e i ther inverting them in water , running cold water over them, or simply stoppering it and letting it sit. And you'll see a similar effect which is not quite as quick, but quite dramatic , because the can is crushed. Those cans are quite strong , and not able to be crushed with ones ' hands . So there are some nice variations or extensions of the demo. (VT #2 , I. 8) E4: There is another variation on [this demonstration ] I just thought of a few seconds ago . You can use a fifty - five gallon drum on a big fire , with some water in the bottom, boil it, get it boiling quite well, get the whole can hot , take away the heat , throw a stopper in the can the same way you do with a gallon duplicator fl uid can, and s it down and wait. And when the pressure differential gets great enough, the fifty-five gallon drum will go "BOOM!!!" The "large-can" variation of the Collapsing Aluminum Can demonstration was discussed by half the novice chemical demons trators . These demonstrators, however, were generally unable to think of additional variations that would be instructive. The experienced demonstrators , on the other hand , were each able to think of at l east one or two other variations of this demonstration as having instructional value. These variations include the use of a f ull or empty aluminum can to conduct the demonstration so that the can' s collapse under these two conditions could be compared with that of t he partial l y- filled can. Some of the experienced demonstrators also suggested the possibility of spraying the can with water instead of inverting it into a pool of ice water to more clearly illus trate the influence of air pressure. Others indicated that the demonstrat i on could be r epeated to see if the same effect would occur us ing an iron can or in cooling the can with its spout right -side up. The novice 160 demonstrators appeared to have little knowledge of these additional variations that could be performed , and thus help students explore and understand the factors influencing the can ' s sudden collapse. The ve rbal r eports , therefore , revealed quantitative and qualitative differences between experienced and novice sub jects regarding chemical demonstration variations . Experienced and novice chemical demonstrators also showed quantitative and qualitative di fferences in their knowledge of variations on the Density Column demonstration. During the semi-structured interview , less than half the novices were able to discuss a variation of this demonstration that differed from the one they observed on videotape. The few novices that did discuss a variation , suggested a more dynamic approach to the demonstration involving the step-wise addition of the liquids to the cylinder to "..,. ., .. i show s tudents how the column was formed. A few others suggested the ?.~,,.. addition of more solids objects to the completed liquid col umn. The following quotes illustrate these novices ' pedagogica l knowledge of dynamic variations on the Density Column Demonstrat ion. Nl (Construct a column): [He could have) taken different samples of liquids , shown them in those little tiny bea kers , that they all had 10 mls and somehow gotten some sort of ma ss for them, and then as k, "Which one do you think is gonna go on the bottom? Which one is gonna go on the top? " and then do it to see if it ' s actually gonna happen. I t hink that would have pulled it all together. All they ' ve seen is that [column ] already being there . They didn't see it actually happe n. (VT #3, I. 8) N8 (Add more solid objects): I think I have done this . I have used more solid objects; rubber , cork, but I can ' t r emembe r what the other one is. [Subject ba s ically reite rates objects observed on videotape ] . (VT #4 , I. 7) 161 The experienced demonstrators also discussed several dynamic var i ations on the Density Column demonstration. In addition , they freely ve rbalized their personal preference for a particular variation , justifying their preference in terms of the impact it ha s on student learning. The following quotes i llustrate how experienced demonstrators responded . (Appendix U provides additional quot es ). El (Construct a column): Instead of trying to introduce three or four liquids , whatever that was in there, maybe starting with two first and then going to four might have made it a l ittle bit easier , or a stepping stone to the larger concept. Or , I don ' t know, I think with that idea I think it's fun for the kids to sit and watch it , to actually do the pouring right there with them using just two liquids and that way they can kind of guess what they think i s going to happen and then they can watch it happen. And I think in watching it happen they might remember it a little bit better. (VT #3, I. 8) E2 (Construct a column ) : I always like them to add , actually add the liquid in front of the students and have the students watch the things separating and layer out. I would have liked to have seen that done. (VT #3, I. 8) ES (Construct a column and add solid object s ): And aga in, I would build a density column in front of the student s and certainly drop in objects. (VT #3 , I. 9) Irrespective of the chemical demonstration obse rved (Coll aps ing Aluminum Can or Density Column) , analysis of the experi enced demonstrators ' transcripts indicate that they di scussed more variations on these chemical demonstrations , di scussed them in great er detail , and discussed more of their implications for l earning t han t he novice demonstrators. Knowledge of Alternative Demonstrations. Besides having a greater awarenes s of variations on specific chemical demons trations , experienced demonstrators also have greater pedagogical knowledge of alt e rnative demonstrations on air pressure and dens ity (see Tables 15 162 and 16, respectively). The verbatim transcripts showed that the experienced subjects were able to suggest significantly more concept-specific demonstrations that could be performed with middle school students than novice chemical demonstrators could suggest. The discourses of novices, when compared to the discourses of experienced demonstrators ' (p. 161-165), revealed their limited breadth and depth of knowledge of alternative air pressure and density demonstrations. The following quotes illustrate the brevity, and in one case the uncertainty, of novices' discourse on air pressure demonstrations. N2 (Vacuum chamber): Yeah, you can take a balloon and put it in a bell jar, and evaporate the air around it and the balloon expands in volume. (VT #3, I. 9) N2 (Barometer): Oh, a barometer. Sure, where you invert a vacuum tube - well, you evacuate air from the tube and invert it upside down; let the air pressure push the liquid up into the tube. (VT #3, I. 9) N3 (Barometer): You could use a mercury barometer, a really , ' ; ' old-fashioned mercury barometer, to let them see that. I ' ~ ' , (VT #3, I. 9) N3 (Cartesian Diver ): Cartesian diver, I guess , wouldn't really necessarily show just air pressure - but it would be , I guess. You could use a Cartesian diver. Interviewer: What part of it would show air pressure? N3: Well, that's what I'm thinking. I'm not sure. I ' d have to think about it. (VT #3, I. 9) Experienced chemical demonstrators also discussed the fir s t two demonstrations mentioned above. The Cartesian Diver demonstration was not discussed by the experienced group probably because it illus trat es changes in density much more clearly than it does the concept of air pressure . In any case , the experienced demonstrators provided much more information on the first two demonstrations by discussing practical variations that could be performed on these familiar 163 demonstrations. The fol l owing quotes il lustrate the depth of pedagogical knowledge held by t he experienced demonstrators. E3 (Vacuum chamber ) : Well , the other way of [showing ] air pressure , or the effect of the absence of [air pressure ], is any number of demonstrations t hat use a vacuum pump , either an el ectric one or a mechanical hand pump , where you withdraw air pressure from a cl osed container. And if one has balloons i nside such a container , one can watch the bal l oon blow up; or marshmallows , or shaving cream, any substance that contains a gas can sort of self- expand, seemingly , until air pressure is allowed to reenter, in which case those items will then return to their previous vo l ume . So , many demonstrations can be used wi th a vacuum pump or water aspirator , or a hand pump. (VT #2 , I. 9) ES (Vacuum chamber ): You can even put - you can soak some dried fruit , and put it into a bell j ar , and the fruit will regain its shape. You can especial ly do it with ... rai s ins , and it will come back into the form of a grape . That ' s something that was done back in the 1800s. Interviewer: I never knew that . ES: I think you need to soak the grapes first so they become a little pliable , and that there ' s enough gas inside that when you put it in a vacuum it will expand to a grape . ES (Barometer) : Well , I used to set up the mercury manometers , but because of the hazards , we don ' t do that. Many of the classrooms have a mercury barometer attached to the wall , and I think it ' s interesting and informative to bring the cl ass around that , look at it , and relate pressure to what they hear on the radio , it being 32 inches of mercury today , pressure ; or ' highs and the lows ', because these ar e things that the kids hear about , and they ' r e directly r el at ed to atmospheric pressure. Perhaps mentioning that the atmospheric pressure changes from day to day would be an important thing to bring into the lesson. Besides discussing these common demonstrations on air pres sure , the experienced chemical demonstrators also di scussed numer ous ot her demonstrations that can be used to teach the concept of ai r pres sure and its influence on objects . The following quotes illust r ate this breadth of pedagogical knowledge. (Additional quotes may be f ound i n Appendix U.) El (Breaking a Stick with a Sheet of Newspaper): Ai r pressure. Oh, there ' s so many of them. Gracious ! You can .,,,- 164 get into things like taking a stick and hanging it over a table and placing newspaper over the top of that , and whacking that , and of course air pressure is going to hold it down so that you can snap it. There ' s a tremendous amount of things you can do with air pressure. (VT #2 , I. 9) E2 (Egg in the Bottle): Oh yeh. I like the Egg in the Bottle. That's always one of my favorite ones . I like it when you [take] a big milk bottle and hard boiled egg; and you light the piece of paper, and you stick it down in there, and the flames shoot up, and the egg sucks down . I love that! That ' s one of my favorite ones for air press ure. E3 (Ammonia Fountain ) : An additional one would be the classic ammonia fountain reaction, where you simply boil a little ammonia in a flask with a stopper - one hole s topper and a tube. And ammonia , or hydrochloric acid, are very very soluble in water, so if you get the steam filling th e container, the gaseous ammonia rather, filling the containe r and invert it, as the gaseous ammonia dissolves in wat er, air pressure on the outside will force the liquid up a tube and cause a fountain type effect. That ' s another example of air pressure . (VT #2 , I. 8) The experienced demonstrators , as a group, discussed several additional chemical demonstrations that illustrate air press ure . A list of these additional air pressure demonstrations can be fo und in Appendix V. When it came to discussing demonstrations dealing with the concept of density , novices mentioned several more than they did with air pressure (compare Tables 15 and 16). However, thes e number s did not nearly approach the number of density demonstration s di scussed by experienced chemical demonstrators. Each novice generally di sc ussed only one other density demonstration during the clinical intervi ew. As a group they mentioned seven different demonstrations that illustrates the concept of density . Each experienced chemi cal demonstator , by contrast, freely recalled and discus sed an aver age of about five density demonstrations. Collectively , they mentioned 17 different chemical demon s trations on density. A li s t of these density 165 and air pressure demonstrations , appropriate for the middle school level, is provided in Appendices V and W. Other distinguishing characteristics observed in the protocols were that the experienced demonstrators were usually more able than novices to specify (1) the kinds of chemical s ubstances that are well-suited for a particular chemica l demonstration as well as (2) the outcomes that are assoc i ated with a given demonstration. Consider the information volunteered by novice N3 and experienced demonstrators E3 and ES in their discussion of a demonstration involving the dens ity of gase s . N3: As he was talking I was trying to think of the dens ity of gases , showing , you know, balloons with various ga ses in them, but other than that, I wish I did know some demos to do [ond ensity). (VT #3 , I . 9) E3 : Somewhere in his discussion he mentions what he r eally wanted to talk about was density of gases. There are a number of ways to show the density of gases. You could i ' l contrast helium versus air versus some dense gas like fr eon ' or s ulfur hexafluoride , all of which can be placed in --- J. , , ! balloons and one can watch the rate of fall or rate of as cent ;' 1 and compare densities that way. (VT #3, I. 8) ,; I :,., j ES: You can fill a balloon with freon and it sinks it ' s so heavy. You can fill a balloon with hydrogen or helium. Thi s is less dense than air so it rises. (VT #3, I. 8) In the general descriptions of the The Great Balloon Race demonstration provided above , the experienced demonstrator' s provided details regarding the number of systems to be compared, the mat er i a l s to be used , and, the kinds of observations that would be apparent to s tudent s . Much of this information was not noticed in novi ces ' protocols . Be s ides differing in the specificity of information provided in their descriptions of chemical demonstrations , the experienced group 166 also discussed (1) the importance of using simple materials/systems and (2) the importance of student involvement and hands-on experiences during a chemical demonstration more so than novices. These patterns are exemplified i n the following discourses taken from a novice and two experienced demonstrators as they discussed demonstrations pertaining to the relative density of solid objects. NS: Let ' s see . Well, certainly there' s a lot that has to do with density when you 're dealing with rocks . I do a uni t on r ocks and certainly some elements are more dense than other s. And I try to get that idea across to the kids , you know that the r ocks that are full of metals and stuff like that are going t o be more dense and are heavier, even if you have two rock samples that are approximately the same volume . You (students? teacher? ) can pick up two rocks; if one ' s ve ry much denser than the another, then you get a handle on it . So indirect things like that. (VT #4, I. 9) Contrast this sel ection of materials and extent of s tudent in - volvement with that of two experi enced chemical demonstrators ' ,, ,I discourses. ' E4: The one that I use i s metal cubes of all the same s i ze , but different metals, ranging from aluminum through s t ee l t o l ead and brass, so that when you try to pick up any two cubes , the difference in mass is obvious . In terviewer : How do you do that as a demo? E4: I lite rally pass them around [and say , J "Tell me about these four metal cubes ." And they're all different met al s . That's obvious without even identifying them. And they will realize that some are heavier than others. (VT #4 , I . 9) ES: Instead of mixing things with different dens ities I would conside r, you know - very often people have thi s pi ece of granit e and a piece of foam that l ooks like granit e and they ar e the same size and they l ook the same and you ca n, passing that around, you can actual ly f ee l the diffe r ence . (VT #3 , I. 9) A compar i son of the variations of the "Hand-Weighing" demonstration discussed by the experienced and novice chemical demons t rat or s s hows that the three subj ect s se l ect ed di f ferent 167 materials to perform this demonstration. The novice selected two objects with complex compositions and different appearances . The experienced subjects selected simpler s ubstances , choosing objects that were either uniform in composition or objects that were visual look-alikes. In other cases , where experienced and novice demonstrators discussed similar demonst r ations , the experienced demonstrators general l y preferred to use simpler systems to demonstrate abstract ideas to middl e school students. In the above discourses , one may also notice that the experienced demonstrators made more passing references to student involvement in a chemical demonstration , even if the demonstration i s very simple and easy to perform . In several instances , the experienced demonstrators would volunteer information about the kinds of sensory or cognitive experiences that students would have during a demonstration. Thi s information was volunteered less often by novices . This differe nce i s somewhat apparent in the last three discourses given above. Extraneous Examples . Novice chemical demonstrators wer e mor e inc l ined than experienced chemical demonstrators to discuss non-demonstration teaching strategies that are suitable for t eaching density and air pressure , (See Tables 15 and 16 , XPOS ). Some of the novices , especially when they were prompted by this researcher to discuss additional demonstrations related to the targeted chemi cal concepts , would frequent l y discuss a lab, class activity , or cl ass discussion that would help students learn about the concept s . The following discourses illustrate this. (Additional example s are provided in Appendix U. ) 168 N8 (Lab): I think what I ' d do is I'd take different sizes and see - I'd figure out the densities for different size objects and different objects that look the same size but they have different densities and I have one that is made out of rubber and one that is made out of aluminum and one made out of steel and one made out of wood. They are all the same size . Then they have to figure out what the density of each one of these even if they are the same size because when you look at the volume, the volume looks the same. I have them use a ruler and then I have them do water displacement and then I have them measure density and measure the grams. Then I think it comes more clear to them. Then I always have them bar graph so they can see which one has the higher density. Then I ask them questions. (VT #4, I. 8) NS (Act ivity ): Well , I did a neat thing with density with the planets, where you take all the planets and you take their sizes and their masses, and then you plot their densities. You can get some interesting relationships going on the re. Predictions you'd make about their densities are oft en different than what their actual densities are . And that's kind of neat. According to them, you know, like Jupite r s huge. The kids - if they have a misconception about dens ity, which a lot of people do, they walk around with how big i t i s when it ' s how dense it is and I try and get that fallacy out of the way. It ' s kind of neat , if you plot density of the planets , and you plot their masses and their vo lumes as well, you start seeing some neat relationships . (VT #4, I. 9) N7 (Discussion ) : What I did myself, I asked student s why i ce floats and we talked about that; and they came up with s ome pretty interesting answers. (VT #4, I. 8) ' I :: j Most of these non-demonstration examples represent l egitimate teaching strategies related to the teaching of density but they di d not address the interviewer's question about demonstrations that illustrate the concept of density (Question I 9. of Interview Guide , Appendix I). These non-demonstration responses may represent brainstorming efforts on the part of the novices to an swer the interviewer's questions when they could not think of actual demonstrations. Experienced demonstrators, after discus s ing several suitab l e demonstrations, openly indicated they could not r espond any further to this researcher's probing of additional demons tra tions . 169 Instead of supplying non-demonstration teaching strategies to illustrate the density concept, they usually made the following comments: "I ' m sure there are many other ones (El )"; "Those are the ones that come to mind right now (E4 )"; "No other ones come to mind right now , but I ' m sure there are many others (E3 )" ; "It' s hard to think off the top of my head. I ' m sure there are hundreds of othe r s (ES ) . " Some of the conceptions that novices had regarding suitable demonstrations on the concepts of density or air pressure were found to be erroneous, or at least pedagogically unsound, (See Tables 15 and 16, XNEG). Such misconceptions represented 43 % of the demonstrati ons (10 of 23 demonstrations ) discussed by the novice chemical demonstrators. These erroneous or pedagogically unsound conceptions represented only 6% percent of the demonstrations (4 of 63 demonstrations ) discussed by the experienced subjects during t he semi-structured interview. Some of these conceptions are now exami ned ' ; in detail . :. j A few novices considered the demonstration of Archimedes principle a legitimate example of a density demonstration . The fo l lowing quotes illustrate this understanding. N2 : The [demonstration ] where you fill the container and then insert something and watch the water displaced . Or you use the container that has the side arm where wat er can go out. And that helps illustrate density by showing volume displaced . (VT #3 , I . 9) N3: ... displacement of water by object s in the wat e r . (VT #3 , I. 9) NS : I think Archimedes principle could have been talked about a little . There ' s a neat story that goes along with Archimedes principle . The guy that has the 'Eureka' experience in the bathtub. My kids enjoy that story. And Archimedes is kind of an interesting name anyway. It seems 170 to stick in peoples minds. Really, when she [the videotaped model) was floating the different things , that ' s really what she was talking about. Things that are less dense are going to float on things that are more dense. (VT 14 , I . 10) The comments above suggest that these three novices conside r the demonstration or discussion of Ar chimedes principle to be synonymous with the demonstration or discussion of the concept of density. Thi s conceptual association influenced their understanding of how to effectively demonstrate the re l ative densities of substances . The experienced subjects did not make this false association. Although i t is true that Archimedes Principle is useful for determining an object' s absolute density and that such a demons tration could al so illustrate the relative density of a solid obj ect with r espect to a liquid , a demonstration that merely illustrates water displ acement by a solid object (an indicator of its volume) is not strictly synonymous with a demonstration on density (mass/volume). The close conceptual association between volume displacement and dens ity apparent ly influenced novices ' understanding of how to effectively demons trat e the dens ity concept . Novices also suggested a few other demonstrations as illustrati ng den s ity , when they, in fact , addressed other phy sical science concept s , such as surface tension and sedimentation rates . The following discourse of NS illustrates an interes ting di screpant eve nt involving density, but does not effectively illustrate the targeted concept. NS (Surface tension ) : What I would like to do i s af t e r we talked about density with chemistry student s , or with higher level elementary s tudents , doing something like showing t hem that [concept) with just a razor blade and just saying , "Obviously , s teel is greater in density than wat e r. So I am going to drop this in this [container of water ) and what i s 171 going to happen?" "Well, it is supposed to sink. " "Why doesn ' t it sink? " So , then you [talk] about that [razor blade] expanding on to that [water surface? ). But it is a little bit too much, I would think , for younger students because you are not trying to confuse them you are trying to straighten them out. It is a twist or so for a little kid. (VT #4, I. 9) Although novice NS had some notion that this demonstration would be a suitable discrepant event on density (demonstrations that are contrary to students ' intuitive understandings of density), the suggested demonstration actually illustrated another basic science concept much better , and was therefore coded as XNEG (extraneous , nega tive) . It basically reflected a pedagogically unsound demonstration on density. Such discrepant event items, though not too often encountered in the protocols , presented some diffi culty f or independent coders , e.g ., whether they should be coded as D (density demonstration) or U (pedagogically unsound demonstration). The clarity of the discourses provided by the demon strators sometimes , '. infl uenced the direction of coding. In any case, novices discussed such discrepant examples as suitable density demonst rations more frequently than experienced demonstrators. One novice demonstrator and one experienced demonstrator discussed a demonstration involving sedimentation layers and sedimentation rates as suitable for illustrating differences in the density of solids. N8 (Sedimentation rates ): I've used a density one where I take some mud and I put it in some water and I shake it. And obviously the gravel and the sand - well , gravel first , sand , and then the clay particles will sit there for a while. That shows different levels of density. I do that. I forgot to tell you that but that's in earth science. (VT #4 , I. 10 ) E4 (Sedimentation rates ): Density of course is just - you can just drop different things into liquids , and wat ch the 172 rate at which they fall, to look at differences in densities . (VT #4 , I . 9) Although the sedimentation rate of particles (and the corresponding sedimentation layers generated) correlates with particle density , other factors , such as the viscosity of the liquid, particle size, and particle shape also have an influence on the overall sedimentation rate and layering effect . Because of the number of variables interacting in this system, this demons tration would probably be considered an unsound demonstration of the concept of r e lative density at the middle school level. A similar pattern of erroneous and pedagogical unsound demonstrations were discussed by several novices and one experienced demonstrator on the topic of air pressure . These s ubj ects suggested that air pressure could be demonstrated by observing the rate of expa nsion of balloons covering the tops of heated flasks containing different liquids. These examples more accurately demonstrate vapor pressure and rates of evaporation of various liquids . Consequently, these demonstrations would also be cons ide red pedagogically unsound methods for illustrating ambient air pressure and its influence on common objects. A few subjects suggested demonstrations involving air foi l s and flight to illustrate air pressure. These demonstrations , however, illustrate Bernoulli's Principle and the laminar f l ow of air around ob j ects much better than it illustrates the s imple presence of air pressure. Although thes e demonstrations have value in their own right, they too were considered unsound selections of demonstrations to illustrate the simpl e presence of air press ure . These examples 173 constitute some of the XNEG entries (Extraneous Examples , Negative) in Table 15. A comparison of experienced and novi'ce chemical demontrator s ' discourses on alternative chemical demonstrations that effect ively i'llustrate air pressure and densi' ty sh ows th at nov?i ces ' concepti,o ns contain many more examples of erroneous or pedagogically unsound demonstrations they associate with the targeted concepts. The pedagogically unsound demonstrat i ons were classified as "unsound" because they better illustrate related chemical and physica l concepts, and not the ones of interest. Having examined experienced and novice chemical demonstrators ' knowledge of alternatives in conducting chemical demonstrati ons (domain c), the discussion next turns to these teachers' Rationale s (domain d) for their suggestions and evaluative judgment s. Experienced and Novice Demonstrators ' Rationales During the clinica l interview , demonstrators frequently gave rationales , or reasons , for their evaluative judgments and suggestions on how to improve a chemical demonstration performance. These rationales were usual l y given after demonstrators r endered a favorable or unfavorab l e judgment of a critical incident or after suggest ing a pedagogical enhancement (P), variation (V), or alternative demons t ra t i,o n (D) . Col lectively , these rationales represent the fourth domain (domain d) of chemical demonstrators ' pedagogical comments. A taxonomic analysis of demonstrators ' rationales showed that rati?o na 1e s cou ld be coded into one of four knowledge areas: (1) most 174 the learner , (2) the demonstration system, (3) the sub j ect matter , or (4) teaching strategies. Rat i onale statements from each of these four areas are exemplified by the fo l lowing quotes and discussion . Knowledge of the Learner Exper i enced and novice chemical demonstrators frequently justified t hei r evaluative j udgments and s uggestions for improvemen t based on their knowledge of middle school students . Thus , whenever demonstrators judged a videotaped teacher as teachi ng above or bel ow the leve l of the student , judgments were usually premi sed on demonstrators ' understanding of learner characteristics . Thi s understanding incl uded an awareness of student s ' (1) prior knowledge of science terms and concepts , (2) scientific r easoning skills , (3 ) motivation to l earn , and (4) attention spans. Two of these knowledge-of-the-learner rationales are exemplified in the following discourses . (Other "learner" discourses are presented in the pages that fo llow ) . E2 (prior know l edge ): He let him (the student ) get away with that [answer ]. I would have said, "Well , t ell me what an ' interface ' is? What do you mean by that ? Can you expl ain that to me? " Maybe not everybody understands the term . (VT #3 ; S: 41) N3 (motivation to learn): Again , with middle school students , [when he says] , "We won't worry about what [phenol phthalein ] is used for ... ," but immediately , kids wou l d want to know and would really want to have a bit more information. (VT #3 , S:84) Knowing what middle school students know and wh at they don't know about science , and knowing the level of student motivation to l earn science , provided demonstrators E2 and N3 , r espect ively, with a ba s i s for making evaluative judgments . Nearly all experienced and novice 175 demonstrators used a knowledge-of-the- learner rationale to s upport a significant proportion of their evaluative judgments regarding effective and ineffective chemical demonstrating. Di scour ses that contained a notably high percentage of rationales reflecting a knowledge of the learner included think-aloud comments on the teacher's use of new science terms (Category 3) and the quality of explanations (Category 4), (see Table 11). Within these categor i es , experienced demonstrators evoked the learner rationale more frequently than novices. Although a quantitative content analysis was not conducted on demonstrators' rationales, the differences between the two groups appeared approximately proportional to the numbe r of critical-stop judgments rendered (see Table 7). Qualitative differences in learner rationales were not discernable between the two groups of demonstrators discussing new terms and explanations . In another context , the experienced demonstrators provided most of the discussions on student misconceptions and how the se misconceptions could arise during a chemical demonstration. With the Col laps ing Aluminum Can demonstration, for exampl e, only the experienced demonstrators discussed the kinds of misconceptions students could develop after seeing the aluminum can suddenly collapse. These comments not only reveal experienced demon s trator s ' understanding of the demonstration and the science behind the demonstration, they reveal a knowledge of middle school student s ' thinking about the demonstration outcomes. The following quot es illustrate this unique knowledge of the learner . E3 (Students' reasoning): Well , one potential factor tha t can confuse students' understanding in the [Collapsing Aluminum Can] demonstration that was done, or the cla ss ic 176 duplicating fluid can demonstration , is that water on the outside of the can ... Interviewer: [You mean] the vapor , the moisture in the air? E3: No , the liquid water on the outside of the can, si nce you ' re inverting it into [a pan of ] water , that somehow liquid water on the outside of the can is crushing the can. Or in the duplicating fluid can experiment , the classic version of this [collapsing aluminum can demonstrati on], they run it under water, a stream of water, and so students can quite easily conclude that it's water press ure that' s crushing the can , and you have a total mi sconception of what ' s going on . (VT #2 , I. 8) El (Student s ' reasoning): Probably a better way to teach that would be , as ha s been mentioned many times , the kid s get confused when you ' re taking those tongs [to grab the ca n] and squashing that can [in a pool of ice water ] . They r eally f ee l like you have done that, [i.e. , squeeze the tongs too hard while trans f erring the can to the ice water]. And probably it ' s a good reason for having a child do that [transfer ] as opposed to you doing that so that they know that it has not been crushed by the instructor [with hi s tongs]. (VT #2 , I. 8) The discourse of experienced chemical demonstrator E3 extended beyond the recognition of potential "misconceptions " that students co uld develop from a demons tration, to ways of addressing the mi sconceptions. Th e following quote illustrates how subject E3 addressed a potential student misunders tanding regarding the Collapsing Aluminum Can demonst ration. A second example is provided in Appendix X. E3 (Dealing with student misunderstandings ): Using the procedures he had, one thing that students often conf use with thi s demo and the classic variation that has been aro und for a l ong time, the duplicating fluid can, is that somehow it is wat er press ure that is crushing the can . And an alternative rout e , after the hea ting, would have been to cap it, simply cap the can and let it cool to room t emperat ure at a normal rat e and that excludes the notion of the external water press ure having anything to do with the crushing of the can . And that could be done either with these soda cans or it could be done with the duplicating fluid can. (VT #2, I . 8) 177 E3's suggestions on how to keep students from developing misconceptions essentially involved a simplification of the origin al demonstration sys tem so that middle school students would not f ocus on the many extraneous vari ables operating on the system, i. e ., vari ables that student s may fal sely attribute to the can's collapse . The experienced demons trators considered simplified variations of a demonstration as a way of reducing student learning diffi culties. Novice demons trators rarely discussed s uch potential sources of s tudent misunders tandings stemming from a demonstration. However, one novice , critiquing the Density Column demonstration, did discuss sources of confus ion for the learner but showed some uncert ainty abo ut the importance of these critiqued features . N4: There are two things that - I don't know if they are important - but as I teach middle school I know the types of things to look for . One is that each one of those l ayer s (liquids in the column) should be an equal amount. And kids would think that wa s a setup. Maybe it' s different [i .e . , the way the liquids order in the dens ity column], beca use this one [liquid] takes up more space than that one . Those are just things that I was looking for that a middl e school kid would look for. They would say, "This wa s n' t fair, because this one [layer] is bigger than the other." [VT #3, S:96] N4: The other thing tha t I'm not too s ure if it' s f ai r t o use a solid because I don't know if that' s - I don't kn ow. I might be wrong about that. That (us ing solids) might not be the right concept to do with density, with the lab (densi ty demonstration) that he was doing. [VT #3, S: 96] The quotes presented above show that experienced demonstrators not only knew of misunders tandings that student s could devel op duri ng a chemical demons tration, they also knew of ways of r emedying t he problem. Their sugges ti ons frequently involved the use of simplified variations of a demonstration that remove potentially conf us ing 178 variables . In cont r ast , s uch suggestions and student rationales were rarely evident in the protocols of novice demonstrators . Knowledge of the Demonstration System Experienced demonstrators often discussed the complexity of the videotaped demonstrations . These comments frequently became the rationa l e for suggesting simpler variations of chemical demonstrations when performing demonstrations with middle school students . The following quotes illustrate this understanding of the complexity of chemical demonstrations held by experienced demonstrators. The quotes are taken from critiques of the Density Column demonstration. They contain both direct and indirect references to the demonstration ' s complexity . (Additional discourses may be found in Appendix X). El (Direct): Weakness. Other than maybe being a litt l e bit too complicated for middle school. The r e are too many things involved with this demonstration that needs to be watered down a little bit. (VT #4, I. 4) E2 (Indirect ): He could make a point with a simple [dens ity column] demonstration and use that as a focal point for quite a whi l e. You see so many people set up maybe four or five [demonstrations]; they do this, this, and this: thi s big impact stuff. (VT #3 , I. 5) E3 (Indirect) : There are several demonstrations that can be used to illustrate the concept of density ... although with the demonstration he did [perform, i.e., the density column demonstration], he did a good job with, but there are other ones which I would probabl y start off with in a clas s room setting if you had total choice . (VT #3, I. 9) ES (Direct): It's very confusing when you start m1 x1ng ideas together. It wasn't clear cut. Maybe at the e nd of the lesson he could show a more complicated situation and ask students to explain it. (VT #3, I. 9) Each of these comments reveals that experienced chemical demonstrators were familiar with the complexities of the videot aped demonstration. These sub jects saw the dens ity column as a complicated 179 chemical system for teaching the concept of density to middle school students. Because of this awareness , these demonstrators were cognizant of the demonstration ' s potential for hindering, rather than fostering , the learning of the density concept. Experienced demonstrators discussed multipl e aspects of a demonstration ' s compl exity. For example, with the density column demonstration , they considered the number of materials used in the demonstration as a major source of complexity. This demonstration typically calls for (1 ) four to five liquids layered one on top of the other and (2) several different solids: one resting at each of the interfaces , one floating on top , and another sitting on the bottom. The experienced demonstrators also discussed the complexity of the Density Column demonstration in terms of the number of chemical concepts that need to be addressed in order to fully explain th e density column system. The experienced demonstrators were aware that the column formed not only because of the density of the liquids , but also beca use adjacent liquids were insoluble in each other. Finally , these demonstrators also considered tangential issues , such as explanations for how the four liquids in the column were color ed , as further adding to the conceptual complexity of the demonstration. Because of the perceived complexity of the density column demonstration , one experienced demonstrator suggested that thi s demonstration would be inappropriate to use as an introductory l es s on on density . Only one novice made any reference to either the simplicity or complexity of the density column system as a rati onal e for accepting or modifying the videotaped presentation. 180 Discourses on the complexity of the Collapsing Aluminum Can xperienced demonstrators demonstration were generally indirect. E ? usually discussed the issue of compl ex demonstration systems in terms of sources of student misunderstandings (see previous subsection). Subject-Matter Knowledge A taxonomic analysis and content analysis of the descriptive knowledge domain (domain b ) has already shown that only the experienced chemical demonstrators identified incorrect science 6 propositions mentioned by the videotaped models (see pp. 143 - 14 ). The rationales that supported their unfavorable judgments of these critical incidents were clearly tied to their knowledge of chemi s try . Subject-matter knowledge also played an important role in experienced demonstrators ' critique of critical incidents involving the application of density and air pressure to students' lives . Subject - matter rationales are illustrated in the following critique s . E3 (Density column ) : And then lastly , what I hoped she would subsequently [do ], if not at this point , is to draw from students other applications of density so it becomes a r eal concept for them and not just something . I mean, ' So what. You know water floats on top of corn syrup. Who cares about that? ' so it would have some application to their lives. .. . The concept of density applies in hund~eds of applications in day to day living that could be drawn in , and eve rything f rom (1 ) submarines to (2) fish swimming to (3 ) density of batte ry acids to many , many, many applications of density . (VT #4, I. 5; VT #3 , I. 9) ES (Collapsing aluminum can) : I like relating that [demonstration ] to the steam engine. As a first treatment on a steam engine , spraying water into the [engine ' s] cylinder, that ' s the same as water going inside the tin (aluminum) can . (VT #2, I. 8) These expe rienced chemical demonstrators possessed ample knowledge of the application of chemis try to student s ' lives . This 181 knowledge of concept applicat i ons is a form of pedagogical content knowledge because it is discussed in the context of teaching chemistry In a non-teaching context, it would be to middle school students. n ei er case , this considered a form of subject-matter knowledge. I 'th specialized knowledge permitted the experienced demonstrators to make evaluative judgments and suggestions for improving the videotaped demonstrations not observed among novice demonstrators. Knowledge of Teaching Strategies Experienced and novice chemical demonstrators evaluated and discussed many critical incidents based on their knowledge of fundamental principles of effective pedagogy and educational psychology . This body of knowledge provided these demonstrators with some justification for the judgments they rendered. A taxonomic analysis of the protocols showed that many critical-stop discourses on effective and ineffective chemical demonstration teaching were supported by statements related to the importance of motivational strategies , wait time , feedback, reinforcement, instructional pace , visual learning, and questioning . Demonstrators also made frequent reference to the value of using analogies, modeling behaviors for students , building on students' prior knowledge, engaging students in discovery learning and other learning strategies as contributing to an effective chemical demonstration presentation. The following quotes illustrate how two demonstrators discussed the value of probing students ' prior knowledge and using motivational strategies to help contribute to an effective chemical demonstration performance at the middle school level. 182 Nl (Prior knowledge): And the fact that he is asking them about their past experience , what it is they know already , so he has a clearer idea of where to start with them, instead of assuming that they all know it (the concept "density ") or assuming that they don't know it. He ' s really getting a hold on , at least, where a few people in the class are [with those questions]. (VT #3 ; S:3 ) El (Motivational strategy ) : That was also good. He didn ' t tell them what he had in the [density] column . Ins tead, he suggested , that I ' m not going to tell you what is here. So it set up some mystery in their (the students' ) mind and probably made them sit up and take a little bit of notice. (VT #3; S: 15 ) Principles of effective pedagogy and educational psychology al so provided demonstrators with rationales to support their sugges tions of how to improve the videotaped chemical demonstrations. Experienced and novice demonstrators both made frequent reference to these educational psychology rationales in their think-aloud discours es . Differences between experienced and novice demonstrators were difficult to discern based on a taxonomic analysis alone . A det ailed content analysis was not conducted in domain c (Category 4: Knowledge of Teaching Strategies ) because resources were not available to train a fourth coder . Such an analysis may have detected subtle quantitative differences i n the number of teaching strategy r ati onal es mentioned by the two groups of demonstrators. Competing Rationales and Evaluative Judgments A sma l l percent (approx . 4-5%) of the critical incidents discussed by two or more demonstrators revealed differences in opinion (i . e ., eval uative judgment ) regarding whether an observed t eacher behavior contributed to or distracted from the chemical demonstration performance. Such divergent judgments were consis tently accompanied by different rationales used to support their judgment s . Explor i ng 183 these differences in teachers' thinking may shed some light on experienced and novice demonstrators' pedagogical judgment s and decisions during preactive and interactive teaching. Differences that emerge may suggest potential areas of growth in science teachers profes s ional knowledge base . Consider the quotes from the foll owing two subjects addressing the critical incident involving the appropriateness of using the term 'phenolphthalein ' during a presentation of the Dens ity Column demonstration at the middle school level . N4 (favorable judgment): I think its good to introduce t hem to things like phenolphthalein, even though they don ' t have to know about it, because they become familar with the words that they'll hear later on if they have more chemistry. And yet , at the same time, he says, "You don't have to worry about it , " so the pressure is off, but the information has been planted. So , I think that's an O. K. technique. (VT #3 , S:85) E3 (unfavorable judgment ) : Here he ' s telling them that he's dyed the water which has changed its vi s ible properties . It i s good that he ' s telling that. Telling them that it' s a dye alone would probabl y be sufficient rather than sayi ng its phe nolphthale in which is jus t some ot her chemical name for people. And telling them that chemical name is not important . I would have preferred just saying to them that the water has been dyed and letting it go at that. By throwing in "phenolphthalein" , is another word that t he students now have to think about, even though he ' s told them it ' s not important. If you ' re going to use words , they are important; use simpler words that you don ' t have to define . (VT #3 , S: 82) Comparing these two di scourses , it i s apparent that the novice demons trator eva luated this critical incident favorably using a ra tionale based on a knowledge of the high school sc i ence curriculum . This teacher considered it beneficial to at leas t minimally familiariz e middle school st udents with new terms they will encount er again in s ucceeding science courses. The experienced demonstra t or , on 184 the other hand, rendered an unfavorable judgment of this incident and provided a supporting rationale based on a knowledge of how midd l e school students learn. This demonstrator considered the new term 'phenolphthalein' excess ive because it places an additional cognitive demand on students that is unnecessary. The use of this term would , presumably, only interfere with the learning of the central concept , density . Such differences in evaluative judgments and supporting rationales were not only observed between experienced and novice demonstrators , but on occasion, competing rationales were also observed between subj ects in the same group. The experienced subjects , however, appeared to show slightly greater agreement in their evaluative judgments and supporting rationales than novices. The Pedagogical Knowledge System Probed by the Critical-Stop Task A theme analysis (Spradley , 1980) was conducted on the protocols for two reasons: to further document experienced- novice chemical demonstrator differences and to examine the extent to which the verbal data obtained during the clinical interview reflected Shulman ' s (1 986 , 1987) two theoretical constructs, pedagogical content knowledge (PCK ) and general pedagogical knowledge (GPK). The entries in Table 17 show that most (over 90 %) of the comments made by the chemical demonstrators during the critical -s top ta s k addressed pedagogical issues important to chemical demonstrati ng (PCK + GPK). About 2% of the critical-incident comments focu sed on s ubj ec t matter knowledge issues (SM), i . e. , incident s involving the conveyance of inaccurate chemistry content. The remaining comment s foc used on 185 i ssues that the coders could not readily classify (NC ) into the three knowledge systems, PCK , GPK, and SM, using literature definitions provided in the codi ng instructions (Appendix P) . The NC comments ge nera lly dealt with the manipulation of demonstration equipment and safety issues. A quantitative theme analysis , summarized in Table 17 , revealed a characteristic difference between experienced and novice chemical demonstrators in terms of the frequency with which the two pedagogical Table 17 Theme Ana lysis of Critical-Stop Discourses According to Pedagogical Knowledge System No. Critical Comments Made by Chemical Demonstrators I AVE. / % TOTAL (per group) Knowledge - - -- --------------------- , --------------- --- - -- - - -- System or Novice Exper. I Novice Expe r . Theme I Pre- Post - I Pre- Post - Workshop Workshop /Workshop Works hop ----------- ---------,' ------------ Pedagogical Content 38.0 41. 2 57 . 0 ** / 71.7 76.0 70.8 Knowledge (PCK) I I General Pedagogical 11. 0 10.0 15.5 * I 20. 8 18 . 5 19.3 Knowledge (GPK) I I Subject Matt e r (SM) 1.0 1. 8 1. 8 I 1. 9 3 . 3 2 . 2 I No Code (NC ) 3.0 1.2 6.2 * I 5.7 2.2 7. 7 ---------------------------/---------- - - --- ---------- Total a: 53 . 0 54.2 80. 5 ** / 100.1 100 . 0 100.0 - -------------------- '- --------- - - Not es . - a- Totals that differ from Tables 7 and 11 reflect the additi onal coding of "neutral " critical incidents . * p . 125 between experienced and pre-workshop novice demons trators . ** p . 06 25 between experienced and pre-workshop novice demonstrators . 186 knowledge systems were evoked during the critical-stop task. In absolute terms, the experienced chemical demonstrators made one _a nd-a- half times more frequent mention of pedagogical content knowledge issues than novices during their critique of the four demonstration These differences videotapes (57 vs 38 PCK comment s , respectively). reached a probability of .0625 using Wilcoxon' s matched-pairs, ranked-sign test. (Probabilit i es~ .187 are considered meaningful using this non-parametric test). Likewise, for general pedagogical knowledge , the experienced demonstrators made 1.4 times mo re frequent mention of general pedagogical issues than novices (1 5.5 and 11.0 GPK comments , respectively; p = .1 25 ). In relative terms , the data in Tabl e 17 show that about 70 % of the comments made by experienced and novice chemical demonstrators critiquing the four demonstration videotapes ref l ected pedagogi cal content knowledge concerns. These data provide evidence that the critical-stop task used in this study effectively probed teachers ' pedagogical content knowledge . To a lesser extent (1 5- 20 % of the discourses ), it al so probed other forms of knowledge important to effective chemical demonstrating. Coder agreement in classifying critical-incident discourses according to theme (PCK , GPK, or SM ) was 87. 2%. Scotts pi was ca l cul ated to be 82 .9% when ad justing for chance coding into these three knowledge categories and NC. several critical-stop discourses are examined next to illustrate the two forms of pedagogical knowledge evident in exper i enced chemical demonstrators' discourses (i .e ., PCK and GPK). Although many examples of PCK have already been provided in previous sections , prior 187 discussions did not explicitly tie demonstrators ' discourses to any particular pedagogical knowledge system reported in the literature. An effort to associate demonstrators ' discourses to a specific knowledge system is the goal of the discussion that follows. The discourses selected to distinguish between PCK and GPK verbalizations come from the critique of Videotape #3, the Density Column demonstration. Selection of this tape to il l ustrate the findings was arbitrary. PCK and GPK discourses were equally evident in the thi nk-aloud critiques of the other three videotapes , but in order to better compare discourses , this researcher decided to keep content and context similar and , therefore , use the critical-stop discourses of only one videotape. The decision to use only experienced demonstrator discourses in this section was based on the clarity of their comments for i llustrating the nature of the two pedagogica l knowledge systems. Examples of Demonstrators ' Pedagogical Content Knowledge , PCK The following quotes reflect the nature of experienced chemical demonstrators' discourses that coded as pedagogical content knowledge (PCK ) during the theme analysis . PCK Example 1: Evaluating the density column demonstration ES: Th' choice of a demonstration at this point i s . is?ate certainl y density is a difficult topic. inappropn not? start with a compl?i cated system whe re you YhaO vU e S ha OmU lidx ture of different dens?i ti' es b ~ t yous h ou ld ma ke sure that [students ] hav~ ~n under~tanding of the ? between densities of single substances such as ac omropcakr iasondn s a piece of styrofoam the same si? ze . (V'I' #3 , S: 15 ) 188 In this example , demonstrator ES mentions that the density column demonstration would not be an appropriate introduction to density at the middle school level. To address this matter, demonstrator ES offered a brief description of a more suitable demonstration to introduce students to the concept of density. The alternative invol ves the simple weighing of two different solids of the same size. (In the follow-up interview, this demonstrator suggested the use of an equal-arm balance to conduct this weighing ) . This alternative demonstration is much simpler to perform and it clearly illustrates Shulman's idea of pedagogical content knowledge, i.e., knowledge of how to teach specific topics, such as density. PCK Example 2: Using visual aids. E3: Here, ... he's identified these layers [in the density column]. It would be preferable if he had a physical drawing on the board or an overhead that indicated the different liquids relative to the actual set - up so people have a direct visual of it, rather than just his words that , "Well, water is here and methanol i s here." (VT.J/:3, S:90) ES: And I think I would like to see some vi sual rather than abstract sketches [words] on the board. The thing he put on the board were pretty much words, whereas a few sketches of how equal masses of material might have varying volumes [would be instructive]. (VT #3 , S:97) PCK Example 2 contains two suggestions for improving the observed chemical demonstration by means of the supplemental use of blackboard visual aids. These suggestions include the use of the blackboard t o (1) diagram the physical system on display and (2) to diagram various objects with equal ma sses occupying different volumes. These "v isual " illustrations provide middle school students with additional 189 representations of the concept of density. They further exemplify Shulman ' s conception of pedagogica l cont ent knowledge. Ot her di scourses coded as pedagogical content knowledge included critiques on the videotaped teachers ' use of new terms. Five types of "new term " discourses qualified as PCK codi ngs. They included discourses in which t he demonstra t or (1) recognizes outstanding explanations of new terms provided by the videotaped teacher , (2 ) provides al t ernative explanat i ons of new terms used in a demonstration , (3) suggests the deletion of certain new terms and their explanations, ( 4 ) suggests the substitution of simpler terms in pl ace of the ones used , and (5 ) identifies the problem of the videotaped teacher focusing on irrelevant terms that detract from th e central i ssue or concept being demonstrated . The foll owing quotes illustrate the thi rd , fourth , and fifth type of di scourse coded as PCK, respectively. The quotes address the i ssue of using the scientific term "phenolphthalein " during a demonstration with middle school students. The quotes refer to a critical f eat ure different from the phenolphthalein incident di scussed on page 144 (a Knowledge Inaccuracy ) but similar to the one discussed on page 155 (Pedagogical Enhancements ) and 182-183 (Competing Rationales ) . Obtaining r epresentative quotes that address the issue of new terms and that also ill ustrate demons trators ' PCK under standings were difficult to find in the protocol s , therefore , a phenolphthalein-related critical incident is presented again. PCK Example 3 : Explaining chemical t erms. E2 (Substituting simpler terms ) : I thought that was a strange comment , [" I added a dye called phenolphthal ein, O.K. And we won ' t worry about what phenolphthalein i s used for right now" ]. I mean, if you're not going to 190 wor ry about it , why br ing it up? Why d on ' t you say, "It ' s coloring "? (VT #3 , S: 83 ) El (Focusing on irrelevant terms; delete term): This was completely irrelevant. The [term] 'phenolphthalein' something he shouldn ' t even have mentioned because fo;as some people that would have just taken them off into another direction. . . . I'm going to stop [the videotape) again because he's saying that he's not goin to talk about phenolphthalein, but this part of the g demonstration i s predominantly on phenolphthalein and what it does and does not do; and it really does not have much to do with the problem of density. (VT #3, S: 81 & 86 ) In the examples provided above, two demonstrators di scuss the 1 appropriateness of using the term ' phenolphthale in' with middle school students during a presentation of the Density Co lumn demonstration. In the first quote given above, demonstrator E2 suggests the use of a simpler term for phenolphthalein, like Thi s 'coloring': a term which most students could readily comprehend. suggestion reflects this demonstrators ' pedagogical content knowledge because it reveals her understanding of effective and ineff ective explanations of chemical phenomena at the middle school l evel . The comments of demonstrator El re late to the probl em of focu s ing students ' attent i on on non-essential terms that detract from the central issue or concept being demonstrated . El's comment s sugges t indirect ly the inappropriateness of discussing 'phenolphthalein' wi t h middle school students when the focus of the demonstration i s on density. such comments reflect demonstrator's pedagogical content knowledge in a more indirect sense because they reveal a knowledge of "ineffective explanations " (the converse of Shulman's definition) . Phenolphthalein is a chemical used to make one of the liquid layers 1 in the density column more visible. 191 These discourses reveal demonstrators' knowledge of the importance of avoiding the use of formal chemical names when teaching density to middle school students. This aspect of pedagogica l content knowledge reflects an understanding of those explanations that interfere with the learning of a particular chemical concept. Discourses on the day-to -day applications of density to students ' lives also code as pedagogical content knowledge, as shown by the fol l owing critical-s top comments . PCK Example 4: Applying density to student experiences E3: This is real good here. He's tying this concept [density ], which they ' ve ... talked about in reference to this particular system [the density column], and now he is applying it to something that the students are more likely to know about , oil on water, [e.g. , oil from a motor] boat in a lake s ituation. (VT #3, S:76) El: This i s excellent. He's using the idea of the concept (density) and applying it to their lives. And again , I find myself listening a little closer. "Gee . Have I ever had this experience?" (VT #3, S:76) The idea of appl ying basic chemical concepts to students' experiences , such as having students think about an experience s imilar to the one of seeing oil float on top of water at a mar ina , al so reflects Shulman's construct of pedagogical content knowledge . In the discourses above , the critical incident discussed actually reflects the videotaped teachers ' knowledge of a useful form of representing the density concept . Even though the videotaped teacher provided the content example, the experienced demonstrators' critique of thi s critical incident illustrates their recognition of the value of integrating such applications of density into a chemical demonstration performance . Critiques of this nature were also considered to be an indirect reflection of chemical demonstrators ' pedagogical content 192 knowledge because the demonstrator clearly recognized the value of the PCK representation supplied by the videotaped teacher. The evidence was indirect because the application elicited by the videotaped teacher may not have been part of demonstrator El ' s and E3's prior PCK representation of density and how to teach it. Both direct and indirect references to PCK comprise the tallies given in Table 17. The quotes given above provide qualitative evidence that the critical-stop task prompts teachers to draw on and verba lize thei r pedagogical content knowledge. This knowledge was evident throughout demonstrators' protocols in the form of useful examples, applications , explanations , and alternative demonstrations that are useful i n the demonstration teaching of density to the middle school students. Examples of Demonstrators ' General Pedagogical Knowledge , GPK About 20% of the critical-stop discourses of experienced and novice demonstrators (see Table 17) reflected a second form of pedagogical knowledge Shulman (1986, 1987) calls general pedagogi ca l knowledge (GPK). This suggests that experienced and novice chemical demonstrators both consider generic teaching ski l l s as contributing in an important way to the demonstration teaching of specific chemical concepts . Examples of this form of pedagogical knowledge are now provided with quotes taken from teachers ' critical-stop discourses . The quot es illustrate the range of contexts discussed by experienced chemica l demonstrators that coded as general pedagogical knowledge and considered important to effective chemical demonstration teaching. ? 193 GPK Example l: Presentation Style E2: He gave a good introduction. Nice and clear. Hi s volume is up. And he's looking around like he is looking at every one in the class and not just talking to himself. (VT #3, S: 2) E4: Also, good interaction with the audience. He i s making eye contact with the audience. He is looking at people . He is acknowledging them. He's probably doing more so, of the tapes we viewed so far , and that ' s good. He is addres sing people. (VT #3, S:37) In this example, two experienced demonstrators discuss the i ss ue of the videotaped teachers possessing good volume and good eye contact while performing the density column demons tration. These comments reflect a knowledge of generic teaching skills that are appli cabl e t o both chemical demonstration teaching and to a variety of other non-chemical demon s tration teaching settings . Di scourses on the communication skills needed to conduct a chemical demon s tration effectively are content-independent issues and therefo r e indicative of general pedagogical knowledge. GPK Example 2: Organization of movement around the demons t ra ti on t able El: It might be minor, but it affected me wh en he s t opped and went around the table. It probably would have been better had he started back there if he wa s going to use the board. (VT #3, S:3) ES : One of the weaknesses i s that he seems t o be continually walking from the front to the back of t he bench and he was as king the questions and he knew he was going write them down . He probably should have s t ayed at the back. (VT #3, S:7) Exampl e 2 talks about an organizational i ss ue relat ed to t he videot aped teacher's excessive movement around a large demons t ration table in the f ront of the room. Because these comment s are not discipline -bound or tied to a knowledge of density, they al so reflect 194 demonstration general pedagogical knowledge associated with chemical teaching. GPK Example 3: Interactive , Participatory Style st E3: Here he's involving students . Rather than him ? talking and showing, he ' s already asked students f orJuth , i,n pu t [ stud ent ob servati,o ns on d ensi, ty column]. And in eir this case he is having one student be a recorder for him which makes the class more invol ved. (VT #3, S:24) E4: O.K. Another good point. Realizing the set-up of the room makes him walk around that [demonstration] table He decided that he was go i ng to work with the audience and he was going to have somebody come up [to the blackboard) and take notes . And that's a good point, a strong point. (VT #3 , S:26) The comments in Example 3 refer to the videotaped teacher asking the class to make observations on the density column while using a student recorder to keep track of students ' observations. This strategy of using a student recorder during a demonstration also reflects general pedagogical knowledge because it can be applied in a variety of teaching contexts. GPK Example 4: Questioning strategies E2: He is good at accepting people's answers all the time . (VT #3, S:77) The discourse in Example 4 talks about the need to be open t o student responses to questions raised by the teacher. This too i s a content-independent issue and reflects general pedagogical knowledge . The four examples given above provide a sampling of some of the critical-stop coJ11iTients made by experienced chemical demonstrators that coded as general pedagogical knowledge {GPKI during a theme analysis of the data . They show that many pedagogical issues related to effective chemical demonstration teaching are clearly content 195 independent. The quotes al so provide qual itative evidence that the critical-stop task probes both teachers ' general pedagogi'cal knowledge (GPK ) as wel l as their pedagogical content knowl edge (PCK). The semi-structured interview that followed the thi' n k -a 1 oud, critical-stop task al so contained discourses r epresenting the two pedagogical knowledge systems . Coding Difficulties Presented by Some Critical - Incident Discourses The reliability of coding teachers ' critical-s t op discourses i nt o t he two ma j or themes , pedagogical content knowledge (PCK) and general pedagogical knowledge (GPK ), was calculated to be 87.2 %. Correcting for chance agreement in coding pedagogical discourses into the two major coding categories, inter-coder reliability was calcula t ed to be 74.4 %. Mos t of the coding difficulties encountered during the theme analysis again involved discourses that would not easily code into one of the two pedagogical categories defined by Shulman (1986, 1987), but represented pedagogical verbalizations that could be coded either way depending on how wel l the coder understood the context of the discourse and whether inferences had to be made in the coding process . The cri t ical -s top discourses that presented coder disagreements during the theme analysis include those related to : (1) the effective manipulation of chemical demonstration equipment and safety issues , (2) the effective use of the blackboard and student volunteers during a demonstration, (3) the use of general communication sdk~lls (~.g., prai~e , encouragemen t) t hat promote class iscussions on sc i ence terms , concepts , or processes , and (4) the use of wait time and fielding questions related to chemical terms , concepts, or procedures . 196 The following two critical-stop discourses I for example , did not different codings by easily code as PCK or GPK, and therefore yiel ded independent coders . Example 1: Mechanics of Demo 0 E3: Here he - I don't think it was intentional for h' have the cylinder sort of covered with his hand 0 rt~~ t it was just the way he was holding it. I don't think 17 ~ was a major problem but in terms of visibility idea11 h should have held it in such that they can see lhe wholy e thing unless he had a distinct purpose in hiding part ~f it . At least it i s not obvious at t his point that he did (VT #3 , S , 45) . In Example 1, a demonst rator suggests the importance of handling the density column without hindering the visibility of the top of the Such a cylinder which contains a liquid that is clear and colorless. comment r elates to procedural knowledge in conducting chemical demonstrations effectively. Procedural knowl edge di scourses were coded several ways during the theme analysis, e ither as PCK, GPK , or NC (No Code ) . E3 ' s suggestion requires some science subject matter knowledge ; knowl edge that some col orless liquids are somewhat more difficult to see from a distance, and especially if they are partially obstructed. Thus , in a limited sense , this particular discourse relates to one of Wilson and Gudmundsdott i r's (1986) definition of PCK , namely, the "types of mat erials to use" to t each a given topic . Therefore , some coders would be inclined to code discourses on the effective or ineffective handling of demonstration equipment as pedagogical content knowledge (PCK ) . on the other hand, the issue of handling demonstration equipment so that it is visible to student s could al so 197 be coded as GPK by other coders who are inclined to think that visibility of teaching aids is a generic teaching issue. Given the confusion in coding this particular discourse , it should probably be coded as NC. In fact, most of the NC tallies in Table 17 involved procedural discourses. Occasionally, non-procedural discourses that contained little or no reference to subject matter content also presented coders with PCK or GPK coding difficulty. Thus, for a few discourses, the degree of specificity a demonstrator chose to provide in discus s ing a critical incident al so influenced coder reliability during the theme analys is. The following critical-stop comments illustrate this dilemma. I t corresponds to one of the density demonstration videotapes and relates to lesson closure. Example 2: Closure N3: O.K. He's doing the generalizing at that point. [ "We could generalize a little bit about densities . Sol ids generally are more dense than liquids . Not all liquids have the same density." (S:90-92)] I think it might be effective to say , 'Seeing what we have seen , looking at this demonstration, what could you generalize ? Can you come up with a generalization you might feel i s valid?' just to see if maybe they might. I think they could. (VT #3, S:91) The quote provided in Example 2 pertains to the videotaped teacher s ummarizing observations and drawing conclusions at the end of the demonst ra tion so students could share a common unders tanding of the concept presented. Because this discourse makes no refer ence to specific science terms or concepts, a coder may occas i onally be inclined to think that these comments reflect a general teaching principle (closure ) and, there fore, code it as GPK . However , the context of the di scour se is made in direct refe r ence to the videotaped 198 teacher's closing corrunents that solids are generally more dense than liquids and not all liquids have the same density. Given this context , the discourse better reflects this teachers ' pedagogical content knowledge of how to present the subject matter to middle school students. It therefore codes as PCK , but was occasionally miscoded as GPK . The discourse given above suggest that the level of specificity a demonstrator chooses to provide during a critical-stop discours e may, at times , influence coding and , thus, coder reliability. This i s e specially true if the coder is not thoroughly familiar with the context of the critique. Increased rater training and familiariz ati on with the demonstration videotapes could minimize such diffe r ences . Closing Comments on the Taxonomic and Theme Analysis Findings of Experienced and Novice Chemical Demonstrator Discourses Research Question 2 asked about the major corrunonalities and differences in experienced and novice chemical demonstrator dis cou r ses on effective chemical demonstration teaching. A taxonomic, component i al , t heme, and content analysis revealed both quantit a t i ve and qualitative differences between the two groups of demonstra t ors i n terms of their knowledge of how to demonstrate specific chemica l concepts to middle school students . Having presented the major differences between experience and novice demonstrators ' pedagogical knowledge discourses , att ent i on is now turned towards the findings that pertain to the influence of an intensive inservice workshop on fostering pedagogical knowledge growth (discourse change) among novice chemical demonstrators. r 199 Influence of Intensive Inserv. icing on Novice. Chemical Dem onst raters' Knowle d ge of Chenu.cal Demonstration Teaching Novice chemical demonstrators pre- and post-workshop responses to the think-aloud task and follow-up questions were audiotaped and transcribed. The resulting protocols provided a written record of their interview verbalizations. Analysis of these protocols helped address Research Question 3 which seeks to understand the influence of an intensive inservice intervention on promoting pedagogical knowledge growth among novice chemical demonstrators. A domain analysis, taxonomic analysis, componential analysis , theme analysis, and content analysis were used to analyz e the pre- and post-workshop protocols. These five methods of analyses helped highlight the commonalities and differences in novices ' pre- and post-workshop verbalizations on chemical demonstration t eaching of two chemical concepts. Changes in Novices' Evaluative Judgments A taxonomic analysis of the evaluative judgment domain showed that post-workshop novices continued to critique the four videotaped demonstrators using three forms of evaluative judgment. Most 11 , ned some form of "strength " or "weakness 1 evaluative statements con t a evaluation. A small percentage of discourses contained an "acceptable -but-could-be-better" judgment. A content analysis within the evaluative judgment domain showed that the inservice workshop intervention did not produce a meaningful l number of critical incidents eva luated by novice change in the _tot_a Table 7 presented earlier in thi s ( ee chemica l demonstrators s 200 A more detailed content analysis using the three taxonomic chapter) . of the evaluative judgment domain showed that the overall categories consistency in critique frequency was accompanied by a small, but non-meaningful, increase in the number of strengths and acceptable judgments rendered and a correspondingly small, and also non-meaningful, decrease in the number of unfavorable judgments rendered as a result of participation in the workshop (Table 7). (Qualitative changes are addressed in the next section ). A comparison of the experienced chemical demonstrators with the post-workshop novices showed that the experienced demonstrators continued to critique more critical incidents during the think-aloud task than the workshop-trained novices (p = .0625, using the Wilcoxon matched-pairs, ranked-signs non-parametric test and n = 4 paired scores; Table 7). In terms of the average number of strengths identified, the experienced subjects continued to identify more critical incidents than post-workshop novices (26 . 8 and 21 . 2, respectively). The novices, however, did identify more critical strengths after the workshop than before the workshop . Their critiquing skill for identifying "strengths ' appeared to be intermediate between the experienced demonstrators and their pre-workshop performance. This change suggests that the works hop had some effect on helping workshop participants become more cogni,ant of factors that contribute to effective chemical demonstrating at the middle school level. The total number of "frequently-cited" critical features identified by novice demonstrators after receiving the works hop ? t , , d sli'ghtlY (i e from 20 to 25 fr equent ly ci ted 1 in ervention increase ? ? 201 features; see Table 10 presented earlier in this chapter). Thi s post-workshop number more closely resembles the number of frequently cited critical features identified by experienced subjects (30). It suggests greater commonality (and decreased variance) in the focus of their think-aloud discourses. Frequently cited critical features continued to comprise a small percentage (13.7 %) of the total number of features identified by post-workshop novices, rising only slightly from the pre-workshop levels of 11.3%. Experienced demonstrators and post - workshop novices generally identified and discussed different sets of frequently cited cri t i ca l incidents. Only 15% of the frequently cited critical incident s identified were mutually discussed by both groups of subject s (see Appendix R). Similar low levels of group agreement wer e obser ved between experienced demonstrators and pre-workshop novi ces (17 % mutually discussed). Pre- and post-workshop novices al so showed a s imilar low l evel of agreement in discussing identical critical incidents (16 % mutual). Changes in Novices' Descriptive Knowled e A taxonomic analysis of the descriptive knowledge domain (doma in b) showed that post-workshop novices addres sed the s ame nine pedagogical issues they addressed during the pre-workshop interview. These issues included the e ffective use of inquiry, ques t ioning , t quality of explanations , interaction, mechanics St ra t egies, new erms, , ? sual aids organization, and presenta t ion Of the d emonstra t ion, vi , e same general issues addressed by Style. They represen t th These experienced demons trators during the critical -s top t as k. 202 pedagogical issues were discussed as contributors to effective chemical demonstration teaching. In quantitative terms , very little change occurred in the number of critiques made by post - workshop novices in eight out of the nine pedagogical categories discussed (Table 11, p. 127). Post-workshop on critiquing subjects placed slightly greater emphasis (p = .187) ? critical incidents related to the quality of explanations provided by the videotaped models . The top three issues discussed by novices after the workshop continued to be questioning strat egies , quality of explanations , and overall organization. The workshop produced a change in discourse focus , with novices giving less attention to issues pertaining to the use of new terms, the use of the blackboard and visual aids, and overall organization (pre-workshop concerns) and greater relative attention to the quality of explanations and questioning strategies (post-workshop concerns) (Table 12, p. 131) . The following discussion summarizes the findings of a componential analysis of novices ' clinical interview discourses , highlighting pre-/post-workshop changes within the nine categories comprising the descriptive knowledge domain. Inquiry, Investigative ~roa~ Prior to the workshop, novices made no direct reference to the term "inquiry" or "inquiry approach" in their analysis of the four demonstration videotapes; however , this situation changed during the post-workshop interview. several of the novice chemical demonstrators began to make reference to the value of this instructional strategy when conducting chemical demonstrations with middle school students. 203 The following quotes taken from the post-workshop transcript s of novice demonstrators illustrate this particular focus. N2: Sometimes he'll make the kids guess what's going on , and that's O.K. using the inquiry approach. But thi s approach is also good because they know how to direct their attention. He asks them questions (video script S:17 ,1 9,21, 26). He i s not giving them any answers but he is telling them how to direct their attention, which can be very helpful, especial ly with the middle school level. It reminds them that they ar e not just being entertained, that they're supposed to be thinking. (VT #1, S: 30 ) N3: She is so excited that something is working right , as we all are, you know, when something works, but t here are so many opportunities for student interactions , an opportunity for students to use inquiry and it ' s not offered t o them, which is a shame. (VT #4, S:95) N7: I can't remember, who said something about the water condensing and causing the air press ure to decrease . Did he give that information? (Interviewer nods ). I thought he s upplied it . I would have given a little mor e inquiry. (VT #2, S:46 ) A descriptive summary of the workshop staff presentations (Appendix G) shows that inquiry demonstrating was frequently model ed by workshop instructors. The term "inquiry " was also mentioned once and alluded to several times during the fir st staff presentation entitled, "What Makes an Effective Demo?" (See quotes in Appendix G). Although novices became more cognizant of the compatability of inquiry me thods of instruction and chemical demonstration t eaching, post-workshop novices did not show any s ignificant change in the to t al number of critical incidents discussed that pertained to inquiry demonstrating (see Tables 11 and 12). In addit i on, the number s continued to fall short of the number of inquiry incidents di scussed by the experienced demonstrators. Within this category , the experienced subject s recognized and discussed more critical i ncidents than pos t-works hop novi ces on the importance of (1) having s tudents 204 on (2) make bas i c site observations duri ng a demonstration and mes. The avoidi ng forecasting demonstration observations and outco of forecasting was addressed i n part i cipants' first wor k shop "error " (Appendix H, Section II I -B) and by at least one workshop handout instructor during the first staf f presentation. "If you forecast ahead of time what is going to happen , you set yourself up for some In fancy footwork when it (the demonstration) doesn't work that well. general , you want to use a more investigative approach. " The critical- incident discourses of post-workshop novices continued to stress the importance of having students attempt to explain or generate hypotheses about the phenomenon demonstrated. The following quotes illustrate this emphasis. (Others are provided in Appendix S) . N3 : O. K. , again it would have been real tempting for him (the videot aped teacher ) to explain to the audience what might have happened [to can ] but he is doing a good job of l eading questions , "Can someone explain what's happening? " ( VT #1 , S : 3 3) Nl : o.K. Good questioning ["What f orce did you use to c r ush that can?" "Do you think that there was possibly a for ce res i sting your hand? "] . He's leading them exactly down the t r ack that he wants them to go, as far as the way he i s asking the question. He is l ooking for specific answers [to how the can was crushed) . (VT #1 , S: 2 0) No noticeable change occurred in the frequency or nature of novices ' discourses regarding the forecasting of demonstration procedures , observations , and explanations . Questioning Strategi~ Novices showed no change in the number of critical incidents di' scussed ? ? t the identification of leading and poorly pertaining o phrased content questions, or to the recognition of the value of 205 x erience demonstrators gave probing students' prior knowledge. E p ? d erview (see these issues particular emphasis during the clinical 1.'nt , pages 139-142). A content analysis , however , did show that the post-workshop novices evaluated a few more critical incidents than pre-workshop novices regarding subject-matter questions asked by the c 1 arity videotaped demonstrators. These critiques focused on question , quo es and appropriateness for middle school students . The followi' ng t exemplify these concerns. Nl : Of course , the way he phrases that question, [ "That the weight of the air , the pressure of the air in the room was greater after I put it in the water?" (S:62 ) ] , really sort of throws them (the students ) off because , obviously, the weight of the air pressure in the room has not changed. (VT #1 , S:64 ) N8: This is sort of a bad question. " [From your observations , what kind of observations can you make involving this cylinder? ] . .. It has something to do with density ." That ' s his clue. I don't know how he can improve it but I think he needs to come up with better questions. Those questions are not something that are going to stimulate a student ' s mind. By now half of my kids would be asleep . (VT # 3, S: 1 7) New Terms Novice demonstrators showed little change in the number of critical incidents discussed regarding attention to new chemistry terms (Tabl e 11 ) . A componential analysis revealed little qualitative change , as well . The workshop did not give special attention to the effective use of new science terms when conducting a chemical demonstration (Appendix G). Quality of Explan~ After the workshop , novices discussed more critical incidents related to the accuracy , clarity , and usefulness of concept 206 The following explanations provided by the videotaped demonstrator. The quotes pertain to a two quotes illustrate this discourse focus. generalization critical incident involving an inappropriate density presented to students by the videotaped teacher. N'.: , Right there, [ ',' So~ids generally are more dense than 1 hquids. Not all liquids have , the same density . " Is: 91 I I didn't - someone could have said, "Well, what about ice'"' d that is just such an obvious everyday example. And that' an really going to confuse kids. I think he just - I don't s think he should make a generalization like that. I think h shoul d just say, you know - That just wasn ' t necessary e (VT t3 , S:92) . NB: Where he starts talking about solids being more dense than liquids, he should have had some examples of some solid that are not as dense as liquids and put them in the middle s layers like we have seen. Put them in the interfaces of the liquids . That would spring more conversation, I think. (VT #3 , S:92 ) Post-workshop novices also focused more on critical incidents involving concept explanations, such as the applications of the concept to students' lives and the effective use of science t erms when presenting a chemical demonstration. (Additional quotes may be found in Appendix S) . N7 (Application) : What he's doing here I think i s good. ["Have you ever had the experience of going out in a boat with an outboard motor and either filling or refilling or watching someone fill the engine with gasoline and the gasoline run off the engine into the water? (S:76)] . He ' s bringing in everyday aspects into w~at density has to do with that [colwnn of liquids], so that kinds of keys them a little bit as to why they're observing what they ' re observing. They've probably seen that [phenomenon] before but they ' r e not aware of it [relating to density]. (VT B, S:7 6) Nl (Weight/mass terminology): She is going to confuse them a little bit because she is using the 'weight ' instead of using the word 'mass'. She tri~d to c~eac that up in the , beginning now she is fallen~ little ~it into the wrong lines as far as using that- That is something she would have to watch in the future. I think that is a pretty common mistake. (VT #4 , s:22) d 207 Given this discourse focus , novices discussed more critical incidents re l ated to concept explanations in their post-workshop s i see also interviews (p ~. 1875 , using a Wilcoxon non-parametric te t Table 11 ). Appendix M indicates that , during the demonstration air workshop , novice demonstrators heard the concepts of density and pressure addressed and explained several times by workshop instructors and workshop participants . Interactive , Participat~ Little qualitative change was observed between pre- and post-workshop novices in their critical-stop discourses on the use of student participants during a chemical demonstration . Before and after the workshop , novices elicited comments very similar to those provided by experienced chemica l demonstrators , albeit , fewer in number (Table 11, category 5)- post-workshop novices identified a f ew more critical incidents involving student-teacher and student -student di alog , a feature frequently discussed by experienced demonstrators . Overall , however, the number of think-aloud discourses novices provided in this category did not appear to be greatly influenced by the workshop intervention even though the topic of effective use of student participants was one issue addressed during the firs t s t af f presentation "What Makes an Effective Demo?" (Appendix G and Appendix H, section III-Bl- This teaching strategy was also mode l ed in severa l staff and participant demonstrations . Mechanics of the oemonstration Little change occurred in terms of novices identifying and disc ? , t, ? ncidents related to the mechanics of the ussing cri ica 1 i 208 ssues such as visibility and safety were demonstration (Table 11). r . emp asis before and after the workshop. discussed with about equal h ? emonstrators yet dd One issue important to the experienced d ' rarely , u e verification of a ressed by pre- or post - workshop novices incl d d . , see PP? 147-148). procedural steps taken prior to the demonstration ( e mechanics of Novices, however , did become more cognizant of how th , earning among the demonstration could hinder or promote concept l , middle school students. Use of Blackboard/Visual Aids After the workshop , discourses on the importance of having blackboard I physical drawings of complex demonstration systems on the The or other visual aids, became more evident among novices. during following quotes come from two novices who addressed this issue This issue was not addressed in the ir the post-workshop interview. pre-workshop interviews. N4: I think before he went on , it would have been a really good time for some diagrams on the board and to go into it a little more where a kid kept repeating that , "Air pressure." "Air pressure [ caused the aluminum can to collapse] . " (VT #1, S:60) N3: There would have been - if she had a diagram [of the density column] for instance, and had a ki~ come up and label Corn syrup where it was, and water, and mrneral oil - it [would hav~ l been clearer- I think that the students are probably confused at this point. I'm confused. I mean I know what should be there - I'm pretty sure? (VT #4, S: 88) The effective use of the blackboard and overhead illustrations during a chemical demonstration was frequently modeled during the workshop by both workshop instructors and workshop participants . This issue was briefly discussed during the first staff presentation 209 terms of (Appendix G) and mentioned in the first workshop handout i'n , ection III-A and B). presenting a data collection table (Appendix H s Overall Organization After the workshop, novices began identifying and discussing a icient few more critical incidents related to the effective and eff' , sequencing of demonstration tasks, an important organizational issue The discussed by experienced chemical demonstrators (seep. 149) . following quotes illustrate novices ' post-workshop concerns for such (Other quotes may be found in Appendix S). issues . N7 (Materials readily available): O.K., later on he doe s it , but he should have had the white background immediately because I couldn ' t tell [what was in the density column ) . He said , "There's three different shades , " and even seeing it on film I couldn ' t tell there were three different shade s unt i l he put the white paper behind it . (VT #3 , S:17) N2 (Sequence in constructing a column ): I ' m glad she did it in that order so that students could see the displacement. I think that is very good because some people just build the column from the bottom up . And uh, I did that; I added the syrup after I had done everything else and watched it go to where Ifwent [through the liquids to the bottom). And I think that ' s fun cause it's good for the kids to see that. (VT #4 , S: 69) Presentation Style In terms of presentation style , the most apparent change observed in novices' pre-/post-workshop critical- stop discourses involved more frequent reference to humor in a demonstration. This issue wa s al so discussed by experienced demonstrators during the critical-stop t ask (p . 150 ) . N3 : He has done an excellent job of getting his ~udience in the palm of his hand. He has a good_humorous begrnnrng, . . . giving enough information to the audrence , to know w~Y he selected Ed [as a student volunteer ] . Its a good rntro. It really is . A good calm approach. (VT #1, S: 9I d 210 N2: He's making good use of humor [with the title of his demo, "The Can-Man Can" lab). (VT #1 , s: 3) e workshop staff An examination of the descriptive summaries of th r was the f ocus presentations (Appendix G) shows that attention to humo of one of the scheduled staff presentations (see also workshop Schedule , Appendix F). Other issues related to presentation style, such as non-verbal ge s tures, courtesy , and rapport were addressed equally during the pre - and post-workshop interviews. Changes in Novices' Knowledge of Alternatives The critical-stop task and semi-structured interview promp ted both experienced and novice chemical demonstrators to discuss ways of improving critical incidents judged as a "weakness ", or hindrance , to overall performance. These suggestions , called pedagogical enhancements (Pl, coded into t he s ame nine pedagogical categor i es identified for discourse s in the descriptive domain , i. e ., Categories 1-9. Table 13 shows that novices made about the same number of suggestions for enhancing the videotaped demonstrations before and after receiving the workshop intervention . Post -workshop novices made a few more suggestions for improving the mechanics of t he demons tration (Category 61 and slightly fewer suggestions for improving inquiry (Category 11, an interactive , participatory style (Category 51, use of the blackboard/visual aids (Cat egory 71, and organizational issues (Category 81 (see Table 14 , P. 153 1 . After viewing videotape 13, post -workshop novices made several suggestions regarding ways in which the videotaped demonst rator could 211 s ra i on. These iml:?:cove the visibility of the Density Column demon t t ? ' a light board suggestions included using larger graduated cylinders aw i e background , and or light box , a display stand for the column , h't se suggestions were also food dyes in place of phenolphthalein. The The is sue of visibility evident in novices' post-workshop discourses. n stressed was one that was consistently modeled during the workshop ad by workshop instructors (Appendix H, Section III-Band C). The following is representative of a post-workshop novice ' s discourse that stressed demonstration visibility. (Other quotes may be found in Appendix U. ) NS: At this point I think he could have used a larger cylinder and he could have used his lighting set which he had up there on the table which would have lit it up. Maybe a white background instead of having to stand there and everything . I think what he did do worked . Everybody saw what he was trying to go after but it just could have been a little cleaner. A little more visual. (VT 13, s,15) Variations on the Observed Demonstration The data presented in Tables 18 and 19 show that novices were able to discuss more variations (V) on the Density Column demonstration and collapsing Aluminum can demonstration after having participated in the chemical demonstration workshop IP < .O S, using the independent t-test and more conservative non-parametric u- test) . The variations discussed represented both simplified and more advanced presentations of the videotaped chemical demonstrations. The following quote represents a variation on the oensity Column demonstration discussed by one post-workshop novice. (Others examples are provided in Appendix U) ? 212 Table 18 Frequency of AIR PRESSURE Demonstrations Cited by Pre - and Post - Workshop Novice Chemical Demonstrators Pre-Workshop Post-Workshop Type of Novices Novices Demonstration t - ratio Citation Mea n SD Mean SD Variation (V ) 0.5 0. 8 1. 6 1. 2 2 . 26 * Demonstration (D) 0.8 1.2 2 . 9 1. 4 2.99 ** Demonstration 0.0 0. 0 0.0 0.0 a Variation (DV ) Extraneous Examples , 0.6 0.7 0. 9 1.0 0.57 Total (XTOT ) Extraneous Examples , 0.1 0.4 0. 4 0 . 5 1.13 Positive (XPOS) Extraneous 0.4 0.5 0. 5 0.8 0.39 Negative (XNEG) Notes. -----a--yndicates scores are significantly different from experi enced demonstrators at the p < .01 level. *p < .05; **p < .01 NS (Density Column variation ): That column can be enhanced. You can do a lot with relative densities given a bunch of knowns in the column and then peg an unknown with your knowns. That is the main variation on that one that I've have seen. If you peg the known densities of the liquids then he could start dropping unknowns in that would float or sink to various levels and people could peg the densit i es of those and maybe even a handout with den s ities on it that would match up with various knowns. Get people thinking about how density could be use to single out an unknown. (VT #3, I. 8) A similar pattern showing novices providing more fr equent discussion on variations of the Collapsing Aluminum Can wer e a l so observed during the post-workshop interviews. The following qu ot es 213 Table 19 P Frequency of DENSITY Demonstrations Cemito ndsbtr ators re- and Post-Workshop Novice Chemical D e Y Post -Workshop Pre-Workshop Novices Novices Type of ---------- ------- ------- t-ratio Demons t ration SD Mean SD Mean Ci tat i on 2 0.7 a 1. 2 .41 * 0 . 4 0.7 Variation (V) 4.0 1. 5 4.50 ** 1. 2 1. 0 Demonst r ation (D) 0. 8 1. 87 o.o o.o 0.5 Demonstration Variation (DV) 1.1 1. 34 1. 8 1. 2 1. 0 Extraneous Examples , Total (XTOT) 0.5 0. 5 1.18 1. 0 1.1 Extraneous Examples , Positive (XPOS ) 0.8 0. 94 0.9 0. 8 0.5 Extraneous Negative (XNEG) -Noate-s.D ifference between exper i enced demonstrators and post- wo r kshop novices is significant at the p < .001 level. *p < .05; **P < .001; represent the demonstration variations discussed by one novice , N7. (Quotes from other post-workshop novices are provided in Appendix u. J N7 (Vacuum pump variation), And the one where you take what the vacuum - you cr eate the vacuum on a vacuum pump and you crush the can. (VT #3, I. 8) N7 (Ditto Fluid can variation ), Or where you just heat it up and cap it and then watch it crush s l owly through the hour. (VT #3 , I. 8) N7 (No water in Al can variation), variation or twist on that one . well, what they (some workshop participant s ) did with one (can ) is [perform the demonstration ] without water inside it to determine the differences, if it wa s really the water that was making a difference; and it does to a point. (VT #3 , I. 8) 214 N7 (Different Types of Cans variation ) : That is the only variation - or you could use different types of cans to see if maybe - you could even run that into an experiment. You could give students two types of cans . I thought this was cute. I started thinking about this. You could take a Pepsi can and a Coke can and you could say , you know , "Americas choice. Which one do you think is better? " and you could use it as a demo. You coul d do that. You could take a vote of the class and see which one is going to be stronger. And you could maybe find a thicker heavier aluminum can or a real light one and see if it makes a difference as to , you know; and you time everything so it ' s exactly the same. So you could turn it into a more of an experiment than it is. (VT #3, I. 8) A few variations of the Collapsing Aluminum Can demons tration were publicl y performed by workshop participants . Others wer e discussed in the workshop sourcebook (Appendix K; Sarquis & Sarqui s , 1987) and by the workshop staff. These training component s appeared to influence novices' post-workshop response to the interview ques ti on regarding demonstration variations (Appendix I , Question I . 8). Alternative Chemical Demonstrations and Extraneous Exampl es Similarly , novices were also able to discuss more alternative chemical demonstrations (D) on air pressure and density as a r esult of the chemical demonstration workshop experience (p < .0 1 and . 001, for the two concepts , respectively ) (Tables 18 and 19). Prior to the workshop , novices generally discussed one other alternative chemica l demonstration on the targeted concepts. After the workshop, they could discuss and critique three or four alternative chemical demons tra t ions on each of the targeted concepts . The foll owing quotes i llustrate the new pedagogical content knowledge acquired by s ubjec t NS on the concept of air pressure. Prior to the workshop, novi ce demonstrator NS provided one pedagogically unsound chemi ca l demonstration on air pressure . After the workshop , thi s indivi dual 215 that suitably freely discussed five different demonstrations Hi s post-workshop discourses illustrated the concept of air pressure. other novice demonstrat ors , (Additional quotes by are quoted below. Nl-N8 , are provided in Appendix U. A complete list of the chemical experienced demonstrations and demonstration variations discussed by and novice subjects is provided in Appendices v and W). NS (Ball o~n Inverted in," Flask) : Balloons, though - all kinds of air pressure things you can do with balloons. Like the one that John did where he heated up the flask and blew the balloon up and then he cooled the flask down and the balloon got sucked into the flask . That was kind of neat And he showed both the effects of heat and the spreading~ t of molecules and the contraction will cool them off again u (VT J2 , I. 81 . N5 (Egg in the Bottle): The egg being pulled into a flask. The egg in the bottle is a classic. (VT #2, r . 8) NS (Vacuum Chamber ) : The balloon and the belljar and the vacuum chamber was a reverse of what was going on there which I thought was kind of neat. God, there is just a zillion . r mean you can go on and on with those. (VT #2, I . 8) N5 (Cartesian Diver): The Cartesian diver for the density. I forgot about that one. That was a good one. Also , the Cartesian diver fits in with this [i.e ., air pressure) to a certain extent , although , I probably ~ouldn ' t use it for that just because that is kind of a confusing model. A tough one . (VT #2 , I . 8) N5 (Not on the 1eve1/u-tube Barometer): Well , the u-tube is another fairly classic one that shows ~ir_pre~sure. You have that u-shaped tube and if you have a liquid wit? al\ the same density it remains at the same ;evel and :h~re_is air pressure on both sides maintaining an equilibrium on the twc ends Then you can do things with that . You can cap off one :f nt to and do some stuff with the other end . I i you wa , d ? -h ? denond ' t know. There is a bunch of things you c~n o wit it . I am trying to think of how you can tie that in exactly though. I suppose you could hook one end up to a hand pump, vacuum pump to show what happens when you stadrt hdeclr~a s7dng, ? n one side of the u-tube an t e iqui is the ? air pressure ? oto the vacuum tube . You are goi?n g t o h ave going to go up in to kee the water from getting up in to figure out som~ way Ph' k of off the 8 top of my 1 your pump . That is all I cant in 2 head . There is a bunch though. (VT * ' I. 216 After participating in the intensive chemical demonstration (XPOS and workshop , novices also discussed fewer extraneous examples (see Tabl es XNEG) of how to demonstrate the targeted chemical concepts When thi s researcher probed novices for additional 18 and 19). demonstrations and variations, they made fewer references to laboratory and seatwork activities , and fewer references to clas s discussions of natural phenomena invol ving air pressure and density I (XPOS). These novices also made fewer references to nebulous erroneous, and pedagogically unsound demonstrations r e lated to the targeted chemical concepts (XNEG I, although, a few persisted (see Appendix u for examples). In general, they provided legitimat e alternative chemical demonstrations on the targeted concept s (D). An example of a pedagogically unsound/erroneous air pressure demonstration that persisted after the workshop with at l east one novice subject, included the following. N2 (Inflating a Balloon) , You canals~ il'.ustrat~ it (air pressure ) by heating a balloon and blowing it up with (a heated Erlenmeyer) flask. That is better. (VT 11 , I. 81 This demonstration actually illustrates the thermal expans i on of gases much better than it illustrates the influence of ambient ai r pressure on flexible objects such as balloons or aluminum cans . Although the suggested demonstration does deal with air pressure in some f ( , ease in the internal air pressure of the orm e.g., an 1ncr balloon through heating ! , it is somewhat removed from the objective demonstrated on videotape which the novices were asked to address . Although such misrepresentations were not directly addres sed during somewhat l es s th the workshop by workshop instructors , ey were prevalent in novices' post-workshoP protocols- - - 217 The da::a in Table 20 summarizes the number of variations , alternative demonstrations , and extraneous demonstrations on dens i t y and air pres sure discussed by experienced chemical demonstrators and pre- and post-workshop novices during the clinical interviews . The table s hows that the intensive workshop intervention had a definite impact on helping novice chemical demonstrators increase the number of chemical demonstrations they coul d discuss on concepts tha t are bas i c to chemistry and the physical sciences (p < .01 ) . Although the number of demonstrations discussed by novices increased substantially , these Table 20 Frequency of AIR PRESSURE.and DEN~ITY Demonstration s Cited by Experienced and Novice Chemical Demonstrators Pre-Works hop Post-Workshop Experi enced Type of Novices Novices Demons tra tors Demonstration - -------- - ---- Citation Mean SD Mean SD pa Mean SD pa 0. 9 2. 8 1.1 ** 5. 2 1. 7 Variation (V) 1. 0 *** 2. 7 *** 10.4 4. 2 ** Demonstration (D ) 2 . 0 1. 6 6.9 o.o 0. 5 0.8 3.0 2.3 * Demonstration 0 . 0 Variation (DV) 0.4 0. 6 * Extraneous 2 . 4 1. 8 1. 9 2.3 Examples , Total (XTOT ) 1. 0 1. 0 0. 2 0.4 * Extraneous 1. 4 1. 2 Negative (XNEG ) 0.9 1. 8 0. 2 0.4 Extraneous 1. 0 1.1 Positive (XPOS ) Not es . pre -workshop novice scores . --a- All probabilitie s in c. ompari*so**np t .187 ) . Prior to the workshop , 71 .7 % of the critical incidents discussed related to pedagogical content knowledge issues . Aft e r the workshop , PCK emphasis increased to 76.0 % of the critical incident s discussed. Discus sion The findings of this study reveal several quantitative and qualitative difference s between experienced and novice chemical 237 demonstrators in terms of their conceptual izations of effective chemical demonstration teaching. The findings further document know l edge growth in teaching among novice demonstrators receiving i ntensive inservicing . These f i ndings support several recent studies contrasting experienced and novice teachers' t hinking about teaching , students , and classroom events (reviewed by Calderhead , 1987; and by Clark & Peterson , 1986 ). They also add to the current r esearch literature by providing data on the nature of science teache rs' pedagogical content knowledge growth in an inservice setting . Finally , this study contributes to the growing body of literature examining the nature and development of pedagogical experti se (see Berliner , 1986 , 1988; Carter et al ., 1987 ; Leinhardt & Greeno, 1986 ). The discus s ion that follows presents a sysnthesis and interpretation of the data obtained in this study. The discussion centers around five distinct areas that highlight the various characteristics of experienced chemical demonstrators' pedagogical content knowledge and factors that contribute to its growth among novices . The five areas of discussion center around demonstrators ' (1 ) skill at critiquing content , (2) knowledge of concep t -specific demonstrations , (3) knowledge of inquiry demonstrating, (4) knowledge of demonstration variations, and (5) skill at processing demonstrati on teaching information . Experi enced and Novice Demonstrator Diffe r e nces In the di scuss i on that follows, all references to novices are to pre-workshop novices unless otherwise indicated. 238 Experienced and Novice Demonstrators' Skill at Critiqing Content A componential analysis of the protocols revealed that experienced chemical demonstrators were considerably more skilled than novices at identifying erroneous and misleading propositions presented in the four videotapes. Shulman (1987) calls this skill "critical interpretation" of content. According to Shulman, this skill represents an important aspect of a teachers' PCK. The findings of this study suggests that this ability appears to be better developed among the experienced demonstrators than among the novices . When teachers prepare a given topic for instruction , they must carefully scrutinize the teaching material and determine whether it i s "fit to be taught." This process of critical interpretation includes (1) detecting and correcting errors of omission and commission in the curriculum texts, and (2) structuring and segmenting the text materials into forms more suitable for teaching. Although the demonstrators in this study were not given textual material to simulate this transformation activity , they performed a s imilar activity (the critical-stop task) in which they critiqued the t eachi ng and content presentation of the videotaped demonstrators . A close examination of the protocols showed that it was primar ily the experienced high school chemical demonstrators, not the experienced middle school chemical demonstrators, that detected errors and misleading propositions presented by the videotaped t eache r s . This difference in error detection can be accounted for, in part , by the prior academic training these demonstrators received . The experienced high school demonstrators all had college degrees in chemistry , whereas the experienced middle school science t eachers 239 completed two to seven semesters of college level chemistry (see Appendix C). These findings suggest that the very first s tep in the transformation process, critical interpretation, requires considerable knowledge of subject matter in order to identify content-matter statements that are erroneous or misleading and thus interfere with the l earning of the targeted concept. In general , this detection skill was evident among experienced chemical demonstrators and , according to Shulman (1 987 ), contributes to their pedagogical content knowledge by informing them of the content "most fit to be taught." Novices did not examine the chemistry content presented in the videotapes as critically as the experienced demonstrators. Evidence for this claim comes primarily from what novices did not say during the critical-stop task , rather than from what they did say. For the most part, novices did not detect or discuss any of the knowledge inaccuracies mentioned dur ing the videotaped presentations . Structuring and segmenting constitutes another component of critical interpretation. Several critical-stop discourses , particularly those that coded as organizational issues (Table 10) , reflected knowledge associated with this transformation subprocess . In this regard , the experienced chemical demonstrators readily identified and discussed critical incidents related to the sequencing of concepts and ideas in order to minimize student confusion about density and air pressure and to maximize the time available for instruction (see quotes provided on page 149 - 150). For the experienced subjects , the issue of structuring and segmenting ex t ended beyond content matter organization , as discussed in Shulman' s model, to include structuring and segmenting the manipulative ta s ks 240 associated with the demonstration presentation. Furthermore, it included the interaction of structuring subject matter content with the manipulative tasks of the demonstration. The experienced chemical demonstrators considered the structuring and segmenting of content and handling of equipment as an important contributor to effective chemical demonstrating. Novices' primary concern regarding the structure and sequence of content during a demonstration involved the issues of pacing, closure , and transitions . Novices did not identify and discuss critical issues related to the sequencing of key ideas and demonstration tasks to make the most of the limited time available for instruction. These findings, when contrasted with the discourse patterns observed among experienced demonstrators, suggest that novice demonstrators could continue to grow in their pedagogical content knowledge of how to optimally structure and sequence the content and demonstration tasks associated with chemical demonstrations. Experienced and Novice Demonstrators' Knowlege of Concept-Specific Demonstrations The verbal responses obtained from experienced chemical demonstrators revealed that these demonstrators possess a large body of knowledge on various chemical demonstrations that could illustrate the concepts of density and air pressure to middle school students . These demonstrators could recall several relevant demonstrations from memory and discuss them in detail . According to Shulman's (1 98 7) model of pedagogical reasoning , these findings suggest that experienced chemical demonstrators possess multiple mental represen t a t ?ons for demonstrating basic chemical concepts This i ? 241 multiplicity in r epresenting basic science concepts reveals a particularly distinctive characteristic of experienced chemical demonstrators ' pedagogical content knowl edge . Given that density and air pressure represent only a smal l sampling of the concepts cove r ed in chemistry (Appendix E) and that demonstration teaching represents only one of several ways of rep r esenting chemical concepts , the findings of this study suggest that experienced chemical demonstrators possess what Shulman (1986 , p . 9) ca l ls "a veritable armamentarium of alternative forms of representation ." By contrast , novice chemical demonstrators lacked the breadth and depth of pedagogical content knowledge exhibited by the experienced demonstrators of how to demonstrate basic chemical concept s to middle school students. It was apparent from the clinical interview discourses that novices could only think of and describe a small number of chemical demonstrations that would help students unde r s t and the concepts of density and air pressure. According to Shulman ' s model of pedagogical reasoning , this indicates that novices possess a limited representational repertoire for demonstrating basic chemical concepts at t he middl e school l evel. This finding is cons i stent with the self-report data gathered on these demonstrators who indicated they do not frequently use chemical demonstrations in their own science teaching (Appendix C). Along with the representations that experienced demonstra tors hold for demonstrating a given chemical concept , they posses an understanding of which representations, e.g ., chemical demonstrations , are most suitable as concept introductions at the middle school l evel. During the clinical inte rvi ew, the experienced chemical demonstrators 242 discussed the instructional value (i .e., the benefits and disadvan- tages ) of the various representations without the prompting of the researcher . For these demonstrat ors this unso l icited information was integral to the ir evaluation of the videotaped demonstrati on presentations . Knowledge of which chemical demonstrations are most instructive for teaching a given concept and their placement in a given unit plan represents two other important characteristics of experienced chemica l demonstrators ' pedagogical content knowledge . Novices , on the other hand , made very little reference to the appropriateness of the videotaped demonstrations for ins tructi on at the middle school level. One novice who did verbalize discontent with the videotaped density demonstration suggested a density lab in place of the Density Column demonstration as a more suitable alternative for transforming the concept of density into a form students could understand. This response is not surprising given that many of the novices were probably more familiar with dens ity labs than with density demonstrations (see Demonstration tallies in Table 16). Another characteristic of novices' pedagogical content knowledge is that they often made reference to erroneous and pedagogically unsound chemical demonstrations for t eaching density and air pressure . Some of the demonstrations novices discussed as suitable illust rati ons of the targeted chemical concepts were actually more appropriate for demonstrating related chemical concepts and principles , such as Archemedes Principle (water displacement ), sedimentation rates , and Bernoulli's Principles (air flow). It appeared that some of the representations novices possessed we r e not optimally associated with the concept of inte rest . These findings further support the notion 243 that novices posses a limited and ill-developed representational r epe rtoire for demonstrating basic chemical concepts . Experienced and Novice Demonstrators ' Knowledge of Inquiry Demonstrating During the clini cal interview , t he experi enced chemical demonstrators made frequent mention of the importance of incorporating inquiry methods of ins truction i nto the chemical demonstration. They recognized and discussed numerous crit i ca l incidents where the videotaped teacher hindered inquiry by forecasting the procedures , outcomes , and explanations of the phenomenon demonstrat ed . These comments ref l ect another aspect of content transformation called instructional selection (Shulman , 1987 ). During instructional selection , a teachers ' representation (e .g ., the Density Column demonstration ) is delivered to students with respect to some teaching s trategy , s uch as lecture , Socratic dialog , discovery l earning , si l ent presentation, seatwork , etc . In this study , the experienced chemical demonstrators frequently discussed i ssues r e lated to instructional se l ection in terms of us ing inquiry methods of instruction when conducting chemical demonstrat i ons . These inqui ry di scourses fu rther depict the nature of exper i enced chemical demonstrator ' s pedagogi cal content knowledge for demonstra t ing basic chemical concepts . During the cl inica l interview , novices made no direct reference to the importance of using inquiry methods of instruction in a chemical demonstration . They also made l ittle r efe rence to some of the important components of inquiry instruction , such as , having s t udents provide support ive evidence and conclusions during a demonstration (Trowbridge & Bybee , 1986 ). Novices did , however , make 244 reference to sel ected character istics associated with good inquiry instruction , such as having students become involved in observing , measuring, predicting, descr ibing, and inferring . Novices were al so reasonably cognizant of how forecasting interferes with inquiry instruction, although novi ces discussed it in fewer critical incident s than did the experienced demonstrators. These findings , collectively, suggest that novices possess adequate knowl edge of instructional selection (Shulman , 1987 ) in terms of presenting a chemical demonstration as a student - centered, inquiry-oriented presentation . However , if experienced demonstrators are used to establish desired standards of pedagogical knowledge and pedagogical reasoning , improvements in novices ' conceptualization of inquiry chemical demonstrating would be appropriate. These findings , when interpreted i n terms of schema theory (Rumelhart & Norman , 1978), suggest that experienced demonstrators' schemata for inquiry teaching and their schemata for demonstration teaching appear to be better developed and interconnected than those of novice demonstrators. By better developed, we mean that the schemata of experienced demonstrators for inquiry teaching and f or chemical demonstrating contain more detail and display more detail when verbally expressed. By interconnected we mean that the networks of knowledge associated with inquiry teaching and chemical demonstrating are more interrelated and that these relationships are more frequently and fluently expressed in verbal discourse among experienced subjects . Other studies , particularly those that have examined teachers ' thinking and decision-making among experienced and novi, ce t eac h ers, have noticed similar differences in teachers' 245 cognitive structures for teaching (Borko & Livingston, 1988; Leinhardt & Greeno, 1986; Peterson & Comeaux, 1987). Experienced and Novice Demonstrators' Knowledge of Demonstration Variations Experienced demonstrators were also quite knowledgeable of chemical demonstration variations. Their discourses revealed that the Density Column demonstration, the Collapsing Aluminum can demonstration, and several other density and air pressure demonstrations could be performed in a variety of different ways. These variations usually involved adjustments in the kinds and number of materials that could be used in a demonstration and the sequence in which these materials could be handled or introduced to student s. These findings suggest that experienced chemical demonstrators are quite capable of adaptation , i.e., fitting a given chemical 1987 demonstration to the characteristics of the learner (Shulman , ). This claim is based on the reasoning that knowledge of demonstration variations provides experienced demonstrators with seve r al options f or adapting a demonstration to student characteristics (e.g., their abilities and motivations). Further evidence of experienced demonstrators' knowledge of adaptations emerged when they discussed the complexity of the In their think- aloud di scou r ses , videotaped chemical demonstrations. identified sources of complex ity the experienced demonstrators often in a demonstration and discussed ways of modifying a given demonstration to minimize student confusion and learning diffi culti es . Experienced demonstrators usually dealt with complex demons trations by suggesting s impler variations of the demonstration that could he lp 246 students focus on the targeted concept rather than on ex traneous or unrelated concepts . Shulman ' s mode l of pedagogical reasoning considers knowledge of what makes the l earning of specific topics easy or difficult another aspect of adaptation. This feature reflects another dimension of experienced chemical demonstrators' pedagogical content knowledge (PCK ). Novices , by contrast , showed l imited understanding of adaptation. The verbal reports indicate that novices possessed minimal knowl edge of chemica l demonstration variations (Tables 15 and 16) , an indicator of their adaptational repertoire. Furthermore , they rarely recogniz ed the complexity of the videotaped chemical demonstrations and situat i ons where adaptation would be necessary, such as with concept introductions. Collectively, critical interpretation , representation , selection , and adaptation represent four actions that permit teachers to transform subject matter content into a form that students can comprehend . The protocols of the experienced chemical demonstra tors showed considerably greater evidence for each of these four actions , than the protocols of novice demonstrators . These actions characterize some of the distinctive commonalities and diffe rences in experienced and novice chemical demonstrators ' pedagogical conte nt knowledge. Experienced and Novice Demonstrators' General Pedagogical Knowl edge The critical - s top discourses of experienced and novice demonstrators included discussions of generic teaching i ss ues that contribute to an ef fective chemical demonstration prese ntation . These verbalizations r ef l ected demonstrators ' general pedagogica l knowledge , 247 GPK , (Shulman, 1987) . This knowledge consisted of an awareness of the importance of integrating an array of generic teaching skills , such as encouraging student participation, promoting classroom discourse, determining prior knowledge, using advanced organizers, providing immediate feedback, using different levels of difficulty during questioning , using wait time , humor , and enthusiasm, and various classroom management skills when presenting a chemical demonstration to middle school students. Although both experienced and novice demonstrators discus sed many of these generic teaching issues during the critica l -stop task, novices discussed them less frequently than experienced demonst rat ors. Experienced and Novice Demonstrators' Skill at Processing Pedagogical Information During the critical-stop task , experienced and novice chemical demonstrators discussed many critical incident s that reflected their pedagogical content knowledge and general pedagogical knowledge of chemical demonstration t eaching . A taxonomic and componential analysis of these discourses revealed that the two groups of demonstrators critiqued a variety of different critical incidents. This patte rn of vari ed response was also observed by Carter et al . (1 988) in a similar think-aloud classroom analysis task. They, too, noticed that individual variance in response to visual mater i a l s (slides ) presented to experienced and novice teachers sometimes appeared as great within groups as it was between groups . In the present study , the variance phenomenon can be accounted for , in part , by the fact that within a typical 7-minute demonstration teaching videotape, experienced demonst rators collectively noti ced as many as 248 70 different critical incidents associated with effective and ineffective chemical demonstrating. This suggests that the variance observed may be partially attributed to the l arge number of critical incidents the demonstrators could choose to process and critique during the critical - stop task. This response variance , and the large number of cr i tical incidents identified for the relatively brief videotaped demonstrations , lends support to Leinhardt and Greeno's (1986 ) notion that teaching , in genera l, is a cognitively complex task. With respect to chemical demonstration teaching, this complexity is suggested by the large number of critical teaching behaviors that chemical demonstrators need to organize and attend to within a limited span of time. In spite of the variance in the data , the frequency with which the experienced chemical demonstrators discussed critical incident s observed on videotape suggests that they were able to process considerably more information pertaining to chemical demonstration teaching than novices. This information processing ability also suggests that experienced chemical demonstrators possess a well-defined knowledge structure or schemata for chemical demonstration teaching that is less-defined among novices (Berliner , 1987a; carter et al, 1987; Gage & Berliner, 1984, pp . 317- 319; Leinhardt , 1983; Leinhardt & Greeno, 1986; Rumelhart & Norman, 197B). In summary , the verbal data can be interpreted to indicate tha t the experienced subjects had a much greater representational repertoire for demonstrating basic chemical concepts than novices . The experienced demonstrators also possessed greater knowledge and skill at critical interpretation , instructional selection, and at 249 adapting chemical demonstrations to match the level of the learner. These components of the transformation process indicate that experienced chemical demonstrators ' pedagogical content knowledge and pedagogical reasoning (Shulman, 1986, 1987) is quantitatively greate r, qualitatively richer in detail , and better integrated with othe r knowledge systems than that of novices. Pedagogical Content Knowledge Growth Through Intensive Inservicing Intensive inservicing produced observable levels of pedagogica l content knowledge growth among science teachers identified as novice chemical demonstrators. This knowledge growth varied, however, with r espect to the form of pedagogical knowledge / skill examined (e .g., skill at critiquing content , knowledge of concept - specific demonstrations , knowledge of inquiry demonstrating , knowledge of demonstration variations). In some pedagogical knowledge areas , novices experienced substantial knowledge growth ; in other areas they experienced very little . Each area contributes to a t eache r' s pedagogical content knowledge . Changes in Novices ' Skill at Critiquing Content During the pre- and post-workshop clinical interviews , novi ces did not identify any of the knowledge inaccuracies present in the videotaped demonstrations . This error detection skill assoc i ated with teachers ' PCK , a skill Shulman (1987) calls critical interpretation of content , appeared to be unaltered by the workshop inte rvention as measured by the critical - stop task. The content er r or s presented to the novice demonstrators, however , were subtle in nature and we r e 250 presented very quickly during the videotape viewing task; thus, possibly eluding novices ' attention. Another process as sociated with the skill of critical interpretation of content is knowledge of how to structure and segment content-specific material for instructional purposes (Shulman, 1987) . This skill also showed little development among novices engaged in the critical - stop task. During the pre- and post-workshop interviews , novices discussed content organizational issues, such as concept introduction and closure, with similar emphasis. Unlike the expe rienced demonstrators, however , pre- and post-workshop novices di d not identify or discuss critical incidents pertaining to the organization of key ideas and demonstration tasks to aid lea r ning , minimize student confusion , and maximize the time available for instruction. Although the novice demonstrators were presented with chemist ry content whenever they observed the experienced demonstrat or s (i. e ., the workshop instructors) perform chemical demonstrations and whenever they observed peer presentations , the workshop did not include f ormal chemistry content lectures as a major component of its des ign. Instead , it sought to provide teachers with additional laboratory t i me to practice the nume rous boxed demonstrations . The fact that novi ces showed no measurable change in terms of critical evaluation skill s i s consistent with the stated goals of the workshop. Both the NSF proposal and ICE brochures (Bell, 1987; Appendix B) make no direct mention of providing ICE Wo r kshop B participant s with ski ll s t hat would allow them t o cr i tically examine chemi s t r y content; al though, it may have , neverthel ess , been a hidden or seconda ry goal of the 251 workshop. The design and duration of the workshop may not have permitted the inclusion of this content goal as a major focus of the workshop. Thus , this study indicates that the critical interpretati on dimension of pedagogical reasoning and pedagogical content knowled ge , as measured by the critical-stop task, showed no observable change among novice chemical demonstrators receiving the workshop intervention. Change s in Novices' Knowledge of Concept-Specific Demonstrations Novices showed an increase in the number of chemical demonstrations on density and air pressure they could discuss and critique as a result of having participated in the workshop. Thi s finding indicates that the workshop produced a significant (p < .Ol) increase in the representational repertoire of novice chemical demonstrators in terms of their knowledge of demonstrations that woul d illustrate basic chemical concepts to middle school students. Given the magnitude of the increase measured (about a 5- fold incr ease in the number of chemical demonstrations discussed, see Tables 15 and 16), and the number of concepts addressed during the workshop (over 30; see Appendix E), novices' knowledge growth in teaching basic chemi ca l concepts may be considered substantial as a result of participating i n the inservice workshop. These changes suggest that post - workshop novices began to develop a representational repertoire for demonstrating basic chemical concepts characteristic of experien ced chemical demonstrators . Cause and effect is difficult to attribute in a study s uch as this , but the increase in novices ' representational repertoire f or demonstrating specific chemical concepts may be attributable t o a 252 number of workshop components including: modeling, observing, reading, practice, and feedback on these concept demonstrations pe rformed by workshop instructors and workshop participants. After the inservice workshop, novices elicited slightly fewer erroneous and pedagogical unsound demonstrations on the targeted concept, density. The decrease, although small and not statistically significant as determined by two statistical tests (at - t est and a non- parametric U- test) , represents a 45 % drop in the number of uns ound concept representations discussed by novice chemical demons trators on this concept (see Table 19). The small sample size available (e .g., eight pre- and post-workshop novices) may account, in part, for the lack of statistically significant change in the number of pedagogically unsound demonstrations novices discussed during the pre- and post-workshop interviews on the concept, density. No change was observed in terms of the number of unsound demonstrations a few novices discussed on air pressure. The numbers cited for thi s concept, however, were initially small. To explain the decrease in the total number of extraneous examples discussed by novices during the clinical interview (see Tab l e 20 showing frequency tallies summed across both concept s , and s ummed across negative and pos itive examples), one could hypothes i ze that as novices gained confidence in conducting and performing vari ous densi t y and air pressure demons trat ions during the workshop, they r e lied l ess on brainstorming effort s and "survival responses" which previous l y lead to discussions of non- demonstration examples and pedagogi cally unsound demons trations on the target concepts. The workshop, it se l f , did not directly address novices' unsound r epresentations for 253 demonstrating specific chemical concepts. Their presence may reflect the existance of stable misrepresentations of how to demonstrate bas i c chemical concepts among novices. As a result of participating in the demonstration workshop , novices became more cognizant of (1) the val ue of using physical drawings of complex demonstration systems on the blackboard , or othe r visual aid , to supplement the demonstration presentation and (2) strategies for increasing the visibility of demonstration systems wh en presented in a large classroom setting . Acquisition of these teaching representations can be attributed to various workshop factors , particularly , the modeling of these representations and direct instruction of their usefulness during the scheduled staff presentations . Learning may have also been reinforced as novices observed these representations modeled in many participant demonstrations. After the workshop , novices discussed more critical incidents pertaining to the quality of the concept explanations provided by the videotaped teachers (Table 11). Novices , howeve r , did not provide more suggestions for pedagogically enhancing the unacceptable explanations they critiqued (Table 14 ) . These finding s sugges t t hat novices may have heard new r epresentations for explaining abstract chemical concepts to middle school students as a r esult of the workshop intervention and, therefore, could identify e ff ec tive and ineffective concept explanations more readily during the cri tica l -s t op task. These new concept explanations , however, may not have been sufficiently instantiated into t eachers ' existing schemata so as to assi s t them in providing more suitable concept explanat i ons wheneve r 254 poor explanations were identified and discussed during the think-aloud task . The pedagogical content knowledge growth that novices did experience in terms of acquiring propositional knowledge for concept explanations may be attributed to the opportunities novice demonstrators had to hear these concepts explained and demonstrated by workshop instructors and the more experienced workshop participants (see Appendix M, Checklist of Density and Air Pressure Demonstrations Observed or Performed by Novices During the Workshop). Changes in Novices' Knowledge of Inquiry Demonstrating The workshop also appeared to help novices become more cognizant of the value of inquiry teaching when conducting chemical demonstrations . This instructional strategy was specifically mentioned by several novice demonstrators during their post-workshop interviews . This increased awareness of the value of inquiry demonstrating may be attributed to (1) the frequent modeling of thi s strategy by workshop instructors , (2) the references made concerning this instructional strategy during the first staff presentation (see Appendix G), (3) the descriptions of effective chemical demonstrati on teaching provided in the first workshop handout (see Appendix H), and (4) the "inquiry tips" participants occasionally received from workshop instructors as feedback to their public presentations . These factors contributed in various ways to novices' enhanced knowledge of instructional selection (Shulman, 1987), i.e. , their increased awareness of an inquiry approach to conducting chemical demonstrations . Although the inservice intervention provided novices with a greater realization of the value of inquiry demonstrating, it appeared 255 to be insufficient to assist novice demonstrators in recognizing a greater number of critical incidents (specific teaching behavior s ) that hindered or promoted inquiry teaching of basic chemical concept s . Before and after the inservice workshop , novices generally identi f i ed and discussed similar inquiry - teaching critical incident s , such as : (1) the appropriateness of having students use their observation skills during a chemical demonstration and (2) the inappropriat eness of having a demonstrator forecast demonstration procedures , observations , and explanations. By contrast , the expe rienced chemical demonstrators gave more frequent and more va ried attention t o inquiry issues than either pre- or post-workshop novices. Changes in Novices' Knowledge of Demonstration Variations Knowledge of chemical demonstration variations provides demonstrators with options that allow them to fit the r epresented material to the characteristics of the students. For thi s r eason, demonstration variations are viewed as the product of the adapt ation process described in Shulman's (1987) model of t eaching. After the workshop intervention, novices were able t o discuss more variations on selected chemical demonstrations (p < .0 5). They also began to recognize the complexity of some chemical demonstrations and how these complex demonstrations could generate confus i on among middle school students. This knowledge prompted novices t o di scus s ways of reducing the complex ity of a chemical demonstration us ing simplified variations. This knowledge of complex chemical demonstrations , and suitable variations on such demons trations , was acquired by novice s partly through the workshop staff ' s emphas i s on simple demonstrations and how such variations pot entially he l p 256 students focus on the central concept. Novices' heightened awareness of demonstration variations can also be traced to some of the f eedba ck workshop participants received immediately after performing a demonstration before their peers. The workshop sourcebook also provided several suggestions for performing variations on chemical demonstrations (see Appendix K for Sourcebook examples). The variations discussed during the workshop and those described in the sourcebook ranged from simpler to more complex approaches for demonstrating the original demonstration. Changes in Novices ' General Pedagogical Knowledge General pedagogical knowledge growth among novice chemical demonstrators during the two-week demonstration workshop was more difficult to detect. Data obtained from the critical- stop tas k revealed no significant increase in the total number of critical incidents discussed involving generic teaching behaviors import ant to chemical demonstrating (Table 17). This lack of change may be accounted for , in part, by the information processing skills that novices possessed for analyzing demonstration teaching epi sodes . (A further discussion of this issue is provided in the section that follows). Issues pertaining to wait time, feedback, and building ra pport with students continued to occupy post-workshop novices' attenti on of general pedagogical issues during the critical-stop ta s k. These issues were not given special attention during the s cheduled s t aff presentations; however, they were addressed in various way s during the public and private feedback sessions that followed the public presentations given by the workshop participants. 257 Novices did give special attention to issues pertaining to the effective use of humor while thinking aloud about effective chemical demonstration teaching (pp. 209 -210). These discourses appeared more frequently in their post - workshop discourses than in their pre-workshop discourses. This apparent change in genera l pedagogical knowledge associated with chemical demonstration teaching may be attributed to one of the workshop staff presentations that specifically gave attention to the use of humor in teaching (see Appendices F and G). This staff presentation may have influenced novice demonstrators in their post-workshop critiques of the four chemical demonstration videotapes. Changes in Novices' Skill in Processing Demonstration Teaching Information The protocols obtained from pre- and post-workshop novice s reflected both general pedagogical knowledge and pedagogical content knowledge. The findings associated with several content analyses performed on this data (e.g . , Tables 7, 10 , and 11) suggest that novices' information processing skills may not have been substant i ally enhanced during the course of the workshop. Given that critical incidents were presented to the research subjects at a rate of one every ten seconds , the critical -stop task apparently requires a well-developed knowledge structures (schemata) for chemical demonstration teaching and well-developed cognitive process ing skill s to effectively perform the think-aloud task of (i) storing and encoding the verbal and visual cues presented via the demons tration videotapes, (ii) processing the information in t erms of ex i s t i ng knowledge structures , and (iii) retrieving the proce ssed informat i on 258 to yield an appropriate verbal response (Gage & Berliner, 1984), i . e ., evaluations of rapidly-occuring critical incidents. Although novices continued to process information provided by the videotaped demonstrations at the same rate before and after the inservice workshop (Tab les 7 and 17), there did appear to be a s l i ght shift in critical-incident focus from generic issues, such as use of the blackboard and sel ected organizational issues (Categories 7 and 8 in Table 11), to content-bound issues, such as the quality of explanations and teacher-directed questions (Categories 2 and 4). In summary , the verbal data can be interpreted to indicate t hat the two-week chemical demonstration workshop had an observable impact on promoting pedagogical content knowledge growth among novice chemical demonstrators . This growth, however, was most evident wi t h respect to novices acquiring multiple representations for demonstrating basic chemical concepts and in terms of adapt ing demonstrations to fit the characteristics of the learne r . Littl e pedagogical content knowledge growth wa s observed in t erms of structuring content materials (critical evaluation) and us ing inqui ry strategies (instructional selection ) during a chemical demons t rat i on. In addition , generic concerns associated with chemical demonstration teaching gave way to more content - bound concerns as a r esul t of the workshop intervention . The findings of this study support the literature that experti se in virtually any complex enterprise , including t eaching , takes considerable time to develop (Bloom, 1985) . Intens ive inservi cing for two weeks, for example , on one science tea ching s trategy (demonstration t eaching) in one subj ect matter di scipline (chemistry) 259 does not suddenly transform novice chemical demonstrators into highly experienced or expert chemical demonstrators. It does, however , produce substantial knowledge growth in teaching that allows teachers to more readily progress through the five hypothesized stages of progression from novice to expert teacher (Berliner , 1988; Dreyfus & Dreyfus, 1986). Evidence for continued professional growth among ICE participants after the workshop comes from O'Brien's (1987) follow-up study of another group of ICE Workshop B participants. This st udy documented increased classroom use of chemical demonstrations as a result of the workshop intervention, a factor that is likely to help perpetuate teachers' continued pedagogical content knowledge growth long after the intensive chemical demonstration workshop exper i ence . Validity Considerations Questions about a qualitative study's internal validity cente r around how well rival hypotheses have been ruled out as alternative explanations f or observed phenomena. In thi s study, this translates to a sea rch for alternative explanations that could account for (1) experienced-novice demonstrator differences and (2) novice s ' pre-/pos t - workshop changes in interview responses . Questions about a qualitative study's external validity center around how well the findings of a s tudy apply to other subj ect s and settings (Bogden & Biklen, 1982; Krathwohl, 1985; Goetz & Lecompte, 1984). Internal validity Response effects and select ion present potential threats t o the interna l validity of th i s s tudy. These two effects provide pl aus i bl e 260 rival hypotheses for explaining the observed differences in experienced and novice chemical demonstrator discourses during the clinical interview. Response effects refer to those factors that influence a respondent (a demonstrator) to provide incomple te or inaccurate data (Borg & Gall, p. 438, 1983). One such response factor includes the predisposition of the interviewer, sometimes r e ferr ed to as an interviewer e ffect. To help minimize errors attributed to interviewer effects , the researcher developed an interview guide (Appendix I) which he used and adhered to closely during each clinical interview. This intervi ew guide followed estabished techniques for conducting interviews (Borg & Gall, 1983 ; Novak & Gowin, 1984). Thi s researcher also implemented many of the strategies for conducting effective clinical intervi ews recommended by Novak and Gowin (1984). Furthermore, this r esearcher piloted the interview guide and recommended interview procedur es with four science education graduate student s and again with eight ICE workshop participants during the first summer session of the workshop . Given that the interviewer conducted thi s study using r ecommended clinical interview procedures and after receiving training and feedback in interviewing science teachers, it would be unreasonable to believe that the observed differences between the two group s of demonstrators were artifacts of the interviewer's skills and biase s . Furthermore, it would be unreasonable to expect that these i nterviewer effects would surface on so many of the variables examined in this study (D, V, XTOT, 9 Categories, Strengths/ Weaknes ses , etc .) . Thus, prior chemical demonstration teaching experience , rather than 261 interviewer effects , provides the more plausible explanation for the observed differences in the two groups' interview responses. Another response effect that could bias the interview data i s the predispositions of the respondents. This effect suggests that the two groups of subjects could be differentially predisposed in their motivation to cooperate with the researcher or in their desire to present themselves in favorable terms. Such a possibility needs to be addressed in light of the fact that this researcher had professional interactions with all the experienced demonstrators and only one of the novice demonstrators prior to the study. One could argue that these differences could account for different predispositions of the respondents and, consequently, generate major differences in interview responses. To help minimize this validity threat, this researcher made every attempt to build a satisfactory rapport with the novice demonstrator s prior to the interview and during the first few minutes of the interview. Novices were contacted by mail reque sting thei r voluntary participation (Appendix D). This letter was written so as to give these subjects some favorable, pre-workshop contact with this investigator. Local participants were also contacted by phone aft e r having received the contact letter. The fact that all the contacted subjects volunteered to participate in this s tudy suggests t hat a favorable rapport had been initiated. Furthermore , those novices that arrived a day before the workshop began , appeared to be eager to schedule time to view and critique the chemical demonstration videotapes and r espond to a f ew fo llow - up questions. 262 Further attempts to build rapport with the novice subjects occurred during the beginning of the clinical interview when the inte rviewer introduced the study, described his role at the workshop, and asked a few non-threateni ng , warm- up questions. Prior to beginning the critical- stop task , novices i ndicated that they had a clear idea of what they had to do and were ready to begin. During the interview , the researcher made an effort to relate to the respondents in a non-threatening fashion , accepting their responses as given and providing a conversational style that communicated trust , confidence , and ease among respondents - conditions necessary for yielding valid data from informants (Goetz & Lecompte , 1984 ) . Thus, with a satisfactory rapport established between the interviewer and respondents, there is sufficient reason to believe that both groups of demonstrators were favorab l y predisposed to the interview ta sks and that group differences reflected prior knowledge of chemical demonstration t eaching. Hence , the threat of differences in respondent predispositions can be ruled out as a plausible hypothesis for explaining experienced/novice chemical demonst rator differences . Piloting this study also permitted a reduction in a third form of response effect that could account for group differences , namely the procedures used in conducting the s tudy. Sources of error in this category include the place where the interview i s held, the l eng th of the interview , the quality of the questi ons , and the instruct i ons presented to the s ubj ects . Although the exper i enced s ubj ects were probably more familiar with the workshop setting, took l onger to complete the inte rview, and were more familiar with terms like "variations" than the novice subj ects , the researche r made an effort 263 to minimize inte rview differences. This researcher conducted the interviews for both groups of demonstrators in rooms that afforded privacy and comfort, kept the interviews to within 50 minutes, yet allowed sufficient time for complete responses , and interacted with the respondents so they could clearly understand all terms , questions, and instructions. During the pilot phase of this study, respondents were also asked to supply their perceptions, feelings , and recommendations regarding the clinical interview to assist this researcher in reducing procedural errors in the actual study. Therefore , errors in interview procedures are hardly likely to account for the many differences in experienced and novice chemical demonstrator performance during the clinical interview. Selection represents another major threat to the internal validity of a qualitative study . In one sense , it was absolutely necessary to identify and select chemical demonstrators who differed considerably in terms of chemical demonstration experience so that pedagogical content knowledge differences could be documented. On the other hand, bias in group composition (Borg & Gall, p . 141 , 19 83 ) could further accentuate or distort these findings. The greatest concerns of this type include differential representation within the two groups regarding the demonstrators' teaching grade level, prior experience in critiquing videotaped teaching episodes , content background , and the ability to verbalize thoughts about chemical demonstrating. The experienced and novice group of chemical demonstrator s both consisted of a mix of elementary , middle school , and high school teachers. These similarit i es are offset slightly by the fact that the 264 experienced group included a community college chemistry instructor who had some high school teaching experience. The novice group had no community college representative. The experienced group also had a slightly higher proportion of high school teachers (3/ 5), the novice group a slightly higher proportion of middle school teachers (5/8). One could argue that these teaching level differences could account for, or skew , group differences in interview response. Although the demonstrators in the two groups were not optimally matched in terms of grade level , this research study operationally defined "novice chemical demonstrators " in terms of self-reported teaching information gathered on the workshop participants. Thus , if another researche r were to replicate this study , similarities and differences could be accounted for in terms of the operational definition given to the term "novice chemical demonstrator." In this study , bias in group composition was therefore "controlled" by gathering quantitative da t a on the demonstrators ' attributes {e .g., grade level, confidence and weekly use in conduct ing chemical demonstrations, chemical demonstration workshop/outreach experience , college chemistry courses , etc .) that become part of the operational definitions for the labe l s "experienced and novice chemical demonstrators. " The background information obtained from the research subjects also provided important data necessary to discuss the generalizability of thi s study ' s findings (see section on External Validity, below) . Another selection threat arises from the possibility that experienced chemical demonstrators are less able to articulate their thoughts on demonstration t eaching than novices because a greater part of their teaching knowledge has become embedded in routine behavior s 265 (knowledge in action). Such factors, i f operating, would generate biased data and, thus, distortions in describing the nature of experienced and novice chemical demonstrators' pedagogical content knowledge. This line of reasoning argues that experienced demonstrators possess more tacit knowledge on the subject of demonstration teaching than novices, a factor that would interfere with efforts at collecting valid data. Tacit knowledge refers to a person's knowledge that has never been verbalized and may not be communicable in verbal form (Calderhead, 1981b). Tacit knowledge becomes a particular problem with behaviors that have reached a level of routinization where behavior is engaged in unthinkingly, a situation that characterizes expertise. In defense of this selection threat to the validity of this study, this researcher observed no s triking difference between the two groups of teachers in terms of their loquaciousness in discussing common critical incidents. The experienced subjects, all of whom had experience at teaching chemical demonstrations to other teachers , appeared very capable of discussing their knowledge of chemical demonstration pedagogy while analyzing videotaped demonstration performances. Furthermore, as Shulman (1 986 ) ha s stated, ''Tacit knowledge among teachers is of limited value if the t eachers are he ld responsible for explaining what they do and why they do it, to their students , their communities , and their peer s ." For this reason , thi s investigator wa s primarily interes ted in examining readily r etri evable pedagogical knowledge , not thoughts which occurred at a low l eve l of awareness among the informant s . 266 Finally, one could argue that differences between experienced and novice subjects may be attributed to their prior experience in critiquing teaching or videotaped models of teaching. To help understand this influence, this researcher asked the research subj ect s (Appendix I) if they had ever critiqued someone else's teaching as observed on videotape. Both groups generally indicated they engaged in a task similar to the critical-stop task where they would carefully critique someone else's teaching on videotape on at least one or two other occasions in their careers. Unfortunately, no information wa s gathered on subjects' history in supervising student teache r s where such critiquing skills could be developed. In any case , it appear s that this threat is minimal and that differences in group pe rformance during the clinical interview is best accounted for in t erms of prior chemical demonstration experience. Other threats to internal validity emerge and need at tenti on when accounting for pre-/post - workshop changes in novi ce demons t rator interview performance. These variables include testing (interview practice) and maturation (changing predispositions). In this study, the threat of testing (interview practice ) suggests that learning during the first interview promotes enhanced responses during the second interview. Evidence for the presence of this extraneous variable comes from a small number of comments elicited by a few subjects during the post-workshop inter vi ew where they referred to one of the pre-workshop videotapes . This researcher made an effort to minimize learning e f fects during pre-testing by using a parallel set of videotapes (same demonstration, di f ferent pe rformance quality) during the pos t - works hop 267 inte rviews. The l ength of time between the pre - and posttes t s (two weeks) may have also worked to this researcher's advantage in keeping this factor under control . It was not possible, however, for this researcher to control the possibility of the pre-workshop inte r vi ew causing novice demonstrators to become sensitized to the workshop treatment, particularly to the two chemical demonstrations obser ved on videotape. The pre-workshop viewing of these videotapes may have piqued novice demonstrators' interest and motivation to l ea rn mor e about these particular demonstrations and, even, to learn mor e abou t chemical demonstrations in general. Some of the novices did appear to be impressed or curious about the concept/phenomenon demons trat ed on videotape. Thus, the influence of testing could have had cons ider ab l e impact on novices' pedagogical content knowledge growth by sens i t i zi ng subjects to the observed demonstrations and to the workshop intervention. Because this factor could not be well controlled , gi ven that this researcher did not have the personal res ources t o conduct clinical interviews with a control group of novices wh o r eceive only post-workshop interviews, it would be best to consider t es ting a part of the workshop treatment for the eight novice demon strator s who participated in the study . This pre-workshop treatment appeared t o have a beneficial, rather than detrimental, eff ect on the works hop participants. Viewed from another pe rspective, it i s hardly likely that a control group of novice demonstrators receiving only the pre-works hop interview would perform as well on the pos t-workshop interview, i . e ., show as much knowledge gain, as those novices who al so experienced the two-week workshop inte rvention. Thus, t he influence of t he wo r kshop 268 becomes the most plausible, empirically grounded explanation for novices pedagogical knowledge growth. Another threat to the internal validity of this study that could influence changes in novices' interview responses is maturation. Given that the workshop was an intensive two -week intervention program, it is not surprising that a few participants appeared to s how some signs of "training fatigue" in terms of being a bit more physically tired than when they arrived for the workshop. This is understandable given the summer temperatures, demanding eight-hour workshop schedul e, several evening sessions , and occasional late nights studying or socializing . During the last two days of the workshop , when most of the post-workshop interviews were he ld , some of the subject were also beginning to mentally prepare for a long trip home, arrange social time with peers, and think about an upcoming summer vacation . These altered physiological and psychological s t at es among post-workshop novices may have prevented these subjects f rom performing even better than they did during the post-workshop interview. It is conceivable that the lack of change in the total number of critical incidents cited by novices after the workshop may be partially attributable to these uncontrollable factor s . Novi ces may have been less alert and less able to focus on the critica l-s t op task. If these extraneous variables actually operated during the post-workshop interviews, this would suggest that novices ' pedagogical knowledge gains are actually even greater than those documented in this study. Having discus sed the various threat s to the internal validi t y of this study , differ ences in experienced and novice chemi cal 269 demonstrator performance appears best accounted for in terms of their prior chemical demonstration teaching experience. Likewise, the obse rved growth in novice demonstrators ' pedagogical content knowledge and general pedagogical knowledge of chemical demonstrating is bes t explained in terms of the workshop intervention. External validity External validity refers to the extent a study ' s findings are generalizable to other educational settings. In this study it refer s to an assessment of the applicability of the present findings to other experienced/novice teacher comparisons and to other teacher inservice programs. The key characteristics describing the setting of thi s study include a group of motivated elementary, middle school, and high school science teachers with limited experience in conducting chemi ca l demonstrations participating in an intensive , two- week chemical demonstration workshop conducted by experienced chemical demonstrators . The workshop was designed to provide direct instruction as well as help participants observe , practice , perform, and receive feedback on numerous chemical demonstrations covering a range of basic chemical concepts. The findings that illustrate the nature of the two groups of chemical demonstrators ' PCK are most generalizable to other el ement ary and secondary physical science teachers possessing either minimal or considerable experience in conducting chemical demonstrations . The findings of this study may also apply to experienced and novice demonstrators in other sc i ence disciplines, such as physics , biology, and earth science , where teachers' pedagogical content knowledge , when defined in terms of knowledge of demonstrating discipline-specific 270 concepts , could be expected to show similar differences in br eadth and depth of knowledge. The findings of this study are consistent with those obtained in other studies examining experienced and novice teacher diffe rences . Previous studies have documented cognitive differences between experienced and novice teachers in terms of their knowledge of students and knowledge of managing routine classroom tasks. The findings in these studies indicate that experienced teachers had amassed and could process a large quantity of information about students and classroom learning environments that distingui s hes them from their less experienced colleagues (Berliner, 1986; Cart er, et al ., 1987; Peterson & Comeaux, 1987; Calderhead , 1983 ; Leinhardt, 1983) . Moving from population validity to ecological validity, thi s study would be most generalizable to other ICE fie ld cente r s across the country that deliver the same workshop to s imilar participant s. Generalizability would also extend to workshop participant s at tending similar programs at new ICE field centers in the future . Besides these specific contexts , the findings of thi s study are expected t o be applicable to other short term inservice programs that focu s intense ly on one teaching strategy applied to a single discipline . Recommendations Recommendations based on the findings of this s tudy can be grouped according to r esearch recommendations , methodol ogi ca l recommendation s , and program-specific r ecommendations . 271 Recommendations for Further Research 1. Conduct follow-up studies on novice chemical demons trator s one year after the workshop to determine whether pedagogical content knowledge continues to grow after the inservice intervention. Thi s could be implemented by having novices critique a set of chemical demonstration videotapes at their leisure and then having them send the audiotaped comments back to the researcher. Follow-up phone conversations using a structured or semi-s t ructured interview could also be conducted to gather additional information on novices ' post-workshop pedagogical knowledge growth. Such studies would hel p assess whe ther the participants have acquired the skill s and motivation that allow for continued growth of their pedagogi cal content knowledge in the realm of demonstrating ba s ic chemical concepts. Local teachers could be contacted directly for a second post-workshop clinical interview. Such follow-up studies would he lp determine whether the novice participants have come closer t o reaching the pedagogical content knowledge levels observed among experienced chemical demonstrators . 2. Determine whether pedagogical content knowledge growth du r ing a workshop intervention interacts with specific teache r ap t itudes , such as prior chemical demonstration experience or prior sub ject- matter knowledge . Such interaction st udies, for example , would r eveal whether pedagogical content knowledge growth during inse rvicing i s greater among demonstrators with extensive content knowl edge or among demonstrators weak in content knowledge. Such finding s would he lp determine whethe r intens ive, skills - oriented workshops have ma r kedly different impact s on participants ' knowledge growth in teachi ng. 272 Interaction effects, such as those described above, should be examined in future studies exploring teachers' pedagogical cont ent knowledge growth resulting from short-term teacher education programs as well as year - long classroom teaching experiences. Such interaction studies would help teacher educators better understand the nature of knowledge growth in teaching science within various learning environments. Such studies would also help program implementors design demonstration workshops that address differences in aptitude among inservice participants. 3. Use the cognitive methods described in this study to take a closer look at the components of inservice training programs most responsible for science teachers' pedagogical content knowledge growth . Workshop components worth examining in an isolated context include PCK studies that focus on the impact of laboratory prac tice , microteaching, observing , peer and videotape feedback, and workshop sourcebooks on knowledge growth in teaching. Methodological Recommendations Several refinements and modifications could be considered for the critical-stop task and semi-structured interview as a probe of teachers' pedagogical content knowledge related to the demonstration teaching of basic chemical concepts. 1. Given that the pedagogical content knowledge (PCK ) and general pedagogical knowledge (GPK) associated with the effective demonstration teaching of density and air pressure constitute a rather extensive knowledge base , it is possible to probe the variou s characteristics of this knowledge base even in greater detai l by focusing only on selected generic or pedagogical content knowledge 273 (PCK ) is sue s . Thi s could be accomplished by simplifying or shortening the critical-stop task and having subjects discuss researcher - identified , rather than subject identified, critical incident s associated with the two knowledge systems. For example, a researcher may seek to focus only on critical incidents related to inqui ry demonstrating and Shulman ' s (1986) construct of instruct i ona l se l ection. By having experienced and novice demons trators discus s their pe rceptions on only these demonstration teaching i ssues , it may be possible to obtain greater insights into experienced and novice demonstrators' conceptualizations of t eaching, especially in area s not voluntarily discussed by subjects during the critical-stop task. 2. To he lp increase the power of the statistical tes ts used t o assess quantitative differences in experienced and novice demonstrators' critical-stop task performance , larger sampl e sizes a re r ecommended. This would permit more powerful parametri c statistical t ests (e.g., the independent t-test) to be run in place of the more conservative non-parametric statistical tests (e .g., t he Wilcoxon Matched-pairs Ranked-signs t est ) where meaningful differences were established at the 0.187 level. The non- parametric tests used in th is s tudy were limited to a sample si ze of n = 4, i. e ., four pair- wi se comparisons between experienced and novice demonstrators with group scores for each of the four videotapes . Non-pa r ametric t est s wer e used in this study because (1) the four videotapes critiqued varied in length and number of critical incident s and (2) the counterbalanced design used in this study had pre-workshop novi ces randomly examine two out of the four available videotapes. With a larger sample s ize , one could consider using individual performance scores rathe r than 274 group scores for each videotape as a unit of analysis , and , thus , not be constrained in using a non-parametric test of significance as part of the content analysis . Larger sample sizes could be achieved by training one or two individuals to conduct the clinical interviews concurrently with the researcher. A single researcher could obtain a larger sample only if the inservice teachers live local to the research site . A larger sample size would be beneficial to a st udy because it would generate more discourses from experienced and novice chemical demonstrators on infrequently cited critical incidents. With the critical-stop task , frequently cited critical incidents typically represented a small percentage of the total number of incident s critiqued (approx. 10%). To counteract this effect and , thus , increase "yield" in terms of the number of discourses addressing the same incident, a larger sample size is recommended. An additiona l advantage of working with a larger sample s i ze includes the possibility of conducting subgroup comparisons (e .g., between experienced and novice middle school chemical demon strators or between experienced and novi ce high school chemical demonstrato rs ) and in effect "control" for selected teacher attributes or background variables. A disadvantage in working with a larger sample s ize , however, involves the time-consuming nature of gathering and ana ly zing verbal reports obtained from individuals regarding their conceptualizations of effective chemical demonstration t eaching. 3 . Modify the interview guide (Appendix I ) to spec ifically probe novices in knowledge areas only di scussed by the experienced chemi cal 275 demonstrators. For exampl e , it was primarily the experienced demonstrators and post-workshop novices who volunteered information on the complexity of the observed chemical demonstration, the appropriateness of the chemica l demonstration for middle school students , and the potential for the demonstration to generate s tudent confusion or misconceptions. A future interview guide may include a series of structured and open-ended questions that specifically probe topics rarely addressed by pre-workshop novices. This would r equire the deletion of other questions less central to teachers' PCK (e .g., I.3 and 1.10 ) in order to help keep the clinical interview unde r 50 minutes. Program-Specific Recommendations Recommendations for the design and evaluation of thi s and othe r skills-oriented chemical demonstration workshops include the following: 1. The pedagogical content knowledge growth observed among novice chemical demonstrators during the two-week workshop sugges t s the importance of continued NSF support of summer institutes that focus on science teaching skills . The pedagogical content knowl edge growth observed in the present study of a NSF-supported chemical demonstration workshop appeared to be more ext ensive than that reported in non - inservice settings (Mason, 1988; Shulman , 1987; Wilson, Shulman , & Richert , 1987). NSF should therefore cont i nue t o support , and even expand , their support of workshops that foc us on science teaching skills (e.g., computer-based laboratories , demonstration t eaching, science - technology -society , scientific instrumentation , etc.) for each of the ma jor scientific di scipl i nes . 276 2. Although substantial pedagogical knowledge gains were observed among novice chemical demonstrators , they did not achieve the levels observed among experienced demonstrators. Although such levels of attainment were not expected for the workshop-trained novices, this finding suggests the importance of providing effective follow-up of workshop participants in the form of newsletters, demonstrat ion materials, and other forms of encouragement that would promote increased use of chemical demonstrations in the classroom and involvement in conducting chemical demonstration workshops . Funding agencies should also consider supporting an "advanced" or "second year " chemical demonstration workshop to help more science teachers achieve the much-publicized need for excellence and expert i se i n science teaching (National Commission on Excellence in Educa tion, 1983; Penick & Krajcik, 1984a, 1984b; Penick & Lunetta , 1984; Yankwich , 1984). The advanced workshop could help t eachers develop more interconnections between the knowledge systems responsible for the growth and development of PCK . 3. Place greater emphasis on issues pertaining to inquiry approaches to chemical demonstrating. The inclusion of this goal is recommended to help bring novice chemical demonstrators further al ong or up to the knowledge level of experienced demonst rators in this important teaching strategy. The implementation of such a works hop ob j ective would help foster the development of teachers' pedagogical content knowledge in the area of instructional selection , a subp rocess in Shulman ' s (1987 ) model of pedagogical r easoning. 277 Swmnary This study has accomplished several objectives. It has (1) provided rich descriptions of the nature of experienced and novice chemical demonstrators' pedagogica l content knowledge system, (2) provided empirical evidence that teachers' pedagogical content knowledge can be enhanced through intensive, short-term inservice workshops, (3) garnered support for the usefulness of cognitive methods for eva luating a s kills - oriented workshop, and (4) generated a set of recommendations for future inservice pr ograms designed to produce professional growth in science teacher s ' knowledge and skill in demonstrating basic chemical concepts. This study addressed an important issue in science education, namely, the goal of achieving excellence in science teaching. Thi s study has demonstrated that the growth and development of pedagogical content knowledge in science teaching can be substantially promot ed through highly focused, short-term, inservice interventions . The results of this study strongly suggests t he importance of inse rvice programs in enhancing science teachers' pedagogical repertoire and thus, the quality of science instruction in this country. 278 APPENDIX A Institute for Chemical Education Brochure/Application Form Institute for Chemical Education 1987 Summer Workshops Refresh and update your background knowledge and add to your repertoire of practical classroom activities and techniques. Nearly 700 elementary, middle, and high school teachers from across the country have taken advantage of the opportunities offered by ICE Workshops. Join us! ? Chemistry Supplements for Pr.,.High ? Chemistry Fundamentals. Streng1hen and ? Chemical Instrumentation Update. Wirk School Classes. learn and prac1ice effec? upda1e your background in selec!ed w ith modern chemica l ins1rument,Hi o1 tive and sc1fe demonstrations, experi- chemistry 1opics 1rea1ed in dep1h. learn and learn from researchers about 1he ments, and act ivi ties appropriate for chemic.i i phenomena. demonstrations. problems 1ha1 are bei ng solved wilh 1hem. younger siudents. Interactive teaching problem solving, and effee1ive ways 10 Prepare spectra and 01 her ma1erials for methods are stressed. presen1 scien t11ic concepts. use in your c lassroom. Elementary, Middle, Teachers of Chemistry High School Chemistry Te,1chers High School Teachers Six Weeks: Si, Credits Two Weeks: Two Credits Two Weeks: Two Credits .. . greatly improved my knowledge of . we were equipped with essentia l the- "So many useful things to take home, so chemistry. The lab and demonmation expe- ory and had 'hands-on' experience w ith many insuuctors willing co assist teachers." riences Jre sure lO make my cl,Hses more over ten insrruments." Barbara Hilli, Slrat~ Paul Thiesfeld!, Overland Park, K.lnsas interesting, exc i11ng, and productive ford, Connecticut Harlan F1,ldt, Wisconsin Dells, Wisconsin Pending comm i tment of fu nds by the National Science Foundation and others. ICE w ill pay fees, expenses , and stipends for participating teache rs. Chemistry Challenges 1. Wi1hout opening unlabeled 12-ounce cans of die! and / ,\4M ?j,seq regular soda, how can you !ell which is wh ich? J Jt> Sl) npoJd su ,uea1) 1sow pue )!PPe a,e S1Jn1spoo1 isow ?suo 11n1os )11eq u, 2. Crush an unflavored Ex-Lax@ 1ablet wi1h 10-15 ml pa, pue 'suo11n1os 1c,1nau pue )!P')C u, ssa1 12-3 tsp.I rubbing alcohol. Add 4 drops of 1his test solu- ?JOIO) '! '~ xe1-x3 u, 1ua!pa.,8u1 ~"l)e a41 tion to 5 ml each of ammonia, vinega r, cit rus juice, ?u1a1e414d1oua4d rsa ,i.,.xlo,d aseq-pp\l) ?t solutions of baking soda, detergen1 , and 01her house- hold subs1ances; record your ob~rvalions. Wha l pal- ?sieou epos 1a!p-H?8ns ou-.1su.>? 'i'i.l l lernlsl can you sce l a41 a1! YM s~u rs epos " I nSa, a41 ?,a1eM P\O) u, sue) q1oq J)Pld rasp,axa ,11, suap \II ?1 If you try these .ictivities, let us know how they worked for you. 279 /m rirure for Chemical Education 1987 Summer \M::Jrkshops Application for Part 1c1pation and Financial Support Name:------------- ---------- - - ------- --- - - - - - -- List First Middle Socio1I Security I _______________ School-- ------------- - Home Add-s ___ ______ _______ Work Add-s _______________ Honw Phone t_____.) _ ___ _ _ _ _ ____ Worl< PhoneL__J _ ___ _____ _ _ lndlco1te tM o1dd-1 to be uwd for cor~nce: Home ___ 'M>rl< ~ Please rank the workshops you .would con sider anending, indi ca ting your first choice with the number 1. (In se lecting 1he field center, please remember 1ha1 compensation (or rravel expenses may be limiled to S250.) Chemistry CMmistry CMmic?I Supplements Fund?ment?ls S1 S2 lnstrumen ..1 1on W isconsi n &12 2-7/3 7/1 ) -7/24 &/29-8/7 7/&-7/17 Af1zona &129-7/10 7/2 0-7/)1 &/2 2-7/3 California &/22-7/3 1 7/6-7/2 4" Maryland &/22-7/3 7/1 J- 7/24 &/22- 7/31 Colorado &/2 2-7/)" 6/29- 7/1 0 ? This session is 1hrtt -weeks, 01hers arr two 'w('eks . .. ? Thi s session is o pen only to Color,1 do Sc ienct Coord inaton . Upon )')Ur acceptance, y,e will contact your Superintendent o( Schools and Principal, or th eir counterparts, 10 ask them 10 assist you in obtaining local public or private funds for your support . Please provide their names and addresses. Superintendent _______________ _ Current o1nd projected teaching ollslgnments: District _________________ _ _ Address ______________ ____ Course title _________________ Number o( sections ___ Enrollment _______ Phone t_____.) ______________ _ Is this a current assignment/ ___________ Do you expect it in '87-tl81 _ ___ _ _ _ ____ PrinciP"'I ________________ _ _ School _________________ _ Course ti tle _ _ _ _______ _ ______ Mdress ____ ______________ Number of sections ___ Enrollment _______ Is this a current assi gnment! _ _ _ _ _ ______ Phone L--....J ______________ Do you expect it in '87-tlSI _ _ ____ _ ____ Course title _ ______________ _ _ SumlTlolrize your coll~e/univenity education. Number of sections ___ Enrollment ____ _ _ _ School __________________ Is this a current assignmenll ______ ____ _ City and State ________________ Do you expect it in '87-tlSI _ _____ _____ Degree ___ _ _ _ ___ Date _ _______ Major ________________ _ _ _ Course title ______ _ _ _ _ ______ _ Minor(s) _________________ _ Number of sections _ __ Enrollment _ _____ _ Is this a current assi gnment/ __________ _ School ________ __________ Do you expect it in '87- 881 ___ _ _ _ _ ____ City and State ____________ ___ _ Degree _________ Date ________ Professional ktivit ies (N STA, ACS, etc.) _ ______ Major ________________ ___ Minor(s) ______ _______ _ _ ___ --- =--~ - ~ -- 280 28 1 In ~dditlon, pleue Mlbmit the following inf0 5 OTHER SCIENCE DEMONSTRATIONS DO YOU USE? 5 DURING THE PAST YEAR, HOW MANY OUTREACH PROGRAMS INVOLVING CHEMICAL DEMONSTRATIONS HAVE YOU DELIVERED TO... (circ le your response to each) STUDENTS OUTSIDE YOUR CLASSROOM? 5 (assembly programs, etc.) OTHER TEACHERS? 5 (inservice workshops, conferences, etc.) Note: the symbol < means "less than" and> means "more than" TURN OVER 28 4 FOR WHAT PURPOSES SHOULD CHEMICAL DEMONSTRATIONS BE USED IN TEACHING? (Rank Order . on al - 7 scale (using each number only once) with: 1 = most important purpose 7 = least important purpose) Entertaining change of pace to the regular class format Substituting for expensive/dange rous labs Illustrating scientific concepts Training students to make careful observations Arousing curiosity when introducing new ideas Stimulating student thought Modeling good laboratory techniques & att itudes WHAT LIMITS THE FREQUENCY OR QUALITY OF CHEMICAL DEMONSTRATIONS IN YOUR TEACHING? (circle your re s ponse for each factor) IMPORTANCE OF FACTOR LOW HIGH Lack of good demonstration ideas l 2 3 4 5 Lack of demonstrating skills l 2 3 4 5 Lack of confidence that demonstrations will work 1 2 3 4 5 Lack of time for preparation and cleanup l 2 3 4 5 Lack of equipment and/or supplies l 2 3 4 5 Lack of time in a overcrowded curriculum 1 2 3 4 5 Inadequate teaching facilities 1 2 3 4 5 Classroom management considerations l 2 3 4 5 Safety considerations l 2 3 4 5 Personal "inertia" 1 2 3 4 5 Other: l 2 3 4 5 DO YOU KNOW ABOUT THESE PUBLICATIONS ON DEMONSTRATIONS ? Sources of Chemical Demonstration Ideas Am Not Am Have (check as appropriate) Aware of Aware of Used Tested Demonstrations in Chemistry (Hubert Alyea & Frank Dutton) Entertaining Science Experiments with Everyday Objects (Martin Gardner) Mr. Wizard's 400 Experiments in Science (Don Herbert) "Elementary" Chemical Demonstrations {Linus Pauling) Chemical Demonstrations (Bass a m Shakhashiri) Chemical Demonstrations: A Sourcebook for Teachers (Lee Summerlin & J ames Ealy) Idea Bank Collation (Irwin Ta lesnick) Idea Bank (a feature of NSTA publication: The Science Teacher) Tested Demonstrations (a feature of The Journal of Chemical Education) Other: THANK YOU 285 APPENDIX C Pre-Workshop Grouping of Participants Scale 1 - 5: Confidence, (Low - High) Use, (0 - 5+ times/week) Participant Confidence Use No. Chem Course / (Chem/Other science) (Chem/Other) Years Teaching More Experienced MA 5/4 5/<1 7 /13 LK 5/4 2/<1 17 /30 BB 4/4 4/ 2 3/10 JW 4/4 3/ 1 22/22 KM 4/4 2/ 2 2/13 JE 4/4 1/ 1 7/15 AH 3/5 2/ 3 3/3 GJ 3/3 2/ 7/1 SY 3/3 2/ 2 4/15 IW 2/4 3/ 4 1/12 JO 2/3 3/ 3 0/24 n = 11 4.2 I 4 2.7 / 1.8 6. 6/14 Less Experienced (Novices) Nl 3/4 <1/ 3 5/1 N2 2/3 <1/ 1 2/1 2 N3 2/3 <1/ 2 0/15 N4 2/5 <1/ 3 0/16 NS 4/4 1/ 3 6/ 2 N6 3/4 2/ 2 2/3 N7 2/4 1/ 2 6/ <1 N8 1/3 <1/ 2 4/7 Nl - N8 2.4 / 3.8 0.5 / 2.2 2.8/7 SD 3/4 1/ 3 2/ 8 LA 2/4 <1/ 3 1/26 BC 2/2 <1/<1 6/9 JA 1/1 <1/ 1 0/ 9 n = 12 2.2 / 3.4 <0.4 /2.0 2.3 / 9 286 Scale 1 - 5: Confidence , (Low - High) Use, (0 - 5+ times/week) Participant Confidence Use No. Chem Course/ (Chem/Other science) (Chem/Other ) Years Teaching Experienced Demonstrators , Trainers El 4/4 3/3 2/15 E2 5/5 3/- 7 /11 E3 5/5 5/5 19/4+5 E4 5/5 >5 / >5 >18/22 ES 5/5 5/ - >12/ >20 El - ES 4.8/4.8 >4.2 I >4.3 >11.6 / >15.4 287 APPENDIX D Letter to Novice Participants Dear The staff of this summer's University of Maryland I.C.E. Demonstration Workshop is completing their final preparations for the NSF-sponsored program you and 23 other teachers from around the country will be attending. As part of the I.C.E. associat e s taff, I share with them the excitment of the upcoming events that are planned. I will be serving you in the chemistry labs where you will be given opportunities to practice and perform numerous chemical demonstrations. In conjunction with the summer workshop, I will be conducting a specia l exploratory research project designed to assess sc i ence methods workshops. You are invited to participate by evaluating two videotaped chemical demonstrations in the beginning and end of the workshop. The videotape viewing will be done on a purely voluntary basis. If you would be willing to view and comment on the two videotaped presentations of popular chemical demonstrations on Sunday any time after you have checked-in, simply return the enc losed re sponse card . Sending the card in no way obligates you once you arrive . It will inform me, however , of par t icipants' general intetntions. If you decide to participate when you arrive, we would be most happy to accommondate you. The videotape viewing will be held in a room adjacent to the lobby area of LaPlata Hall dormatory where you will be arriving. The staff members at the check- in table can direct you to the lobby area where you can view and comment on the videotaped demonstrations . Signs to direct you the videotape equipment room will also be posted. The videotaped demonstrations are each about five minutes long and the entire session lasts about 50 minutes. The viewing room will be conveniently set up on Sunday afternoon as well as Sunday evening , after the official check-in . 288 I hope you may find the occasion to participate in this optiona l viewing sess ion and research study. I would also like to wish you safe travels on your way to the University of Maryland. The ent ire staff look forward to your arrival. We hope the workshop will be a most pleasant and highly rewarding experience for you. Respectfully yours, Chris Clermont --- _____-,,...~. .. -----? ~--- - 289 APPENDIX E List of 50 Boxed Chemical Demonstrations Set Up for the workshop Number of Subjects Practicing Boxed Demonstrati on Demonstration, During Workshop 1 Concept Demo# ----------- - ------------- 1. Air Pressure 12 Balloon Inverted in a Flask #2 9 The Collapsing Aluminum can #21 ( 2 ) Egg into the Bottle #38 2. Density 9 9 Density Bobbing #31 4 Dens ity Column #32 Density: "Sink or swim" #35 11 Density: Not on the Level " #34 7 2 Mouth-to-Mouth Bottles #57 Water Fountain #83 3 . Gases and Their Properties 4 Carbon Dioxide - Three Activities #8 9 Carbon Dioxide Rocket Launch #11 9 7 Density Bobbing #31 2 MProoudthu-cttoi-oMn ooufth HBydortotgleens v#ia5 7t he Oxidation of Aluminum #67 4. Acids and Bases 8 5 Cabbage Patch Detective, #7 Red Cabbage Juice as Acid Base Indicator #72 7 1 Mouth-to-Mouth Bottles #57 A Tornado Show #80 5 . Boiling Point 5 Boiling Water at Low Temperatures #5 6. Capillary Action 12 Chromatography and Capil lary Action-Paper and Chalk 117 1 Total number of workshoP participants is 22 , wi th about hal f being novices . 290 7. Catalysis Blue Bottle #4 13 Rate of a Chemical Reaction #69 1 Oxidation of a Tartrate Salt #64 2 8. Chemical Reaction Blue Bottle #4 13 Carbon Dioxide - Three Activities #8 4 Carbon Dioxide Rocket Launch #11 9 Change in Oder (Esters) #12 5 Changing Hard Water to Fire Water (Sterno) #13 11 Dehydration of Sugar #29 5 Density Bobbing #31 9 Egg into the Bottle #38 2 Endothermic Reaction Ba(OH) 8H O and NH Cl #40 Evidence for Chemical Reactions #41 5 An Experiment in Alchemy - "A Penny for Your Thoughts" 1142 6 Production of Hydrogen Via the Oxidation of Aluminum Jr67 2 Rate of a Chemical Reaction #69 1 A Tornado Show #80 1 Lemonade or Grape Juice #52 6 Oxidation of a Tartrate Salt #64 2 Reaction of Zinc and Iodine #71 1 9. Chemcial Substances Carbohydrates: Dehydration of Sugar #29 5 Carbon Dioxide: Carbon Dioxide - Three Activities #8 4 Carbon Dioxide Rocket Launch #11 9 Esters: Change in Oder (Esters ) #12 5 Metals: An Experiment in Alchemy - "A Penny for Your Thoughts" #42 6 Flame Test #44 9 Proteins: Coagulation of Milk With Rennin #20 10 Polymers: An Amazing Pin Cushion #1 12 Soaps: Surface Tension , Waterproofing and Soap #77 3 291 10. Colloids 11 Changing Hard Water to Fire Water (Sterno) #13 10 Coagulation of Milk With Rennin #20 11. Color Changes 8 Cabbage Patch Detective #7 5 Red Cabbage Juice as Acid Base Indicator #72 9 6 Flame Test #44 Lemonade or Grape Juice #52 1 Magic Signs #54 12. Combustion 11 3 Cloud in a Jar #19 Burning candle #50 10 Inve s tigating the 4 Non-burning Towel #58 Ripple Your Chips Complex Ion Formation 13. 1 Magic Signs #54 14. Condensation 11 Cloud in a Jar #19 15. Crystals Crystallization from supersaturated Solutions of Sodium Acetate #27 6 16. Dehydration 5 Dehydration of Sugar #29 17. Discrepant Events 12 An Amazing pin cushion #1 13 9 Corn Starch Putty #24 Mini-Tornado #55 18. Endothermic/Exothermic Endothermic Reaction Ba(OH) 8H O and NH Cl #40 5 Dehydration of sugar (Exothermic) #29 1 Reaction of zinc and Iodine (Exothermic) #71 19. Energy 11 Changing Hard wate r t o Fire water (Sterno) 113 5 Dehydration of sugar #29 292 Endothermic Reaction Ba (OH) 8H O and NH Cl #40 Non-burning Towel #58 10 Reaction of Zinc and Iodine #71 1 Investigating the Burning Candle #50 3 20. Heat/Kindling Temperature Changing Hard Water to Fire Water (Sterno) #13 11 Dehydration of Sugar #29 5 Investigating the Burning Candle #50 3 The Paper Pot (Kindling Temperature) #65 5 21. Foods Cabbage Patch Detective #7 Red Cabbage Juice as Acid Base Indicator #72 5 Coagulation of Milk With Rennin #20 10 22. Indicators Cabbage Patch Detective, Red Cabbage Juice as Acid Base Indicator #7 , 72 5 Changing Hard Water to Fire Water (Sterno) #13 11 Magic Signs #54 1 Mouth-to-Mouth Bottles #57 7 A Tornado Show #80 1 23. Kinetics & Reaction Rates Blue Bottle #4 13 Coagulation of Milk With Rennin #20 10 Rate of a Chemical Reaction #69 1 Oxidation of a Tartrate Salt #64 2 24. Mixing and Mixtures Chromatography and Capillary Action- Pape r and Chalk Jr17 12 Density: Not on the Level" #34 11 A Tornado Show #80 1 25. Oders Change in Oder (Esters) #12 5 26 . Phase Change Balloon Inverted in a Flask #2 12 Boiling Water at Low Temperatures #5 5 The Collapsing Aluminum Can #21 9 The Paper Pot #65 5 Which Will Evaporate First #85 3 293 27. Physical Properties and Physical Changes An Amazing Pin Cushion #1 12 Boiling Water at Low Temperatures #5 5 Corn Starch Putty #24 13 Density Column #32 9 Density: Not on the Level" #34 11 Density: "Sink or Swim" #35 4 Egg into the Bottle #38 2 Flame Test #44 9 Mini-Tornado #55 9 Water Fountain #83 2 Which Will Evaporate First #85 3 28. Polarity Surface Tension, Waterproofing and Soap #77 3 29. Reaction Mechanisms Blue Bottle #4 13 30 . Redox (Oxidation-Reduction Reactions) Production of Hydrogen Via the Oxidation of Aluminum j/ 67 2 31. Scientific Reasoning An Amazing Pin Cushion #1 12 32. Separation Techniques Chromatography and Capillary Action - Paper and Chalk U7 12 33. Solid State Reactions Endothermic Reaction Ba (OH) 2 8H2o and NH 4Cl #40 Evidence for Chemical Reactions #41 5 34 . Sublimation Reaction of Zinc and Iodine #71 1 35. Solubility Crystallization from Supersaturated Solutions of Sodi um Acetate #27 6 Evidence for Chemical Reactions #41 5 Mouth- to -Mouth Bottles #57 7 294 36. Surface Tension Surface Tension, Waterproofing and Soap #77 3 37. Suspensions ** Corn Starch Putty #24 13 38. Tornados Mini-Tornado #55 9 A Tornado Show #80 1 39. Vacuum Balloon Inverted in a Flask #2 12 The Collapsing Aluminum Can #21 9 Production of a Chemical Foam A-23 7 * Rate of a Chemical Reaction #69 1 * Oxidation of a Tartrate Salt #64 2 * Which Will Evaporate First #85 3 * Reaction of Zinc and Iodine #7 1 1 * Investigating the Burning Candle #50 3 * Water Fountain #83 lp 2 * Lemonade or Grape Juice #52 6 * Flame Test #44 9 * An Experiment in Alchemy - "A Penny for Your Thought s " #42 6 * The Paper Pot #65 5 * Evidence for Chemical Reactions #41 5 ? Genie in a Bottle 6 * Endothermic Reaction Ba(OH) 8H O and NH Cl #40 2 2 4 6 * Density: "Sink or Swim" #35 4 * Density: Not on the Level" #34 11 * Ripple Your Chips 4 * Surface Tension, Waterproofing and Soap #77 3 * Density Column #32 9 * Dehydration of Sugar #29 5 * Density Bobbing #31 lp 9 * Crystallization from Supersaturated Solutions of Sodium Acetate #27 6 * Corn Starch Putty #24 13 * The Collapsing Aluminum Can #21 lp 9 * Coagulation of Milk With Rennin #20 10 * Cloud in a Jar #19 11 * Chromatography and Capillary Action-Paper and Chalk #17 12 * Carbon Dioxide Rocke t Launch #11 9 * Change in Ode r (Esters ) #12 5 * Changing Hard Water to Fire Wate r (Ste rno) #13 11 * Carbon Dioxide - Three Activities #8 4 * Cabbage Patch Det ective , Red Cabbage Jui ce as Ac id Base 295 Indicator #7, 72 5 * Boiling Water at Low Temperatures #5 5 * Blue Bottle #4 13 * Balloon Inverted in a Flask #2 12 * An Amazing Pin Cushion #1 12 * Magic Signs #54 1 * Dehydration of sugar by Sulfuric Acid #29 7 # ' sand Names mismatch Unnamed Demo 1 (Crushing Cans Reversibly, #25) 3 Unnamed Demo 2 (The Water Exchange: An Equilibrium Model #82 ) 2 Unnamed Demo 3 2 Unnamed Demo 4 6 Unnamed Demo 5 4 * Egg into the Bottle #38 (2) * Non-burning Towel #58 10 * Mouth-to-Mouth Bottles #57 7 * Mini-Tornado #55 (9) 296 APPENDIX F Two-Week Workshop Schedule WEEK !/ICE WORKSHOP B/SESSl()-.1 VJULY 13 - JULY 17 Monday Tuesday Wednesday Thursday Friday --- - -------- - -- ----- ------ -- -- - - ----- - --- ------------------------7:00 I <----------------- Breakfast (7:00 - 8:00) ---------------> - - - -- - ------ - --- -------- - -- -------- ----- -- 8:30 Participant / I What Ma kes !Demonstration I Participant I Cos. t , Ti me , St a ff I a r, E ff,:, ct i ?J e I Sh oc.,Js. w i th a I s,,,,a 1::o Sh op I Sc<.fety, & 9:00 Intros. , a.n d I Demo ~' I Theme I& Individual I Other Cour?se I I I ' .! i deot a pe "L imitir,g 9:30 P,:,gistrat1onsl I An a l ysis f;' ':' :tgE- n t S 11 C- 32 19 I C- 3 219 C- 3 219 I C- 32 19 I 10.001 ------- - - ------------ - ------------- ---- - -- - ------ - ----- - -- - - -- ----- - - - - I Intro to I I Intro to Chem l I !0:301Demonstr?a.ti onsl Demo Pr'O\cticel Camp/C-3219 I Demo Pract i ce! Demo I C-321'? I in Labs I Demo Practic e ! in Labs I Pr act i ce 11: 00 I - ------- - - - ---1 i n La bs I I in Labs I Lc,g i s. t i cs & I 11:301 Expect atio ns I C-1 208./ 1228 C- 1208/ 1228 I C- 1208/1228 IC-1208/ 1228 I C-3219 I I I I 12.00l----------- - -- - --- - - -------- - ---------- - ---- - - ---- - --- - -- - - - - - ------ - - - 1 <-------------------- Lunch (12:15 - 1:00) -------------------> 1------------- ---- - ----------------- ----- --- - - ------- - --- -- ---- - - -- - - 1 : 40 I Ir, tr o tc, Labs.I Pres.er, ters. Public F'r?esen ter?s I I g?.ther supplies I Pr?es.en tat ion s I gather s.upp l i es 2:00 I 1---------------1 of Ass igned 1- -- -- -- -- ---- -1 -- - - - - -- -- - C-1208/1228 I !Demonstr at ions! Mond ,e.;,' Camp I Tues Camp 2:30 I - - - - - - - - - - - - - I F' Ubl I C I I Te am p i l O t I Te am p i I O t ID,;,mo Practicel Pr?esentations 11 2 pa rticipants l Th e ir De mo s lT heir Demos 3:00 I in Labs I of Assigned I I C- 140 2 C-1402 Demonstrations! - -- -- - -- - - - - 3 :30 C-!208/ 122 13 I C- 14 02 1- - ------ - --- -- 1 F'u bl ic 112 p a rticip a nts! I Less Magic IP resentatio 4:00 I 1- ---- --- -- - ---IMore ln st ru c- l of Second C-1402 I Videotape I t i on I D l- - --- - -- --- ---- -- -- ---- - - - - - - - -- -- -- --- - -- ---- -- -- - - -- -- - -- - - --- -- - - - - - 7:00 Vid eot a pe I Ch emi stry Can l Gue st Lect ure ! An a l ysis Be Fun ! De mo nstrat i on I 7:30 ( individu a ll y 8, Shan,ashiril R. Perkins. s c he duled ) I 8:00 !Optional: Demo C-1402 I C-1402 I I Pr? a ct Ice in La b I I I 8:30 1-------------- - ------ - --------------- - --- - -- - -- -- - --- - -- - - - - -- ------ - - - INSTITUTE FOR CHEMICAL EDUCATICN/FACULTY AND STAFF Ann Benbow: Univ. of Mary land ( Inst ructor / Chem Camp Dir ector ) Mi ke Bel l ama: Uni v . of Man? l an d ( I CE F iel d Ce-n ter Director) Chris Cler mont: Un iv. of Ma r y l an d Sc hools ( Instruct or / S t ockroom Contact> Ka th y Mose r: Univ, of Mary l an d (G r aduate St ud e nt Ass i stant) Tom O' Br i en: SUNY/ Binghamton 8 : 30 Humor i n t he IOu ick1 Ki ds l--- - --- ---- - - - -- ---- - --- - - --- - -- --- -- ------ ---- -- ---------- ----- - ---- - -- 1: 40 I Pre =, e,r, t ers. gat her su pp 1 i es Pub l ; c I Pr ese n ters I fo r? Pub 1 i c F'r? e s e r, t ?. t , on s. I Pr? es e r, ;_ ;_t I c.,n s. I gc>. the r s.u pp 1 i es 2 ? 00 1- - - - --- - - - ---1--- - - - - - -- - - --- 1 o f Outr e,c h 1- - --------- - -- 1 I I IDemon s.tr?,._t, onslPub l i c I 2; :;o ? cd:,l ic ?re~e nt a t ion ! l- --------- ---------- ---- -- -- ---- -- -- -- - - - - - ------------- - - -1 7 : 0 0 I I I Pr ofessiona l I An I CE I I De mo nst r ato r s l Ou t in g/ Dinn e r 7: 30 I on \,'ideotape I I Opti ona l: I 8:00 De mo Pr act i ce ! I I I i n La bs. I I 8 : 30 1---------------- -- -- -- - ---- -- ---------- - ------- - ----- - ------ -- ----- - - - - INSTITUTE FOR CHEMICAL EDUCATION/ FACULTY AND STAFF Ann Benbow: Un iv . o f Mary l and ' , and i nvestiga ti?Je , Accurate, conci<::.e, gr?a de level appropri a te e x pl anations should evolve as part of the presentation. As much as possible, "let the demonstration do the talking" - ask more, t e l l l e '=? '=? ? Se e t h e Dem c, n '=? tr? a t i o r, An c1 l v '=? i '= F c, r m an d a c c om p a n y i n g handout for specific points to consider. d. Pr?e<:.entation<:. ma y include v aria tion<: that in c r?ease the utilit y a r,d a.pplicabilit y of the demon<:.tra.tion, but if the y differ signific antl y from that described in the ICE Guidebook ( other than shortening a demonstration series>, the y s hould be approved by one of the instructors. Note: This demonstration, and all subsequent public present at ion<::. will be videotaped for later ind i vidua l and instructor provided ar,alysis, The intent is to pr?ovide b o th positive support and cr-itical feedb ack to a<:s i <:.t each ind i vidua l presenter to a higher level of performance competenc y and conceptual ur,der? <::. tanding, A ll par-ticipant<::. are expecte d to attend a n indi vidually scheduled evening vi d eota pe a n a l ys i s <::e<::sion for this first public demonstration . Su b<::.equent ana l ys is sessions for other- pr-esentations w il 1 be optional. 3 . Perfor-m as many of the bo x ed demonstrations from the I CE Guidebook as possible gi v en that for each dem onstration you per-for-m y ou should: a. Mar-k the Demon s tr a tion Log in each bo x to indic a te yo u ha v e completed it a nd c a n serve a s a po ss ib l e re source per son 301 for others. A Key ingredient to the anticipated success of the 1,1or?kshop is the 1A1i 11 in gnes.s of participants to share thi: ir? e x pi:rience a nd Knowledge with others, b, Keep a 1,,,ritten, persona l record that in cludes: (1) several crit i cal pr?e- and post-demc, questions. you would ask yo ur class or aud i ence ( 2) real-1 ife ctppl ications or e x amples of the principle / concept being demonstrated 0:3) additionctl ?Jariations (opt i o n a l) (4) s.uggestions for improvement ( i f thes.e are necessary or highly desirable, please also leave a brief note in the bo x ed demonstrc1. t i o r,), 4, Work w ith yo ur assigned team to develop a 30 minute Demonstration Program for the 5-8th grade youngsters attending the Chem Camp which runs concurrently with the teacher workshop during the mornings of the second week. A typed handout that des.cribes at -home extensic,r,s. , rea l-1,Jor ld app li cat i ons. , related questions or puzzles, etc ., should also be prepared for distribution to the youngsters , your own "favorite" demonstration you brought with you, or from other available demonstration sourcebooks. Again, if the demonstration setup i s not a l ready bo xe d, a Demonstration Approv al For m mu s t be turned in at least 24 hours before the anticipated trial run to a1 101A1 time to as.sembl e the necessary s.uppl i es (when poss. i bl e 302 articipants are encouraQed to qather their own household-t ype upplies) and to provide-a quality check. Once appr?oved, a ormal writeup that matches the ICE guidebook style is required in the case of a demonstration from the ICE Guidebook, a cr itical review v,,ill take the place of a formal ..,,_,riteup). Yo u \'-Jill be alloted 10 minute s to perform tt-,e dem onst r?ation w ith 5 minutes for group discussion a nd a nal ysis. The final draft of the writeup wi l 1 be due on Thursday morning, July 23 so that copies can be xero xed for al 1 workshop participants. 6. De?,,elop ar,d ,;ubmi t b >? l,Je dnes.da.,?, July 22, an outline for a 1-hour? Ou tr?ec>.ch ~r-ogr eo.m to be car-r i ed out in one's. home district. a. Include: genera.I the me, intended audience, goals., and the sequence of activities and demonstrations including ti ties and source citations .dily a?,,ailable chemicals. and equipmer,t that you were able to obtain from "garbage," grocery stores, etc. A brief (one-ha lf to one page) wri teup should be turned into a n instructor by Thursd eo.Y morning, Jul y 23. 8, Help the ICE staff check the conditions of the pre - bo x ed demonstrations and complete evaluation materials before le aving the workshop to return home. FRANK & ERNEST BOB THINE$ AH, i1-'Ell-e '(Ou ~E,, Wlr.itgu:--? X. WANi you To fc:EAP THl,S" ~opo..rN-. N,JP TEL.I.. ME IT~ A G~~AT IDEA. 303 DEM~STRAT I ~S: AN INTERACTIVE INSTRUCTICNAL STRATEGY FOR THE SCIENCE CLASSR001 Demonstrations - A Part of Science and A Part of Learning The dictionary defines the word demonstrate as to manifest or reveal - to display, oper?ate, and e x plain. In science instruction this occurs when the teacher manipulates models, chemica.ls, apparatus, etc in such a way a.s to introduce, illustra.te, or cha.llenge students' understanding of an important concept by engaging their a.bility to observe, question, and rea.son. This process is of course, the essence of science - science is not just a ~,ighly organized body of unalterable facts and theories to be memorized; it is also a method of inquiry, and a wa.y of looking at the world. In addit i on to reflecting the nature of the scientific enter?prise, us1 r. g demonstrations as a.n instructional strategy reflects. what common sense and educational research tel 1 us a.bout the process of learning. Learning is an active process the learner's mind must be engaged to opera.te on a.nd tra.nsform sensory data into concepts and models within his/her own particula.r mental framework know ledge can not be poured out (or injected under pressure ) from the teacher ' s mind into the learner ' s, For lea.rners who a.re still at a concrete operational stage with r?es.pect to science concepts (which includes most high school students accor?ding to a. number of Pia.getia.n type studies>, words alone a.re not an adequate sensory e x perience to promote reasoning and understanding. Thus, if one ' s objective a.s a teacher is to promote more than "regurgitation" type learning, it is essenti a l that you provide opportunities for your students to "see" scienc e in a.ct i on. II Demon~tra.tions vs Laboratory Experiments: Not a Question of Either / Or, But Rather When The Chinese proverb "A picture is worth a thousand words" is logically e x tended to another Chinese prov erb: "I hear &: I forget - I see&: I remember - I do&: I understand,' Clearly, laboratory e x periences should be an important pa.rt of science classes. Given this belief, why not just haOJe student laboratory experience s a.nd foroet a.bout demonstrations? Beyond the obvious fa.ct that reliance on any single "right" instructiona.1 stra.tegy is educationally unsound, demonstrations can open up new possibilities not a.va.ilable through direct student laboratory experiments, Practical considerations such as safety, a.va.ila.bility and/ or e x pense of equipment / supp 1 i es, and ti me 1 i mi t both the number and type of experiments students can perform within a given clas s. Also, teacher demonstrations provide a.n environment where: (a) student ideas/ hypotheses can be ,immediately tested fluid). Intended Audience: Kindergarten through adult. Presentation: Each of these demonstration/experiments must be seen so volumes may need to be adjusted for audience size. Time Required: Takes less than 15 minutes to set-up if the materials are readily available. Demonstration and discussion time will depend upon grade level and complexity of explanation/discussion. Clean-up time is usually under 5 minutes. Materials: 1 -~ 00 ml or larger cylinder 10 to 20 ml each of these liquids: mercury (optional) glycerin water (with optional food coloring) cooking oil or mineral oil alcohol (isopropyl or ethanol) samples of these solids sized to fit into the cylinder: lead sinker or steel bolt rubber stopper piece of plastic piece of oak or other substance with a density between 0.9 and 0.8 glee. cork NOTE: You may want to try other materials such as corn syrup, motor oil , saturated salt solution, ditto fluid, aluminum, brass, rocks or chalk. I II 330 Density: "Density Column" pg.-2 Safety and Disposal: Mercury is extremely toxic and should be handled with care to aboid prolonged or repeated exposure to the liquid or vapor. Continued exposure to the vapor may result in severe nervous disturbance, insomnia, and depression. Continued skin contact also can cause these effects as well as dermatitis and kidney damage. Mercury should be handled only in well-ventilated areas. Mercury spills should be cleaned up immediately by using a capillary attached to a trap and an aspirator. Small amounts of mercury in inaccessible places should be treated with zinc dust to form a nonvolatile amalgam. The density column may be kept for months if handled carefully. A separatory funnel might be used to recleaim the mercury when the density column is dismantled. DO NOT DISCARD MERCURY! Save for reuse or collect cleaning! Procedure: 1. Place the liquids in the cylinder. If immiscible liquids are used the order of addition is not important. Miscible liquids will require addition in the order of densities from the greatestto the smallest (See Figure 1) . Caution should be used when pouring mercury (See Figure 2). 2. Add one or more of the solids and note where they come to rest. 3 . Students note the relative densities of each substance. 4. If given a list of the names of the substances and a list of the densities represented, they can match the substances, in order, with their densities. Explanation: All materials have characteristic densities. This demonstration allows for a discussion of relative densities. The less dense materials float above the more dense. Each substance sinks into another fluid until it displaces its own weight (Archimedes principle) . For example, a lead sinker, which has a mass of 11 .3 grams, would displace only 1 gram of water. This is because lead atoms weigh about 11 times as much as water molecules. We see that water cannot support the lead sinker because 11 .3 grams of water cannot be displaced as the sinker sinks into the water. 112.. 331 Density: "Density Column" pg.?3 When the lead sinker reaches the mercury, however, it sinks until about 80% of its volume is in the mercury. At that point the sinker has displaced 11 .3 grams of mercury and is supported by the mercury. Curriculum Integration: This study is part of the characterisitcs of matter unit or in a density unit. References: Herbert, Don, Mr. Wizard's Supermarlset Science., Random House, New York, NY, 1980, pp.14. Windholz, Marcia, The Merck Index, 10th edition, Merk and Company, Rahway, NJ, 1983. Weast. R.C., CRC Handbook of Chemistry and Physics., 49th edition, CRC Press, Cleveland, OH, 1968. Smot, Robert C., Modern Chemistry, Merrill Publishing Company, Columbus, OH, 1983, pp.35. Contributors: Debe Dankel, Antioch Upper Grade School, Antioch, ILL 60002. Jane Kotlewski, Custer High School, Milwaukee, WI 53209 . Don Morton, Kellogg High School, Little Canada, MN 55117. Reviewer: /13 332 APPENDIX L chemical Demonstration Log - . ur na.rne indicating tnat you have D~Met-JSTRATl Cl'l T !TLE/'*: Pl???? pl?'' a ,o,cK macK ,,o,od ,o ., 1, ,,,,, ?? ? ,,,oocc? c=o"''' ,,. 0a,moostc?'"" ,ad ,.o P '"00 ,;, ,am? a,moo