ABSTRACT Title of Dissertation: A MODERATED MODEL OF THE JOB SATISFACTION AND TURNOVER INTENTIONS OF AFRICAN-AMERICAN EMPLOYEES: INVESTIGATING THE ROLE OF PERCEPTIONS OF FIT AND RACIAL CLIMATE IN THE THEORY OF WORK ADJUSTMENT Heather Zoe Lyons, Doctor of Philosophy, 2003 Dissertation directed by: Professor Karen O?Brien Department of Psychology The work lives of African Americans are understudied (Brown, 1995; Fitzgerald & Betz, 1994). The purpose of the present study is to address this dearth in the literature in three ways: (a) providing descriptive data related to the work experiences of African- American employees, (b) incorporating racial climate into the Theory of Work Adjustment, a theoretically-based model that predicts employee job satisfaction and turnover intentions, and (c) attempting to address methodological limitations raised in this body of literature. Self-report data were collected from African American employees (n=212), assessing their levels of perceptions of fit, sense of departmental racial climate, job satisfaction, and intentions to quit. In two separate regression equations, perceptions of fit predicted job satisfaction (R 2 =.43) and intentions to quit (R 2 =.17). However, racial climate did not emerge as a moderator, as hypothesized. Implications for practice and research as well as limitations are addressed. A MODERATED MODEL OF THE JOB SATISFACTION AND TURNOVER INTENTIONS OF AFRICAN-AMERICAN EMPLOYEES: INVESTIGATING THE ROLE OF PERCEPTIONS OF FIT AND RACIAL CLIMATE IN THE THEORY OF WORK ADJUSTMENT By Heather Z. Lyons Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2003 Advisory Committee: Professor Karen O?Brien, Chair Professor Cheryl Holcomb-McCoy Professor Marylu McEwen Professor Robert Steele Professor Franklin Westbrook ? Copyright by Heather Zoe Lyons 2003 ii Acknowledgements Thank you to my mentors, friends, family, and others who helped with this project: Karen M. O?Brien and the rest of my dissertation committee (Professors Cheryl Holcomb-McCoy, Marylu McEwen, Robert Steele, and Franklin Westbrook); Professor Janet E. Helms and The Research Divas (Jamila Codrington, Regine Talleyrand, Stacey Jackson, Rahsheda Perine, and Miriam Phields); my coders and friends (Seth Hayes, Bradley Brenner, and Lisa Baker); Adolphus Belk; Rhome Anderson; Denise Dove; Osei Appiah; Colanda Howard; Mom and Dad and family; my intern class at Georgia State University (Qinza Najm and Carmenlita Sayles); The Afro American Studies Program; Betty Padgett in the Psychology Department; Monica Jackson, Paul Hanges and Amy Nicole Salvaggio for their help with data analysis; The Black Graduate Association at the University of Maryland; Professor Elizabeth Cole; staff and faculty at Georgia State University; Northwest Family Center; Atlanta Public Schools; Richmond Public Schools; United States Department of Education; United State Department of HUD; County of Essex; United States Department of Justice; D.C. Office of Corporate Counsel; Sprint; the Centers for Disease Control; the State of Louisiana; and Progress Energy. iii TABLE OF CONTENTS List of Tables iv List of Figures v Chapter 1: Introduction 1 Chapter 2: Review of Literature 17 African Americans and Work: Historical Context 17 Theory of Work Adjustment 21 Race and Job Satisfaction 43 The Theory of Work Adjustment and Race 50 Racial Climate 57 Chapter 3: Statement of Problem 73 Chapter 4: Method 83 Participants 83 Measures 83 Procedure 93 Chapter 5: Results 96 Preliminary Analyses 96 Research Questions 99 Hypotheses 108 Chapter 6: Discussion 118 Describing the Work Experiences of African Americans 118 Racial Climate and the Theory of Work Adjustment 120 Perceptions of fit & African American Employees 125 Creating a Theoretically-Based Measure of Fit 129 Limitations 131 Implications 134 Appendices Appendix A: Consent Form 143 Appendix B: Biodata Sheet 144 Appendix C: Perceptions of Fit 145 Appendix D: Workplace Compatibility 147 Appendix E: Racial Climate Scale 149 Appendix F: Job Satisfaction 151 Appendix G: Intentions to Quit 153 Appendix H: Debriefing Form 154 References 155 iv List of Tables 1. Biodata 84 2. Descriptive Statistics for Study Variables 101 3. Intercorrelation Matrix for Study Variables 101 4. Categorization of Responses to the Open-Ended Question 102 5. Summary of Past Research with European American Participants 105 6. Regression Analysis with Job Satisfaction as Outcome Variable 111 7. Regression Analysis with Intentions to Quit as Outcome Variable 114 v List of Figures 1.Figure Depicting the Hypothesized Relationship between Study Variables 141 2.Figure Depicting Resulting Relationships between Study Variables 142 THEORY OF WORK ADJUSTMENT 1 Chapter 1 Introduction The work lives of African Americans are understudied (Brown, 1995; Fitzgerald & Betz, 1994). Ryan (1999), in a metanalysis of research focused on career outcomes (e.g., congruence, career choice, vocational identity), found that only 21% of studies in this area reported the race of their participants. Of the studies that reported race, 68% of participants were European American compared to 21% who were African American. Examination of the vocational histories and present day realities indicate that differences exist in the work experiences between African Americans and European Americans which may mean that research on the former group may not generalize to African Americans (Berry & Blassingame, 1982; Williams, 1999). The purpose of the present study was to address this dearth in the literature in three ways: (a) providing descriptive data related to the work experiences of African-American employees, (b) incorporating the role of culture into, the Theory of Work Adjustment (i.e., TWA), a theoretically-based model that predicts African- American employees? job satisfaction and turnover intentions, and (c) attempting to address methodological limitations raised in this body of literature. Perhaps because research on the work experiences of African Americans is relatively underdeveloped, much of the existing scholarship in this area has not been empirical. That is, a good deal of scholarly writing related to work and African Americans is focused on outlining propositions and hypotheses for future research THEORY OF WORK ADJUSTMENT 2 (e.g., Cheatham, 1990; Hackett & Byars, 1996; Helms & Piper, 1994). Therefore, the first purpose of the present study was to collect descriptive data that provide information about African-American employees? work experiences. In an attempt to gain a better understanding of African Americans? perceptions of racial climate, perceptions of fit, job satisfaction, and turnover intentions, the present study presented means, ranges, and standard deviations of these variables for the present study?s sample. The little empirical research that exists on African Americans? work behaviors and work lives has been criticized for being largely atheoretical (see Byars, 1999) or focused on simplistic, nominal conceptualizations of race (see Helms & Piper, 1994). In light of these criticisms, the second purpose of the present study was to build on past research and theory to advance knowledge regarding the perceptions of the job satisfaction and turnover intentions of African-American employees. Unlike past studies that are either atheoretical or use between-group designs (i.e., comparing African Americans and European Americans) to examine vocational outcome variables with African-American employees, the present study utilizes a theoretical model, the Theory of Work Adjustment (TWA; Dawis, England, & Lofquist, 1964; Dawis & Lofquist, 1984). Within the framework of the TWA, the role of within- group variables, specifically perceptions of fit and of racial climate were used to predict job satisfaction and turnover intentions. Additionally, as prescribed by the TWA, the ability of job satisfaction to predict turnover intentions was tested. THEORY OF WORK ADJUSTMENT 3 Addressing these limitations of past research, the present study contributes to our understanding of African-American employees? job satisfaction and turnover intentions. Job satisfaction refers to the extent to which employees? jobs fulfill their needs and values (Dawis, England, & Lofquist, 1964). Turnover refers to the departure of employees from an organization either voluntarily or involuntarily (e.g., termination; Jewell, 1998). Turnover intentions (also referred to as intentions to quit) differ from actual turnover, in that turnover intentions are employees? self-report of their intentions to quit or stay in the workplace (Mobley, 1982). In the present study, voluntary turnover intentions were of interest. A related concept, tenure, is the length of time that employees remain in an organization (Dawis & Lofquist, 1984). In the TWA literature, measures of turnover intentions are often used as a proxy for actual tenure or in addition to the measurement of actual tenure (e.g., Saks & Ashforth, 1997; Taris & Feji, 2001). Job satisfaction and turnover intentions are important vocational outcomes because of the variables? connections to constructs of importance to employees and employers. Namely, job satisfaction is thought to lead to tenure (Dawis & Lofquist, 1984), life satisfaction (Near, Rice, & Hunt, 1983), and lower levels of absenteeism (Goodman & Atkin, 1984). Turnover intentions have been found to lead to actual turnover (Jewell, 1998; Mobley, 1982). Turnover can result in expenses to organizations who must engage in selection and training of new employees (Darmon, THEORY OF WORK ADJUSTMENT 4 1990; Jewell, 1998). Because of the paucity of research with African Americans, little is known about the determinants of job satisfaction and turnover intentions with this population. Job Satisfaction Existing research uncovered inconsistencies in the comparison of the job satisfaction of African-American and European-American employees (Alexander & Kauffman, 1999; Austin & Dodge, 1992; Brenner & Fernstein, 1984; Greenhaus, 1990; Mueller, Finley, Iverson, & Price, 1999; Tuch & Martin, 1991). Some of this research suggested that African Americans were not as satisfied with their jobs as their European-American counterparts (e.g., Austin & Dodge, 1992; Greenhaus, 1990; Tuch & Martin, 1991). Other research indicated that African Americans actually reported higher levels of job satisfaction than European Americans (e.g., Brenner & Fernstein, 1984; Mueller et al., 1999). Still, other research found no differences in the job satisfaction of African-American and European-American employees (e.g., Alexander & Kauffman, 1999). Clearly, additional research is needed to explicate the comparative levels of job satisfaction between these groups. Furthermore, this research should attempt to identify the determinants of job satisfaction for African Americans. Turnover Inconsistencies also have been reported in research comparing the actual turnover rates of African Americans and European Americans. No research was THEORY OF WORK ADJUSTMENT 5 located comparing levels of turnover intentions between African Americans and European Americans or other racial groups. Borjas (1984) found that in comparisons of young African-American and European-American men, African-American men were more likely to be laid off than European-American men but less likely to quit their jobs than European American men, suggesting differing turnover rates for these two groups. Flanagan (1978) reported that pay discrimination between African- American and European-American employees resulted in a higher quit rate for African-American employees. These findings suggest the need to investigate the determinants of these behaviors for African Americans. Therefore, as stated previously, the second purpose of the present study was to advance knowledge regarding the predictors of the job satisfaction and turnover intentions of African- American employees. Theory of Work Adjustment One of the most widely used theories to explain job satisfaction and tenure is the Theory of Work Adjustment (TWA; Dawis et al., 1964; Dawis & Lofquist, 1984). The TWA describes the ways that person and environmental factors interact to predict employees? satisfaction with their jobs and workplaces? satisfaction with their employees. The interaction between person and environment factors is conceptualized in terms of person-environment fit. In the TWA research, this fit is called correspondence and is described along two dimensions: skills and needs. According to the TWA, the greater the correspondence between the skills that employees possess THEORY OF WORK ADJUSTMENT 6 and those required by the work environment, the more satisfied the workplace is with their workers. The TWA researchers also hypothesized that employees are satisfied with their work environments when the needs that they possess related to work correspond with the reinforcements offered by their work environments. Furthermore, when employees are satisfied with their workplaces and workplaces are satisfied with their employees, employees will remain in the workplace, thus, satisfaction is hypothesized to predict tenure. Methodological Limitations in TWA Research One purpose of the present study was to address the methodological limitations recognized in measuring person-environment fit by using a newly developed perceptions of fit measure. In the past, the use of difference scores (i.e., scores created by subtracting employees? values, needs, or ability assessments from the values, needs, or ability requirements of their workplaces) dominated measurement of fit. However, using difference scores resulted in a variety of measurement and conceptual problems including lack of reliability (Edwards, in press) and the ambiguous meaning of a construct (i.e., fit) that resulted from the combination of two distinct elements (Edwards, in press; Kristof, 1996). Literature in this area suggested that new techniques are needed to measure fit (Edwards, 1991; Edwards, in press; Kristof, 1996). One of these suggestions was to use employees? direct perceptions of fit to assess this construct (Kristof, 1996). That is, employees are asked directly about their perceptions of the extent to which they feel that they fit in THEORY OF WORK ADJUSTMENT 7 an organization (e.g., ?To what extent are the values of the organization similar to your own values??; Saks & Ashforth, 1997). However, the literature has yet to embrace a particular method to measure or an actual instrument to assess fit. Furthermore, none of the measures commonly used are theoretically based. Therefore, the present study utilized a perceptions of fit measure designed by the present author and based on the TWA. A pilot study of this direct perceptual measure of fit indicated that this measure has adequate psychometric properties (Lyons & Brenner, in preparation). Culture and the Theory of Work Adjustment The tenets of the TWA have been tested in many empirical studies, and findings support the theory?s ability to explain job satisfaction and tenure, often measured as turnover or turnover intentions (Breeden, 1993; Bretz & Judge, 1994; Rounds, 1990; Rounds, Dawis, & Lofquist, 1987; Saks & Ashforth, 1997; Taris & Feij, 2001). However, only one empirical study investigated the ability of the TWA model to explain job satisfaction and turnover intentions with an African-American sample (Lovelace & Rosen, 1996). The work of Lovelace and Rosen indicated that African-American employees reported lower levels of person-environment fit than European-American and Latino/a employees. Additionally, the authors found that perceptions of fit were positively related to job satisfaction and negatively related to turnover intentions, findings that are consistent with the TWA. Because of the lack of THEORY OF WORK ADJUSTMENT 8 research testing the applicability of the TWA with an African-American sample, additional research in this area is needed. In addition to the aforementioned study, researchers have begun to draw attention to the possible importance of cultural variables in the TWA model (Dawis, 1994; Dawis & Lofquist, 1993; Fitzgerald & Rounds, 1993; Rounds & Hesketh, 1994; Tinsley, 1993). Specifically, researchers offered suggestions for the hypothesized role of race in the TWA. These suggestions generally have taken three directions: consideration of a mediated model of the TWA (Dawis, 1994; Dawis & Lofquist, 1994; Swartz-Kulstad & Martin, 2000), race or other cultural variables as moderators (Fitzgerald & Rounds, 1993; Rounds & Hesketh, 1994; Tinsley, 1993), and race or other cultural variables as independent variables in the TWA model (Rounds & Hesketh, 1994). Heeding these suggestions, the present study investigated the role of culture in the TWA in an attempt to advance knowledge regarding the relevance of the model for African-American employees. More specifically, the present study tested the ability of racial climate to moderate the relation between perceptions of fit and job satisfaction as well as the relation between perceptions of fit and turnover intentions. Mediated model. Dawis (1994) and Dawis and Lofquist (1993), original TWA theorists, predicted that cross-cultural research on the TWA model with women and people of color would result in validation of the TWA with these populations. They suggested that culture would not change the nature of the relationships that they THEORY OF WORK ADJUSTMENT 9 outlined in their model. Instead, they endorsed a mediated model where cultural variables may influence correspondence which in turn would influence the TWA outcome variables (i.e., job satisfaction, satisfactoriness, tenure). For example, cultural variables may be related to different experiences and opportunities which would make some individuals more correspondent with particular environments and, in turn, more satisfied with and more likely to remain in particular environments. Therefore, based on these authors? suggestions, correspondence may adequately capture culture?s influence on job satisfaction and turnover intentions. Moderated model. Other authors suggested that the TWA model needs to be amended to include the influence of employees? culture (Fitzgerald & Rounds, 1993; Rounds & Hesketh, 1994; Tinsley, 1993). Two amendments have been proposed: culture as a moderator in the model and culture as direct influence on job satisfaction. More specifically, Fitzgerald and Rounds (1993), Rounds and Hesketh (1994), and Tinsley (1993) proposed that culture might indirectly influence the TWA variables by moderating relationships in the model. For example, Rounds and Hesketh hypothesized that workplace discrimination can either mediate or moderate the relation between fit and job satisfaction. Rounds and Hesketh stated that this hypothesized relationship can be extrapolated to account for the influence of discrimination on the relation between other TWA variables including the relation between fit and tenure (which is most often studied in TWA literature as turnover intentions). Moderation in this example would mean that the relation between fit and THEORY OF WORK ADJUSTMENT 10 turnover intentions would become harder to predict in the presence of workplace discrimination. Rounds and Hesketh advised researchers to investigate these relationships. Unlike the original TWA authors (Dawis, 1994; Dawis & Lofquist, 1993), Fitzgerald and Rounds (1993), Rounds and Hesketh (1994), and Tinsley (1993) hypothesized that correspondence alone would not account for the influence of culture in the TWA model. Culture as a direct influence. Alternatively, Rounds and Hesketh (1994) offered the possibility that cultural variables have a direct influence on job satisfaction. Again, Rounds and Hesketh suggested that future research consider the ways that cultural variables can influence other variables in the TWA model directly (e.g., tenure). Researchers have presented several cultural variables that may be relevant to the TWA: experiences with sexual harassment (Fitzgerald & Rounds, 1993), background characteristics (Tinsley, 1993), and on-the-job discrimination (Rounds & Hesketh, 1994). Related to on-the-job discrimination, researchers found a link between racial climate and job satisfaction (Ensher et al., 2001; Holder & Vaux, 1998; Hughes & Dodge, 1997) and discrimination and turnover (Flanagan, 1978) and turnover intentions (Shaffer et al., 2000). Racial Climate Racial climate is defined as the atmosphere in an organization as it relates to the ways in which (a) interpersonal processes (e.g., warmth, acceptance, support), (b) decision making processes (e.g., involvement in decision making), and (c) reward THEORY OF WORK ADJUSTMENT 11 systems (e.g., promotion) are determined by and address organizational racism and employees? race(s) (Barbarin & Gilbert, 1981; Watts & Carter, 1991). In the Watts and Carter study, a workplace that received a high racial climate score was described as having a hostile racial climate; low scores corresponded to positive racial climates. Most research on racial climate has made connections between racial climate and job satisfaction (Ensher et al., 2001; Holder & Vaux, 1998; Hughes & Dodge, 1997). Less research exists on the relation between racial climate or experiences with discrimination and turnover or turnover intentions. When incorporating culture into job satisfaction research, Holder and Vaux (1998) found a negative relation between racial climate and job satisfaction for African-American employees. That is, the more negative the employees? perceptions of the workplace racial climate, the less satisfied employees were with their workplaces. The authors were unsuccessful in their attempt to identify variables (i.e., internal locus of control, work-related social support) that moderated the relation between racial climate and job satisfaction. Other research offered similar findings. Hughes and Dodge (1997) examined the ability of two aspects of racial discrimination, institutional discrimination and interpersonal prejudice to predict job quality, a variable that was assessed with a job satisfaction measure, for African-American employees. The authors found that racial discrimination at work accounted for 11% of the variance in job quality. THEORY OF WORK ADJUSTMENT 12 Ensher et al. (2001) also offered support for the relation between racial climate and job satisfaction. These authors operationalized racial climate as a combination of the following three organizational characteristics: discrimination from the co-worker, discrimination from the supervisor, and organization level discrimination (e.g., discriminatory policies). These authors found that for a mixed-race sample of employees, racial climate was related to job satisfaction as well as other variables of importance to employers (i.e., organizational commitment, organizational citizenship behaviors). Organizational-level discrimination was a better predictor of the dependent variables than supervisor and coworker discrimination. Other research indicated that experiences with workplace discrimination may have consequences for employees? turnover intentions, such that when employees experience workplace discrimination, they may be more likely to report intentions to leave that workplace (Shaffer et al., 2000). Although this research has not examined racial discrimination but has focused on gender discrimination, the findings can be used to predict the possible relation that may emerge in future research on racial climate perceptions and turnover intentions. Given the research on the relation between racial climate and job satisfaction and discrimination and turnover intentions, inclusion of racial climate in the TWA model may serve to enhance the prediction of job satisfaction and turnover intentions for African-American employees. Extrapolating from the postulations of Fitzgerald and Rounds (1993) and Rounds and Hesketh (1994), the present study investigated THEORY OF WORK ADJUSTMENT 13 whether racial climate moderated the relation between perceptions of fit and job satisfaction as well as the relation between perceptions of fit and turnover intentions. The decision to test the ability of racial climate to moderate, as opposed to mediate, the relations between perceptions of fit and job satisfaction and turnover intentions was informed by two factors. First, Baron and Kenny (1986) described moderators as having the ability to attenuate the relation between the independent variable and the dependent variable while maintaining the properties of an independent variable. Results of previous racial climate research (e.g., that racial climate as an independent variable predicted job satisfaction) suggested that this variable possesses these qualities. Second, the TWA scholars who have hypothesized the role that culture plays in the TWA model have drawn attention to the possibility that cultural variables (e.g., racial climate) more often moderate, as opposed to mediate, the relation between fit and other TWA variables (Fitzgerald & Rounds, 1993; Rounds & Hesketh, 1994). In summary, the present study proposed that the model presented by the TWA theorists may need to be altered for African Americans. Implications for Counseling Psychology Because the present study aimed to (a) offer a greater understanding of African Americans? perceptions of work and working behaviors with descriptive data, (b) test an augmented theoretical model of job satisfaction and turnover intentions that incorporates culture, and (c) address methodological limitations of the TWA, the THEORY OF WORK ADJUSTMENT 14 present study offers counseling psychologists guidance for practice, advocacy and research. These implications are addressed below. Implications for practice. Gelso and Fretz (2001) noted that counseling psychologists have been slow to engage in organizational consultation activities. One of the reasons provided to explain our professions? lack of involvement in consultation is a lack of sufficient knowledge in this area. The present study provided information from the perspective of a counseling psychologist which could help bridge the gap in literatures related to organizational consulting and counseling psychology. Specifically, if some of the hypotheses of the present study are supported, counseling psychologists could counsel individuals to consider the role of perceptions of fit and racial climate in their work experiences. This would mean assisting African- American employees to recognize and select the environments which they feel would meet their needs and provide a positive racial climate. Furthermore, interventions may involve educating managers about the consequences of their employees? perceptions of racial climate and fit. If counseling psychologists can work to match employees with workplaces based on their needs and perceptions of racial climate, employees can expect to be more satisfied (Dawis & Lofquist, 1984), experience longer tenures (Dawis & Lofquist, 1984), greater life satisfaction (Near, Rice, & Hunt, 1983), and less job absenteeism (Goodman & Atkin, 1984). Implications for advocacy. Although workplace discrimination is illegal (Shaffer et al., 2000), employees still perceive discrimination in their places of THEORY OF WORK ADJUSTMENT 15 employment (Ensher et al., 2001; Holder & Vaux, 1998; Hughes & Dodge, 1997; Shaffer et al., 2000; Watts & Carter, 1991). Research from the present study could be brought to the attention of politicians and organizations to encourage these two entities to recognize that (a) regardless of legislation that aims to equalize work environments, certain groups still perceive workplace discrimination and (b) these perceptions have negative consequences for employees and employers (Cohn, 1997; Ensher et al., 2001; Holder & Vaux, 1998; Hughes & Dodge, 1997; Shaffer et al., 2000). Fassinger and O?Brien (2000) drew attention to the responsibility that counseling psychologists have to engage in advocacy for their clients which could translate into helping organizations to more closely heed current legislation. Counseling psychologists also can assist organizations by training management to understand employment law and identifying discriminatory behaviors, policies, and procedures in their organizations. Furthermore, counseling psychologists working in these settings can educate managers in organizations to understand consequences of discrimination and take a results-oriented approach to organizational change. Because of the legal implications of workplace discrimination, this matter can extend beyond interpersonal issues, to become an organizational problem. Implications for research. The present study could assist researchers in their understanding of African-American employees? perceptions of their work environments and work behaviors by presenting descriptive information related to THEORY OF WORK ADJUSTMENT 16 perceptions of fit and racial climate as well as job satisfaction and turnover intentions. This investigation is of importance because of the lack of information related to African Americans? work experiences (Brown, 1995; Fitzgerald & Betz, 1994). Additionally, the present study provided a cross-cultural test of a model of job satisfaction and turnover intentions, the TWA, (Dawis, England, & Lofquist, 1964; Dawis & Lofquist, 1984) that previously has been confirmed with European (e.g., Taris & Feji, 2001) and European-American samples (e.g., Breeden, 1993; Rounds et al. 1987; Saks & Ashforth, 1997). However, research comparing levels of job satisfaction and turnover intentions between African Americans and European Americans suggested that these variables may be experienced differently in these two groups (Alexander & Kauffman, 1999; Austin & Dodge, 1992; Mueller et al., 1999), supporting the need for an augmented model of job satisfaction and turnover intentions for African Americans. Furthermore, the present study advanced research in perceptions of fit by using a theoretically-based measure of perceptions of fit. Use of this measure may increase the interpretability of empirical findings in future research. Finally, the present study integrated the influence of culture into the TWA model to test for the relevance of racial climate in the TWA model when explaining African Americans? work experiences. This should assist future researchers to more accurately predict the work experiences of African Americans. THEORY OF WORK ADJUSTMENT 17 Chapter 2 Review of Literature The present literature review summarized and critiqued research related to (a) the relation between person-environment fit and job satisfaction, (b) the relation between person-environment fit and tenure intentions, (c) race and job satisfaction, (d) the proposed role of culture in the TWA, and (e) racial climate. However, I began with a summary of the vocational histories and present day realities that mark the differences in the work experiences between African Americans and European Americans. Then, I presented an overview of the Theory of Work Adjustment (the TWA; Dawis et al., 1964; Dawis & Lofquist, 1984). African Americans and Work: Historical Context ?One very important difference between white people and black people is that white people think that you are your work?Now, black people think that my work is just what I have to do to get what I want? (Gwaltney, 1981, pp. 173-174). This quote, taken from a blue-collar worker interviewed by Gwaltney, has been endorsed in many counseling psychologists? critiques of current career theories (e.g., Fassinger, 2001; Peterson & Gonzalez, 1999; Richardson, 1993) and indicated that differences may exist in the ways that some African Americans and European Americans perceive, experience, and use work. Past research helped explain why these differences may exist. In this section I provided a brief overview of the vocational history of African Americans as well as African Americans? contemporary situation in the world of THEORY OF WORK ADJUSTMENT 18 work in an attempt to contextualize the research summarized and the hypotheses offered in the present study. Overwhelmingly, research on the work experiences of African Americans found that this group has experienced discrimination in selection and hiring, pay, and layoffs (Berry & Blassingame, 1982; Peterson & Gonzalez, 1999; Williams, 1999; Williams, 2000). Historically, African Americans have been disproportionately segregated in low-paid, unskilled jobs (Berry & Blassingame, 1982). Even when African Americans possessed other skills, many times they were denied access to other professions or jobs because of discrimination in the selection process (Berry & Blassingame, 1982; Williams, 1999). Unfortunately, this discrimination was determined to be legal because of Civil Rights rulings in 1883 that found that the 14 th Amendment protections were applicable only to public spaces, allowing private employers to hire whomever they pleased. This ruling was not overturned until overt job discrimination in hiring was stopped by the 1964 Civil Rights Act and less obvious forms of job discrimination were ended by the Griggs v. Duke Power Co. (1971) ruling (Williams, 1999). In part for reasons stated above, African Americans have typically had unemployment rates that are twice those of European Americans (Plous, 1996). Since 1954, African Americans? unemployment rates have typically risen above 7.5% (Berry & Blassingame, 1982). Furthermore, African Americans are most often underemployed. They have been more likely than European Americans to be occupied THEORY OF WORK ADJUSTMENT 19 only part-time and also tend to be unemployed for longer periods of time than European Americans (Berry & Blassingame, 1982). African Americans have also been found to experience precarious job situations because of their greater susceptibility to ?downsizing? (Peterson & Gonzalez, 1999; Williams, 1999). Williams (1999, 2000) noted that between the late 1970s and the early 1990s the wage gap between African Americans and European Americans increased. In the 1990s when African Americans started to regain economic ground, college-educated African American males experienced wage increases that were twice that of their European-American peers. Still by 1997, college-educated African-American males were earning 75% of what college-educated European-American males were earning (Williams, 1999). Berry and Blassingame (1982) estimated that in the period of time between 1950 and the time of publication of their book, African Americans lost between $30 and $40 billion dollars because of pay disparity. Today, attempts to remedy past instances of discrimination or attempts to increase diversity in the workplace with affirmative action policies, have been met with resistance (Williams, 2000). Research indicated that two out of three Americans oppose affirmative action policies. Heilman, Simon, and Repper (1987) suggested that this resistance to affirmative action may have negative consequences for employees who are assumed to be targets of affirmative action (e.g., African Americans). THEORY OF WORK ADJUSTMENT 20 Considering these realities related to the cultural, political, and economic influences informing African Americans? vocational experiences, Peterson and Gonzalez (1999) urged career counseling professionals to attend to the historical contexts of clients? cultural groups because these histories may inform employees? work-related behaviors and vocational development. Indeed, research considering scarcity of employment and cultural tendencies found that some of the relations between work-related behaviors previously found with European and European- American employees do not hold true with all populations (Nyambegera, Daniels, & Sparrow, 2001; Parkes, Bochner, & Schneider, 2001). In fact, because of culture and economic realities (e.g., the ability to secure employment) in Asian and sub-Saharan countries it is harder to predict organizational commitment, job satisfaction, tenure, and job involvement (Nyambegera, Daniels, & Sparrow, 2001; Parkes, Bochner, & Schneider, 2001). This highlights the importance of contextualizing African Americans? vocational experiences when considering the applicability of past research. The information provided in this section indicated that African Americans may have different experiences of work than European Americans, on whom most vocational research is conducted. Therefore, it may not be possible to apply findings of past research on European Americans? work experiences to African Americans? work experiences. For that reason, the present study?s purpose was to promote an understanding of African Americans? work experiences in light of previous research THEORY OF WORK ADJUSTMENT 21 on the variables of interest in the present study and African Americans? unique vocational history. Theory of Work Adjustment The Theory of Work Adjustment (TWA) was first formulated to provide a framework to explain the existing scholarship on individuals? on-the-job work experiences. First iterated by Dawis, England, and Lofquist (1964) and later more fully developed by Dawis and Lofquist (1984), the TWA describes the ways that person and environment factors interact to predict employees? satisfaction with their jobs and workplaces? satisfaction with their employees. Central to this individual differences model is the idea of correspondence between employees and their work environments. Dawis and Lofquist defined correspondence as the extent to which an environment and individual meet the requirements of the other. Work environments and employees make demands of each other and for employees to remain in work environments and feel satisfied, these demands must be met. Correspondence is conceptualized as a set of stimulus-response conditions. Correspondence is thought to occur along two dimensions: the correspondence between the reinforcements offered by work environments and employees? needs (which is hypothesized to predict satisfaction) as well as the correspondence between the requirements of work environments and employees? skills (which is hypothesized to predict satisfactoriness; Dawis & Lofquist, 1984). THEORY OF WORK ADJUSTMENT 22 Satisfaction. Employees? needs are defined as the demands that they make of their work environments. When employees? needs are fulfilled, they elicit responses that the work environment requires. Employees? needs preferences are thought to originate from their experiences in previous work environments (Dawis & Lofquist, 1984). Dawis and Lofquist differentiate between needs and values, stating that needs are expressive, measurable traits in the worker while values are the dimensions underling needs. Values can also be referred to as the latent variables while needs are the manifest variables (Dawis, 2000). Dawis and Lofquist (1984) present six value areas defined by a variety of needs. They name achievement as the first value which includes consideration of the following needs: ability utilization (e.g., ?I could do something that makes use of my abilities?) and achievement. Activity, independence, variety, compensation, security, and working conditions make up the comfort value. The status value is defined by opportunity for advancement, recognition, authority, and social status. Employees? needs related to their co-workers, moral values, and social service make up the altruism value. Safety is defined by employees? reactions to company policies and practices, the human relations aspects of supervision, and technical aspects of supervision (e.g., training). Finally, the need for creativity and responsibility define the autonomy value. Just as employees? values can be considered along these six value dimensions, work environments can be assessed for the extent to which they provide employees with satisfaction of these values. THEORY OF WORK ADJUSTMENT 23 Dawis and Lofquist (1984) define workplace reinforcers as stimulus conditions that are meant to respond to employees? needs. Workplace reinforcers that meet employees? needs are effective. Workplace reinforcers can be classified in the same six areas that employees? values are classified: achievement, comfort, status, altruism, safety, and autonomy. Examples of workplace reinforcers are bonuses or certificates of appreciation. Satisfactoriness. Just as needs correspond with values, skills correspond with abilities. That is, abilities are the dimensions underling skills. According to Dawis (1994), skills represent those stimulus responses indicative of employees? capabilities. Dawis and Lofquist (1984) describe nine ability areas that are an interaction between three content areas related to work and three psychological processes related to work. The content areas are derived from the U.S. Department of Labor?s (1977) classification of work tasks related to data (symbolic abilities), people (interpersonal abilities), and things (sensorimotor abilities). The psychological processes are afferent abilities related to the way employees take in information, mediational abilities which describes employees? information processing capabilities, and efferent abilities or employees? actions and output in the work environment. The resulting abilities areas are: perception of data (afferent ability x symbolic ability), data processing (mediational ability x symbolic ability), fluency (efferent ability x symbolic ability), interpersonal perception (afferent ability x interpersonal ability), interpersonal reasoning and processing (mediational ability x interpersonal ability), communication THEORY OF WORK ADJUSTMENT 24 and leadership (efferent ability x interpersonal ability), sensorimotor perception (afferent ability x sensorimotor ability), sensory judgement (mediational ability x sensorimotor ability), and manual dexterity (efferent ability x sensorimotor ability). Workplace requirements are the skill and ability demands that the workplace makes of its employees. The correspondence between needs and reinforcers and skills and requirements are thought to comprise work adjustment. Dawis and Lofquist (1984) suggested that work adjustment was not a static process but one in which the environment and employee are constantly responding to each other in an attempt to reach and sustain correspondence. Satisfaction and satisfactoriness mark the degree of success that workplaces and employees have in responding to one another. The greater the needs correspondence, the greater job satisfaction that employees will report. Skills correspondence is positively related to workplaces? satisfaction with their employees. This type of satisfaction is called satisfactoriness. When employees and workplaces are satisfied with each other, the employee remains in the work environment. Therefore, satisfaction and satisfactoriness are thought to predict tenure (Dawis & Lofquist, 1984). Propositions of the Theory of Work Adjustment. Dawis and Lofquist (1984) offered ten propositions, constituting the TWA. Proposition I stated that work adjustment can be determined by employees? satisfaction and satisfactoriness. THEORY OF WORK ADJUSTMENT 25 Proposition II stated that satisfactoriness is determined by the correspondence between employees? abilities and workplaces? requirements. Whereas, satisfaction was determined by the correspondence between employees? needs and workplaces? reinforcements, which is outlined in Proposition III. In Proposition IV, the authors suggested that the relation between satisfactoriness and the ability/requirement correspondence was moderated by satisfaction. In the same way, Proposition V suggested that satisfactoriness moderated that relation between satisfaction and the needs/reinforcer correspondence. Proposition VI outlined the inverse relation between satisfactoriness and the likelihood that the employee would be asked to leave to the workplace and Proposition VII outlined the inverse relation between satisfaction and the likelihood that an employee would want to leave the workplace. Proposition VIII predicted that satisfaction and satisfactoriness determine employees? tenure. The authors contended in Proposition IX that correspondence increased with tenure. In Proposition X, the authors asserted that employees? personality style and work environment style moderated the relation between work adjustment and correspondence. In the present study, I was interested in testing Proposition III of the TWA. Proposition III stated that the needs/reinforcer correspondence between employees and their workplaces was related to job satisfaction such that a greater degree of correspondence was related to higher levels of job satisfaction. I also examined THEORY OF WORK ADJUSTMENT 26 Proposition VII, the relation between satisfaction and an employees? intention to quit their jobs. Perceptions of fit and job satisfaction. Several studies provided support for the ability of the needs/reinforcer correspondence outlined by the TWA (Dawis & Lofquist, 1984) to predict job satisfaction for employees in a variety of work settings. Rounds et al. (1987) used the TWA to design a study that tested the predictors of job satisfaction. The authors also were interested in the different ways that person- environment correspondence could be conceptualized and measured and the ways that these measurement differences influence the ability of correspondence to predict job satisfaction. The authors predicted that the correspondence between workers? needs and the workplaces? reinforcers would predict job satisfaction. Testing this hypothesis in two different studies, the authors found support for their hypothesis (Rounds et al., 1987). However, the samples used and the way in which the idea of fit was measured influenced the strength of the relationship between fit and job satisfaction. In study 1, Rounds et al. (1987) considered the relationship between fit and job satisfaction of tenured employees using 19 different indices of fit and their correlation to job satisfaction. The participants in study 1 were employees (n=635) from six occupations: cashiers, checker-marker, rehabilitation counselor, salesperson, telephone service representative, and telephone operator. No other identifying information (i.e., race, sex, age, income) was offered for this sample. THEORY OF WORK ADJUSTMENT 27 Rounds et al. (1987) found that when measuring fit, it was best to use the product moment index (which measures the similarity between two profiles in terms of elevation, shape, and scatter), and the D 2 prime index (which measures shape and scatter of the relationship). The use of other indices resulted in relationships that were either nonsignificant or in the opposite direction of that predicted by the TWA. While a relationship between fit and job satisfaction was found in Study 1, the correlation between these two variables was low. Postulating that the low correlations between fit and job satisfaction was caused by a lack of range in fit scores provided by tenured employees (i.e., tenured employees should report high correspondence and high job satisfaction), the authors attempted to increase the range in fit scores in study 2 by sampling vocational counseling clients (n=405). The female participants in this sample (n=187) ranged in age from 17 ? 58, with a median age of 29.8 years. The male participants (n=218) ranged in age from 17 ? 62, with a median age of 29.8 years. Information related to participants? race was not provided. The authors postulated that this sample may include members who may be experiencing a lack of correspondence in P-E factors. Again, the authors found support for Proposition III of the TWA. However, in Study 2, the relations between fit and job satisfaction were moderate to high depending on participants? sex and fit index used. Overall, the correlations in the female sample were stronger than in the male sample. In terms of the indices used, when the D 2 index was used, stronger correlations for females than for males resulted. THEORY OF WORK ADJUSTMENT 28 Further, the use of D 2 prime rather than D 2 resulted in stronger correlations for participants of both sexes. However, the use of D 2 double prime, which measures only shape, resulted in the strongest correlations for both males and females. Also interested in testing Proposition III of the TWA, as well as investigating the relationship between need-reinforcer fit and career success (i.e., salary, job level), Bretz and Judge (1994) tested three hypotheses. First, the authors predicted that correspondence would positively predict tenure. Next, the authors predicted that correspondence would be positively related to satisfaction. Finally, the authors predicted that correspondence would influence career success. Bretz and Judge (1994) surveyed alumni from industrial relations programs to collect data related to their salary and job level (i.e., indicators of career success), job satisfaction, perceptions of organizational climate and preferences for various organizational climates related to values, needs, personality, and culture. The last two constructs were measured with two surveys designed by the authors to measure fit. The authors used data from 873 alumni (63% of whom were male, average age was 34.80 years, and standard deviation was not reported), which reflected a 40% response rate. The average participant reported that they had been in their current job 3.91 years (SD = 4.90) and that they were making $66,508 yearly (SD = $43,450). Most participants reported working in a job 4.21 levels above the entry level (SD = 1.77). The race of the participants was not reported. THEORY OF WORK ADJUSTMENT 29 Using a regression equation, the authors tested the extent to which P-E fit predicted job satisfaction and tenure. In steps 1-4 of the regression equations, the authors entered demographic variables (e.g., race, gender), human capital variables (e.g., degree, hours per week worked), job and organizational variables (e.g., access to a mentor, salary), and industry variables (e.g., service, public administration), respectively. P-E fit was entered as step 5 in both equations. Race did not significantly predict any of the variance in either job satisfaction or tenure. P-E fit predicted a significant amount of the variance in both job satisfaction and tenure, above what was predicted by steps 1 through 4. P-E fit was also predictive of the career success variables: salary and job level attained. Building upon the work of Holland (1985) and Dawis and Lofquist (1984), Rounds (1990) attempted to test the unique and combined ability of interests and values to predict job satisfaction. Both Holland (1985) and Dawis and Lofquist (1994) postulated that person-environment fit leads to job satisfaction. However, Holland posed his idea of fit, which he calls congruence, in terms of interests (e.g., artistic, investigative, enterprising), whereas Dawis and Lofquist considered the work values fit, which they call correspondence, between employees and their work environment. Rounds (1990) surveyed male (n=218) and female (n=187) vocational counseling clients to collect data. Female participants ranged in age from 17 ? 58 years (median 29.8 years). Male participants ranged in age from 17 to 62 years (median 29.8 years). The race(s) of the participants was not reported. THEORY OF WORK ADJUSTMENT 30 Rounds (1990) collected data related to participants? Holland interest code types, work values, job satisfaction, job title and job status. He also collected information related to the Holland Occupational Codes, which was used with the Holland interest code types to determine interest fit, and reinforcer patterns which were used with work values to determine values fit for each participants? job. Fit was determined by the correlation between employees? work values and the reinforcement patterns offered in the jobs in which participants were employed, resulting in a product moment coefficient to measure the profile shape. Then, each measure of fit (i.e., interest and value) became the independent variable in a multiple regression analysis intended to evaluate the relationship between interest and value fit and job satisfaction. The author was also interested in the amount of variance explained by values over and above interests. Data from males and females were analyzed separately. Results confirmed Rounds? (1990) hypotheses. Interest and value fit predicted job satisfaction. Additionally, the values fit predictor explained variance in job satisfaction over and above that which was explained by interest fit. Individually, values fit accounted for approximately twice the amount of job satisfaction as interest fit in the female sample (15% and 6%, respectively). In the male sample, values fit accounted for 29% of the variance in job satisfaction and interest fit accounted for 2% of the variance in job satisfaction. In the female sample, the interest and value fit were related, sharing about 12% of the variance in job satisfaction. The two types of fit THEORY OF WORK ADJUSTMENT 31 were not related in the male sample. Based on these findings, Rounds stressed the importance of both interests and values in the determination of job satisfaction. Taris and Feij (2001) completed another study of the relationships between needs/reinforcer correspondence and two dependent variables (job satisfaction and intention to leave the job). This study was unique in its focus on three aspects of work values: intrinsic (e.g., self-expression), extrinsic (e.g., salary), and social (e.g., interactions with peers) values. The authors also were interested in the shape of the relationship between values (main effect), values fit (interaction) and the dependent variables. Finally, the authors tested the TWA using a longitudinal design. Data related to participants? work values, organizational reinforcers, job satisfaction, intention to leave, and psychological well-being were collected from participants at two points, four years apart. The participants were female (46%) and male young adults (18-26 years old) living in the Netherlands. Similar to other research in this area, Taris and Feij (2001) found that, for the most part, the correspondence between an employee?s values and workplace reinforcers was related to job satisfaction and intention to leave. A relation was also found between correspondence and psychological well-being but this relation was not as strong as the relations between correspondence and job satisfaction and correspondence and intention to leave the workplace. When the intrinsic values correspondence was considered, the relationship between correspondence and intention to leave was consistent across both data points but the relation between THEORY OF WORK ADJUSTMENT 32 correspondence and job satisfaction was found only at the second data point. When examining the extrinsic and social value correspondence, a relation between correspondence and intention to leave was found only at the second data collection point. In other words, at the second data collection point those employees? experiencing greater extrinsic and social correspondence reported less intention to leave. Workplace reinforcers also predicted job satisfaction and intention to leave such that those who reported that their workplace supplied more reinforcers (i.e., higher salaries) were more satisfied and less likely to leave. This was true at the first and second data collection points. Taris and Feji (2001) explained that as employees remain in their jobs longer they become more aware of their values and the reinforcers offered by their jobs. Therefore, at the second data collection point, they were better able to make statements about values and reinforcers that can offer useful information about their job satisfaction and intention to leave than they were able to at the first data collection point. Therefore, at the second data collection point, stronger relations between the variables were found. Breeden (1993) recognized both the importance of and lack of longitudinal research on the TWA propositions. Also, she was interested in examining the role of values correspondence and job satisfaction in predicting job and occupational change. Job change was described as employees? change of work environment but not tasks. Occupational change was defined as employees? change of job type and tasks. These THEORY OF WORK ADJUSTMENT 33 two variables are similar to tenure, which is one of the constructs described in the TWA. Breeden hypothesized that values correspondence and job satisfaction would be lower for employees before a job change than after job/occupational change. Furthermore, she predicted that values correspondence would be correlated with job satisfaction. Data were collected from 436 employed adults who sought vocational counseling over the course of nine years. The participants ranged in age from 17 to 62 (M = 32.0, standard deviation was not provided). Of those who participated, 60% were male, 98% were European American, 63% of them had at least a bachelor?s degree, 87% were employed full time. Their mean income was $16,699 (standard deviation was not provided). Participants were given two measures of job satisfaction: a one-item measure of global job satisfaction and the Minnesota Satisfaction Questionnaire short form (Weiss et al., 1967). The Minnesota Importance Questionnaire (Rounds et al., 1981) was used to measure participants? occupational values. When participants were interviewed at the second data collection point, two years after they were originally interviewed, they were asked whether they had changed jobs. Raters then coded these changes as either a job or occupational change. Data related to employees? vocational interests were collected with the Strong- Campbell Interest Inventory (Campbell & Hanson, 1981). Analyzing these data using a one-way analysis of variance, Breeden (1993) did not find differences in values correspondence when comparing employees before and THEORY OF WORK ADJUSTMENT 34 after job or occupational changes. However, she did find that job satisfaction was higher after a job or occupational change when compared to before change rates of job satisfaction. She found that values correspondence and job satisfaction were correlated in three of the four possible calculated correlations. At intake, correlations were found between values correspondence and both measures of job satisfaction. However, at the two year follow-up, a correlation was found between values correspondence and the one-item measure of global job satisfaction, but a correlation was not found between correspondence and the Minnesota Satisfaction Questionnaire. Attempting to explain the results that did not confirm the author?s hypotheses, Breeden (1993) cited possible measurement problems as one explanation. Specifically, she hypothesized that problems related to range restriction and correspondence measurement might explain the nonsignificant results. Range restriction may have occurred because the sample was drawn from a population receiving career counseling. The author suggested that future research on correspondence employ different measures of correspondence. The author also recognized the possibility that the hypothesized relations between variables may not have existed as she hypothesized. More recent studies of correspondence used alternate ways of determining correspondence between employees? values and workplace reinforcers. Rather than using difference scores to determine the correspondence between employees? values and workplace reinforcers, Saks and Ashforth (1997) investigated employees? THEORY OF WORK ADJUSTMENT 35 perceptions of correspondence. However, similar to Breeden (1993), these authors used a longitudinal design to examine the relationship between perceptions of correspondence and job satisfaction, intentions to quit, and turnover at three data points: prior to entering a work environment, and again at 4 and 10 months on the job. Saks and Ashforth (1997) hypothesized that applicants? perceptions of person- job and person-organization correspondence would be positively related to job satisfaction and negatively related to intentions to quit and turnover. Person-job correspondence is the fit between employees? knowledge, skills, and abilities and those knowledge, skills, and abilities required by the job. Person-organization correspondence is the fit between an employees? values, beliefs, and personality traits with those values, beliefs, and norms of an organization. Saks and Ashforth (1997) collected data from 231 participants who were graduating seniors who then progressed into their first job post-graduation. The mean age of the participants was 23.4 years (SD=3.1). The authors did not report the races of the participants. However, they did report that 60% of the participants were female. The response rate for this study was 66%. Participants? perceptions of correspondence were measured using a measure developed by the authors which tapped into participants? perceptions of person-job and person-organization fit. Job satisfaction was assessed using a 3-item, global measure of job satisfaction authored by Cammann, Fishman, Jenkins, and Klesh (1983). Participants? intention to quit was measured using Colarelli?s (1984) intention THEORY OF WORK ADJUSTMENT 36 to quit scale, a three-item measure. Finally, turnover was operationally defined as a participant?s departure from her or his workplace by the time of the 10-month follow- up. Saks and Ashforth (1997) found that perceptions of person-job and person- organization correspondence explained a significant amount of the variance in job satisfaction, intentions to quit, and turnover at 4 and 10 months. When the relations between the subscales and the outcomes were analyzed separately, the authors found that the beta weight was significant in the relation between person-job correspondence and job satisfaction. The resulting beta weights revealed significant relationships between both subscales and intention to quit. A significant relationship was also found between the person-organization correspondence subscale and turnover. Only perceptions of person-organization correspondence was negatively related to turnover. Additionally, the authors found that person-job correspondence was positively related to organizational commitment and organizational identification and negatively related to participants? reports of stress symptoms. Person-organization correspondence was positively related to organizational identification. Summary. Taken together, these studies provided support for Proposition III of the TWA, which stated that the correspondence between employees? needs and the reinforcements provided by the workplace predict employees? satisfaction with their jobs. However, the strength that researchers found between correspondence and job satisfaction seemed to be dependent on two variables: the way that fit was measured THEORY OF WORK ADJUSTMENT 37 and the population sampled. Furthermore, needs/reinforcer correspondence seemed to be a better predictor of job satisfaction than interests. Finally, the research of Saks and Ashforth (1997) also supported the notion that fit was negatively related to turnover intentions and tenure. Person-environment fit and turnover intentions. Proposition IX of the TWA has linked fit and turnover, stating that a positive relation should exist between fit and tenure (Dawis & Lofquist, 1984). Additionally, other research (e.g., O'Reilly, Chatman, & Caldwell, 1991; Saks & Askforth, 1997; Van Vianen, 2000) and theory (Schneider, 1987) have made this connection. Namely, Schneider?s Attraction- Selection-Attrition (ASA) model has served as the framework for much of the empirical literature that sought to investigate the relation between fit and tenure intentions. According to the ASA model, person characteristics (e.g., values, personality) determine the culture of organizations (i.e., ?The people make the place.). Futhermore, Schneider?s model posited that individuals in organizations become more similar to one another over time through the processes of attraction, selection, and attrition. Related to the present study, Schneider postulated that those who are not similar to an organization (i.e., those who do not fit), will leave that organization. Often, when turnover or tenure has been investigated in the TWA research, measures of turnover intentions were used in place of or in addition to measures of actual tenure or turnover (e.g., Saks & Ashforth, 1997; Taris & Feji, 2001). Tenure is the length of time that employees remain in an organization (Dawis & Lofquist, THEORY OF WORK ADJUSTMENT 38 1984). Turnover refers to the departure of employees from an organization either voluntarily or involuntarily (e.g., termination; Jewell, 1998). Turnover intentions differ from actual turnover, in that turnover intentions are employees? self-report of their intentions to quit or stay in the workplace (Mobley, 1982). O?Reilly et al. (1991) supported Schneider?s (1987) contentions with their research on the relation between employees? fit with an organization, in terms of congruence of organizational cultural preferences, and turnover intentions and tenure. Organizational culture is conceptualized as the values and assumptions shared by members of an organization (e.g., decisiveness, taking initiative, fairness). The authors hypothesized that person-organization fit would be negatively related to turnover intentions and actual turnover. Also, person-organization fit would be positively related to job satisfaction and organizational commitment. Five different participant groups were included in the O?Reilly et al. (1991) study. However, information on one of the sample groups (i.e., new accountants) is summarized here because this group and the hypotheses tested using this group were relevant to the present study. This group consisted of new accountants (n=171) working in eight consulting firms. Forty-seven percent of the new accountant group were men. Their mean age was 25 years (standard deviation was not reported). Race was not reported. These participants were followed for two years to collect longitudinal data on this group. THEORY OF WORK ADJUSTMENT 39 Data on person-organization fit were collected using a Q-sort data method where new accountants were asked to rank their preferences for organizational culture. Then, senior accountants used the Q-sort method to rank their perceptions of organizational culture. The discrepancy between new accountants? preferences and senior accountants? perceptions constituted new accountants? person-organization fit where greater fit was indicated by less discrepancy. Data related to participants? organizational commitment, job satisfaction, turnover intentions, and actual turnover also were collected. O?Reilly et al. (1991) found support for their hypotheses. More specifically, they learned that person-organization fit was negatively related to turnover intentions and actual turnover. Person-organization fit was positively related to job satisfaction and organizational commitment. Van Vianen (2000) investigated the fit between newly hired employees? reported values and a) the values that they perceived that their organization endorsed (i.e., person-organization fit) and b) the values reported by other members of the organization (i.e., person-person fit). Because Schneider (1997)?s ASA theory asserted that person characteristics determine organizational characteristics (i.e., ?the people make the place?), Van Vianen determined that it was important to consider person-organization fit as well as person-person fit. For both conceptualizations of fit, Van Vianen hypothesized that fit would be negatively related to turnover intentions. THEORY OF WORK ADJUSTMENT 40 Participants were recruited from 68 organizations located in the Netherlands. The participants were new hires (n=154), peers of the new hires (n=104), and supervisors of the new hires (n=101). Of the participants who reported their sex, 99 were female and 254 were male. Race of the participants was not reported. The new hires had been working in their current organizations for a mean of six months (SD=3.9 months). Participants were given a measure of cultural preferences which measured their values related to human relations (i.e., positive feedback, peer cohesion, human development, participation, conflict tolerance) and goals (i.e., reward, work pressure, effort, competitions). A comparable measure of perceived organizational reinforcers was also given to participants. Turnover intentions were measured with a three item measure created by Ten Horn (1983; e.g., ?When I have the opportunity to work in another organization, I will take that opportunity?). Van Vianen (2000) utilized polynomial regression analyses to test the hypotheses presented. Results did not support Van Vianen?s (2000) hypothesis that person- organization fit would be related to turnover intentions. However, person-person fit was related to turnover intentions when supervisors? values preferences were considered, but not when peers values preferences were considered. Van Vianen pointed to possible methodological flaws (i.e., the way fit was operationalized) when explaining why the person-organization fit hypothesis was not supported. However, Van Vianen concluded that generally, the results confirmed Schneider?s (1997) THEORY OF WORK ADJUSTMENT 41 contention that person characteristics were more influential than organizational characteristics. Rather than considering the influence of values or needs fit in Schneider?s (1997) ASA theory, Jackson et al. (1991) constructed a study to test the possible influence of fit in terms of organizational demography on turnover. Organizational demography concerns the demographic characteristics (i.e., age, tenure, sex, race, socioeconomic background, religion) of members that comprise organizations. Jackson et al. hypothesized that demographically heterogeneous working teams would have higher turnover rates than demographically homogeneous teams. Jackson et al.?s (1991) study was of relevance to the present study because of the assumptions that they offered for the cause of turnover rates of demographically heterogeneous teams. They conjectured that demographic categories (e.g., age) may be related to certain psychological attributes such as attitudes. The present study goes beyond these conjectures to examine the possible psychological attributes (i.e., perceptions of person-organization fit, perceptions of racial climate) that might be influenced by organizational demography patterns and in turn, influence turnover intentions. Participants were executives (n=625) who were employed in bank holding companies. All participants were members of management teams that ranged in size from three to 18. Because only two percent of the resulting sample was female, Jackson et al. (1991) eliminated them from the sample. Race of the participants was THEORY OF WORK ADJUSTMENT 42 not mentioned. While Jackson et al. mentioned that race and sex were two of the organizational demographic characteristics of interest in the organizational demography literature, and in the introduction of their study, these demographic characteristics were not part of their analyses. Information related to organizational demography characteristics (i.e., age, education level, college curriculum, military experience, tenure) and turnover rates were acquired through public archival data. Organizational demography was determined by entering organizational demography data into a Euclidean distance equation. Jackson et al. (1991) found support for their hypothesis that organizational dissimilarity was predictive of turnover. Heterogeneity along the lines of the previously stated organizational demographic characteristics explained 22% of the variance in turnover. Jackson et al. concluded by urging future research to consider the role of other organizational demography variables such race, sex, and religion in the prediction of turnover rates. Summary. Research (O?Reilly et al., 1991; Saks & Ashforth, 1997; Van Vianen, 2000) and theory (Dawis & Lofquist, 1984; Schneider, 1987) suggested that person-organization fit was negatively related to employees? turnover (O?Reilly et al., 1991; Saks & Ashforth, 1997) and turnover intentions (O?Reilly et al., 1991; Saks & Ashforth, 1997; Van Vianen, 2000). This research investigated the role of fit in terms of values as well as organizational demography (Jackson et al., 1991). THEORY OF WORK ADJUSTMENT 43 Race and Job Satisfaction To date, research on the differences in job satisfaction of African Americans and European Americans has been equivocal (Alexander & Kauffman, 1999; Austin & Dodge, 1992; Brenner & Fernstein, 1984; Greenhaus, 1990; Mueller et al., 1999; Tuch & Martin, 1991). That is, some of this research suggested that African Americans were not as satisfied with their jobs as their European-American counterparts (e.g., Austin & Dodge, 1992; Greenhaus, 1990; Tuch & Martin, 1991). Other research suggested that African Americans reported higher levels of job satisfaction than European Americans (e.g., Brenner & Fernstein, 1984; Mueller et al., 1999). Finally, other research found that no differences existed in the job satisfaction of African-American and European-American employees (e.g., Alexander & Kauffman, 1999). This body of research created confusion related to the comparative levels of job satisfaction between these groups. Furthermore, this research has not furthered understanding of the determinants of job satisfaction for African Americans. Before attempting to identify the determinants of job satisfaction for African- American employees, the research that presented between-group comparisons of job satisfaction is described. Tuch and Martin (1991) designed a study to determine whether differences existed in African Americans? and European Americans? reports of job satisfaction. Furthermore, these authors were interested in the cause of the difference if a difference existed. They postulated two explanations for the possible difference. The THEORY OF WORK ADJUSTMENT 44 first explanation focused on structural features of the job and workplace (i.e., rewards given by certain workplaces to certain employees to make workplaces more or less satisfying). The second explanation assumed that dispositional differences between African Americans and European Americans exist. By this the authors meant that people vary by race in terms of the values and needs that they require of workplaces. The authors did not address potential within group variables that might influence employees? reports of job satisfaction (e.g., racial climate perceptions). Tuch and Martin provided support for the importance of job satisfaction by citing its negative relationship with a number of outcomes including employee theft, work attendance, work commitment, and life satisfaction. Tuch and Martin (1991) analyzed data provided by the National Opinion Research Center?s General Social Surveys (Davis, 1989). Data related to employees? job satisfaction, demographic characteristics, workplace characteristics, and work values were collected using a full-probability sampling procedure. African- Americans? (n=271) and European-Americans? (n=2,246) data were analyzed separately by regressing job satisfaction on the relevant demographic and work variables. T-tests were also conducted to examine between-group differences. African-American respondents reported an average age of 38.90 years (SD= 12.89) and European-American respondents reported an average age of 39.58 years (SD= 12.56). Data related to the numbers of males and females who participated in the sample were not provided. Results revealed a racial difference in reports of job THEORY OF WORK ADJUSTMENT 45 satisfaction with African Americans reporting lower levels of job satisfaction than European Americans (Tuch & Martin, 1991). Upon examination of the data, it appeared that African Americans reported lower levels of the structural workplace variables that were thought to lead to job satisfaction. The racial groups did not report differences in the dispositional variables (e.g., work-related values) that were also thought to lead to job satisfaction. This research suggested that African Americans were not receiving reinforcers at the same level of European American workers. Greenhaus, Parasuraman, and Wormley (1990) intended to explain the way that race may influence employees? job satisfaction. More specifically, the authors proposed that race would lead to certain organizational influences (e.g., mentoring, supervisor support, acceptance) which, in turn, would influence job performance evaluations. Job performance evaluations were then thought to influence career outcomes such as opportunity for advancement and career satisfaction. The authors hypothesized that because of discrimination, African-American employees would have less access to positive organizational experiences, positive job performance evaluations and would, therefore, report lower levels of career satisfaction. In summary, the authors hypothesized that race indirectly influenced job satisfaction. This model was tested with data collected from African-American (n=373) and European-American (n=455) managers employed at three companies. The sample included 228 African-American women, 140 African-American men, 221 European- American women, and 231 European-American men. An attempt was made to match THEORY OF WORK ADJUSTMENT 46 the managers in terms of age, tenure, job function, and organizational position. The mean age of managers was 38.72 (SD=8.20). Most managers earned between $30,000-$39,000. The participants were given several measures of organizational experiences and career outcomes, including career satisfaction (e.g., ?I am satisfied with the progress I have made toward meeting my overall career goals?). The managers? supervisors were also given measures intended to provide a performance evaluation of the managers and a measure of the managers? opportunities for advancement. Results indicated that African-American managers reported less positive evaluations of organizational experiences (e.g., less acceptance) than European- American managers. Furthermore, African-American managers received lower performance evaluations and fewer opportunities for advancement from their supervisors than European-American managers. As Greenhaus et al. (1990) expected, African-American managers were more dissatisfied with their careers than their European-American counterparts. The authors found that the model explained 22% of the variance in career satisfaction. Contrary to the authors? predictions, the direct relationship between race and career satisfaction (.14) was stronger than the indirect relationship between race and career satisfaction (.04). The authors did not discount the ability of an indirect model to explain racial differences in career satisfaction. Rather, the authors encouraged future research to investigate other variables that may better explain career satisfaction. THEORY OF WORK ADJUSTMENT 47 Kaye, Alexander, and Kauffman (1999) provided a unique study of the work experiences of low-income senior citizens. The authors attempted to describe various aspects of the work experiences of this population. Relevant to the present critique was the authors? examination of employees? perceptions of on-the-job discrimination and job satisfaction. Interview data were collected from Asian-American (n=1), Latino/Latina (n=36), African-American (n=65), and European-American (n=60) employees. Of the 164 participants who were part of the study, two did not report their race. Males comprised 33.5% (n=55) of the sample. Women made up 66.5% (n=109) of the sample. The participants ranged in age from 54-86 (no means or standard deviations were provided). These participants were randomly selected from job placement programs designed for senior workers. Of those selected, 63.5% participated in the study. Comparisons between those selected from the study and the general population taking part in the job placement programs revealed that the two groups were similar except for the fact that Latinos were overrepresented in the study?s sample. The interview protocol was designed for the study. However, ?previously developed indices? were also included in the measure. No psychometric information was provided for the measure. A Spanish version of the instrument was available. The methodological procedures were not presented completely. Because of this, it is unclear whether the participants were interviewed verbally or given paper and pencil THEORY OF WORK ADJUSTMENT 48 measures. Also missing from this study were hypotheses or testable research questions. Overall, job satisfaction was high for all participants. However, differences in job satisfaction did exist. European-American employees reported lower rates of job satisfaction than African-American or Latino employees. Rates of job satisfaction did not differ for gender. Overall, participants did not report significant problems with racial discrimination at their places of work. Only five participants (3.1%) reported experiencing racial discrimination at work. Of these participants, three of them reported that it was a ?slight problem? and two reported that it was ?no problem at all? for them. Because workplace racial discrimination was never defined in this study, the reader does not know what aspect of work was discriminatory (e.g., policies, pay, interpersonal interactions). Unfortunately, the race of participants who provided these responses was not made known to the reader. Austin and Dodge (1992) analyzed national data from the General Social Survey from 1973, 1974, 1976, 1977, 1980, 1982, 1984, 1985, and 1987 in an attempt to uncover possible attitudinal differences between African Americans and European Americans. Their research was not specific to job attitudes but an examination of job satisfaction was included among the authors? variables of interest. The authors predicted that experiences with discrimination would negatively influence African Americans? satisfaction with various life domains including work. Unfortunately, the THEORY OF WORK ADJUSTMENT 49 authors? description of the sample was limited to an indication of approximate sample sizes. That is, each year data from approximately 1,500 participants were collected. Job satisfaction was measured with only one item of the questionnaire, which read: ?On the whole, how satisfied are you with the work you do ? would you say you are very satisfied, moderately satisfied, a little dissatisfied, or very dissatisfied??. Background information was also collected on the participants including gender, age, and socioeconomic status (SES). Austin and Dodge (1992) found that African Americans reported greater job satisfaction than European Americans in five of the nine years that they were surveyed. The authors also found that this greater job dissatisfaction was present not only for lower-SES African Americans but also for upper-SES African Americans. When the authors used the interaction between race and sex to predict job satisfaction, the authors found that African American women reported the highest rates of job dissatisfaction. Perhaps because the authors were sociologists, they attempted to explain the differences in job satisfaction as an indication that the presidential administration influenced reports of job satisfaction for African American and European American employees. Specifically, the authors noted that the differences between African Americans? and European Americans? reports of satisfaction were greater during Reagan?s administration than Carter?s administration. Summary. The job satisfaction research reviewed above was not informed by the TWA model because researchers did not report on the interaction between THEORY OF WORK ADJUSTMENT 50 employees? needs and environmental reinforcers, but examined each factor separately. Regardless, this research raised important questions regarding racial differences in the reports of job satisfaction. Researchers offered various reasons to explain these differences in job satisfaction including organizational experiences and presidential administrations. The findings in cross-racial studies of job satisfaction were equivocal. That is, some research reported that African Americans had lower levels of job satisfaction than European-American employees, the subject of most vocational research. Other research suggested otherwise. These findings indicated the need for investigation into the unique variables that may account for these differences. The TWA and Race Cross-cultural considerations in the TWA research are limited to brief postulations by a handful of researchers. The TWA?s original authors encouraged researchers to study the implications of race in the TWA, but they believed that race would not change the nature of the relations that they have outlined in their model (Dawis, 1994; Dawis & Lofquist, 1993). Related to the role of cultural diversity in the TWA research, Dawis (1994) stated that because the TWA is an individual differences model, research should focus on individuals rather than groups. Furthermore, he purported that cultural group membership can not be used to predict any of the TWA variables. Dawis and Lofquist (1993) predicted that cross-cultural research on the TWA will result in the ability to confirm the TWA research with racial groups other than European Americans. THEORY OF WORK ADJUSTMENT 51 Tinsley (1993) disagreed. She suggested that personal and familial characteristics of employees should be considered in the determination of job satisfaction because of the potential for these types of variables to moderate or mediate relations between variables in the TWA model and, thus, influence job satisfaction. However, Tinsley never mentioned race or racial climate specifically as variables that might play a role in the TWA model. Furthermore, she did not specify where within the TWA model personal and familial variables might influence relations between existing TWA variables. However, she advised researchers to partial out such cultural variables as job discrimination and sexual harassment before attempting to determine the relation between correspondence and satisfaction and satisfactoriness. In this statement, she recognized the potential of climate to influence the relations between the TWA variables. Regarding gender and related discrimination, Fitzgerald and Rounds (1993) hypothesized that these cultural variables might influence person-environment correspondence and satisfaction in the TWA model. Specifically, they urged researchers to consider how company policies handle sexual harassment when measuring female employees? needs as well as workplace reinforcers and the resulting correspondence between the two. They also stated that job satisfaction measures should include questions related to sexual harassment. Rounds and Hesketh (1994) interpreted Fitzgerald and Rounds? (1993) propositions regarding the role of sexual harassment to mean that harassment may THEORY OF WORK ADJUSTMENT 52 moderate the relation between needs/reinforcer correspondence and job satisfaction. Extrapolating from Fitzgerald and Rounds? predictions, we can assume that racial climate can also be integrated into the concept of correspondence in similar ways. That is, it might be important to measure African-American employees? needs related to racial climate (e.g., discrimination policies, experiences with racial discrimination) and the racial climate found in the workplace as well as employees satisfaction with the racial climate. Furthermore, researchers might also assume that racial climate moderates the relation between correspondence and job satisfaction. Rounds and Hesketh also advised that discrimination might influence the relation between fit and tenure. Moderation in this example would mean that the relation between fit and tenure intentions would become harder to predict in the presence of workplace discrimination. Rounds and Hesketh (1994) presented what might be the most comprehensive explanation of the ways in which culture may factor into the TWA research. They considered the following two aspects of culture: group differences and experiences with discrimination. Agreeing with Dawis (1994), the authors warned researchers away from conducting studies focused on group differences. Because between-group studies do not offer information about how culture influences work adjustment and in the past have been methodologically flawed, the authors advised researchers to conduct within-group, longitudinal research when interested in cultural variables. THEORY OF WORK ADJUSTMENT 53 They proposed investigating discrimination as a more viable approach to studying culture in the context of the TWA than group differences. According to Rounds and Hesketh (1994), discrimination may influence job satisfaction in the following three ways: indirectly, by influencing the correspondence between needs and reinforcers; by directly influencing TWA variables; and by moderating the relationship between TWA variables. The way that discrimination may affect TWA variables has implications for the TWA model. If discrimination exerts its influence on TWA outcomes variables (e.g., turnover intentions) through correspondence (i.e., a mediated model), then the TWA accounts for discrimination through correspondence and the theory is not in need of modification related to this variable. However, if discrimination directly accounts for a proportion of the variance in outcomes variables or if it acts as a moderator between TWA variables, then the TWA model is in need of modification to more accurately represent the role of discrimination in work adjustment. Swartz-Kulstad and Martin (2000) used the idea of person-environment fit to explain the process of cultural adaptation. They suggested that those who are culturally different from their environments may experience greater discordance than those who are culturally similar. Particularly, they hypothesized that this discordance might influence the skills/requirement dimension of fit though the authors do not specify their environment of focus (e.g., workplace, community). In the TWA model, skills/requirement correspondence is thought to predict an environment?s satisfaction THEORY OF WORK ADJUSTMENT 54 with those who inhabit it (Dawis & Lofquist, 1984). Therefore, based on Swartz- Kulstad and Martin (2000)?s assumptions, cultural dissimilarity can cause a lack of person-organization fit which may, in turn, cause lowered job satisfaction for employees who are culturally different from their workplaces. Lovelace and Rosen (1996) presented the only empirical study of the influence of race in person-organization fit which also provides a test of the postulations of Swartz- Kulstad and Martin (2000). Considering research that suggests that people of color and European-American women have different standards of career success, possess different interests, and have less access to on-the-job mentors, Lovelace and Rosen hypothesized that European-American women and people of color (i.e., Latinos/Latinas and African Americans) would report lower levels of perceptions of fit than European-American males. The authors also hypothesized that perceptions of fit would be positively related to job satisfaction and negatively related to intentions of quit. Furthermore, they hypothesized that perceptions of fit would mediate the relation between race and sex and job satisfaction and intention to quit. Lovelace and Rosen (1996) tested their hypotheses using a sample of MBA graduates (n=366) who were currently employed as managers. The participants were European- American males (24%), European-American females (43%), African-American males (12%), African-American females (11%), Latinos (6%), and Latinas (3%). Participants were surveyed using a previously established measure of job satisfaction, the three-item Michigan Organizational Assessment Questionnaire. A one-item THEORY OF WORK ADJUSTMENT 55 measure of intention to leave (?I am looking for a job in a different organization?) and a 14-item measure of perception of fit were constructed by the authors based on literature on the subject matters. The perception of fit measure included items that tapped into sense of employees? organizational compatibility related to values, goals, skills, interests, evaluations of career success, and participation in activities outside of work. Initial analyses revealed a difference in perceptions of fit, job satisfaction, and intention to quit for race but not for sex or the interaction between sex and race. Therefore, data for African American males and females and Latinos and Latinas were combined. Results suggested that African-Americans reported lower levels of perceptions of person-organization fit than European-American males. However, European-American women and Latinos/Latinas did not report lower levels of perceptions of fit than European-American males. Supporting the authors? hypotheses, which were informed by the TWA (Dawis & Lofquist, 1976, 1978), perceptions of fit were found to be positively related to job satisfaction and negatively related to intention to quit. Additionally, results suggested that perceptions of fit act as a mediator between race and job satisfaction and intention to quit, suggesting the need to consider psychological variables when attempting to explain differences in job satisfaction. The authors also reported that perceptions of fit were negatively related to reports of job stress. THEORY OF WORK ADJUSTMENT 56 These findings concerned Lovelace and Rosen (1996) because (a) African-American employees were reporting lower levels of perceptions of person-organization fit than employees from other racial groups and (b) many consequences are present for employees who report low levels of person-organization fit (i.e., lowered job satisfaction, increased intentions to quit, increased job stress). The authors encouraged organizations to implement socialization practices that may help increase African-American employees? person-organization fit perceptions (e.g., education, affirmative action programs). Summary. Few researchers have commented on the role of culture in the TWA research. Among the suggestions for considering culture in the TWA is the possibility that cultural variables (e.g., racial climate) moderate the relationship TWA variables or that they directly influence TWA variables (Fitzgerald & Rounds, 1993; Rounds & Hesketh, 1994; Tinsley, 1993). This suggestion is in opposition to other researchers who assume that race will not change the nature of the relationship between TWA variables (Dawis, 1994; Dawis & Lofquist, 1993). These researchers believe that cultural variables (e.g., racial climate) may play a role in an employee?s indication of correspondence, which in turn would influence TWA outcome variables (e.g., job satisfaction). Lovelace and Rosen (1996)?s research provided support for this view of how culture might influence the relationship of the TWA variables. THEORY OF WORK ADJUSTMENT 57 Racial Climate Racial climate is defined as the atmosphere in an organization as it relates to the ways in which a) interpersonal processes (e.g., warmth, acceptance, support), b) decision making processes (e.g., involvement in decision making), and c) the reward systems (e.g., promotion) are determined by and address organizational racism and employees? race(s) (Barbarin & Gilbert, 1981; Watts & Carter, 1991). Based on Schneider and Hall?s (1972) research on organizational climate, Watts and Carter (1991) assumed that racial climate is determined by objective environmental characteristics and subjective person characteristics. In other words, racial climate is based on reality and perception. Holder and Vaux (1998) defined racial climate simply as perceptions of racism in the workplace. Watts and Carter (1991) were interested in the relation of racial identity to various racial variables when they surveyed African American workers. The authors were most interested in African Americans? perceptions of racism in their working environment as it related to institutional racism, racial climate, and personal discrimination. In this study, institutional racism was defined as the use of institutional policies and procedures that result in systematic advantages for the majority group and disadvantages to members of the minority group. The authors conceptualized racial climate as a unit-level phenomenon related to racial interactions, race, and the interpretations of institutional policies created by those in the majority and those with power in the organization. The authors predicted that THEORY OF WORK ADJUSTMENT 58 racial climate would have more of a direct influence on employees? workplace experiences than institutional racism. Finally, the authors described personal discrimination as individual employees? perceptions of job selection, promotion, performance evaluation, and interactions with other individuals in the workplace related to discrimination. In their study, Watts and Carter (1991) described racial identity as a psychological variable related to race. Racial identity is the psychological process by which individuals come to define themselves culturally, react to their respective racial groups, as well as relate to other cultural groups and locate themselves within a predominately White society (Helms, 1990). Watts and Carter were interested in the racial identity of African-American employees and describe four racial identity statuses. The first status is Preencounter which is characterized by a rejection of African-American culture and identification with European-American culture or ignorance of the influence of race. Those endorsing the Encounter status wish to identify with African-American culture but still feel ambiguous about their racial group membership. In the Immersion-Emersion status, African Americans ?immerse? themselves in African-American culture while rejecting European-American culture. When African Americans endorse the Internalization status they no longer reject European-American culture, but recognize the positive and negatives of both cultural groups. Those in this status possess a strong identification with the African-American cultural group. THEORY OF WORK ADJUSTMENT 59 The relationship between racial identity and organizational climate was assessed with the following measures: institutional-racism items, the racial climate scale (Watts & Carter, 1991), the personal discrimination scale (Watts & Carter), and the racial identity scale (Helms & Parham, 1990). The racial climate scale consisted of two subscales: experience and intensity of racism subscale, which measured participants? perceptions of racism in interpersonal interactions and as well as at the institutional level and, the management power and policy subscale, which measured respondents? perceptions of whether power is used to better African Americans? experiences in the workplace. This scale also tapped into whether African Americans maintain powerful positions in the workplace. Data were collected from African-American employees (n=142) in a civil service agency. Most of the participants had worked in the agency for at least 10 years and most made between $25,000 and $35,000. Sixty-four percent of those surveyed were women. The median age of the group was 44. Eighty-five percent of the participants were African American. The remainder of the sample identified as West Indian or African. Watts and Carter (1991) found that racial identity was related to racial climate such that those endorsing the Preencounter status reported more positive evaluations of their organization?s racial climate and those endorsing Immersion/Emersion and Internalization statuses reported less favorable evaluations of the experience and intensity of racism subscale of the racial climate scale (i.e., more frequent and intense THEORY OF WORK ADJUSTMENT 60 experiences with racism). Immersion/Emersion and Internalization statuses were unrelated to perceptions of management power and policy. Immersion/Emersion was related to personal discrimination. That is, those endorsing the Immersion/Emersion status reported more experiences with personal discrimination. For the most part, the authors? hypotheses were confirmed. That is, those operating in the Internalized and Immersion/Emersion statuses were more likely to report experiences with racism than those in the Preencounter status. Building on the research of Watts and Carter (1991), Holder and Vaux (1998) attempted to determine the relationships between job satisfaction and several predictor variables including job stressors (race-related and general), workplace spirituality, internal locus of control, and social support (work- and non-work related). The authors were also interested in the ability of social support, workplace spirituality, and internal locus of control to moderate the relationship between stressors and job satisfaction. Holder and Vaux (1998) surveyed African American professionals (n=112) employed in workplaces with 20% or fewer other African Americans. The sample consisted of males (n=68) and females (n=41). Most participants reported earning above $40,000 annually and had worked in their current professions 10.9 years on average (standard deviation was not reported). Data related to participants? perceptions of occupational stress and racial climate were collected using Watts and Carter?s (1991) racial climate scale as well as a Racial Treatment Scale constructed by THEORY OF WORK ADJUSTMENT 61 Holder (1994) and a Personal Discrimination Scale constructed by Watts and Carter. The authors also collected data related to internal locus of control, workplace spirituality, social support, and job satisfaction. Preliminary analyses were performed to test for the effects of work setting and gender on the dependent variables before a test of the major hypotheses were performed. Work setting and gender were not related to the dependent variables (Holder & Vaux, 1998). The authors reported a negative relationship between job stressors (race-related and general) and job satisfaction. Internal locus of control and work- related social support were related positively to job satisfaction. Race-related stressors, racial climate, and personal discrimination were negatively related to job satisfaction. Internal locus of control and work-related social support did not emerge as moderators of the relation between stressors and job satisfaction. Furthermore, non-work related social support and workplace spirituality were not directly or indirectly related to job satisfaction. In summary, Holder and Vaux (1998) provided confirmation for the ability of racial climate and racial stressors to predict the job satisfaction of African-American employees. Hughes and Dodge (1997) investigated the extent to which perceived racial discrimination at work predicted job quality. Considering past research on racial discrimination, the authors hypothesized that two types of workplace discrimination would predict job satisfaction: institutional discrimination and interpersonal prejudice. Institutional discrimination was similar to Watts and Carter?s (1991) THEORY OF WORK ADJUSTMENT 62 institutional racism (i.e., the use of institutional policies and procedures that result in systematic advantages for the majority group and disadvantages to members of the minority group), except that Hughes and Dodge (1997) specified African Americans as the minority group. Interpersonal prejudice was defined as behaviors and interactions informed by negative perceptions of African Americans. The authors did not define job quality. Of secondary importance to the authors was the identification of factors that led to perceptions of workplace discrimination. Data were collected from 79 African American women employed in full-time work. The sample ranged in age from 21 to 53 years, with a mean of 37 years (the standard deviation was not reported). The women had been employed in their current jobs for an average of 7.5 years (the standard deviation was not reported). Twenty- three percent reported that they worked with ?no? or ?few? other African Americans. The participants? median income was between $10,000 - $24,999 per year. Once a portion of the sample was identified, this smaller sample was asked to identify other potential participants. Preestablished self-report measures were used to measure all but one of the variables in this study, workplace racial discrimination. As is common in other studies of workplace racial climate and discrimination, the authors constructed a new measure for this study. Items were generated from a series of focus groups with African American workers. Factor analysis revealed two subscales similar to conceptualizations of racial discrimination in past research: institutional THEORY OF WORK ADJUSTMENT 63 discrimination and interpersonal prejudice. A global job satisfaction measure was used to measure job quality. A strength of this study was Hughes and Dodge?s (1997) consideration of past variables that have been related to job quality when conducting analyses. These authors controlled for organizational demography (i.e., the number of women and people of color in the organization), salary, job category (i.e., managerial/professional, clerical/sales, and unskilled/service), and non-racial work stressors. After controlling for these variables, the authors found that interpersonal prejudice and the combination of interpersonal prejudice and institutional discrimination were predictors of job satisfaction, accounting for 11% of the variance after controlling for the previously mentioned variables. Related to the second hypothesis, Hughes and Dodge (1997) found that job category and organizational demography were related to institutional discrimination and interpersonal prejudice. More specifically, job category and organizational demography accounted for 25% of the variance in institutional discrimination after controlling for age and education. Women in clerical and sales jobs reported more institutional discrimination than women in other job categories. Organizational demography accounted for 12% of the variance in interpersonal prejudice after controlling for age and education such that women who worked with fewer African Americans reported more interpersonal prejudice than women who worked with more THEORY OF WORK ADJUSTMENT 64 African Americans. These findings provide important information about the factors related to workplace racial discrimination. Interested in conducting a national study of work experiences of people of color in the United States, Brown, Minor, and Jepson (1991), in collaboration with the National Career Development Association, hired the Gallup Organization to randomly survey 1,350 working adults over the telephone. African Americans (n=232), Asian Americans (207), and Latinos (n=199) were oversampled. However, the final sample was weighted to reflect population as reported by the 1988 Bureau of Census Current Population Survey. Weighting followed a three-step process involving correcting for the disproportion created by oversampling to make the sample proportional to the number of those in the U.S. with phones. Next, the authors attempted to resolve discrepancies in other demographic characteristics (e.g., age, sex, education). Finally, the sample was adjusted to the racial proportions represented in the Census data. In the resulting sample, 53% of the participants were male and 47% were female. Participants were surveyed using a questionnaire developed for the Brown et al. (1991) study. The survey was constructed by researchers in the field of vocational education and career development, reviewed by experts and pilot tested. The survey contained questions related to perceptions of workplace discrimination, turnover intentions, and other vocational experiences questions. After data were collected, cross-tabulations were calculated. The cross-tabulations provided descriptive information related to how certain racial and ethnic groups THEORY OF WORK ADJUSTMENT 65 answered certain questions. Brown et al. (1991) found that African-American employees were more likely than other groups, with the exception of Asian-American employees, to report with greater frequency that race- and sex-based discrimination existed in their workplaces. Of the African-American employees sampled, 48% felt that no discrimination existed in their workplaces. When asked whether minorities but not women experienced discrimination at work, 11% agreed. When asked whether women but not minorities experienced discrimination, 4% agreed. Of the African- American employees sampled, 31% reported that both women and minorities experienced discrimination in their workplaces. When participants were asked, ?Thinking ahead for the next three years, which one of the following best describes what is most likely to happen with regard to your present job?? Fifty percent of African Americans surveyed expected that they would still be working for the same employer, whereas 63% of Latinos and Latinas and 57% of European Americans anticipated remaining with their employers. Unfortunately, no analyses were run that would allow for investigation of the relation between experiences with discrimination and turnover intentions. Also interested in the personal consequences of perceived workplace discrimination, Mays, Coleman, and Jackson (1998) investigated the relationship between perceived racial discrimination and demographic variables on job stress in a sample of African American women. Job stress was never defined but the reader is THEORY OF WORK ADJUSTMENT 66 informed of the variable?s relation to other work variables (e.g., absenteeism, productivity). Data were obtained from a national probability sample of African Americans. This sampling procedure resulted in a sample size of 2,107 with a response rate of 69%. Because only employed female participants were surveyed as part of the study, the final sample size was 753. Participants were engaged in a 2.5 hour interview with an African-American interviewer. Rather than relying on existing measures of perceived racial discrimination and job stress, an interview protocol was designed for the Mays et al. (1998) study. The authors reported that their protocol was based on a literature review of the ?ethnic-racial concerns of African Americans? (p.321). Interview data were dummy- coded to be entered into regression equations. The measure was also reviewed by experts and pilot tested. Mays et al. (1998) found that perceived racial discrimination was a significant predictor of job stress in the lives of African American women. Perceived racial discrimination remained a significant predictor of job stress even after demographic variables were entered in the first step of a regression equation. Age and education were also significant predictors of job stress. Younger participants and participants with more education were more likely to report job stress. Cohn (1997) intended to examine the influence of racial discrimination on the workplace attitudes and career development process of people of color. She used THEORY OF WORK ADJUSTMENT 67 Super?s (1990) model of career development as a guide. This is a developmental stage model which begins with the growth stage spanning from ages 4-14. During this stage children begin to develop an awareness of their interests and abilities through play. In the exploratory stage (ages 14-25), young adults begin to make choices that will influence their later career choices. The next stage is the establishment stage which lasts from approximately age 25 through 45. During this stage, individuals? career choices begin to become more concrete. During the maintenance stage, ages 45 through 65, individuals attempt to ?maintain? their career identities by remaining current. Alternatively, adults can experience the maintenance stage as a period when they slow down. Finally, during the decline stage, after age 65, most individuals retire or identify alternative career or recreational activities. In Cohn?s (1997) study, qualitative data related to participants? experiences with racial discrimination at work within the context of Super?s (1990) model were collected using interviews. She provided data from nine of her 22 participants who she described as ?minority persons.? These nine represented the only participants who reported discrimination which later influenced their process of career development. While the data from these nine participants may be useful in helping to identify the ways in which racial discrimination may influence career development, they do not represent the rest of the voices in her sample (e.g., did others experience positive racial climates?). THEORY OF WORK ADJUSTMENT 68 Among the experiences that Cohn?s participants reported was lack of a ?critical mass? in terms of the numbers of other employees who were also people of color, inability to conform to the organizational culture, and having to deal with superiors who were uncomfortable interacting with people of color. The author reported that these experiences had short-term influences (e.g., decreased self-esteem, negative influence on outcome expectations related to school) and long-term influences. These long-term influences included doubt which influenced feelings of person-environment fit and an inability to express and utilize their talents fully. On a somewhat positive note, Cohn reported that her participants learned to become their own source of encouragement when it came to career goals. This study was important in its suggestion that workplace racial discrimination may influence more than employees? on-the-job experiences. Workplace racial discrimination may interrupt the career developmental process. Surprisingly, little direct research exists to explain how racial climate impacts variables relevant to organizations. In a unique study that does acknowledge the possibility that racial climate may influence organizational variables, Ensher et al. (2001) examined the role of perceived discrimination on job satisfaction, organizational commitment, organizational citizenship behaviors, and grievances. Organizational commitment is ??the psychological bond that an employee has with an organization?? (Ensher et al., 2001, p.56). Presumably, this construct was related to the congruence between employees? goals and values as well as employees? THEORY OF WORK ADJUSTMENT 69 intention to remain in an organization (Mowday, Porter, & Steers, 1982). Related to organizational commitment were organizational citizenship behaviors (OCBs). OCBs described employees? efforts to ?go the extra mile?; that is, the prosocial, extrarole behaviors that employees engaged in to assist their peers and better the organization (Organ, 1988). Finally, in the present study, grievances were defined as formally made complaints related to reported experiences with race- or sex-based discrimination. The authors predicted that workplace discrimination would predict job satisfaction, organizational commitment, OCBs, and grievances. More specifically, they predicted that discrimination at the organizational level would predict more of the variance in the outcome variables than discrimination from supervisors and coworkers. Ensher et al. (2001) were unique in that they presented a theoretical foundation that guided their research. More specifically, they suggested that social identity theory might explain why discrimination may be related to negative work behaviors (i.e., workers may feel alienated because of unfair treatment and want to react). Also, this theory explained the importance of considering discrimination from supervisors and coworkers because social identity theory considers an individual?s social groups. A racially mixed sample of employees (n=366) provided researchers with their perceptions of their experiences of the variables of interest to the researchers (i.e., THEORY OF WORK ADJUSTMENT 70 discrimination, job satisfaction, organizational commitment, and OCBs) and reported whether they had filed grievances with the companies in which they worked. The participants were 70% women. They were 45% Latino, 15% African American, 10% Asian American, and 22% European-American. The participants represented a variety of organizations, working in a variety of jobs. A measure of discrimination was created by the authors that aimed to measure workplace discrimination at three levels: discrimination from the co-worker, discrimination from the supervisor, and organization level discrimination (e.g., discriminatory policies). Reliability estimates for the subscales ranged from .65 to .90. Validity information for the scale was not provided. Surprisingly, Ensher et al. (2001) found no racial group differences in perceptions of discrimination. Perceptions of discrimination were related to job satisfaction, organizational commitment, and organizational citizenship behaviors. As expected, organizational discrimination was a better predictor of these variables than supervisor and coworker discrimination. Because of the lack of research investigating the relation between racial climate and turnover intentions, it is necessary to make inferences about this possible link based on the little research that exists relating workplace gender discrimination and turnover intentions. Shaffer et al. (2000) offered research investigating the relation between gender discrimination (i.e., sexual harassment and gender evaluation) and various outcome variables including turnover intentions and job THEORY OF WORK ADJUSTMENT 71 satisfaction. Other outcome variables of interest to Shaffer et al. were organizational commitment and life stress. Participants in the Shaffer et al. (2000) study were women (n=583) who were employed in the United States (n=126), mainland China (n=103), and Hong Kong (n=354). Other demographic information on the participants was not provided. Participants were given questionnaires measuring job satisfaction (Hackman & Oldham, 1975), organizational commitment which measured both affective commitment and normative commitment (Meyer, Allen, & Smith, 1993), turnover intentions (Hom & Griffith, 1991), life stress (Quinn & Staines, 1979), sexual harassment (comprised of gender harassment, unwanted sexual attention, and sexual coercion; Fitzgerald et al., 1988; Gelfand, Fitzgerald, & Drasgow, 1995). The final variable measured was gender evaluation, measured with a scale constructed by Joplin and Shaffer (1997). In the Shaffer et al. study, gender evaluation was defined as participants? perceptions of the degree to which gender informs organizational decision-making (e.g., ?Gender played a role in the last performance evaluation I received.?). In the United States participants completed English versions of the questionnaire. In Hong Kong bilingual participants were randomly selected to complete either English or Chinese versions of the questionnaire. In mainland China, all participants completed Chinese versions of the survey. Chinese versions of the questionnaire were translated and back translated. Also, responses on the English and THEORY OF WORK ADJUSTMENT 72 the Chinese versions of the questionnaire were compared to ensure the equivalency of the English and Chinese versions of the questionnaire. In partial support of their hypotheses, Shaffer et al. (2000) found that gender harassment was positively related to turnover intentions. Unwanted sexual attention was negatively related to job satisfaction. Gender evaluation was negatively related to job satisfaction, affective commitment, and positively related to turnover intentions. Gender discrimination was not predictive of life stress. Summary. According to this research, workplace characteristics and worker characteristics may influence perceptions of racial climate. In turn, research suggested that racial climate has implications for African-American employees. Holder and Vaux (1998), Hughes and Dodge (1997), and Ensher et al. (2001) found that racial climate predicted employees? job satisfaction. Additionally, racial climate may influence African-American employees? self-esteem (Cohn, 1997), outcome expectations (Cohn, 1997), perceptions of person-environment fit (Cohn, 1997), job stress (Mays et al., 1997), organizational commitment (Ensher et al., 2001), and organizational citizenship behaviors (Ensher et al., 2001). Extrapolating from Shaffer et al.?s (2000) research on the effects of on-the-job gender discrimination on turnover intentions, future research might expect to find that on-the-job racial discrimination might also predict turnover intentions. THEORY OF WORK ADJUSTMENT 73 Chapter 3 Statement of Problem Counseling psychologists know little about the work adjustment of African Americans (Brown, 1995; Fitzgerald & Betz, 1994). Reflecting this dearth in the literature is our lack of understanding of the determinants of job satisfaction and turnover intentions for African Americans. Existing research on job satisfaction tends to focus on between-group examinations of the job satisfaction of African Americans and European Americans. However, this research has uncovered inconsistencies in the comparisons of job satisfaction differences (Alexander & Kauffman, 1999; Austin & Dodge, 1992; Brenner & Fernstein, 1984; Greenhaus, 1990; Mueller et al., 1999; Tuch & Martin, 1991). Similarly, research on the comparative levels of turnover in African and European Americans has proven to be equivocal (Borjas, 1984; Flanagan, 1978). This research is largely atheoretical and focused on nominal conceptualizations of race that do not allow for an understanding of the determinants of job satisfaction or turnover intentions for either group. Previously, research based on The Theory of Work Adjustment (TWA; Dawis & Lofquist, 1984; Dawis et al., 1964), a person-environment fit model of work adjustment, successfully explained European-American employees? job satisfaction and turnover intentions (e.g., Bretz & Judge, 1994; Breeden, 1993; Rounds et al., 1987; Saks & Ashforth, 1997). However, only one study to date has supplied research to support the ability of the TWA to explain the work adjustment of African- THEORY OF WORK ADJUSTMENT 74 American employees (Lovelace & Rosen, 1996). Recently, TWA theorists offered suggestions about how culture (i.e., gender, race) may play a role in the TWA model. These suggestions included the possibility that correspondence mediates the relation between culture and job satisfaction (Dawis, 1994; Dawis & Lofquist, 1993), culture moderates the relations between TWA variables (Fitzgerald & Rounds, 1993; Rounds & Hesketh, 1994; Tinsley, 1993), and that culture directly exerts its influence on TWA variables as an independent variable (Rounds & Hesketh, 1994). Guided by the work of Fitzgerald and Rounds (1993) and Rounds and Hesketh (1994), and by research that suggests that discrimination (Shaffer et al, 2000) and perceptions of fit may be related to turnover intentions and job satisfaction (Jackson et al., 1991; Saks & Ashforth, 1997), the present study examined the role of culture in the TWA. More specifically, the present study attempted to determine whether racial climate moderates the relation between perceptions of fit and job satisfaction as well as perceptions of fit and turnover intentions. Additionally, the ability of job satisfaction to predict intentions to quit, a relationship predicted by the TWA and supported by previous research with European American samples, was tested. See Figure 1 for the model of job satisfaction and tenure intentions for African-American employees that was tested in the present study. The hypotheses were also informed by Baron and Kenny?s (1986) work related to the determination of moderators in psychological research. THEORY OF WORK ADJUSTMENT 75 Research Question 1: What are the means, standard deviations, and ranges of African-American employees? perceptions of fit, perceptions of racial climate, job satisfaction, and intentions to quit? Because so little is known about the work experiences of African American (Brown, 1995; Fitzgerald & Betz, 1994), the first purpose of the present study was to provide descriptive data related to the work experiences of African-American employees. In an attempt to gain a better understanding of African Americans? perceptions of racial climate (Watts & Carter), perceptions of fit (Saks & Ashforth), job satisfaction (Weiss et al.), and turnover intentions (Colarelli), the present study presented means, ranges, and standard deviations of these variables for the present study?s sample. Research Question 2: What on-the-job factors play a role in job satisfaction for African-American employees? Existing research has uncovered inconsistencies in the comparison of the job satisfaction of African-American and European-American employees (Alexander & Kauffman, 1999; Austin & Dodge, 1992; Brenner & Fernstein, 1984; Greenhaus, 1990; Mueller et al., 1999; Tuch & Martin, 1991). Some of this research suggested that African Americans were not as satisfied with their jobs as their European- American counterparts (e.g., Austin & Dodge, 1992; Greenhaus, 1990; Tuch & Martin, 1991). Other research indicated that African Americans actually reported higher levels of job satisfaction than European Americans (e.g., Brenner & Fernstein, THEORY OF WORK ADJUSTMENT 76 1984; Mueller et al., 1999). Still, other research found no differences in the job satisfaction of African-American and European-American employees (e.g., Alexander & Kauffman, 1999). Because the data in this area were equivocal and need further exploration, this research question attempted to explicate the determinants of job satisfaction for African Americans. Responses to this open-ended question were coded for compatibility with the six values of the TWA and racial climate. An intra-class correlation was run to determine the inter-rater reliability of the rater?s responses. The mean of raters? codes were used to represent participants? responses. Research Question 3: What differences exist between the relationships between perceptions of fit, job satisfaction, and intentions to quit for African Americans? in the present study and Europeans and European Americans? in past research? The purpose of this research question was to allow for a comparison between existing research on the variables of interest in this study in European-American samples and those produced by an African-American sample in the present study. To answer Research Question 3, correlation coefficients that specified the relationships between perceptions of fit, job satisfaction, and intentions to quit were recorded from previous studies that sampled European and European American samples (Breeden, 1993; Bretz & Judge, 1994; O?Reilly, Chatman, & Caldwell, 1991; Rounds, 1990; THEORY OF WORK ADJUSTMENT 77 Saks & Ashforth, 1997; Van Vianen, 2000) and compared to the correlation coefficients in the present study. Research Question 4: Are perceptions of fit correlated with years in position? Theory (Dawis & Lofquist, 1984; Schneider, 1987) has hypothesized a positive relation between fit and tenure. Furthermore, research supported this hypothesized link (Saks & Ashforth, 1997; Taris & Feji, 2001). However, historical data presented suggested that work is more difficult for African Americans to secure (Berry & Blassingame, 1982; Williams, 1999). This may mean that once African Americans find work, they may be less likely to leave their job than European Americans, regardless of their degree of fit, possibly weakening the relation between fit and tenure. A correlation was used to determine the ability of fit to predict years in position. Research Question 5: Are racial climate and perceptions of fit correlated? Dawis (1994) and Dawis and Lofquist (1993) predicted that cultural variables may influence employees? perceptions of fit. Swartz-Kulstad and Martin (2000) endorsed this idea suggesting that those who are culturally different from their environments may experience greater discordance than those who are culturally similar. These postulations were used to inform Research Question 5. Research Question 5 was tested with a correlation. THEORY OF WORK ADJUSTMENT 78 Hypothesis 1: Perceptions of fit will account for a percentage of the variance in job satisfaction, and participants with a greater degree of fit will experience greater job satisfaction than those reporting a lesser degree of fit. Dawis and Lofquist (1984) outlined a number of propositions that correspond with the TWA. Among these propositions is Proposition III which predicted that need/reinforcer correspondence leads to job satisfaction (Dawis & Lofquist, 1984). Research provided support for this relationship with European Americans (e.g., Bretz & Judge, 1994; Rounds, 1990; Rounds et al., 1987; Taris & Feij, 2001). This hypothesis was tested with an African American sample using regression analysis in the present study. Hypotheses 1 through 3 were tested in a manner consistent with Baron and Kenny?s 3-step procedure for testing for the existence of a moderator. To test Hypothesis 1, perceptions of fit was entered in the first step of a regression analysis where job satisfaction was the dependent variable. To test Hypothesis 2, racial climate was entered in the second step of the same regression equation and to test Hypothesis 3, the interaction between perceptions of fit and racial climate was entered in step 3. Hypothesis 2: Racial climate will account for a percentage of the variance in job satisfaction, and participants reporting positive racial climates will experience greater job satisfaction than those reporting negative racial climates. Although this is a new line of research, Holder and Vaux (1998)?s study suggested that racial climate was related negatively to job satisfaction. That is, THEORY OF WORK ADJUSTMENT 79 African-American employees who reported positive racial climates, expressed as lower scores on a racial climate scale (Watts & Carter, 1991), reported higher levels of job satisfaction than those African-American employees who reported more negative racial climates at work. The present study tested this relationship using regression analysis. Hypothesis 3: The interaction between perceptions of fit and racial climate will account for a percentage of the variance in job satisfaction, and higher levels of fit and positive racial climates will result in higher levels of job satisfaction. This hypothesis suggested that racial climate acts as a moderator of the relation between perceptions of fit and job satisfaction. Baron and Kenny (1986) advised that the presence of a moderator is determined by the significant relationship of the interaction between the proposed independent and moderator variables and the dependent variable. In other words, this was tested by investigating when perceptions of fit x racial climate was predictive of job satisfaction. Hypothesis 4: Perceptions of fit will account for a percentage of the variance in turnover intentions, and participants with a greater degree of fit will experience lower turnover intentions than those reporting a lesser degree of fit. The direct relationship between perceptions of fit and turnover intentions was not outlined in the TWA (Dawis & Lofquist, 1984). However, empirical research supported the direct relation between these variables (Jackson et al., 1991; Saks & Ashforth, 1997). Namely, Jackson et al. (1991) and Saks and Ashforth (1997) THEORY OF WORK ADJUSTMENT 80 presented research that drew a direct link between employees? perceptions of fit and turnover intentions (Jackson et al., 1991; Saks & Ashforth, 1997) and actual turnover (Saks & Ashforth, 1997). Theoretical postulations hypothesized a positive correlation between fit and tenure (Dawis & Lofquist, 1984; Schneider, 1987). Similar to the first three hypotheses, Hypotheses 4 through 6 were tested using Baron and Kenny?s 3-step procedure for testing for moderator effects. A second regression equation was used to test these hypotheses where perceptions of fit was entered in the first step, racial climate was entered in the second step, and the interaction between perceptions of fit and racial climate was entered in step 3. Intentions to quit was the dependent variable. Hypothesis 5: Racial climate will account for a percentage of the variance in turnover intentions, and participants reporting positive racial climates will report lower levels of turnover intentions than those reporting negative racial climates. Unfortunately, there is a lack of research in this area. However, Flanagan (1978) reported that pay discrimination between African-American and European- American employees resulted in a higher quit rate for African-American employees. Research by Shaffer et al. (2000) found that gender discrimination resulted in increased turnover intentions for American and Chinese women. This research was used to investigate the possible relation that may be found in research on racial climate perceptions and turnover intentions in African Americans. THEORY OF WORK ADJUSTMENT 81 Hypothesis 6: The interaction between perceptions of fit and racial climate will account for a percentage of the variance in turnover intentions, and higher levels of fit and positive racial climates will result in lower levels of turnover intentions. This hypothesis suggested that racial climate acts as a moderator of the relation between perceptions of fit and tenure intentions. This hypothesis was tested by examining the interaction between perceptions of fit and racial climate accounts for a percentage of the variance in turnover intentions (Baron & Kenny, 1986). Hypothesis 7: Job satisfaction will be correlated with turnover intentions, and participants reporting lower levels of job satisfaction will report higher levels of turnover intentions than participants with higher levels of job satisfaction. This hypothesis is consistent with Proposition VII of the TWA (Dawis & Lofquist, 1984) which hypothesized an inverse relationship between job satisfaction and tenure intentions. In other words, employees who experience greater satisfaction with their work will be less likely to want to leave. This Proposition has been supported by several studies (e.g., Breeden, 1993; Bretz & Judge, 1994; Saks & Ashforth, 1997; Taris & Feji, 2001). The relation between these variables in this study was tested with a correlation. Hypothesis 8: The reliability and validity estimates of the perceptions of fit for the measure designed for the present study will be comparable to the reliability and validity estimates of Saks and Ashforth?s (1997) measure of perceptions of fit. THEORY OF WORK ADJUSTMENT 82 The present study attempted to address the methodological limitations recognized in measuring perceptions of fit (Edwards, 1991; Edwards, in press; Kristof, 1996) by using a perceptions of fit measure designed by the present author (Lyons & Brenner, in preparation) and based on the TWA (Dawis & Lofquist, 1984). In an attempt to determine the measure?s usability, reliability and validity estimates of this measure were compared to an existing measure of fit (Saks & Ashforth, 1997). THEORY OF WORK ADJUSTMENT 83 Chapter 4 Method Participants Participants (n=212) were African-American employees of large government agencies (n=88), a public university in the Southeast (n=88), two public middle schools in the Southeast (n=22), one cellular phone chain (n=10), and an electric utilities company (n=4). The sample included 159 females (71.3%) and 52 males (23.3%). One participant did not indicate sex. Most participants identified as Black (n=107; 48.0%) or African American (n=95; 42.6%). Participants ranged in age from 18-70 years old (M=37.57, SD=10.56). In terms of education, most participants attended some college (n=54; 24.2%) or above (n=134; 59.6%). Years in position ranged from .10 years to 35 years (M=5.02, SD=5.82). Years in the organization ranged from .10 years from 41 years (M=7.10, SD=7.43). Table 1 includes additional biodata reported by participants. Measures Biodata. The biodata sheet (See Appendix B) was constructed by the author to collect information that described the sample including race, gender, age, country of origin, length of time in U. S., socioeconomic status, and occupational status. Perceptions of fit. Perceptions of the correspondence between employees? needs and workplace reinforcers were measured using Saks and Ashforth?s THEORY OF WORK ADJUSTMENT 84 Table 1. Biodata Category Total Percentage Mean Standard Deviation Range Gender Female 159 71.3% Male 52 23.3% Race/Ethnicity Black 107 48.0% African American 95 42.6% Negro 2 .9% Multiracial 2 .9% Caribbean 1 .4% Age 37.57 10.56 18-70 18-30 68 30.2% 31-50 117 52% 51-70 23 9.2% Highest Degree Earned Less Than High School 1 .4% High School Diploma 20 9% Some College 43 19.3% Associate?s Degree 11 4.9% Bachelor?s Degree 65 29.1% THEORY OF WORK ADJUSTMENT 85 Graduate/Professional School 68 30.5% Hours/Week Worked 39.21 6.04 8-60 8-30 Hours 13 5.5% 31-39 Hours 12 5.4% 40 Hours 160 71.7% 41-60 Hours 23 11.2% Years in Position 5.06 5.82 .10- 35 Less Than 5 Years 152 67.8% 5-15 Years 57 25.2% 16-25 Years 7 3% 26-35 Years 5 2.1% Years with Organization 7.10 7.43 .10- 41 Less Than 5 Years 122 54.1% 5-15 Years 68 30% 16-25 Years 17 7.3% 26-41 Years 10 4.3% Born in U.S.? Yes 193 86.5% THEORY OF WORK ADJUSTMENT 86 No 17 7.6% THEORY OF WORK ADJUSTMENT 87 (1997) perceptions of fit measure (See Appendix C) and a measure constructed by the author (Lyons & Brenner, in preparation; Appendix D). The Saks and Ashforth (1997) measure is a global measure of employees? perceptions of fit with two subscales: person-job fit (e.g., ?To what extent does the job fulfill your needs??) and person- organization fit (e.g., ?To what extent are the values of the organization similar to your own values??). Each subscale contains 5 items. Participants rated their agreement with each item using a 5-point rating scale where 1= ?To a very little extent? and 5 = ?To a very large extent.? Responses were summed for a total score. High scores on this scale represented that participants report a better fit. The person- job fit (coefficient alpha = .89) and person-organization fit (coefficient alpha = .92) subscales displayed adequate internal consistency (Saks & Ashforth). The total scale was used for the present study. Alpha reliability in the present study was .92. Saks and Ashforth (1997) found support for the construct validity of the measure with a confirmatory factor analysis. A two-factor (i.e., person-job fit, person- organization-fit) solution was confirmed for the measure. Furthermore, the Saks and Ashforth (1997) scales were correlated with two two-item scales that measured person-job fit and person-organization fit that the authors developed. The authors found a greater correlation between the two measures of person-job fit and the two measures of person-organization fit than between the person-job and person- organization fit measures, providing support for convergent and discriminant validity. Validity data on this scale were collected from participants in the Saks and Ashforth THEORY OF WORK ADJUSTMENT 88 (1997) study as well as a separate sample that was utilized solely for validity data. This measure of fit was used in all research questions and hypotheses involving fit in the present study. Total scores were used. The second measure of perceptions of fit (Lyons & Brenner, in preparation) is a measure that assessed employees? perceptions of values fit based on the needs and values outlined in the Theory of Work Adjustment (Dawis & Lofquist, 1984). It was used only when testing Hypothesis 8 in order to provide reliability and validity information for this scale. This measure consists of six hypothesized values (i.e., achievement, safety, autonomy, altruism, comfort, status) that comprise the values identified by the TWA to be important (Dawis & Lofquist). The instructions read: ?The following items ask that you indicate the extent to which you value some elements of your workplace compared to how your workplace values these elements.? Participants rated their agreement with each item (e.g., achievement, independence) using a seven-point scale as follows: ?1-2-3-4-3-2-1,? where the first ?1? equaled ?My organization values this more than I?, ?4? represented ?Similar level of value?, and endorsement of the second ?1? indicated ?I value this more than my organization.? Therefore, ?4? represented the greatest degree of fit employees could perceive between themselves and their organizations. Responses to this scale were recoded as ?1-2-3-4-5-6-7? before entering the data in order to test for linear and quadratic equations. Total scores were used in the present study. The resulting alpha coefficient (.85) from a pilot study using data from a national study of gay, lesbian, THEORY OF WORK ADJUSTMENT 89 and bisexual employees indicated that the scale displays adequate internal consistency reliability (Lyons & Brenner, in preparation). Support for the measure?s construct validity was supplied by results of a pilot study which indicated that the measure was positively correlated with the Saks and Ashforth (1997) measures of person- organization and person-job fit and with the Minnesota Satisfaction Questionnaire (MSQ; Weiss et al., 1967), a measure of satisfaction based on the TWA. The alpha coefficient for the present study was .92. Racial climate. Racial climate was measured with a revised version of Watts and Carter?s (1991) racial climate scale (See Appendix E). The measure was revised to evaluate racial climate at a group level rather than measuring individual employees? perceptions of racial climate by modifying the stem to state ?Below are statements about the behaviors and policies that may or may not apply to your workplace. When completing the questionnaire, pretend that you are an ?on site reporter? reporting what you observe in your workplace.? The anchors were also changed from measuring agreement to measuring frequency with the anchors ?Never,? ?Sometimes,? and ?Always.? The racial climate scale consists of 18 items measuring decision-making processes, reward system, perceptions of ?adverse impact,? and interpersonal processes. For example, one item asks: ?Black and White employees generally have good working relationships here.? Item responses were obtained using a 5 point scale ranging from 1 = ?never? to 5 = ?always.? In the present study item responses were THEORY OF WORK ADJUSTMENT 90 summed for a total score where high scores represent more positive perceptions of racial climate than low scores. Watts and Carter (1991) reported an alpha of .83 and Holder and Vaux (1998) reported an alpha of .91 for this scale. For the present study, an alpha of .88 was found. Hypotheses involving racial climate were confirmed in both studies (Holder & Vaux, 1998; Watts & Carter, 1991), thus lending support to the scale?s construct validity. More specifically, using the racial climate scale, Watts and Carter (1991) found support for the hypothesis that African-American employees? psychological perceptions of the importance of race in their lives and how they interact with members of other races (i.e., racial identity status) was predictive of racial climate. Furthermore, Holder and Vaux (1998) confirmed their hypothesis that racial climate would be negatively related to job satisfaction for African-American employees. In the present study, racial climate scores were collected at the individual level and analyzed at the departmental level. To do this, individual total scores were aggregated (i.e., averaged) by department resulting one racial climate score per department. Job satisfaction. Job satisfaction was measured with the Minnesota Satisfaction Questionnaire (MSQ; Weiss et al., 1967; See Appendix F). Although this questionnaire was developed many years ago, it remains a well-used measure of job satisfaction, especially when collecting data related to the TWA (e.g., Bizot, 1993; Holder & Vaux, 1998). This scale assesses 20 dimensions in the consideration of job THEORY OF WORK ADJUSTMENT 91 satisfaction. The 20 dimensions of job satisfaction are: ability utilization, achievement, activity, independence, variety, compensation, security, working conditions, opportunity for advancement, recognition, authority, social status, co- workers, moral values, social service, reactions to company policies and practices, the human relations aspects of supervision, the technical aspects of supervision, creativity and responsibility. Participants were asked to rate their satisfaction with various aspects of their work, along a 5-point Likert-like scale which ranges from 1 = ?Very Dissatisfied? to 5 = ?Very Satisfied.? Sample items include: ?The way my job provides for steady employment?, ?The chance to do something that makes use of my abilities?, and ?The freedom to use my own judgement.? Items were summed to determine a total score that was used in analyses. High scores on this scale represent greater job satisfaction than low scores. The 20-item measure was developed by the TWA authors with colleagues (Weiss et al., 1967) and is consistent with the theory?s conceptualization of job satisfaction. A full scale reliability coefficient of .90 has been reported. In the present study, an alpha coefficient of .92 was found. Corroboration of the measure?s construct validity has been provided by studies that have used the measure to support the relationship of the TWA variables (e.g., Bizot & Goldman, 1993). For example, Bizot and Goldman found that values fit was related to job satisfaction. In other research, THEORY OF WORK ADJUSTMENT 92 Holder and Vaux (1998), using the scale, provided support for their hypothesis that racial climate was related to job satisfaction. Turnover intentions. Employees? turnover intentions were measured with Colarelli?s (1984) 3-item intention to quit measure (See Appendix G). The items read as follows: ?If I have my own way, I will be working for my employer one year from now,? ?I frequently think of quitting my job,? and ?I am planning to search for a new job during the next 12 months.? These items were rated along a 1= ?strongly disagree? to 5= ?strongly agree? Likert scale. Item 1 was reverse scored, then responses were summed to determine intentions to quit. Total scores were used in the present study and high scores represented strong intentions to quit. Colarelli (1984) found an alpha reliability of .75 for this scale. Saks and Ashforth (1997) found an alpha reliability coefficient of .86 for this measure. An alpha coefficient of .83 was found in the present study. Support for the scale?s construct validity was provided by Saks and Ashforth (1997). Using Colarelli?s intention to quit Scale to provide support for the TWA model, the authors found that perceptions of person-job and person-organization correspondence explained variance in intentions to quit 4 and 10 months after working at an organization. Qualitative Question. The present study?s author constructed an open-ended question to gather data related to the determinants of job satisfaction for African- American employees. The question was stated as follows: ?What on-the-job factors THEORY OF WORK ADJUSTMENT 93 play a role in job satisfaction for African-American employees?? Responses to this question were coded for compatibility with the six values of the TWA and racial climate. Responses were coded by three graduate students, two Counseling Psychology graduate students and one Industrial / Organizational Psychology graduate student, who are familiar with the TWA. They were trained by the researcher to use the coding system developed for the present study which included a delineation of TWA values and racial climate. Training included refamiliarizing the raters with the TWA by giving them definitions of each TWA value. The author then presented the raters with a sample response, authored by and scored by the author. Raters asked for clarification when necessary before beginning rating. Most training took place via email and phone because of geographical distance between the authors and her raters during this phase of the research. Procedure Participants (n=212) were recruited from large government agencies, a public university in the Southeast, two public middle schools in the Southeast, one cellular phone chain, and an electric utilities company. Between 4 to16 participants from each department in each of the organizations were surveyed. The author identified sites through three methods: professional and personal contacts, snowball sampling, and a list of Black faculty and staff at the large, public university that was sampled. When data was collected at sites where professional or THEORY OF WORK ADJUSTMENT 94 personal contacts worked, the author gave data packets to the contact and asked that contact to distribute questionnaires along with a request for participation to potential participants. These participants had the option of returning questionnaires to the study?s author or the contact person. In the second sampling method, contacts or past participants provided the author with the names and contact information of employees in potential data collection sites. The study?s author contacted these sites to inquire about interest in participation. In the final sampling method, the author contacted participants via email to request their participation. Interested participants were sent data packets via campus mail and asked to send the completed questionnaire back to the author. On a few occasions, participants asked to meet with the author to discuss the study and/or asked her to pick up the questionnaire, which was done. Participants first provided informed consent. Participants also completed the biodata sheet and the following open-ended question: ?What on-the-job factors play a role in job satisfaction for you?? in an attempt to gain further insight into the factors that contribute to job satisfaction for African Americans. To preserve the construct validity of supposed cause and effects in the present study (e.g., responses to one measure will not influence responses to another; Cook & Campbell, 1979), data from 164 participants were collected using a split sample methodology. That is, approximately one-half of the participants in one department received the racial climate scale (Watts & Carter, 1991) and the remaining participants received the perceptions of fit (Lyons & Brenner, in preparation; Saks & THEORY OF WORK ADJUSTMENT 95 Ashforth, 1997), job satisfaction (Weiss et al., 1967), and intentions to quit (Colarelli, 1984) measures. Data from an additional 48 participants were collected without using the split- sample methodology. These 48 participants received and completed all measures. Preliminary analyses were conducted to ensure that these different methods of data collection did not influence the relationships between the study variables. These analyses are described in the results section. Depending upon the location and availability of the participants, the researcher either collected data packets on the same day, collected them on a date arranged with the participant(s), or the surveys were distributed and collected via post. Participation took approximately 30 minutes. THEORY OF WORK ADJUSTMENT 96 Chapter 5 Results Preliminary Analyses Preliminary analyses were conducted to determine the effects of collecting data from (a) multiple sites and (b) those who differ in their country of origin on outcome variables (i.e., job satisfaction, intentions to quit) and the moderator, racial climate. Additionally, data were analyzed to ascertain the influence of using split sample methodology on relationships between study variables. Finally, analyses were conducted to justify aggregation. Data collection site. Intentions to quit (F(23, 112)=.96, p=.52) and job satisfaction (F(23, 112)=1.01, p=.47) scores did not differ significantly across departments. Departments were defined as units, offices, or schools within a work site or organization. For example the Law School is a department within a particular site, the large, public university. However, racial climate scores did differ by department (F(22, 105)=2.00, p<.01). A post-hoc test of department means revealed that all departments significantly differed from each other with the exception of two departments (i.e., Department of Communication, Law School) in the large, public university. Aggregated departmental racial climate scores ranged from 43.80-70.25 (M=55.64, SD=7.32). The possible range on the racial climate measure was 18-90. Because variance in the moderator variable was desirable, this range in scores was kept. THEORY OF WORK ADJUSTMENT 97 Country of origin. The mean levels of intentions to quit (F(1, 133)=.08, p=.77), job satisfaction (F(1,133)=.48, p=.49), and racial climate (F(1,125)=.27, p=.61) did not differ significantly by country of origin. Split sample methodology. Data from 164 participants were collected using a split sample methodology (i.e., approximately half of the participants in one site received the racial climate scale and the rest of the participants received the remaining measures). Data from an additional 48 participants were collected without using the split-sample methodology. That is, these 48 participants received and completed all measures. To ensure that these different methods of data collection did not influence the relationships between the study variables, two sets of correlations were run on the relationships between the study variables resulting in correlation coefficients. These correlation coefficients were converted to z-scores using Fisher?s r to z transformation. The resulting z-scores are compared using formula 2.8.5 from Cohen and Cohen (1983). Resulting z-scores greater than the absolute value of 1.96 are generally accepted as significantly different. None of the comparisons in the present study resulted in significant z-scores greater than the absolute value of 1.96. More specifically, the comparison of the correlations between racial climate (Watts & Carter) and intentions to quit (Colarelli) in the two samples produced a z score of 1.19, p=.23. The comparison of correlations between perceptions of fit (Saks & Ashforth) and racial climate scores resulted in a z-score of 1.33, p=.19. When comparing the correlation between job satisfaction (Weiss et al., 1967) and racial THEORY OF WORK ADJUSTMENT 98 climate (Watts & Carter) between the two groups, a z score of -.22, p=.83 resulted. The z-score yielded from the comparison of the correlations between perceptions of fit and intentions to quit was .84, p=.41 for the two groups. For the comparison of the correlations between job satisfaction and intentions to quit the z-score was 0, p=1. Finally, the comparison between perceptions of fit and job satisfaction also was not significantly different (z=-1.19, p=.23). Therefore, the two groups were combined in all analyses. Justifying aggregation. Racial climate scores were collected at the individual level and aggregated at the work department level. Before aggregating racial climate scores by work department, a test of within-group agreement (i.e., ., r wg(j) ; James, Demaree, & Wolf, 1984) was conducted. R wg(j) calculates the agreement or similarity in racial climate scores for each department, individually, without consideration of between-group variance (Bliese, 2000; Cohen, Doveh, & Eick, 2001; James, Demaree, & Wolf, 1984, 1993). R wg(j) scores above .70 are desirable (James, Demaree, & Wolf, 1984, 1993). Present study r wg(j) scores ranged from .85-.99. Intra-class correlation estimates (i.e., ICC(1), ICC(2)) also were collected but are of less significance because they measure reliability rather than agreement. More specifically, ICC (1) is used to determine the amount of variance due to department membership (Bliese, 2000). In the present study, the ICC(1) was .31, suggesting that 31% of the variability in individual racial climate scores can be explained by group membership. The F-value associated with the intra-class correlation tests was 3.25, THEORY OF WORK ADJUSTMENT 99 p<.05, suggesting that variability in group means was adequately explained by group membership. ICC(2), which provides an estimate of the reliability of department means (i.e., determines whether means can be used to distinguish group membership) was .69. ICC(2) values are considered adequate when they reach .70. Because the present study was more interested in within-group agreement (i.e., similarity of department members? scores) than ability to distinguish departments by racial climate scores, aggregation was justified by r wg(j) scores (Bliese, 2000). Research Questions Research question 1: The first research question asked: What are the means, standard deviations, and ranges of African-American employees? perceptions of fit, perceptions of racial climate, job satisfaction, and turnover intentions? The means, standard deviations, and ranges of perceptions of person-organization and person-job fit, racial climate, job satisfaction, turnover intentions were calculated and are displayed in Table 2. Participants? (n=136) mean score of perceptions of fit (Saks & Ashforth) was 32.14, SD=8.36. Participants? range of scores on this variable was 11-49 out of a possible range of 10-50 points. On the racial climate scale (Watts & Carter), participants? (n=128) mean score was 55.64, SD=7.32. Participants? scores ranged from 43.80-70.25. The possible range for this variable was 18-90. An examination of the job satisfaction (Weiss et al.) descriptive data revealed a mean score of 66.89, SD=14.47. Participants? scores ranged from 26-100, out of a possible range of 20- THEORY OF WORK ADJUSTMENT 100 100. Finally, participants? mean score for intentions to quit (Colarelli) was 8.57, SD=3.69. Participants? scores ranged from 3-15, which represented the possible range for this variable. An intercorrelation matrix of these variables is presented in Table 3. Research question 2. The second research question sought to answer the question, ?What on-the-job factors play a role in job satisfaction for African-American employees?? Responses to this open-ended question were coded for compatibility with the six values presented in the TWA (i.e., achievement, comfort, status, altruism, safety, autonomy) and racial climate (i.e., do participants indicate that racial climate plays a role in their job satisfaction?). Responses were coded by three graduate students, two Counseling Psychology graduate students and one Industrial / Organizational Psychology graduate student, who are familiar with the TWA. They were trained by the researcher to use the coding system developed for the present study which included a delineation of TWA values and racial climate. An intra-class correlation was run to determine the inter-rater reliability of the raters? responses. The responses displayed adequate reliability (alpha = .88). Because the inter-rater reliability of the scores was high, raters? scores were averaged to produce descriptive data TWA values (See Table 4). Of the coded responses, the comfort value represented 33.23% of the responses. Safety was endorsed 22.34% of the time. The status value was coded in 19.51% of the responses. The achievement value was endorsed 4.42% of the time. The altruism value accounted for 10.43% of the coded responses. Autonomy was represented by 5.99% of the coded THEORY OF WORK ADJUSTMENT 101 Table 2. Descriptive Statistics for Perceptions of Fit, Racial Climate, Job Satisfaction, and Intentions to Quit Variable Mean Standard Deviation Range Possible Range Perceptions of Fit (n=136) 32.14 8.36 11-49 10-50 Racial Climate A (n=128) 55.64 7.32 43.80-70.25 18-90 Racial Climate I (n=128) 59.99 11.04 28-85 18-90 Job Satisfaction (n=136) 66.89 14.47 26-99 20-100 Intentions to Quit (n=136) 8.57 3.69 3-15 3-15 Note. Racial Climate A = aggregate scores, Racial Climate I = individual scores Table 3. Intercorrelation Matrix for Perceptions of Fit, Racial Climate, Job Satisfaction, and Intentions to Quit Variable  1 2 3 4 1. Perceptions of Fit .92 1.00 2. Racial Climate .88 .40** 1.00 3. Job Satisfaction .92 .65** .16 1.00 4. Intentions to Quit .83 -.45** -.03 -.38** 1.00 Note. ** p<.01 THEORY OF WORK ADJUSTMENT 102 Table 4. Categorization of Responses to the Question: ?What on-the-job factors play a role in job satisfaction for African-American employees?? Category Percentage of Responses Endorsed 1. TWA: Comfort Value ?More money for the work I do.? 33.23% 2. TWA: Safety Value ?STOP changing policy and procedures every other week.? 22.34% 3. TWA: Status Value ?Recognition for hard work (staying late, coming in early, missing lunch), appreciation, respect, and consideration.? 19.51% 4. TWA: Altruism Value ?Being able to use professional knowledge and skills in an effective way to help support and empower the disenfranchised segments, the community, particularly in relation to criminal justice issues.? 10.43% 5. TWA: Autonomy Value ?Authority to do my job the way I think best with my supervisors absent approval.? 5.99% 6. TWA: Achievement Value ?A chance to do something to make use of my abilities?? 4.42% 7. Racial Climate ?One of my big issues is how race is used here. All of the low paying jobs go to blacks and from what I have heard similar jobs that are being done by whites are making more money. There is only 1 black tenure track faculty in my department which is huge. We have about 30 tenure track/tenured faculty and only one of them is black, probably because they offer less money to the very few black faculty candidates we get. If race wasn?t treated as a reason to undercut a person?s value and their pay, that would make me more satisfied. I don?t think it is intentional but I do think that treatment of blacks here are tainted by racist subconsciousness.? 4.08% THEORY OF WORK ADJUSTMENT 103 responses. Racial climate accounted for 4.08% of the coded responses. TWA values and racial climate were endorsed in all but an average of 15.33 of the open-ended responses (SD=10.41, average total responses = 187). Examples of participants? responses for each domain are included in Table 4. Research question 3. The third research question investigated the differences that exist between the relationships between perceptions of fit (Saks & Ashforth), job satisfaction (Weiss et al.), and intentions to quit (Colarelli) for African Americans? in the present study and Europeans and European Americans? in past research. Table 5 contains a summary of the study variables and results from past research with European and European American samples. To answer Research Question 3, correlation coefficients that specified the relationships between perceptions of fit, job satisfaction, and intentions to quit were recorded from previous studies that sampled European and European American samples (Breeden, 1993; Bretz & Judge, 1994; O?Reilly, Chatman & Caldwell, 1991; Rounds, 1990; Saks & Ashforth, 1997; Van Vianen, 2000) and compared to the correlation coefficients in the present study. Comparison studies represent all identified studies that investigated the relationships between study variables in European American and European samples. For the comparison between the present study and other studies identified, the square-roots of the regression coefficients (i.e., R 2 ) in past studies were transformed to z-scores and compared to the z-scores resulting from the transformed correlation coefficients from the present study. THEORY OF WORK ADJUSTMENT 104 Formula 2.8.5 from Cohen and Cohen (1983) was used to compare resulting z-scores to determine whether significant differences between the correlations exist. By convention, z-score differences reach significance when the absolute values rise above 1.96. In most comparisons, the relationship between perceptions of fit and job satisfaction for the present study significantly differed from studies of European and European American employees (Breeden, 1993; O?Reilly, Chatman, & Caldwell, 1991; Saks & Ashforth, 1997; Van Vianen, 2000) with more of the variance explained with the African American sample. Of the ten relationships compared, absolute z-score values ranged from 1.37 - 7.11, resulting in only two non- significant z-scores. Significant differences included comparisons between the present study and relationships between fit and job satisfaction in Saks and Ashforth (1997), the source of the present study?s perceptions of fit measure. Significant differences did not exist when comparing the present study to Bretz and Judge (1994), depending on their measurement of fit, and Rounds (1990), who employed the product moment coefficient when measuring fit. In all cases, perceptions of fit in the present study (i.e., 42.25%) explained more variance in job satisfaction than in compared studies (i.e., .42%-32.00%). THEORY OF WORK ADJUSTMENT 105 Table 5. Summary of Past Research with European and European American Participants Past Research Relevant Relations Measures Results Absolute Z-score Differences Breeden (1993) 1. Fit  Job Sat (Hoppock JSB) at intake 2.Fit  Job Sat (MSQ) at intake 3.Fit  Job Sat (Hoppock JSB) at 2 year follow up 4. Fit  Job Sat (MSQ) at 2 year follow up Fit: Minnesota Importance Questionnaire / Minnesota Occupational Classification System III Job Sat: the first item of the Hoppock Job Satisfaction Blank, Minnesota Satisfaction Questionnaire 1.Fit explained 2.89%, p=.001 2. Fit explained 11.56%, p=.001 3. Fit explained .64 %, p=.05 4. Fit explained .49%, p>.05 1. Z= 6.09 2. Z= 4.13 3. Z= 7.01 4. Z= 7.11 Bretz & Judge (1994) 1. Fit  Job Sat Fit: Authors created a measure, Job Sat: G.M. Faces Scale, 1. Fit explained between 12%-32%, p<.01 (depending on method of analysis) 1. Z scores between 1.37 ? 4.28 O?Reilly, Chatman, & Caldwell (1991) 1. Fit  Job Sat 2. Fit  Int to Quit Fit: Authors created a measure Job Sat: Single-item Faces Scale (Kunin, 1955) Int to Quit: 4-item scale created for the study 1. Fit explained 12.25% of the variance in Job Sat, p<.01 2. Fit explained 13.69% of the variance in Int to Quit, p<.01 1. Z= 3.29 2. Z= .77 Rounds 1. Fit  Job Sat Fit: Minnesota Importance 1. Fit explained 29% of 1. Z= 1.71 THEORY OF WORK ADJUSTMENT 106 (1990) Questionnaire / Occupational Reinforcer Patterns and Minnesota Occupational Classification System II Job Sat: Hoppock Job Satisfaction Blank variance in Job Sat w/ Occupational Reinforcer Patterns and 2.4% w/ Minnesota Occupational Classification System II 2. Z= 5.72 Saks & Ashforth (1997) 1.Fit  Job Sat 2. Fit  Int to Quit Fit: Authors created a 2-item measure Job Sat: 3-item scale by Cammann et al. (1983) Int to Quit: Colarelli (1984) 1. Fit explained 17%, p<.01 at 4 mnths and 10% at 10 mnths, p<.01 2. Fit explained 30% at 4 mnths, p<.01 and 21% at 10 mnths, p<.01 1a. Z= 3.11 1b. Z= 4.07 2a. Z= 1.23 2b. Z= .12 Van Vianen (2000) 1. Newcomers? cultural preferences compared w/ perceptions of organizational culture (P-O Fit)  Int to Quit 2. Newcomers? cultural preferences compared w/ supervisors? and peers? preferences of organizational culture (P-P Fit)  Int to Quit P.O Fit: Based on difference scores derived from commensurate measures previously validated in a Dutch sample (Van Vianen & Kmieciak, 1998) P.P. Fit: Based on difference scores derived from commensurate measures previously validated in a Dutch sample (Van Vianen & Kmieciak, 1998) Intentions to Quit: 3 item-scale (Ten Horn, 1983) 1. P-O Concern for People subscale explained 20% of the variance in Int to Quit, p <.01. Authors indicated that the relationship between the P-O Goal Accomplishment subscale and Int to Quit was not sig. 2. P-P Concern for People subscale explained 16% of the variance in turnover intentions, p<.05; 3. PP Goal 1. Z= 0 2. Z= .52 3. Z= .43 THEORY OF WORK ADJUSTMENT 107 Accomplishment subscale explained 17.4%, p<.05 Note. Abbreviations were used in this table. Fit= perceptions of fit, Hoppock JSB = Hoppock Job Satisfaction Blank, MSQ=Minnesota Satisfaction Questionnaires, Job Sat = job satisfaction, Int to Quit = Intentions to Quit/Tenure Intentions, P- O Fit = Person Organization Fit, P-P Fit = Person-Person Fit THEORY OF WORK ADJUSTMENT 108 Six relationships were compared when examining differences between the relationship between fit and intentions to quit in African American and European or European American samples (Breeden, 1993; Bretz & Judge, 1994; O?Reilly, Chatman & Caldwell, 1991; Rounds, 1990; Saks & Ashforth, 1997; Van Vianen, 2000). No significant z-score differences occurred in these six comparisons. The absolute values of z-scores ranged from 0-1.23, p>.05. Research question 4. Research Question 4 asked: Are perceptions of fit (Saks & Ashforth) correlated with years in position? Employees? perceptions of fit was not correlated with the amount of time in their current positions (r=.15, p=.09). However, employees? perceptions of fit was correlated with years with their current organization (r=.23, p<.01). Research question 5. The fifth research question asked: Are racial climate and perceptions of fit scores correlated? When the correlation between racial climate scores and perceptions of fit was computed, a significant and positive relationship was found where r=.40, p<.01. Hypotheses As discussed in Chapter 3, eight hypotheses were tested and explored. These hypotheses are discussed below. Hypotheses 1 through 6 are presented in a manner consistent with Baron and Kenny?s (1986) model for testing for moderator effects. Baron and Kenny?s (1986) three-step procedure was used to investigate whether racial climate moderated the relationship between perceptions of fit and job satisfaction and THEORY OF WORK ADJUSTMENT 109 perceptions of fit and intentions to quit. Two hierarchical regressions were conducted, one for each dependent variable (i.e., job satisfaction and intentions to quit). As described by Baron and Kenny, each hypothesis represented one of the steps in each of the two regression equation. Hypothesis 7 predicted the relationship between job satisfaction and intentions to quit and was tested with a correlation. Hypothesis 8 made predictions regarding the ability of the Lyons and Brenner (in preparation) measure to adequately assess perceptions of fit. This hypothesis was tested with reliability and validity estimates. Hypothesis 1. Hypothesis 1 predicted that perceptions of fit (Saks & Ashforth) would account for a percentage of the variance in job satisfaction (Weiss et al.), and participants with a greater degree of fit would experience greater job satisfaction than those reporting a lesser degree of fit. The ability of perceptions of fit to predict job satisfaction was tested by entering perceptions of fit in the first step of the first hierarchical regression analysis of variance as instructed by Baron and Kenny for tests of moderation. In the test of Hypothesis 1, perceptions of fit was the independent variable and job satisfaction (M=66.89, SD=14.47) was the dependent variable. As summarized in Table 6, the results of the regression analysis revealed a positive relationship between perceptions of fit and job satisfaction where fit accounted for 43.2% of the variance in job satisfaction (F(1,126)=95.71, p<.01). Hypothesis 1 was supported. THEORY OF WORK ADJUSTMENT 110 Hypothesis 2. Hypothesis 2 predicted that racial climate (Watts & Carter) would account for a percentage of the variance in job satisfaction (Weiss et al.), and participants reporting positive racial climates would experience greater job satisfaction than those reporting negative racial climates over and above the variance accounted for by perceptions of fit. Racial climate was entered in the second step of the first regression equation to analyze the relation between racial climate and job satisfaction after controlling for perceptions of fit. Racial climate did not significantly account for variance in job satisfaction over and above the variance accounted for by perceptions of fit (Incremental F(2,125)=2.70, p>.05; R 2 = .01). Therefore, Hypothesis 2 was not supported (See Table 6). Hypothesis 3. Hypothesis 3 predicted that the interaction between perceptions of fit (Saks & Ashforth) and racial climate (Watts & Carter) would account for a percentage of the variance in job satisfaction, and higher levels of f it and positive racial climates would result in higher levels of job satisfaction. As advised by Baron and Kenny (1986), the ability of the interaction between fit and racial climate to predict job satisfaction would be tested in the third step of the first hierarchical regression equation, after controlling for perceptions of fit and racial climate. If this third step is significant, a moderator exists (Baron & Kenny, 1986). That is, a significant interaction would suggest that racial climate moderates the THEORY OF WORK ADJUSTMENT 111 Table 6. Summary of the Hierarchical Multiple Regression Analysis Testing Racial Climate as a Moderator of the Relationship between Perceptions of Fit and Job Satisfaction Variable R p F F Ch R 2 R 2 Ch p Perceptions of Fit .66 .00 95.75 .43 .00 Racial Climate .67 .00 2.71 .44 .01 .10 Interaction: Perceptions of Fit x Racial Climate .67 .00 .24 .45 .00 .63 THEORY OF WORK ADJUSTMENT 112 relation between fit and job satisfaction. The interaction between perceptions of fit and racial climate was not significant when job satisfaction was the dependent variable (Incremental F(3,124)=.24, p>.05; R 2 =.00). Hypothesis 3 was not supported. Hypothesis 4. Hypothesis 4 predicated that perceptions of fit (Saks & Ashforth) would account for a percentage of the variance in turnover intentions (Colarelli), and participants with a greater degree of fit would experience lower turnover intentions than those reporting a lesser degree of fit. A second regression equation was analyzed to determine the ability of the independent and moderator variables to explain the variance in intentions to quit (M=8.57, SD=3.69). The first step in this regression equation was used to analyze the relation between fit and intentions to quit. Perceptions of fit explained 17.20% of the variance in intentions to quit (F(1,126)=26.16, p<.01). This relationship was negative and significant. Therefore, hypothesis 4 was supported (See Table 7). Hypothesis 5. Hypothesis 5 predicted that racial climate (Watts & Carter) would account for a percentage of the variance in intentions to quit (Colarelli), and participants reporting positive racial climates would rep ort lower levels of intentions to quit than those reporting negative racial climates over and above the variance accounted for by perceptions of fit. To test this hypothesis, racial climate was entered in the second step of the second regression equation to analyze the relation between racial climate and intentions to quit after controlling for perceptions of fit. Racial climate did not THEORY OF WORK ADJUSTMENT 113 significantly predict a unique proportion of the variance in intentions to quit (Incremental F(2,125)=3.35, p=.07; R 2 = .02) and Hypothesis 5 was not supported (See Table 7). Hypothesis 6. Hypothesis 6 predicted that the interaction between perceptions of fit (Saks & Ashforth) and racial climate (Watts & Carter) would account for a percentage of the variance in intentions to quit, and higher levels of fit and positive racial climates would result in lower levels of intentions to quit. The ability of the interaction between fit and racial climate to predict intentions to quit was tested in the third step of the second hierarchical regression equation, after controlling for perceptions of fit and racial climate. A significant interaction was not found (Incremental F(3,124)=1.27, p>.05; R 2 =.00). Therefore, Hypothesis 6 was not supported (See Table 7). Hypothesis 7. Hypothesis 7 predicted that job satisfaction (Weiss et al.) would be correlated with intentions to quit (Colarelli), and participants reporting lower levels of job satisfaction would report higher levels of intentions to quit than participants with higher levels of job satisfaction. A correlation coefficient was computed to test whether job satisfaction was correlated with intentions to quit. In the test of Hypothesis 7, job satisfaction was the independent variable and intentions to quit (M=8.57, SD=3.69) was the dependent variable. As summarized in Table 3, the correlation between job satisfaction and intentions to quit was negative and significant (r=-.38, p<.01). Therefore, Hypothesis 7 was supported. THEORY OF WORK ADJUSTMENT 114 Table 7. Summary of Hierarchical Multiple Regression Analyses Testing Racial Climate as a Moderator of the Relationship between Perceptions of Fit and Intentions to Quit Variable R p F F Ch R 2 R 2 Ch p Perceptions of Fit .42 .00 26.16 .17 .17 .00 Racial Climate .44 .00 3.35 .19 .02 .07 Interaction: Perceptions of Fit x Racial Climate .45 .00 1.27 .20 .01 .26 THEORY OF WORK ADJUSTMENT 115 Hypothesis 8. Hypothesis 8 predicted that the reliability and validity estimates of the perceptions of fit for the measure designed for the present study (Lyons & Brenner, in preparation) would be comparable to the reliability and validity estimates of Saks and Ashforth?s (1997) measure of perceptions of fit. Reliability was estimated for the present study?s measure of perceptions of fit using a Cronbach?s alpha, a measure of internal consistency. The internal consistency for this measure was .92, which suggests that this measure possesses adequate reliability. Support for construct validity for the perceptions of fit scale created for the present study was provided by analyzing the measure?s relationship to the present study?s outcome variables, job satisfaction and intentions to quit as well as with Saks and Ashforth?s perceptions of fit measure. Regression equations were used to analyze these relationships. Because the middle anchor, 4, represented the highest degree of fit, I expected that quadratic relationships would be stronger than linear relationships in equations using this measure. To test for a quadratic relationship, regression equations are needed. The total scores for the new fit measure were added in the first step and in the second step the total score as well as the total fit score squared were added. Regression equations conducted suggest that a quadratic relationship, F(2,133)=13.13, p<.01; R 2 change=.13, explains more variance in the relationship between fit and job satisfaction, than a linear relationship, F(1,134)=5.10, p<.05; THEORY OF WORK ADJUSTMENT 116 R 2 =.03. A quadratic equation explains 12.82% of the variance in job satisfaction over and above a linear equation. When examining the relationship between fit and intentions to quit, a linear equation, F(1,134)=4.53, p<.05; R 2 =.03, explained more variance than a quadratic equation, F(2,133)=2.28, p=.11. The quadratic equation explained no variance over and above that explained by the linear equation. When comparing the perceptions of fit measure designed for the present study (Lyons & Brenner) with Saks and Ashforth?s measure of fit, the relationship between the two measures was strongest when the present study?s measure of fit (Lyons & Brenner) was compared using a quadratic equation, F(2,133)=10.32, p<.01; R 2 change=.04, rather than a linear one F(1,134)=14.36; R 2 =.10. Analyses were conducted to compare the two measures? ability to predict job satisfaction and intentions to quit. The Saks and Ashforth measure explained 42.6% of the variance in job satisfaction (F(1.134)=99.51, p<.01). The Lyons and Brenner perceptions of fit measure explained an additional 5.6% of the variance in job satisfaction over and above that explained by the Saks and Ashforth measure (F(3,132)=7.11, p<.01). However, the Lyons and Brenner measure does not add predictive ability to intentions to quit over and above the Saks and Ashforth measure. When Saks and Ashforth was entered first it predicted 20.30% of the variance in intentions to quit (F(1,134)=34.23, p<.01). No significant variance in intentions to quit was explained THEORY OF WORK ADJUSTMENT 117 above and beyond the Saks and Ashforth measure by the Lyons and Brenner measure (F(2,133)=.31, p>.05; R 2 change=.00). Because reliability and validity statistics of the Lyons and Brenner perceptions of fit measure are comparable to the established Saks and Ashforth measure of perceptions of fit, Hypothesis 8 was supported. THEORY OF WORK ADJUSTMENT 118 Chapter 6 Discussion The purpose of the present study was to address the lack of knowledge of African Americans? workplace experiences in three ways: (a) providing descriptive data related to workplace variables for African-American employees, (b) testing an attenuated model of job satisfaction and turnover intentions incorporating racial climate, and (c) attempting to address methodological limitations raised in this body of literature. Describing the Work Experiences of African Americans An examination of descriptive data in the present study provided an understanding of African-American employees? sense of whether they share similar values with their place of work, levels of job satisfaction, perception of racial climate, and to what extent they consider remaining with their organization. Perceptions of fit scores ranged from ?very little? to ?very strong.? On average, African-American employees reported that their workplaces provided an adequate fit with their sense of their needs, values, and abilities as they related to their job duties and the organization. Racial climate scores were restricted to moderate levels indicating that African-American employees perceive ?sometimes? having access to rewards, power, and decision making in the organization (versus ?never? and ?always?) in addition to experiencing a moderate amount of respect and sensitivity in their interactions with European Americans at work. On average, African-American THEORY OF WORK ADJUSTMENT 119 employees also reported moderate levels of job satisfaction. However, job satisfaction scores displayed a wide range, from ?very dissatisfied? to ?very satisfied.? Finally, the mean for intentions to quit fell at about the middle of the range for this measure suggesting that, on average, African-American employees occasionally consider quitting their jobs and planning a search for a new job. Employees sampled reported a wide range on this variable. Reponses to the open-ended question ?What on-the-job factors play a role in job satisfaction for you? In other words, what makes you or would make you a satisfied employee?? provided data to test the relevancy of the TWA values and racial climate in describing the job satisfaction of African-American employees. Incorporating a qualitative component to this study provided an opportunity to hear directly from this understudied group. Cross-methodological validation has importance for this group because relation of the TWA to this population was previously based on assumption. The TWA was not, or at least not explicitly, informed by the cultural values, vocational histories, and present day work experiences of African Americans. Additionally, past research suggested that these aspects of African Americans? work adjustment may differ from European Americans, on whom most vocational theory was formulated and tested (Brown, 1995; Fassinger, 2001). Therefore, without this validation it has been difficult to understand the relevance of the TWA to this population and the meaning of empirical results associated with African Americans. THEORY OF WORK ADJUSTMENT 120 In all (n=187) but approximately 15 responses, TWA values or racial climate were endorsed suggesting that African-American employees recognize the significance of these variables in their work experience. Close examination of the distribution of endorsed items suggested that some values are more salient for African-American employees than others. The comfort value which includes staying busy, independence, variety, compensation (e.g., salary, benefits), and security appeared to be the most significant value for African-American employees in determining job satisfaction. Dawis (1996) predicted that this value would be the most important value for employees. Comfort was followed by status, which includes the opportunity for advancement and recognition, and safety, which includes access to training, relationship with supervisor, and perceptions of organizational policies and procedures. Statements related to the racial climate of African-American employees were made in very few (4.08%) of the responses. Therefore, analysis of the qualitative portion of the present study provided grounded validation for the centrality of TWA?s core values in the determination of satisfaction for in this population. This validation, if replicated, can indicate that researchers and practitioners may purposefully use the TWA in future research and practice focused on African Americans. Racial Climate and the Theory of Work Adjustment Results of the present study indicated that racial climate did not moderate the relationship between perceptions of fit and job satisfaction and perceptions of fit and intentions to quit. This finding was consistent with the contention of TWA theorists, THEORY OF WORK ADJUSTMENT 121 Dawis (1994) and Dawis and Lofquist (1993), who anticipated that cultural variables would not alter the nature of relationships in the TWA model. Rather, they anticipated that culture variables would influence fit, which in turn would influence workplace outcome variables. Closer examination of specific results in the present study suggested that this may be the case. More specifically, racial climate was related to perceptions of fit in the present study but was not related to the other study variables. It is possible that racial climate may indirectly influence other TWA variables through its impact on fit perceptions. Future research can seek to provide further support for this hypothesis by testing the ability of fit to mediate the relationship between racial climate and TWA outcome variables, as suggested by Dawis (199) and Dawis and Lofquist (1993). However, in the present study, because racial climate was not statistically related to TWA outcome variables, a statistical mediation would not occur. For fit perceptions to statistically mediate the relationship between racial climate and TWA outcome variables, racial climate would need to be significantly related to job satisfaction and intentions to quit and as the ability of fit to explain variance in the outcome variable is controlled for, the ability of racial climate to explain variance in the outcome variables would be diminished (see Baron & Kenny, 1986). In other words, fit would need a relationship to mediate which the present study did not provide. Furthermore, Dawis (1994) stated that the TWA seeks to explicate individual differences and thus, sociological and cultural group factors exert their influence only THEORY OF WORK ADJUSTMENT 122 when they affect certain individual differences variables outlined in the model. In the case of the TWA, perceptions of fit or the values endorsed by particular employees can be seen as within group variables that are informed by the employees? vocational histories and individual experiences with vocational reinforcements. Dawis anticipated that these experiences would vary based on individual opportunity and exposure. However, other results shed light on the relationship between TWA variables and racial climate for African-American employees. For example, African-American employees were overwhelmingly more likely to offer TWA values as determinants of job satisfaction when asked by the open-ended question at the end of their data packet. Racial climate was named as a determinant of job satisfaction in 4.08% of the responses. Therefore, this qualitative result provided support for the quantitative finding that TWA values, and perhaps, consequently, values fit played a larger role in the job satisfaction of African-American employees than racial climate. Also, the resulting limited range in racial climate scores which tended to hover in moderate levels, may suggest that racial climate in the departments sampled were not problematic enough to exert the hypothesized influence on workplace outcome variables. Alternatively, those who did experience more hostile racial climates may not have participated in the present study or terminated work simply because of the negative influence of the work environment meaning that the present study may not have benefited from their input. THEORY OF WORK ADJUSTMENT 123 Finally, range restriction may also reflect the attempt of some employees to ease cognitive dissonance. Cognitive dissonance theorists suggest that dissonance arises when people are confronted with two inconsistent pieces of information. This dissonance is thought to create a tension which is usually subject to attempts at alleviation (Festinger, 1957). Hypothetically, in the present study, participants may be in hostile work environments but at the same time recognize a desire for respectful work environments or the benefits of work (e.g., health insurance). In response to these conflicting thoughts, participants experiencing cognitive dissonance may either deny or rationalize their hostile climates to ease the tension. Therefore, negative reports of racial climate would not be recognized and would not be reported in the survey. Because there was not a range in these scores it was not possible, with this sample, to understand the influence of very hostile and alternatively, very respectful racial climates on job satisfaction and intentions to quit. Therefore, participants may have been experiencing a ?tolerable? level of frustration in terms of their racial climate which may be a level that does not lead to thoughts about termination or dissatisfaction. Alternatively, because of past experiences with racism or comparisons with racism experienced in other life domains, African Americans may come to expect a certain level of workplace racism and therefore, rated their workplace racial climate relative to other experiences with racism. THEORY OF WORK ADJUSTMENT 124 Consideration of the makeup of this sample may lend support to this explanation. Data were collected from participants who worked in departments with between 4 and 16 other African-American employees. It may be that the presence of other African Americans buffered negative racial interactions in the workplace. We know from Kanter?s (1977) work on tokenism in the workplace that being ?the only one? of your kind along gender and/or racial dimensions can have negative consequences for the token and create dominant group solidarity which may, theoretically, lead to more hostile climates. Because of the nature of data collection (i.e., collecting from a minimum of 4 African Americans in a work environment), African-American employees in the present study were not reporting from the experience of a token. Future researchers planning to collect racial climate data with adequate variance might collect data from a variety of work sites including those with only one African American or African Americans who are the sole person of color in their department or organization. Also, in an attempt to alleviate the possible influence of turnover on racial climate scores, researchers should attempt to collect information on racial climate perceptions during exit interviews or ask participants to reflect on racial climate in past work environments. Finally, to avoid the possible effects of cognitive dissonance, future research might focus on applicants? perceptions of racial climate, rather than asking those already in the work environment who might be using cognitive dissonance to cope with their work environments. THEORY OF WORK ADJUSTMENT 125 The Role of Perceptions of Fit in the Work Lives of African-American Employees Previously, in the absence of empirical support, vocational psychologists predicted that fit would play a relatively insignificant role in the lives of African- American employees because of the competing significance of discrimination which may act as a barrier to jobs that could provide a sense of fit (Brown, 1995; Fassinger, 2001). Results of the present study suggested that African-American employees experience a range of fit with their jobs and organizations. Furthermore, perceptions of fit took on greater significance with African Americans compared to European and European American employees when predicting job satisfaction. The contrasting results in the present study suggest some possible explanations: (a) opportunities for African Americans to use work as a means of value and ability expression may be increasing, (b) hypotheses about the ability of fit to explain job satisfaction in the lives of African Americans were based on available research which tends to employ European Americans who express weaker relationships between these variables, and/or (c) fit may take on particular importance because of the nature of African Americans? stigmatized status. When comparing findings in the present study with past research on Europeans and European Americans, discrepancies resulted in the groups? abilities to predict job satisfaction from relative levels of fit. African-American employees? reports of perceptions of fit consistently explained larger amounts of variance in job satisfaction (43.2%) when compared to European and European American employees THEORY OF WORK ADJUSTMENT 126 (i.e., on average, fit explained 14.70% of the variance in job satisfaction, SD=9.74%). In all but two of the cases examined, these differences were significant. Comparable results were found when examining the ability of perceptions of fit to predict job satisfaction in a sample of lesbian and gay employees. A recent study of the relationship between fit and job satisfaction for lesbian and gay employees (Lyons, Brenner, & Fassinger, 2003) revealed the ability of fit to explain 48.2% of the variance in job satisfaction. Together, these findings may suggest that fit for these populations is relatively more important given their stigmatized statuses. It may be that membership in stigmatized groups forces these groups to be more highly attuned to and, therefore, influenced by their environments when making evaluations of job satisfaction. The relationship between perceptions of fit and intentions to quit was comparable to the relationship between these variables for European and European American employees (i.e., on average fit explained 19.68% of the variance in intentions to quit, SD=5.71%). In the present study, fit accounted for 17.2% of the variance in intentions to quit. Therefore, while perceptions of fit are statistically and practically significant for African Americans considering leaving their workplace, other variables may prove to play a more significant role in predicting whether African Americans stay with their employers and should be investigated in future research. Variables to explore in future research include perceptions of job security, ability to secure alternative employment, geographical allegiance, and sociological THEORY OF WORK ADJUSTMENT 127 and political concerns such as national unemployment rates. Furthermore, future research should aim to uncover why fit is of greater relative significance in it?s relationship with job satisfaction but is of relative comparable significance in it?s relationship with intentions to quit when comparing African American and European American samples. It is possible that African-American employees may allow fit to influence job satisfaction or dissatisfaction but for other reasons (e.g., need for financial security, ability to obtain other employment) do not allow fit, to the same extent, to lead them to consider quitting their jobs. Consideration of within group variables that may be responsible for distinctions in African Americans? perceptions of fit and its relationship with workplace outcomes may shed light on this discrepancy between Fassinger (2001) and Brown?s (1995) assertions and findings in the present study. For example, the present study?s sample was well-educated. Approximately 19.3% of the sample attended at least some college, 59.6% of the sample obtained at least a college degree, and many (30.5%) completed a graduate or professional degree. Attaining this level of education might have granted this sample greater access to opportunities to find jobs and organizations that match their values and perceptions of abilities. Perceptions of fit was also related to the number of years African-American employees remained with an organization but not the numbers of years they occupied their current positions. Perhaps a sense of fit is more connected with the community and benefits within particular organizations rather than with a specific job or set of THEORY OF WORK ADJUSTMENT 128 duties. This may point to the fact that particular considerations such as geographical location, organizational demography (e.g., racial makeup of the organization), and health insurance, may play an important role in employee retention. Organizations carry with them particular characteristics that may remain stable across positions that may strengthen the relationship between fit and the amount of time African Americans remain with particular organizations. For example, Human Resources policies tend to be relevant to entire organizations rather than department specific. Explanation for these findings may also lie in African-American employees? experiences with workplace discrimination. Because historically, it has been difficult for African Americans to secure work (Berry & Blassingame, 1982; Williams, 1999), factors other than fit may be related to years in a position, obscuring the relationship between fit and years in a position. Practical concerns such as job security may be more significant than an employees? sense of fit. This can be investigated in future research. For example, future research should include surveys of African-American employees? sense of job security, ability to obtain other work if they desired, financial need, and knowledge of the impact of the national economy on their work experiences. Finally, the lack of significant relationship between years in position and fit contrasted with the significant relationship between years with organization and fit may be explained by employees? hopes of entering a particular organization with the intention of ?moving up the ladder.? Therefore, they might not experience high levels THEORY OF WORK ADJUSTMENT 129 of fit with particular positions that they hold before reaching their ideal positions within an organization. At the same time, they remain with an organization with which they fit until they are able to obtain the position in which they can express an ideal fit between their values and those of the positions held. Creating a Theoretically-Based Perceptions of Fit Measure Construction of a new perceptions of fit measure (Lyons & Brenner, in preparation), represented the attempt to contribute a theoretically-based, empirically supported measure of fit to the vocational psychology literature. Analyses with this measure provided mixed results though they were in support of the measure?s ability to reliably and validly measure perceptions of fit. The Lyons and Brenner perceptions of fit measure was designed to assess employees? subjective sense of fit with their organizations. Participants rated their agreement with each item (e.g., achievement, independence) using a seven-point scale as follows: ?1-2-3-4-3-2-1,? where the first ?1? equaled ?My organization values this more than I?, ?4? represented ?Similar level of value?, and endorsement of the second ?1? indicated ?I value this more than my organization.? Therefore, 4 represented the greatest degree of fit employees could indicate between themselves and their organizations. Responses were then coded as ?1-2-3-4-5-6-7? for data entry and analysis. Because this rating scale was used, the nonlinear relationship resulting between this measure and job satisfaction was consistent with TWA theory (Dawis, THEORY OF WORK ADJUSTMENT 130 England, Lofquist). What appeared to matter most in the relationship between perceptions of fit and job satisfaction was the match between employees? and organizations? value such that both parties? needs were perceived as being met. However, when the relationship between fit and intentions to quit was considered, a different pattern emerged. In that case, the relationship between fit and intentions to quit was not highest when the relationship was quadratic (i.e., when employees were reporting the greatest degree of fit) but rather when the relationship was linear. This suggested that African-American employees are most likely to consider leaving their organizations when they perceive that their needs were not being met. The role of fit took on secondary importance in this case. These results may help to explain ambiguous findings in research on the link between fit and turnover intentions. Van Vianen (2000), for example, did not find a link between person-organization fit (i.e., fit with values they perceived that their organization endorsed) and turnover intentions, as hypothesized. However, the author did find a significant link between person-person fit (i.e., fit with values reported by either coworkers or supervisors) and turnover intentions when employees? values were compared with their supervisors? values. They explain this finding by citing Schneider?s (1997) contention that person characteristics are more influential in organizations than organizational characteristics. Indeed, results in the present study related to turnover intentions, using the newly constructed measure of fit were consistent with Schneider?s theory. A linear relationship between fit and intentions to THEORY OF WORK ADJUSTMENT 131 quit suggest that person characteristics do become more important than perceptions of values endorsed by the organization or even the fit between organizational and employees? values when employees consider leaving their organizations. In the case of the Lyons and Brenner scale a linear relationship indicated that as employees express stronger endorsement of particular workplace values their intentions to leave the workplace increase. Although this measure did not share as much of the variance with job satisfaction and intentions to quit as the Saks and Ashforth measure did, results from the Lyons and Brenner measure have the potential to yield more sophisticated information. Because the measure is multi-dimensional, it takes into account not only the strength of the fit but the direction (i.e., Do employees endorse a value more strongly than the workplace or does the workplace endorse a value more strongly than employees?). As seen in the case of intentions to quit, it cannot be taken for granted that the concern is only whether or not employees fit with their workplaces. Limitations In addition to the limitations addressed throughout the discussion, methodological limitations in the present study included the possible reduced power of the moderator variable, racial climate and bias in the sample. These limitations will be addressed. The ability of racial climate to moderate the relationships between perceptions of fit and job satisfaction and perceptions of fit as well as perceptions of fit and THEORY OF WORK ADJUSTMENT 132 intentions to quit may have been limited. Range restriction, found in racial climate scores, can contribute to Type II error making it more difficult to observe significant interaction and direct relationships (Tracey, 2000). Range restricted might be avoided in future research by providing more response options for participants (e.g., 7 rather than 5 points). Additionally, future researchers should consider removing the midpoint by giving participants the option of a 6 or 8 point response scale. Additionally, because of possible complications in interpretation, Baron and Kenny (1986) warn that the moderator variable should not be correlated with the independent variable. In the present study, racial climate was significantly correlated with the independent variable, perceptions of fit. Finally, particularly with job satisfaction, robust variance was explained by perceptions of fit, making it more difficult to uncover other relationships. In fact, Baron and Kenny suggest that moderators should be investigated when attempting to improve the relationship between two variables or when a relationship is inconsistently found. Because approximately 43% of job satisfaction was explained by fit and because previous studies consistently uncovered a relationship between these two variables, it may have been more difficult to find a moderator in this case. When considering the possible limitations of the present study, it is important to remind readers of the level of education of participants. Perhaps, because many of the participants were drawn from an institution of higher education, they are highly educated. Nationwide, 17.2% of the African American population holds an THEORY OF WORK ADJUSTMENT 133 undergraduate degree or greater (U.S. Census Bureau, Education and Social Science Stratification Branch, 2002) compared with 59.6% in the present study. Because of the participants? educational status, they may represent a unique segment of the population who have had greater access to employment opportunities that offer employees opportunities to express their values and abilities as well as find satisfaction. Additionally, this sample may have had more experience interacting with European Americans through educational experiences and have acclimated to racially mixed or isolating work environments and therefore, already possess the skills and coping strategies necessary to succeed in these environments. Therefore, the generalizablity of findings is limited to formally educated African-American employees. Future research should aim to collect data from African-American employees representing a more diverse range in educational experiences. Findings resulting from the intentions to quit measure (Colarelli, 1984) should be viewed cautiously. Although this measure is well-established, it has only 3 items. A review of past research of factor analysis requirements suggested that a 3 item measure is not sufficient in capturing psychological constructs. Idazak, Bottom, and Drasgow (1988) found that using 3 items was insufficient in the detection of factors within a measure of job characteristics. They found that they needed 6 items per factor to adequately capture each factor. Furthermore, Lawley and Maxwell (1971) stipulated a minimum of three items per factor. Future researchers should use larger numbers of items when attempting to measure psychological constructs. THEORY OF WORK ADJUSTMENT 134 In terms of sampling, readers should note that the sample was a volunteer sample. This sampling method may have influenced results. For example, dissatisfied participants or those afraid to report inadequate racial climates might not have participated in the present study. Future researchers should randomly select participants for studies of African-American employees? work experiences. Implications The present study can offer counseling psychologists guidance for practice and research. These implications are addressed below. Implications for practice. Dawis and Lofquist (1984) encouraged counseling psychologists to use the TWA to guide three aspects of vocational counseling: career choice counseling, on-the-job career adjustment, and job change counseling. If replicated, the results of the present study can be used to inform these suggestions. When counseling African Americans for job choice and job change, particular attention should be paid to the potential of African Americans to fit with particular organizations. To do this, their values should be assessed so that organizations and fields of employment with similar values can be identified. This might also be of interest to Human Resources professionals in their efforts to recruit African-American employees. Additionally, attempts to increase access to and enrollment in secondary education for African Americans should be encouraged because of the possibility that education may provide African Americans with the opportunity to find careers with which they perceive fit. THEORY OF WORK ADJUSTMENT 135 When counseling African Americans to help them increase their on-the-job work adjustment, they may be made more aware of the ways that their current organization and job meet their needs. If these needs are not being met, counselors can work with employees to find ways of increasing the congruence between their needs and those provided by the organization and job (e.g., increasing positive relationships with colleagues, assuming increasing responsibilities, negotiating with supervisor for greater flexibility and personal control) or helping employees to find other work environments that may better meet their needs. Although racial climate did not prove itself as a moderator in the present study, it was related to perceptions of fit for African-American employees. Correlates of fit take on particular importance in the work lives of African Americans because of the ability of fit to predict job satisfaction and intentions to quit. Because of this, employees and employers should be attuned to opportunities for African-American employees to have non-racist, empowering and respectful interactions with others in the workplace. This may involve educating supervisors about the significance of racial climate for the perceptions of fit in the work experiences of African-American employees as well as assisting African-American employees to recognize and select the environments which they feel would provide positive racial climates. Career counselors and employers should also consider the possibility that racial climate perceptions may contribute to a sense of fit or that employees filter perceptions of fit through their organizational experiences with discrimination. For THEORY OF WORK ADJUSTMENT 136 example, when African Americans are answering the following question: ?To what extent are the values of your workplace similar to your own values?? (Saks & Ashforth, 1997) or asked to compare how strongly they value the following two dimensions in relation to how strongly their organization values them, ?amiable (friendly) working relationships? and ?fair company policies and procedures,? they may taking into account cultural values and interactions. In fact, Rounds and Hesketh (1994) encouraged researchers to incorporate elements of employees? perceptions of fairness and discrimination into TWA measures to account for the ability of these factors to influence TWA variables. Implications for research. Results of the present study can be used to direct future research in at least three ways: exploration of fit as a mediator in the TWA model, inclusion of other cultural variables in the TWA model, and identification of the predictors of fit for African-American employees. Original TWA theorists predicted that fit may mediate the relationship between cultural variables and TWA outcomes so that the TWA model would not need to be altered when applying it to People of Color (Dawis, 1994; Dawis & Lofquist, 1993). Results of the present study suggested that this may be the case when considering the role of racial climate. However, a more formal investigation of this proposition should be undertaken which would include collecting data from a wider range of African-American employees expressing a range of racial climate and fit perceptions as well high and low levels of job satisfaction and intentions to quit. THEORY OF WORK ADJUSTMENT 137 Efforts should be made to collect data from participants with a range of educational backgrounds and levels of power within an organization. This collected data can be analyzed for mediation using Baron and Kenny?s (1986) test of mediation which allows for the determination of whether fit can fully mediate the relationship between fit and TWA outcome variables. This is consistent with the findings in Lovelace and Rosen (1996) who found that fit mediated the relationship between demographic group (i.e., racial category, sex) and job satisfaction as well as between demographic group and intentions to quit. Future research may uncover when fit is able to adequately explain levels of job satisfaction and when inclusion of cultural variables becomes additive and significant even in the presence of fit. Other cultural variables (e.g., access to power, presence of critical mass which means that American Americans are working in environments with other African Americans versus feeling isolated) should be considered that may explain the inability in the present study to uncover a moderating relationship between fit and the TWA outcome variables. Additionally, researchers should attempt to measure and analyze the influence of individual-level measures of culture rather than at the group-level as analyzed in the present study. Perhaps culture exerts its influence on the TWA model at an individual level rather than at the group level. Therefore, racial identity status and race-related stress should be examined in future research for their potential to attenuate the TWA model. THEORY OF WORK ADJUSTMENT 138 Thomas and Alderfer (1990) made a case for the importance of assessing critical mass within an organization. According to Thomas and Alderfer experiences of social-group minority members are influenced not only by their experiences in society but by their relation to the demographical majority group within an organization. In the present study, organizational demography was not assessed. Although data were not collected from predominately Black organizations it is possible that data were collected from predominately Black departments. Therefore, the role of racial variables that assess racism, such as racial climate, may become less significant than other workplace variables. In this environment it may be more possible for the relationships outlined in the TWA model to emerge. In addition to the possible influence of other African Americans working in the same department as the participants in the present study, the participants in the present study were drawn from sites that historically have employed African Americans (e.g., government, education settings). It is possible that African Americans may be drawn to these particular settings because of a history of social and racial compatibility which might have influenced the results of the present study. Therefore, future research should investigate other settings and industries (e.g., Wall Street, natural sciences, engineering) that have employed few African Americans historically. The ability of perceptions of fit to explain so much of the variance in job satisfaction begs the question, ?Why is fit so important in this population?? Future THEORY OF WORK ADJUSTMENT 139 research can seek to answer this question by attempting to identify the predictors of fit for African-American employees. Results of the present study suggested that fit may be influenced by racial climate. Other possible variables that influence fit include the presence of critical mass which is supported by Thomas and Alderfer (1990) and Kanter (1977). Racial identity may also emerge as a predictor of fit. Results explained in Watts and Carter (1991) suggested that racial climate shapes African-American employees? assessments of their work environment in terms of racially salient stimuli. It stands to reason then that those endorsing different dominant racial identity statuses may differentially perceive likeness between themselves and their organization as it pertains to racially salient work variables (e.g., the role of race in hiring and advancement decisions). Finally, practical and economic necessities which may include employees? desire to reside in particular geographical locations, proximity to family, and the ability of a job to meet employees? financial responsibility can be researched to determine the impact of this variable on fit. Thus, incorporating the suggestions offered in this section, future research should aim to place an emphasis on context. More specifically, research on fit perceptions for African-American employees might incorporate many ecological layers including national concerns (e.g., perceptions of the national economy), community factors (e.g., experiences with racism outside of the workplace), lifestyle (e.g., educational opportunities), workplace dimensions (e.g., THEORY OF WORK ADJUSTMENT 140 presence of critical mass, job security), and personal characteristics (e.g., racial identity). Conclusion. Results of the present study suggested that the TWA may have relevance for African-American employees because it may allow for the consideration of context, and more specifically, African Americans? perception of their environment within their assessment of fit. Indeed, the results of the present study pointed to the increased need to attend to the perceptions of fit for African-American employees because of the statistical significance of this variable in the prediction of job satisfaction and intentions to quit. Additionally, it is possible that perceptions of fit act as a buffer against hostile racial climates. The ability of perceptions of fit to mediate the relationship between other workplace variables (e.g., perceptions of critical mass and job satisfaction) should be attended to in research. In terms of practice, attempts to help African Americans gain access to workplaces in which they can express their values and abilities as well as have these needs met becomes important. Future research and practice should focus on creating better conditions to make fit more likely to happen for African-American employees. THEORY OF WORK ADJUSTMENT 141 Figure 1. Hypothesized model portraying racial climate as a moderator of the relationship between perceptions of fit and job satisfaction and perceptions of fit and turnover intentions Perceptions of Fit PREDICTOR Racial Climate MODERATOR Perceptions of Fit X Racial Climate PREDICTOR X MODERATOR Job Satisfaction Intention to Quit a d f c b e g THEORY OF WORK ADJUSTMENT 142 Figure 2. Model presenting resulting relationships between racial climate, perceptions of fit, job satisfaction, and turnover intentions Racial Climate Job Satisfaction Intention to Quit Perceptions of Fit THEORY OF WORK ADJUSTMENT 143 Appendix A Consent Form: Workplace Experiences Study Your involvement in the Workplace Experiences study will involve the completion of a questionnaire, which will take roughly 30 minutes. The purpose of the Workplace Experiences Study is to help to understand the work experiences of African-American employees. This consent form will be kept separately from all forms that you complete today to maintain the anonymity of your responses. In addition, your participation will be kept confidential. For reporting and presentation, the data you provide will be grouped with data that others provide. Your employer will not have access to the individual data that you provide. Furthermore, all data will be given to or sent directly to Heather Z. Lyons. You may, at any time, withdraw from this study. However, all of your answers are very valuable to this study. I understand that there are no known risks associated with participation in this study. The benefits of this study are not intended to help me personally, but rather to help the investigators to develop an understanding of the work experiences of African-American employees. I state that I am over 18 years of age and wish to participate in a program of research conducted by Heather Z. Lyons, M.A. and Karen M. O?Brien, Ph.D. in the Department of Psychology at the University of Maryland, College Park. If you have any questions during the course of your participation, please feel free to ask. Also, should you have any questions after you leave, please contact one of the researchers. Furthermore, should you have any questions regarding the rights of a research participant please contact Professor Hal Sigall at hsigall@psyc.umd.edu. Date: _______________ Signature: ______________________________ Print name: _____________________________ Heather Z. Lyons, M.A. Karen O?Brien, Ph.D. Department of Psychology Department of Psychology Biology-Psychology Building Biology-Psychology Building University of Maryland University of Maryland College Park, MD 20742 College Park, MD 20742 (301) 405-2923 (301) 405-5812 hlyons@psyc.umd.edu kobrien@psyc.umd.edu THEORY OF WORK ADJUSTMENT 144 Appendix B Biodata Sheet Please complete the following demographic information. 1. Gender: Female / Male 2. Race : ________________________________ 3. Organization/Employer: __________________________ 4. Department/Bureau: ___________________________ 5. Position: ______________________________ 6. Years with organization: __________________ 7. Were you born in the United States? Yes / No If not, a. How long have you lived in the U.S.? ________________ b. In what country do you hold citizenship? __________________________ 8. Age: _______________ 9. Hours per week worked: ______________________ 10. Highest degree earned (circle one): a. Less than high school b. High school diploma c. Some college d. Associate?s degree e. Bachelor?s degree f. Graduate/professional school g. Other: ___________________________ 11. Years in position: ________________________ THEORY OF WORK ADJUSTMENT 145 Appendix C Perceptions of Fit Please answer the following questions by circling what you feel is the best response. In this questionnaire, your workplace is the actual organization/setting where you work. Your job is the actual tasks that you perform. To a Very Little Extent 1 2 3 4 To a Very Large Extent 5 1. To what extent does your workplace measure up to the kind of workplace you were seeking? 1 2 3 4 5 2. To what extent are the values of your workplace similar to your own values? 1 2 3 4 5 3. To what extent does your personality match the personality or image of the organization? 1 2 3 4 5 4. To what extent does your workplace fulfill your needs? 1 2 3 4 5 5. To what extent is your workplace a good match for you? 1 2 3 4 5 THEORY OF WORK ADJUSTMENT 146 6. To what extent does your job measure up to the kind of job you were seeking? 1 2 3 4 5 7. To what extent do your knowledge, skills, and abilities match the requirements of your job? 1 2 3 4 5 8. To what extent does your job fulfill your needs? 1 2 3 4 5 9. To what extent is the job a good match for you? 1 2 3 4 5 10. To what extent does the job enable you to do the kind of work you want to do? 1 2 3 4 5 THEORY OF WORK ADJUSTMENT 147 Appendix D Workplace Compatibility Please answer the following questions by circling what you feel is the best response. Compare your values with what you believe your workplace values. My organization values this more than I Similar level of value I value this more than my organization 1. Using your talents at work 1 2 3 4 3 2 1 2. Achievement 1 2 3 4 3 2 1 3. Work production 1 2 3 4 3 2 1 4. Staying active 1 2 3 4 3 2 1 5. Autonomy 1 2 3 4 3 2 1 6. Independence 1 2 3 4 3 2 1 7. Variety of tasks 1 2 3 4 3 2 1 8. Financial compensation 1 2 3 4 3 2 1 9. Job security 1 2 3 4 3 2 1 10. Working conditions 1 2 3 4 3 2 1 11. Advancement opportunities 1 2 3 4 3 2 1 12. Recognition for work 1 2 3 4 3 2 1 13. Expressing authority 1 2 3 4 3 2 1 14. Garnering status 1 2 3 4 3 2 1 THEORY OF WORK ADJUSTMENT 148 15. Amiable (friendly) working relationships 1 2 3 4 3 2 1 16. Expression of moral values 1 2 3 4 3 2 1 17. Helping others 1 2 3 4 3 2 1 18. Fair company policies and procedures 1 2 3 4 3 2 1 19. Superiors who support their subordinates 1 2 3 4 3 2 1 20. Investment in training 1 2 3 4 3 2 1 21. Creativity at work 1 2 3 4 3 2 1 22. Responsibility at work 1 2 3 4 3 2 1 THEORY OF WORK ADJUSTMENT 149 Appendix E Racial Climate Scale Below are statements about the behaviors and policies that may or may not apply to your workplace. When completing this questionnaire, pretend that you are an ?on site reporter? reporting on what you observe in your workplace. Never 1 2 Sometimes 3 4 Always 5 1. This organization goes out of its way to make Blacks feel at home. 1 2 3 4 5 2. Blacks are given the respect they deserve from Whites at work. 1 2 3 4 5 3. Blacks are discriminated against through hiring practices. 1 2 3 4 5 4. Race determines who gets the most desirable work/assignments. 1 2 3 4 5 5. It?s just as hard for Whites to get ahead here as for Blacks. 1 2 3 4 5 6. In general, organization-wide racism is a problem here. 1 2 3 4 5 7. The racism here has caused me to consider quitting/transferring. 1 2 3 4 5 8. An important job of management here is to help Blacks and other groups get along. 1 2 3 4 5 9. There are Black people in positions of power here. 1 2 3 4 5 THEORY OF WORK ADJUSTMENT 150 10. Blacks have little to say about decisions affecting this agency. 1 2 3 4 5 11. There are enough Blacks in powerful positions here. 1 2 3 4 5 12. Black and White employees generally have good working relationships here. 1 2 3 4 5 13. There is a very sensitive understanding and acceptance of differences about ethnic or racial groups here. 1 2 3 4 5 14. Blacks get the promotions they deserve. 1 2 3 4 5 15. Many changes have been made to make services (resources) available to Black people. 1 2 3 4 5 16. Little has been done to change services or functioning to serve the culture of Black professionals. 1 2 3 4 5 17. The opinions of Black people are respected here. 1 2 3 4 5 18. Racism is not tolerated here. 1 2 3 4 5 THEORY OF WORK ADJUSTMENT 151 Appendix F Job Satisfaction For each item below ask yourself: How satisfied am I with this aspect of my job? Indicate your level of satisfaction by circling one of the responses for that item. Very Dissatisfied 1 2 3 4 Very Satisfied 5 1. Being able to keep busy all the time 1 2 3 4 5 2. The chance to work alone on the job 1 2 3 4 5 3. The chance to do different things from time to time 1 2 3 4 5 4. The chance to be "somebody" in the community 1 2 3 4 5 5. The way my boss handles his/her workers 1 2 3 4 5 6. The competence of my supervisor in making decisions 1 2 3 4 5 7. Being able to do things that don't go against my conscience 1 2 3 4 5 8. The way my job provides for steady employment 1 2 3 4 5 9. The chance to do things for other people 1 2 3 4 5 10. The chance to tell people what to do 1 2 3 4 5 11. The chance to do something that makes use of my abilities 1 2 3 4 5 THEORY OF WORK ADJUSTMENT 152 12. The way company policies are put into practice 1 2 3 4 5 13. My pay and the amount of work I do 1 2 3 4 5 14. The chances for advancement on this job 1 2 3 4 5 15. The freedom to use my own judgment 1 2 3 4 5 16. The chance to try my own methods of doing the job 1 2 3 4 5 17. The working conditions 1 2 3 4 5 18. The way my co-workers get along with each other 1 2 3 4 5 19. The praise I get for doing a good job 1 2 3 4 5 20. The feeling of accomplishment I get from the job 1 2 3 4 5 THEORY OF WORK ADJUSTMENT 153 Appendix G Intentions to Quit Please rate the extent to which you agree or disagree with the following statements. Strongly Disagree 1 2 3 4 Strongly Agree 5 1. If I have my own way, I will be working for my current employer one year from now. 1 2 3 4 5 2. I frequently think of quitting my job. 1 2 3 4 5 3. I am planning to search for a new job in the next 12 months. 1 2 3 4 5 THEORY OF WORK ADJUSTMENT 154 Appendix H Debriefing Statement Thank you for taking part in the Workplace Experiences Study. This study is intended to determine whether your perceptions of particular aspects of your workplace climate influence your level of job satisfaction and intentions to stay in or leave your workplace. To do this, you were given measures of job satisfaction, racial climate, intention to remain in the workplace, and your perceptions of the compatibility between your values and the values endorsed in your workplace. Because knowledge of the purpose of this study can influence the ways in which people take tests, please do not divulge the purpose of this study to anyone who plans to participate in this study. If you have questions about the study, I can be reached at: Heather Z. Lyons, M.A. Karen O?Brien, Ph.D. Department of Psychology Department of Psychology Biology-Psychology Building Biology-Psychology Building University of Maryland University of Maryland College Park, MD 20742 College Park, MD 20742 (301) 405-2923 (301) 405-5812 hlyons@psyc.umd.edu kobrien@psyc.umd.edu Please feel free to contact me if you would like results of this study. Thank you again for your participation! THEORY OF WORK ADJUSTMENT 155 References Austin, R. L., & Dodge, H. H. (1992). Despair, distrust and dissatisfaction among Blacks and women, 1973-1987. Sociological Quarterly, 33, 579-598. Barbarin, O. A., & Gilbert, R. (1981). Institutional racism scale: Assessing self and organizational attributes. In O. A. Barbarin, P. R. Good, O. M. Pharr, & J. A. Siskind (Eds.) Institutional Racism and Community Competence. Rockville, MD: U. S. Department of Health and Human Services. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Berry, M. F. & Blassingame, J. W. (1982). Long memory: The Black experience in America. New York: Oxford University Press. Bizot, E. B., & Goldman, S. H. (1993). Prediction of satisfactoriness and satisfaction: An 8-year follow up. Journal of Vocational Behavior, 43, 19-29. Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability. In S. W. Koz.owski & K. J. Klein (Eds). Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 349-381). San Francisco: Joseey-Bass. Breeden, S. A. (1993). Job and occupational change as function of occupational correspondence and job satisfaction. Journal of Vocational Behavior, 43, 30- 45. THEORY OF WORK ADJUSTMENT 156 Brenner, O. C., & Fernsten, J. A. (1984). Racial differences in perceived job fulfillment of white collar workers. Perceptual & Motor Skills, 58, 643-646. Bretz, R. D., & Judge, T. A. (1994). Person-organization fit and the theory of work adjustment: Implications for satisfaction, tenure, and career success. Journal of Vocational Behavior, 44, 32-54. Brown, D., Brooks, L., & Associates. (1996). Career choice and development. San Francisco: Jossey-Bass Publishers. Brown, D., Minor, C. W., & Jepsen, D. A. (1991). The opinions of minorities about preparing for work: Report of the second NCDA National Survey. The Career Development Quarterly, 40, 5-19. Brown, M. T. (1995). The career development of African Americans: Theoretical and empirical issues. In F. T. Leong (Ed.), Career development and vocational behavior or racial and ethnic minorities (pp.7-36). Mahwah, NJ: Lawrence Erlbaum Associates. Byars, A. M. (1999, August). Racial and ethnic influences on social cognitive career- related constructs. In R. Lent (Chair), Applying Social Cognitive Career Theory to Black Americans: Challenging Prospects. Symposium conducted at the annual meeting of the American Psychological Association, Boston, MA. Cheatham, H. E. (1990). Africentricity and career development of African Americans. The Career Development Quarterly, 38, 334-346. THEORY OF WORK ADJUSTMENT 157 Cohen, A., Doveh, E., & Eick, U. (2001). Statistical properties of the r wg(j) index of agreement. Psychological Methods, 6, 297-310. Cohen, J. & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum. Cohn, J. (1997). The effects of racial and ethnic discrimination on the career development of minority persons. In H. S. Farmer & Associates (Eds.), Diversity and women?s career development. (pp. 161-171). Thousand Oaks, CA: Sage Publications. Colarelli, S. M. (1984). Methods of communication and mediating processes in realistic job previews. Journal of Applied Psychology, 69, 633-642. Cook, T. D. & Campbell, D. T. (1979). Quasi-experimental designs and analysis issues in field settings. Chicago: Rand McNally. Darmon, R. Y. (1990). Identifying sources of turnover costs: A segmental approach. Journal of Marketing, 54, 46-56. Davis, J. A. (1989). General social surveys, 1872-1989. Chicago: National Opinion Research Center. Dawis, R. V. (1994). The theory of work adjustment as convergent theory. In M. L. Savikas & R. W. Lent (Eds.), Convergence in career development theories: Implications for science and practice. (pp. 32-43). Palo Alto, CA: CPP Books. THEORY OF WORK ADJUSTMENT 158 Dawis, R. V., England, G., & Lofquist, L.H. (1964). A theory of work adjustment (a revision). Minnesota Studies in Vocational Rehabilitation, 15. Dawis, R. V., & Lofquist, L. H. (1976). Personality style and the process of work adjustment. Journal of Counseling Psychology, 23, 55-59. Dawis, R. V., & Lofquist, L. H. (1978). A note on the dynamics of work adjustment. Journal of Vocational Behavior, 12, 76-79. Dawis, R. V., & Lofquist, L. H. (1984). A psychological theory of work adjustment: An individual differences model and its applications. Minneapolis, MN: University of Minnesota Press. Dawis, R. V., & Lofquist, L. H. (1993). Rejoinder: From the TWA to PEC. Journal of Vocational Behavior, 43, 113-121. Edwards, J. R. (1991). Person-job fit: A conceptual integration, literature review, and methodological critique. In C.L. Cooper & I.T. Robertson (Eds.), International review of industrial and organizational psychology, 6. Chichester: Wiley. Edwards, J. R. (2002). Alternatives to difference scores: Polynomial regression analysis and response surface methodology. In F. Drasgow & N. W. Schmitt (Eds.), Measuring and analyzing behavior in organizations: Advances in measurement and data analysis. (pp. 350-400). San Francisco: Jossey-Bass. Ensher, E. A., Grant- Vallone, E. J., & Donaldson, S. I. (2001). Effects of perceived discrimination on job satisfaction, organizational commitment, and THEORY OF WORK ADJUSTMENT 159 organizational citizenship behavior, and grievances. Human Resource Development Quarterly, 12, 53-72. Fassinger, R. E., & O'Brien, K. M. (2000). Career counseling with college women: A scientist-practitioner-advocate model of intervention. In D. A. Luzzo (Ed.)., Career development of college students: Translating theory and research into practice (pp. 253-265). Washington, DC: American Psychological Association. Fassinger, R. E. (2001). Diversity at work: Research issues in vocational development. In D. Pope-Davis & Coleman, H. (Eds.) The intersection between race, class, and gender: Implications for multicultural counseling. (pp. 267-288). Thousand Oaks, CA: Sage. Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson. Fitzgerald, L. F., & Betz, N. E. (1994). Career development in cultural context: The role of gender, race, class, and sexual orientation. In M. L. Savikas & R. W. Lent (Eds.), Convergence in career development theories (pp.103-118). Palo Alto, CA: CPP Books. Fitzgerald, L. F., & Rounds, J. (1993). Women and work: theory encounters reality. In W. Walsh & S. Osipow (Eds.), Career counseling for women (pp. 327-354). Hillsdale, NJ: Erlbaum. Fitzgerald, L. F., Shullman, S., Bailey, N., Richards, M., Swecker, J., Gold, A., Ormerod, A. J., & Weitzman, L. (1988). The incidence and dimensions of THEORY OF WORK ADJUSTMENT 160 sexual harassment in academia and the workplace. Journal of Vocational Behavior, 32, 152-175. Gelfand, M. J., Fitzgerald, L. F., & Drasgow, F. (1995). The structure of sexual harassment: A confirmatory analysis across cultures and settings. Journal of Vocational Behavior, 47, 164-177. Goodman, P. S., & Atkins, R. S. (1984). Absenteeism. San Francisco: Jossey-Bass. Greenhaus, J. H., Parasuaraman, S., & Wormley, W. M. (1990). Effects of race on organizational experiences, job performance evaluations, and career outcomes. Academy of Management Journal, 33, 64-86. Gwaltney, J. L. (1981). Drylongso: A self-portrait of Black America. New York: Vintage. Hackett, G., & Byars, A.M. (1996). Social cognitive theory and the career development of African American women. Career Development Quarterly, 44, 322-340. Hackman, J. R., & Oldham, G. R. (1975). Development of the job diagnostic survey. Journal of Applied Psychology, 60, 159-170. Heilman, M. E., Simon, M. C., & Repper, D. P. (1987). Intentionally favored, unintentionally harmed? Impact of sex-based preferential selection on self- perceptions and self-evaluations. Journal of Applied Psychology, 72, 62-68. THEORY OF WORK ADJUSTMENT 161 Helms, J. E., & Parham, T. A. (1990). Black racial identity attitude scale. In J. E. Helms (Ed.), Black and white racial identity. (pp.245-247). Westport, CT: Praeger. Helms, J. E., & Piper, R. E. (1994). Implications of racial identity theory for vocational psychology, Journal of Vocational Behavior, 44, 124-138. Holder, J. C. (1994). Workplace racial climate scale. Unpublished measure, Southern Illinois University. Holder, J. C., & Vaux, A. (1998). African American professionals: Coping with occupational stress in predominantly white work environments, Journal of Vocational Behavior, 53, 315 ? 333. Holland, J. L. (1985). Making vocational choices. Englewood, NJ: Prentice-Hall. Horn, P. W., & Griffeth, R. W. (1991). Structural equations modeling test of a turnover theory: Cross-sectional and longitudinal analyses. Journal of Applied Psychology, 76, 350-366. Hughes, D., & Dodge, M. A. (1997). African American women in the workplace: Relationships between job conditions, racial bias at work, and perceived job quality. American Journal of Community Psychology, 25, 581-599. Idazak, J. R., Bottom, W. P., Drasgow, F. (1988). A test of the measurement of equivalence of the revised job diagnostic survey: Past problems and current solutions. Journal of Applied Psychology, 73, 647-656. THEORY OF WORK ADJUSTMENT 162 Jackson, S. E., Brett, J. F., Sessa, V. I., Cooper, D. M., Julian, J. A., & Peyronnian, K. (1991). Some differences make a difference: Individual dissimilarity and group heterogeneity as correlates of recruitment, promotions, and turnover. Journal of Applied Psychology, 76, 675-689. James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interater reliability with and without response bias. Journal of Applied Psychology, 69, 85-98. James, L. R., Demaree, R. G., & Wolf, G. (1993). An assessment of within-group interrater agreement. Journal of Applied Psychology, 78, 306-309. Jewell, L. N. (1998). Contemporary industrial/organizational psychology. Pacific Grove, CA: Brooks/Cole Publishers. Joplin, J. R. W., & Shaffer, M. A. (1997). Harassment, lies, theft, and other vexations: A cross-cultural comparison of aggressive behaviors in the workplace. Report presented at the National Meetings of the Academy of Management, Boston, MA. Kanter, R. M. (1977). Men and women of the corporation. New York: Basic Books. Kaye, L. W., Alexander, L. B., & Kauffman, S. (1999). Factors contributing to job quality and satisfaction among ethnically diverse, lower income, elderly part- timers. Journal of Gerontological Social Work, 31, 143-166. THEORY OF WORK ADJUSTMENT 163 Kristof, A. L. (1996). Person-organization fit: An integrative review of its conceptualizations,measurement, and implications. Personnel Psychology, 49, 1-46. Lawley, D. & Maxwell, A. E. (1971). Factor Analysis as a Statistical Method. New York: Elsevier. Lovelace, K., & Rosen, B. (1996). Differences in achieving person-organization fit among diverse groups of managers. Journal of Management, 22, 703-723. Lyons, H. Z. & Brenner, B. R. (in preparation). Developing a theoretically-based measure of perceptions of fit. Lyons, H. Z., Brenner, B. R., & Fassinger, R. E. (2003). Testing the role of experiences with heterosexism in the Theory of Work Adjustment. Paper presented in the symposium, Results of the National Gay and Lesbian Experiences Study, American Psychological Association Multicultural Summit, Los Angeles, CA, January, 2003. Mays, V. M., Coleman, L. M. & Jackson, J. S. (1996). Perceived race-based discrimination, employment status, and job stress in a national sample of Black women: Implications for health outcomes. Journal of Occupational Health Psychology, 1, 319-329. Meyer, J. P., Allen, N. J., & Smith, C. A. (1993). Commitment to organizations and occupations: Extension and test of a three-component conceptualization. Journal of Applied Psychology, 78, 538-551. THEORY OF WORK ADJUSTMENT 164 Mowday, R.T., Porter, L. W. & Steers, R. M.. (1982). Employee-organization linkages. New York: Academic Press. Mueller, C. W., Finley, A., Iverson, R. D., & Price, J. L. (1999). The effects of group racial composition on job satisfaction, organizational commitment, and career commitment: The case of teachers. Work & Occupations, 26, 187-219. Near, J. P., Rice, R. W., & Hunt, R. G. (1983). Job satisfaction and nonwork satisfaction as components of life satisfaction. Journal of Applied Social Psychology, 13, 126-144. Nyambegera, S. M., Daniels, K., & Sparrow, P. (2001). Why fit doesn?t always matter: The impact of HRM and cultural fit on job involvement of Kenyan employees. Applied Psychology: An International Review, 50, 109-140. Parkes, L. P., Bochner, S., & Schneider, S. K. (2001). Person-organization fit across cultures: An empirical investigation of individualism and collectivism. Applied Psychology: An international review, 50, 81-108. Peterson, N. & Gonzalez, R. (1999). The role of work in people?s lives. Boston: Brooks/Cole. Plous, S. (1996). The affirmative action Debate: What's fair in policy and programs?. Journal of Social Issues, 52, 25-31. Quinn, R., & Staines, G. L. (1979). The 1977 quality of employment survey. Ann Arbor, MI: Survey Research Center. THEORY OF WORK ADJUSTMENT 165 Richardson, M. S. (1993). Work in people?s lives: A location for counseling psychologists. Journal of Counseling Psychology, 40, 425-433. Rounds, J. B. (1990). The comparative and combined utility of work value and interest data in career counseling with adults. Journal of Vocational Behavior, 37, 32-45. Rounds, J. B., Dawis, R. V. & Lofquist, L. H. (1987). Measurement of person- environment fit and prediction of satisfaction in the theory of work adjustment. Journal of Vocational Behavior, 31, 297-318. Rounds, J. B., & Hesketh, B. (1994). The theory of work adjustment: Unifying principles and concepts. In M. L. Savikas & R. W. Lent (Eds.), Convergence in career development theories (pp. 177-186). Palto Alto, CA: CPP Books. Ryan, N. E. (1999). Career counseling and career choice goal attainment: A meta- analysis dervied model for career counseling practice. Unpublished doctoral dissertation. Loyola University, Chicago. Saks, A. M., & Ashforth, B. E. (1997). A longitudinal investigation of the relationships between job information sources, applicant perceptions of fit, and work outcomes. Personnel Psychology, 50, 395-427. Schneider, B. (1997). The people make the place. Personnel Psychology, 40, 437-454. Schneider, B. (2001). Fits about fit. Applied Psychology: An International Review, 40, 141-152. THEORY OF WORK ADJUSTMENT 166 Schneider, B., Smith, D. B., Taylor, S., & Fleenor, J. (1998). Personality and organizations: A test of the homogeneity of personality hypothesis. Journal of Applied Psychology, 83, 462-470. Shaffer, M. A., Joplin, J. R. W., Bell, M. P., Lau, T., & Oguz, C. (2000). Gender discrimination and job-related outcomes: A cross-cultural comparison of working women in the United States and China. Sharf, R. S. (1997). Applying career development theory to counseling. Boston: Brooks/Cole. Taris, R., & Feji, J. A. (2001). Longitudinal examination of the relationship between supplies-values fit and work outcomes. Applied psychology: An international review, 50, 52-80. Ten Horn, L. A. (1983). Behoeften, werksituatie en arbeidsbeleving. Pijnacker: Dutch Efficiency Bureau. Thomas, D. A> & Alderfer, C. P. (1990). The influence of race in career dynamics: Theory & research on minority experiences, pp. 138-158. In M. Arthur, D. T. Hall, & B. Lawrence (Eds.). Handbook of Career Theory. Cambridge: Cambridge University Press. Tinsley, D. J. (1993). Extensions, elaborations, and construct validations of the theory of work adjustment. Journal of Vocational Behavior, 43, 67-74. Tracey, T. (2000). Issues in the Analysis and Interpretation of Quantitative Data: Deinstitutionalization of the Null Hypothesis. In S. D. Brown & R. W. Lent THEORY OF WORK ADJUSTMENT 167 (Eds.). Handbook of Counseling Psychology. Hoboken, NJ: John Wiley & Sons. Tuch, S. A., & Martin, J. K. (1991). Race in the workplace: Black/White differences in the sources of job satisfaction. The Sociological Quarterly, 32, 103-116. U.S. Census Bureau, Education and Social Stratification Branch (2002). Current Population Survey. TabA-2. Retrieved August, 30, 2003. www.census.gov/population/socdem/education/tabA-2.pdf. U.S. Department of Labor. (1977). Dictionary of occupational titles. (4 th ed.). Washington, D.C.: U.S. Government Printing Office. Van Vianen, A. E. M. (2000). Person-organization fit: The match between newcomers? and recruiters? preferences for organizational cultures. Personnel Psychology, 53, 113-150. Watts, R. J. & Carter, R. T. (1991). Psychological aspects of racism in organizations, Group & Organization Studies, 16, 328-344. Weaver, C. N. (1977). Relationships among pay, race, sex, occupational prestige, supervision, work autonomy, and job satisfaction in a national sample. Personnel Psychology, 30, 437-445. Weiss, D.J., Davis, RV., England, G.W., & Lofquist, L.H. (1967). Manual for the Minnesota Satisfaction Questionnaire. Minneapolis: University of Minnesota, Work Adjustment Project Industrial Relations Center. THEORY OF WORK ADJUSTMENT 168 Williams, R. (1999). Unfinished business: African-American political economy during the age of ?color-blind? politics. In W. Spriggs (Ed.), The state of Black America. (pp. 137-152). New York: National Urban League. Williams, R. (2000). If you?re black, get back; if you?re brown, stick around; if you?re white, hang tight: A primer on race, gender, and work in the global economy. A Working Paper for the Preamble Center.