ABSTRACT Title of Disertation:A STUDY OF SELF-AWARENES, SELF-EFICACY, AND SOJOURNER ADJUSTMENT OVER TIME Bradley R. Brenner, Doctor of Philosophy, 2003 Disertation Directed by:Profesor Robert W. Lent Department of Counseling and Personnel Services Most models of intercultural adjustment rely on the asumption that adjustment changes systematicaly over the course of a sojourn. However, studies in the intercultural transitions literature generaly have not found evidence of systematic change, most likely hampered by their cross-sectional methodology. In the present study, changes in psychological and sociocultural intercultural adjustment were examined using data collected over time. Two additional constructs, intercultural adjustment self-eficacy beliefs and self-awarenes, also were studied for their change over the course of a sojourn. Intercultural adjustment self-eficacy beliefs also were examined for their temporal precedence over intercultural adjustment. Moreover, intercultural adjustment self-eficacy was tested as a possible moderator in the relationship betwen self- awarenes and adjustment. To examine the proces and progresion of intercultural adjustment, 52 U.S. university study abroad students reported their intercultural experiences. They provided responses to measures just prior to departure, and again at one-month intervals during their first thre months overseas. Latent growth modeling with factor means estimation was the primary method of data analysis, while path analysis was used as a secondary method. It was found that sociocultural adjustment, intercultural adjustment self-eficacy, and self-awarenes changed in a systematic fashion. The trajectory of sociocultural adjustment was found to be similar to a ?U?- curve, self-eficacy demonstrated upward growth, and self-awarenes was found to be similar to an inverted ?U?-curve. No systematic change in psychological adjustment was found. Self-eficacy beliefs were shown to demonstrate temporal precedence over changing sociocultural beliefs. Lastly, intercultural adjustment self-eficacy was shown to moderate the relationship betwen self-awarenes and sociocultural adjustment, such that the relationship betwen self-awarenes and sociocultural adjustment was stronger at higher versus lower levels of intercultural adjustment self-eficacy. Future areas of research, both in the international and domestic spheres were discussed, as wel as the implications of the current study for counseling interventions and its limitations. A STUDY OF SELF-AWARENES, SELF-EFICACY, AND SOJOURNER ADJUSTMENT OVER TIME by Bradley R. Brenner Disertation submited to the Faculty of the Graduate School of the University of Maryland, College Park in partial fulfilment of the requirements for the degre of Doctor of Philosophy 2003 Advisory Commite: Profesor Robert Lent, Chair Profesor Ruth Fasinger Profesor Margaretha Lucas Profesor Marylu McEwen Profesor Karen O?Brien ?Copyright by Bradley R. Brenner 2003 i TABLE OF CONTENTS LIST OF TABLES....................................................ii LIST OF FIGURES...................................................iv Chapter 1: Introduction.................................................1 Change in Sojourners? Experience Over Time......................................2 Prediction of Sojourner Experience...............................................4 Chapter 2: Review of the Literature........................................7 Positive and Negative Growth of Sojourner Characteristics...........................8 Change in Sojourner Adjustment Over Time...............................................8 Change in Sojourner Self-Eficacy Beliefs Over Time......................................16 Change in Sojourner Self-Awarenes Over Time...........................................17 Correlates of Sojourner Adjustment.............................................21 Self-Eficacy and Sojourner Adjustment.................................................23 Self-Awarenes and Sojourner Adjustment...............................................25 Intercultural Adjustment Self-Eficacy as Moderator of the Relationship betwen Self-Awarenes and Sojourner Adjustment................................................................29 Statement of the Problem.......................................................31 Hypotheses and Research Questions..............................................33 Research Question 2: Do sojourner intercultural adjustment self-eficacy beliefs change in a systematic manner?..................................................................33 Research Question 3: Does sojourner self-awarenes change in a systematic manner?............34 Chapter 3: Method....................................................37 Participants.........................................................................37 Measures...........................................................................40 Procedure..........................................................................46 Chapter 4: Results....................................................49 Chapter 5: Discusion.................................................71 Prediction of Sojourner Adjustment.....................................................76 Future Research.....................................................................78 Implications for Counseling Interventions................................................82 Limitations.........................................................................85 Appendix A..........................................................87 Appendix B..........................................................94 Appendix C..........................................................95 Appendix D..........................................................96 Appendix E..........................................................97 REFERENCES......................................................98 ii LIST OF TABLES Table 1: Intercorrelations, Means, and Standard Deviations of Four Waves of Psychological Adjustment, Sociocultural Adjustment, Intercultural Adjustment Self- Eficacy, and Self-Awarenes............................................53 iv LIST OF FIGURES Figure 1: Hypothesized Moderating Efects of Self-Eficacy on the Relationship Betwen Self-Awarenes and Sojourner Adjustment...........................36 Figure 2: Structural Model of Latent Growth of Psychological Adjustment.........56 Figure 3: Psychological Adjustment Sample Means...........................56 Figure 4: Structural Model of Latent Growth of Sociocultural Adjustment..........59 Figure 5: Latent Growth of Sociocultural Adjustment Factor Means..............60 Figure 6: Structural Model of Latent Growth of Self-Eficacy...................61 Figure 7: Latent Growth of Intercultural Adjustment Self-Eficacy Factor Means....62 Figure 8: Structural Model of Latent Growth of Self-Awarenes.................63 Figure 9: Latent Growth of Self-Awarenes Factor Means......................64 Figure 10: Structural Model of Efect of Self-Eficacy at Time 1 on Latent Growth of Sociocultural Adjustment...............................................66 Figure 11: Efects of Self-Eficacy at Time 1 on Latent Growth of Sociocultural Adjustment..........................................................66 Figure 12: Standardized Path Analysis Model of Moderating Efects of Self-Awarenes on the Relationship Betwen Self-Awarenes and Psychological Adjustment........68 Figure 13: Standardized Mean Centered Path Analysis Model of Moderating Efects of Self-Awarenes on the Relationship Betwen Self-Awarenes and Sociocultural Adjustment..........................................................70 Figure 14: Moderating Efects of Self-Eficacy on the Relationship Betwen Self- Awarenes and Sociocultural Adjustment...................................70 Sojourner Adjustment 1 Chapter 1: Introduction Living in a foreign country is often a life-altering event (Harison, Chadwick, & Scales, 1996; Jerusalem & Mitag, 1997; Mak & Tran, 2001). For some people it is the chance of a lifetime to learn and develop a lasting appreciation for other cultures, gain a beter understanding of their own values and abilities, and begin friendships with foreign nationals. These people could be described as having succesfully adjusted and thrived during an intercultural transition. Other people?s experiences are far diferent from those just listed. For them, life in a foreign country is entirely disimilar. It is a time of anxiety and interpersonal frustration. Instead of a pleasurable experience, it is a period that may engender negative beliefs and atitudes about other cultures, decreased confidence in one?s ability to handle dificult situations, or dislike of people from other countries or diferent cultural backgrounds. Understanding this basic diference in experience has been at the forefront of intercultural transition research. Though the study of adjustment to life in a foreign country has a relatively long history (Church, 1982), litle progres has been made in identifying variables that consistently predict succesful and unsuccesful sojourns. However, recent advances in the intercultural transition literature have opened new avenues for researchers to study sojourners? experiences in a foreign country. Chief among these advances is a relatively new conceptualization of sojourner adjustment. Ward and her colleagues have developed a bipartite definition of sojourner adjustment Sojourner Adjustment 2 that acounts for two primary outcomes of intercultural transition (Ward, 1996). The first is psychological adjustment. This type of adjustment acounts for a person?s emotional and cognitive reaction to life in a foreign country. The second type of adjustment is sociocultural adjustment, a behavioral and interpersonal form of adjustment. Ward and colleagues? program of research generaly has supported their bipartite model of sojourner adjustment (Searle & Ward, 1990; Ward & Kennedy, 1992; Ward & Searle, 1991). Change in Sojourners? Experience Over Time Other aspects of a sojourner?s overseas experience may be approached with more precision now that researchers have an empiricaly supported definition of sojourner adjustment at their disposal. One of the most enduring questions about intercultural transition is the nature of change during a sojourn. This topic itself often has spurred much speculation (Church, 1982; Zapf, 1991). Most theories of changing adjustment are based on a fundamental notion that sojourners progres through an identifiable patern of adjustment whereby they enter the new culture with energy and excitement, followed by periods of increasing distres and psychological disintegration, ending with re-atainment of positive felings. Defined as the U-curve hypothesis (Lysgaard, 1955), it remains an active point of inquiry despite dismal empirical support (Ward, 1996). The lack of empirical support may be atributable in part to inappropriate research methodologies. Research with overseas participants is logisticaly dificult, and until recent advances were made with web-based questionnaires (Smith & Leigh, 1997), gaining aces to sojourners? in vivo experiences has been excesively expensive and time consuming. Because of these logistical dificulties, cross-sectional research designs Sojourner Adjustment 3 are much more the norm than is change over time methodology. In the case of identifying change during intercultural transitions, change over time methodology sems worth the costs. One ses a clear diference in findings when examining the diferences betwen studies that utilized cross-sectional and those that employed data collected over time. Change over time studies appear more adept at uncovering changing sojourner adjustment (e.g., Kealey, 1989), whereas cross-sectional designs do not (e.g., Ward & Kennedy, 1992). With the advent of web-based questionnaires, data collection over time of intercultural phenomenon is much les cumbersome. Data collection over time also afords researchers the opportunity to explore other theorized changes that occur during intercultural transitions. For instance, Adler (1975) speculated that intercultural transition may engender heightened self-awarenes that may be the cause of psychological dificulty commonly experienced by sojourners. Self- awarenes is defined as a state of self-focus where the self becomes the object of one?s atention (Davis & Franzoi, 1999; Duval & Wicklund, 1972; Fenigstein, Scheier, & Buss, 1975; Silvia & Duval, 2001). Collection of data through online means alows researchers to beter track relatively dynamic shifts in self-focused atention over the course of a sojourn. Self-eficacy is another construct that has atracted increased visibility in the intercultural transition literature (Brenner, 2001; Harison et al., 1996; Jerusalem & Mitag, 1997; Mak & Tran, 2001). Self-eficacy is defined as beliefs that one holds about his or her capabilities to perform at a specified level (Bandura, 1986). As with adjustment and self-awarenes, litle is known about the positive or negative growth of self-eficacy beliefs during sojourns. Sojourner Adjustment 4 Prediction of Sojourner Experience The search for variables that serve as efective predictors of sojourner adjustment is a historical focal point of the intercultural transition literature (Benson, 1978; Church, 1982; Ward, 1996). This aspect of the literature has been criticized for a lack of theoretical coherence and an absence of studies that clearly state and test postulates derived from theory (Furnham & Bochner, 1986). Consistently using a bipartite definition of sojourner adjustment (e.g., psychological and sociocultural adjustment) partialy addreses this concern (Ward, 1996). In utilizing an empiricaly sound outcome criterion, researchers can be les concerned with defining adjustment and can go about theorizing and testing models of how best to predict it, which was first suggested over two decades ago as a much needed, yet problematic aspect of the literature (Benson, 1978). Two theories that may prove to be particularly promising in developing efective predictive models of sojourner adjustment are self-eficacy theory and objective self- awarenes theory. Self-eficacy theory suggests that heightened eficacy beliefs are positively related to higher levels of general adjustment (Bandura, 1997). Recent atempts to study the phenomenon in intercultural setings generaly have supported this contention (Harison et al., 1996; Jerusalem & Mitag, 1997). However, those inquiries utilized generalized self-eficacy measures. Context-specific measures may yield even more convincing results (Lent, Hacket, & Brown, 1998). Objective self-awarenes theory offers an explanation of the proces whereby increased self-focused atention may impact sojourner adjustment (Duval & Wicklund, 1972; Silvia & Duval, 2001). The theory holds that environmental cues cause a person to Sojourner Adjustment 5 become objectively aware of oneself. The theory postulates that self-awarenes is an inherently self-evaluative proces with implications for afective reactions (e.g., adjustment). Two studies were identified that empiricaly explored a variable similar to self-awarenes during intercultural transitions, but their results were contradictory (Harison et al., 1996; Kealey, 1989). Part of the confusion may be due to their choice of self-monitoring which is a static, personality-like variable defined as an individual?s capacity to adapt behavior to met situational demands (Snyder, 1974). Recent research emphasizes the dynamic nature of self-awarenes (Govern & Marsch, 2001), which likely needs to be taken into acount when studying sojourner adjustment. It is dificult, however, to make definitive statements about the efects of self- awarenes on sojourner adjustment. The original formulation of objective self-awarenes (OSA) theory suggested that self-awarenes universaly leads to negative self-evaluations (Duval & Wicklund, 1972). Yet, self-awarenes has been found to be related to positive and negative intrapersonal and interpersonal outcomes (Davis & Franzoi, 1999), and OSA theory has been revised to state that self-focused atention can result in both negative and positive states (Silvia & Duval, 2001). Studies of people who are anxious demonstrate that heightened self-awarenes is an important correlate of their problematic afective states (Alden, Bieling, & Walace, 1994). Alden et al. demonstrated that outcomes of self-focused atention (e.g., negative self-evaluation) were related to self- eficacy beliefs. With those participants, individuals with high confidence in their abilities, as measured by self-eficacy beliefs, did not exhibit problematic afective states. If these results were to generalize to intercultural contexts, it may be that the efect that Sojourner Adjustment 6 self-awarenes has on sojourner adjustment may be influenced (e.g., moderated) by intercultural adjustment self-eficacy. Intercultural transition represents a unique area of inquiry that can be efectively studied by counseling psychologists. Recent trends in the intercultural transition literature have transformed deficit-based models of sojourner adjustment to models of transition and development. Counseling psychology offers much to the study of these phenomena. Formulating sojourner adjustment as an outcome of eficacy beliefs and environmentaly cued self-focused atention wil alow researchers and practitioners to draw upon a wealth of knowledge previously acumulated in the counseling psychology literature. For instance, practitioners could help develop instructional programs to bolster pre-departure sojourners? intercultural adjustment self-eficacy to beter prepare them for their upcoming trip. Likewise, a practitioner could draw upon the multiculturalism literature for guidance in helping a sojourning client to recognize the power that the social and political environment has on perceptions of one?s worth. This study, therefore, examined the development of sojourner intercultural adjustment self-eficacy beliefs, self-awarenes, and adjustment over time. Moreover, by adopting an empiricaly supported definition of sojourner adjustment (e.g., psychological and sociocultural adjustment), it was possible to derive predictive models based on self-eficacy theory and objective self-awarenes theory. Models that examine the efects over time of intercultural adjustment self-eficacy beliefs and self-awarenes on adjustment were tested. Sojourner Adjustment 7 Chapter 2: Review of the Literature Two questions remain a source of much discussion and inquiry in the intercultural transition literature. The first is the enduring question of change in sojourner characteristics over the course of a sojourn. Much theorizing and many empirical investigations have been devoted to the study of the postulated changes that occur from the beginning to the end of an intercultural experience. Work in this area has been prolific at times. Even when it is a les presing isue, it remains an important secondary isue, and is often examined within the broader context of studies designed to explore other aspects of intercultural transition. To best grapple with the questions of change over time of sojourner characteristics an understanding of theory and research findings is needed. Studies that atempted to study change in sojourner adjustment with cross- sectional and change over time research methodology are presented. Two much les copious bodies of knowledge, change over time of sojourner self-eficacy beliefs and self-awarenes, are then presented. The second major point of inquiry in the intercultural transition literature is the search for variables that predict sojourner adjustment. Beginning with studies of misionaries and Peace Corp volunters, intercultural transition researchers have developed a tradition of study in this area (Benson, 1978). Much, however, remains to be learned about the proces and ultimate outcome of intercultural transition. An overview of the intercultural adjustment literature wil review the role of two variables in predicting sojourner adjustment. Included here is an examination of the relationship Sojourner Adjustment 8 betwen intercultural adjustment self-eficacy beliefs and sojourner adjustment, and the potential moderating role that self-eficacy may place on the relationship betwen self- awarenes and adjustment. Positive and Negative Growth of Sojourner Characteristics Intercultural transition researchers commonly study growth and decline of sojourner psychological functioning, emotional wel being, or interpersonal functioning -- al indicators of sojourner adjustment. A review of studies that examined changes in sojourner adjustment over the course of a sojourn in a foreign country through cross- sectional and change over time methods is presented below. Next, a review of studies that examined a bipartite structure of sojourner adjustment (psychological and sociocultural) is reviewed. These studies demonstrate that sojourner adjustment is not a unitary construct; rather, there are two interelated aspects that should be acounted for when studying adjustment over time. Though the study of changes in sojourner adjustment is the most commonly scrutinized aspect of change over time in sojourner characteristics, other psychological variables have been studied. Self-eficacy, though a relative newcomer to the intercultural transition literature, also has been studied for change over time. One study examining change in self-eficacy over time wil be presented. Lastly, although no empirical work has been completed in the intercultural transition literature on self-awarenes, theoreticaly relevant works are reviewed that discuss self-awarenes. Change in Sojourner Adjustment Over Time Change over time in sojourner adjustment has been the subject of extensive theorizing. The most enduring notion of growth in sojourner adjustment is known as the Sojourner Adjustment 9 U-curve hypothesis of adjustment (Lysgaard, 1955). As implied by the name of the hypothesis, adjustment supposedly follows a curvilinear trajectory. Acording to this hypothesis a sojourner enters the foreign country with euphoric excitement. This changes, however, to increasing unhappines and distres. Finaly, the sojourner begins to recover and regains his or her positive felings and atitudes. This basic idea has been incorporated into many stage models of sojourner adjustment. In reviewing stage models from 1954 to 1985, Zapf (1991) identified a total of 19 that generaly conform to the U- curve trajectory. Unfortunately, the copious theorizing about change over time in sojourner adjustment has not been matched by empirical support. Atempts to uncover this phenomenon have resulted in contradictory results. Results generaly have found either (a) the opposite (e.g., an inverted ?U?-curve), (b) linear or stable trajectories, or (c) no evidence of changing sojourner adjustment. Selected studies are presented here that demonstrate methodological weakneses that may be responsible for the lack of empirical clarity. Cross-sectional designs. One of the most commonly employed methodologies for studying changing sojourner adjustment is cross-sectional data analysis. Using this methodology, researchers collect data from sojourners at one point in time. Samples are then divided into groups of sojourners based on varying lengths of time in a foreign country (e.g., 1-3 months = group 1, 4-6 months = group 2). These groups are then examined for diferences in adjustment. Hsiao-Ying?s (1995) study of sojourner adjustment in Japan utilized this strategy. Data were collected from 321 sojourners from 44 countries. Participants ranged in age from 17 to 66 years. Relevant themes of Sojourner Adjustment 10 adjustment in this study were generated by examining leters to the editor sent to English- language newspapers published in Japan. A measure was constructed using these themes, and was subjected to a principal components analysis. It suggested a four-factor structure of adjustment consisting of atitudes toward Japan, nativism, host culture as model, and alienation. To examine change over time, participants were divided into 6 groups based on their amount of time in Japan. Participants were divided into period 1 (0-6 months), period 2 (7-12), period 3 (13-24), period 4 (24-36), period 5 (37-60), and period 6 (more than 61 months). The author contended that the later groups are composed of longer time spans because the adjustment proces presumably slows as sojourners live longer in a foreign country. Utilizing the cross-sectional methodology they found decreasing adjustment over time. Maladjustment appeared to increase up to year thre of the sojourn and then flatened out to a consistently low level of adjustment. Though a wide range of time in Japan was covered by Hsiao-Ying?s (1995) study, as with cross-sectional data in general, it is hard to draw definitive conclusions from the data. Adjustment appears to decrease, but that trend may be due more to an artifact of cohort membership than time in a foreign country. For example, the people who have been in Japan for 37-60 months (period 5) could have experienced a particularly traumatic entry into the country because of economic conditions in Japan at the time of their entry. Their entry into Japan under those conditions may influence their adjustment more than time in the country per se. The inability to rule out cohort efects is a criticism that can be leveled against any study utilizing cross-sectional designs, and is especialy problematic when only Sojourner Adjustment 11 studying the experience of sojourners in one country. If one broadens the scope of study to include many countries in diferent world regions, the possibility of cohort efects may be lesened, but not eliminated (e.g., international events could stil afect most participants). Janssens (1995) studied the adjustment of sojourners on 6 continents. They were Belgian-based managers (96% male) with an average age of 39 years. Janssens studied sociocultural adjustment of the managers, relying on an author- constructed measure of intercultural interaction to operationalize the construct. Time in the country was measured with one item asking about time (in months) in the foreign country. The author tested adjustment for linear and curvilinear growth as a function of time. Both were significant for positive growth. Adjustment in this sample generaly appeared to increase with time. The positive curvilinear efect was likely a result of a relative dip in adjustment found in participants who had been in their respective countries for 18 months. Although one is able to make somewhat stronger inferences about the efects of time with these data because many countries on 6 continents are represented in the data, it is stil not possible to definitively state whether individuals change with time during their sojourn. Change over time designs. Kealey (1989) atempted to gain a beter understanding of growth in individual adjustment over time with a mixed change over time/cross-sectional design. He sampled 277 Canadians working in four regions of the world including: Asia, Anglophone Africa, Francophone Africa, and the Caribbean. Eighty-nine participants were part of the change over time data collection sub-sample, and 188 respondents comprised the cross-sectional sub-sample. The cross-sectional sample included participants who responded to the questionnaires at only one point in Sojourner Adjustment 12 time. Analysis of variance indicated that change over time and cross-sectional respondents were not significantly diferent on the variables and were, thus, combined. Using Cantril?s (1965) satisfaction scale, Kealey found evidence that the combined sample (change over time and cross-sectional) had significantly diferent mean satisfaction scores at diferent points while they were overseas (e.g., 1-3 months, 4-6 months, 7-12 months, 13-24 months). He found that the longer that the participants were overseas, the more satisfaction they reported. Kealey also atempted to examine changes in adjustment by dividing the sample into groups based on diferent growth trajectories. He found that approximately 35% of the sample showed a positive linear increase in satisfaction. Approximately 55% of the sample maintained the same level of satisfaction that they reported prior to departure. Only approximately 10% of the sample demonstrated the much-hypothesized U-curve patern. Although informative, these groups of sojourners were formed by the author in an atempt to analyze trends in the data, and do not represent statisticaly defensible categories. Additionaly, combining the change over time sample with the cross- sectional sample, though increasing the sample size, efectively negates the advantages of a change over time research design. Because this is the only example of change over time work that could be identified, it is clear that further studies that employ a change over time design are needed. Measuring change in multidimensional aspects of sojourner adjustment. Deciding how best to operationalize sojourner adjustment is a critical feature of the study of sojourner adjustment growth or decline. Sojourner adjustment has a long history of diverse definitions (Benson, 1978). Recent work has suggested, however, that most Sojourner Adjustment 13 definitions of sojourner adjustment can be grouped into one of two types ? psychological and sociocultural (Ward, 1996). Thus, most measures of adjustment tend to tap sojourner?s (a) afective or cognitive reactions (psychological) to the foreign country, or (b) interpersonal functioning, cultural knowledge, or intercultural communication skils (sociocultural). Adopting a bipartite structure of sojourner adjustment may alow researchers to more clearly study changes in sojourner adjustment. From two of the just highlighted studies it is evident that the two types of adjustment may not follow the same trajectory. Hsiao-Ying?s (1995) study, relying on a measure of psychological adjustment, demonstrated a decreasing adjustment patern for those participants. Janssens? (1995) examination of intercultural interaction (sociocultural adjustment) demonstrated the opposite. Sociocultural adjustment increased for the participants in her study. Similarly, in a recent review of their measure of sociocultural adjustment, Ward and Kennedy (1999) reported a general patern of improvement in sociocultural adjustment with time. Ward and her asociates have extensively studied the bipartite model of sojourner adjustment. Their results support the division of adjustment into two constituent components. In one of their first studies, they examined the experience of 105 Malaysian and Singaporean students with a mean age of 21.2 years (53% female) in New Zealand (Searle & Ward, 1990). The participants had been in New Zealand for an average of 27 months, with a range of six months to six years. They used the Zung Self-Rating Depresion Scale as a measure of psychological adjustment, and an author-constructed measure of self-rated management of everyday social interactions and living in a foreign country as an indication of sociocultural adjustment (Cronbach?s alpha = .81). The Sojourner Adjustment 14 authors included many measures of independent variables that were used to build models to predict psychological and sociocultural adjustment. The independent variables included expected dificulty with the new culture, perceived cultural ?distance? betwen their home and host culture, the frequency and satisfaction of contact with co-nationals and host-nationals, atitudes towards New Zealand, extraversion, and level of reported stres. In addition, sociocultural adjustment was used to predict psychological adjustment, and vice versa. Lower psychological adjustment was significantly related to poorer sociocultural adjustment (?=.28), more experienced adjustive stres (?=.28), les satisfaction with host-national contact (?=-.28), and higher extraversion (?=-.30). Lower sociocultural adjustment was predicted by lower psychological adjustment (?=.31) and two variables diferent from those that were significant predictors of psychological adjustment (greater expected dificulty with a foreign culture, ?=41, and more perceived cultural distance, ?=.17). This patern of results suggests that psychological and sociocultural adjustment are predicted by mutualy exclusive sets of independent variables, and that they are distinct but moderately related aspects of sojourners? experiences (r=.38). Notably, Searle and Ward (1990) narowly defined psychological adjustment as the absence or presence of depresion. Other forms of psychological reactions were not asesed, limiting conclusions to the mood of depresion. Their findings cannot speak to other types of negative psychological adjustment (e.g., anxiety) or to indications of positive adjustment (e.g., wel-being). Moreover, it is dificult to draw inferences about the origins of participants? depresion because their level of depresion was not asesed prior to departure. Participants who reported felings of depresion may have felt Sojourner Adjustment 15 depresed prior to leaving for their sojourn, and their elevated self-report of depresion may be unrelated to their experience as a sojourner. Support for Ward and asociates? bipartite model of sojourner adjustment has been found in other studies. For instance, Ward and Searle (1991) studied a culturaly diverse sample of participants from 42 countries. The total sample was composed of 155 university students (110 men) ranging in age from 18 to 41. In this study psychological adjustment was measured by the Profile of Mood States (McNair, Lorr, & Droppleman, 1971). Ward and Searle (1991) found that 27% of the variance in poor psychological adjustment was acounted for by measures of lonelines and cultural distance. Cultural knowledge and cultural identity acounted for 14% of the variance in sociocultural adjustment. This patern of results where psychological and sociocultural adjustment are predicted by diferent variables is further evidence of the utility of distinguishing betwen the two types of adjustment. Notably, psychological adjustment and sociocultural adjustment were significantly related (r=.35) so they are not entirely independent constructs. As with their previous study, it is dificult to make definitive statements about the origins of psychological adjustment while in a foreign country. Had they collected pre-departure data, the researchers could have asesed whether psychological adjustment difered from their pre-sojourn level of adjustment. It is important to note that the authors made atempts to search for diferent levels of adjustment based on time in a host country (Searle & Ward, 1990; Ward & Searle, 1991). In their study of sojourners to New Zealand they divided their sample into five periods of residency that alowed them to analyze their data for cross-sectional Sojourner Adjustment 16 diferences in adjustment (Searle & Ward, 1990). They found no diferences for either type of adjustment. Using a similar procedure, the authors examined the data from their second study for cross-sectional diferences in adjustment (Ward & Searle, 1991). Again, they did not find any diferences for either type of adjustment based on time in the host-country. Unlike studies where she and her colleagues utilized a data set collected over time (e.g., Ward & Kennedy, 1999), cross-sectional designs appear to fail to find efects due to time in the host country, suggesting that they are inadequate for detecting growth or decline in adjustment. It sems that reliance on cross-sectional methodology may make it dificult to find growth or decline in sojourner adjustment. However, these studies do suggest that it is best to study sojourner adjustment as a multidimensional construct. Psychological and sociocultural adjustment appear to be distinct but related constructs, and examinations of changing levels of adjustment should acount for each type. Change in Sojourner Self-Eficacy Beliefs Over Time Although sojourner adjustment has received much atention in the theoretical and empirical intercultural transition literature, other aspects of sojourners? experience deserve atention. Self-eficacy and self-eficacy related variables (e.g., locus of control) have been increasingly studied aspects of intercultural transition (Brenner, 2001; Harison et al., 1996; Jerusalem & Mitag, 1997; Ward & Kennedy, 1992). Self-eficacy beliefs have been studied in terms of their theorized influence on other variables important to intercultural transition but, curiously, have been infrequently studied for their own change over time. This is surprising given that they can be highly maleable beliefs (Bandura, 1997). Sojourner Adjustment 17 One study was identified that examined changing sojourner self-eficacy beliefs over time. Utilizing a change over time design, Jerusalem and Mitag (1997) measured generalized self-eficacy at thre points in time. Participants were 124 adults (69 males) with an average age of 25 years who were living through the reunification proces in Germany. East Germans moving to West Germany were asesed one month after the fal of the Berlin Wal (T1), six months later (T2), and a year after the second asesment (T3). At each point in time, participants were asked to respond to the German Generalized Self-Eficacy Scale (Jerusalem & Schwarzer, 1986). Participants? generalized self-eficacy did not show significant mean diferences across any of the time periods. As measured in their study, self-eficacy beliefs sem unafected by moving to a new cultural environment. The stability of beliefs may be largely due to the type of self- eficacy belief measured, however. Generalized self-eficacy beliefs are global and trait- like in nature and, thus, may not be overly sensitive to changing residence. Additionaly, they may not acurately reflect the type of context-dependent eficacy beliefs conceptualized by Bandura (1997). Self-eficacy beliefs specific to intercultural transition are likely more relevant when trying to understand the experience of sojourners (Brenner, 2001; Mak & Tran, 2001). Change in Sojourner Self-Awarenes Over Time Several authors have made arguments for the inclusion of self-awarenes as an important aspect of sojourners? experience. Hormuth (1990) postulated that self-focused atention is an important consideration because people?s self-concepts are destabilized, which ultimately engenders self-awarenes, when they find themselves in novel physical Sojourner Adjustment 18 and psychological environments (e.g., a foreign country). Adler (1975) posited that heightened self-awarenes is an inherent aspect of any intercultural transition. However, it is a variable that is rarely studied. Self-awarenes may be an element of a sojourner?s experience that changes greatly over the course of a sojourn. Deviations from a pre-departure baseline level of self-awarenes may be related to important changes in critical social identity-related fedback. For instance, new social environments (e.g., a foreign country) may drasticaly change fedback received about one?s relative social position and ability to cope within the environment (Elemers, Spears, & Doosje, 2002). Similarly, American sojourners may experience a heightened sense of personal distinctivenes in a foreign country. Distinctivenes in relation to the social environment has been shown to be an important determinant in how people define themselves (McGuire, McGuire, Child, & Fujioka, 1978). Furthermore, distinctivenes theory suggests that people are conscious of themselves insofar as they perceive themselves to be diferent from those around them. Social psychology experiments demonstrate that perceptions of feling diferent may cal atention to one?s self. In a controlled laboratory seting Frable, Blackstone, and Scherbaum (1990) demonstrated that certain groups of people are much more mindful during social interactions. Forty-four participants were selected from a pool of over 600 female university students who reported that they posses a master status condition. Acording to Goffman (1963), a master status condition is a culturaly valued or devalued group membership that occurs infrequently in the general population (e.g., being a gifted poet, a lesbian, or very wealthy). Master status conditions are either visible or concealable. Frable et al. were particularly interested in the experience of those who Sojourner Adjustment 19 reported devalued, concealable master status conditions. In their study, participants who met this definition were 8 females who were bisexual, rape survivors, or incest survivors. That is, 8 of the 44 participants possesed a devalued and concealable master status. Subjects were matched with female confederates. Participants were led to believe that they were to take a questionnaire in the company of the confederate. In reality, the confederate and participant were video tapped for five minutes while the experimenter left the room to retrieve the supposedly mising questionnaires. After informing the participants about the real nature of the study the researchers obtained consent to review the videotape with the participants? help. The participants were asked to remember al that they could about the five-minute interaction with the confederate, in addition to thoughts and felings that they had about the confederate during the interaction. Research asistants coded the participants? responses for the amount of thoughts and felings listed, diferentiating betwen direct perspectives and meta-perspectives. Direct perspectives are thoughts and felings believed to have originated from the participant, while meta-perspective thoughts and felings are those that the coders believed were adaptations of the confederates? verbalized felings and thoughts. Frable et al. (1990) found that those with a concealable, devalued status (bisexual, rape, or incest survivor) took the perspective of the confederates much more often as compared to the other participants in the study. This supported a hypothesis that those with culturaly-defined devalued diferences would be significantly more mindful of their self and social status during social interactions. That is, concealable distinctivenes was related to being highly aware of one?s diferences, and carefully atending to others in an efort to manage one?s identity. Sojourner Adjustment 20 In real world situations, arousal of self-awarenes is undoubtedly more complex than when elucidated by social psychology experiments, but studies such as these suggest that perceiving oneself to be diferent in social situations raises self-awarenes. Self- awarenes theories share a common premise that environmental factors are principaly responsible for generating heightened focus on one?s self (Duval & Wicklund, 1972; Fenigstein et al., 1975; Scheier & Carver, 1985; Silvia & Duval, 2001). These theories basicaly state that in the presence of certain environmental stimuli people become more aware of their personal characteristics, performances, and idiosyncratic histories. It has been shown that sojourners who find themselves in new sociocultural environments experience solidification, refinement, or redefinition of identity as wel as heightened self-awarenes (Church, 1982). The environmental cues found in a foreign country, therefore, likely arouse increased atention to the self as sojourners go about sensing and integrating their status as a foreign national (Sussman, 2000). Objective self-awarenes theory offers a conceptualization of self-awarenes that may offer much to the study of intercultural transitions. Duval and Wicklund (1972) conceived that conscious atention alternates dynamicaly betwen one of two possible targets. One is the external environment. The other is one?s self, conceived as an observable entity. In other words, OSA hypothesizes that the self becomes the object of one?s own consciousnes. The theory also specifies that changes betwen self-focused and environment-focused atention are dependent on the situational context, and are contingent on cues in one?s environment (Duval & Wicklund, 1972; Silvia & Duval, 2001). Self-focused atention is, thus, conceived as a fluid state. Sojourner Adjustment 21 Two intercultural transition studies have examined variables akin to self-focused atention (Harison et al., 1996; Kealey, 1989). These studies examined self-monitoring, which is a stable, personality-like trait defined as an individual?s ability to adjust his or her behavior to situational demands (Snyder, 1974). By choosing self-monitoring, Harison et al. (1996) and Kealey (1989) recapitulated a patern often sen in the intercultural transition literature, namely, that personality variables are inconsistently related to adjustment (Church, 1982). Self-monitoring was a significant predictor of sojourner adjustment in one instance (Harison et al., 1996), but not in the other (Kealey, 1989). Turning to a les static, more dynamic conceptualization of self-focused atention (e.g., self-awarenes) may, therefore, prove valuable, and ultimately lead to more eficient models of sojourner adjustment. Correlates of Sojourner Adjustment The search for correlates and predictors of sojourner adjustment has been a central feature of the intercultural transition literature. A review of the sojourner adjustment literature by Church in 1982 surveyed many approaches taken in trying to predict a person?s adaptation in a foreign country. Atempts at diferentiating betwen those who do wel overseas and those who do not have looked at personality variables, language proficiency, and previous intercultural experiences as potential significant predictors. Church concluded at the time that sojourner adjustment could not be acurately predicted, since there was litle consistency and great contradiction across studies. Intercultural transition researchers have suggested ways to addres the shortcomings in the sojourner adjustment literature. Furnham and Bochner (1986) pointed to the lack of a central theory that guides study of sojourner adjustment. They Sojourner Adjustment 22 suggested that intercultural researchers could best advance the understanding of sojourner adjustment if they were to adopt a theory-testing approach. Others have suggested that clearer definitions of sojourner adjustment would help advance the field (Benson, 1978). Recent work in sojourner adjustment has made atempts to addres both of these isues, especialy concerns with arbitrary definitions of adjustment (Ward, 1996). Much work remains to be completed, however, to gain a beter understanding of significant predictors of sojourner adjustment. Although progres has been made in more clearly defining sojourner adjustment as a bipartite construct (psychological adjustment and sociocultural adjustment), the study of adjustment often relies on exploratory techniques, such as step-wise regresion, to build predictive models (e.g., Ward & Kennedy, 1992). Incorporating more theory into the study of sojourner adjustment wil likely help researchers to develop clearly testable postulates and to refine theoretical models of sojourner adjustment. Studies that atempted to examine the relationship betwen self-eficacy beliefs and sojourner adjustment wil be reviewed. Their results generaly support the notion that self-eficacy theory (Bandura, 1997) may be useful in developing testable models of sojourner adjustment. Self-awarenes theory (Duval & Wicklund, 1972; Silvia & Duval, 2001) and studies of the relationship betwen adjustment and constructs similar to self- awarenes (e.g., self-monitoring and self-consciousnes) are then presented. These studies ilustrate the semingly inconsistent relationship betwen self-awarenes and adjustment. Finaly, a rationale wil be presented for hypothesizing that intercultural adjustment self-eficacy beliefs may moderate the relationship betwen self-awarenes Sojourner Adjustment 23 and sojourner adjustment, which may explain the inconsistent findings regarding the self- awarenes/adjustment relationship. Self-Eficacy and Sojourner Adjustment Few studies have examined the specific relationship betwen self-eficacy beliefs and sojourner adjustment. Generaly, studies of self-eficacy and sojourner adjustment suggest that further exploration of their relationship is waranted. Though lacking in sophistication, these studies generaly support the notion that self-eficacy beliefs play a role in the proces of sojourner adjustment. Harison et al. (1996) studied self-eficacy as a correlate of sojourner adjustment. Their sample consisted of 99 expatriates living in Europe who were mostly males (62%). The authors used the general and social subscales of the Self-Eficacy Scale (Sherer et al., 1982) as indicators of expatriates? self-eficacy beliefs. They correlated the subscale scores with measures of adjustment (general, work, and interaction adjustment). They found that generalized self-eficacy was significantly related to al thre types of adjustment, and that social self-eficacy was significantly related to interaction adjustment. These results are encouraging as they suggest that self- eficacy is related to sojourner adjustment outcomes. A study reviewed above for its search for changes in sojourner self-eficacy beliefs over time also studied the relation of self-eficacy to sojourner adjustment. Jerusalem and Mitag (1997) administered the German Generalized Self-Eficacy Scale (Jerusalem & Schwarzer, 1986) and the anxiety subscale from the German version of the State-Trait Personality Inventory (Hodapp, Schwarzer, Schwenkmezger, Laux, & Spielberger, 1988) at thre points in time. As described above, 124 East Germans migrating to West Germany were asked to complete measures of self-eficacy and Sojourner Adjustment 24 adjustment (anxiety subscale) one month after the fal of the Berlin wal (T1), six months later (T2), and a year after the second asesment (T3). Jerusalem and Mitag (1997) investigated self-eficacy?s hypothesized efect on adjustment by using the ratings of self-eficacy at Time 1 to create two groups of migrants, those with high self-eficacy and those with low self-eficacy. They found that those participants with high self-eficacy relative to those with low eficacy were les anxious at Times 2 and 3, suggesting that generalized self-eficacy afects adjustment. Studies in the domestic literature corroborate these findings. Holahan and Holahan (1987) studied the experiences of 52 elderly individuals (26 female, 26 male), who ranged in age from 65 to 75 years. The authors were concerned with the efect of changing roles and life transitions on adjustment. To study this they measured participants? self-eficacy beliefs twice, one year apart, using a measure of self-eficacy for dealing with daily hasles (Holahan, Holahan, & Belk, 1984), in addition to symptoms of depresion, as measured by an author constructed index. Path analysis revealed that self-eficacy beliefs at Time 1 had a direct negative relationship to depresion at Time 2 (?= -.31), in addition to indirect influence through social support at Time 2. The authors further strengthened their causal argument by controlling for depresion at Time 1. Specificaly, they calculated the partial correlation betwen self- eficacy at Time1 and depresion at Time 2, controlling for depresion at Time 1. They found a partial r of -.42 (p<.01), suggesting that self-eficacy is a viable predictor, even when the efects of on-going depresion are controlled. In esence, they argued that higher self-eficacy beliefs lead to les symptoms of depresion a year later as older adults cope with the changes in life caused by aging. Sojourner Adjustment 25 Harison et al. (1996) and Jerusalem and Mitag (1997) used measures of generalized types self-eficacy in their studies, whereas Holahan and Holahan (1987) used domain specific self-eficacy. Generalized self-eficacy beliefs are self-referent beliefs that hold across many diferent situations, and can be conceived of as stabile, personality-like traits. As sen in the Holahan and Holahan study, domain specific eficacy beliefs had a medium efect on ratings of depresive symptoms, such that a standard deviation change in self-eficacy is predictive of a .31 standard deviation decline in depresion symptoms. The self-eficacy research literature has generaly found that domain specific self-eficacy beliefs, as compared to generalized self-eficacy beliefs, are more eficient predictors of domain specific outcomes (Lent et al., 1998). In the current study, therefore, an examination of the relationship betwen intercultural adjustment self- eficacy (Brenner, 2001) (defined as a situation-specific belief about one?s ability to adjust to a foreign culture) and sojourner adjustment wil be undertaken. Self-Awarenes and Sojourner Adjustment Being aware of one?s self has implications for adjustment. It may be that those who are aware of themselves are more adept at managing their felings and adjusting their behavior to make the best of demanding situations. However, it may be just as likely that self-awarenes causes a person to be constantly aware of and preoccupied with his or her shortcomings in relation to others, or with internalized cultural expectations. Early theories of self-focused atention postulated that self-awarenes is an inherently evaluative proces (Duval & Wicklund, 1972). Moreover, acording to the original formulation of objective self-awarenes theory, evaluations that one makes about him- or herself are almost exclusively negative. This position has since been revised Sojourner Adjustment 26 (Silvia & Duval, 2001), as research has shown that self-focused atention can result in positive outcomes as wel (Davis & Franzoi, 1999). In social situations, standards of correctnes theoreticaly stipulate aceptable behavior, atitudes, and traits (Duval & Wicklund, 1972). Intercultural transitions place sojourners in social and cultural environments where learning and adapting to the country?s social and cultural norms are a necesary aspect of adjustment (Furnham & Bochner, 1986). Learning to navigate and expres the acepted culture-specific social norms likely produces heightened awarenes of behavior, atitudes, and traits. Furthermore, intercultural transitions are believed to be situations rife with bewildering environmental and cultural cues (Adler, 1975). Intercultural transitions should, therefore, theoreticaly heighten self-awarenes. The intercultural transition literature generaly has not studied the efects of self- focused atention on sojourner adjustment. The only relevant empirical study that could be identified studied self-monitoring, finding that self-monitoring and sojourner adjustment were positively related (Harison et al., 1996). As described earlier, participants in this study consisted of 99 expatriates living in Europe (62% male) with an average age of 45 years. The authors utilized the Self-Monitoring Scale (Snyder, 1974) and a measure of general, work, and interaction adjustment (Black & Stephens, 1989). They found that self-monitoring was positively related to general and interaction adjustment. Though this is an interesting and potentialy useful finding, self-monitoring is a stable, personality-like trait that has litle in common with dynamic self-awarenes as conceptualized by objective self-awarenes theory (Duval & Wicklund, 1972; Silvia & Duval, 2001). As an aside, they found that self-monitoring also was significantly related Sojourner Adjustment 27 to general and social self-eficacy, such that higher self-monitoring was related to higher levels of self-eficacy. Self-consciousnes is another construct that is somewhat similar to self- awarenes. Like self-monitoring, self-consciousnes is believed to be a personality trait. It is defined as a relatively permanent tendency to spend more or les time in the state of dynamic self-focused atention (Davis & Franzoi, 1999). Studies of self-consciousnes have suggested that it can have both beneficial and detrimental efects on psychological health (Reves, Watson, Ramsey, & Morris, 1995; P. J. Watson, Hickman, Morris, Stutz, & Whiting, 1994). For example, in times of significant life change, like an intercultural transition, self-consciousnes has been shown to have significant health promoting features (Mullen & Suls, 1982). Eighty-eight undergraduates (approximately 50% female) took part in a change over time study examining self-consciousnes and self- reported physical ilnes. At Time 1 students completed the private self-consciousnes subscale of the Self-Consciousnes Scale (Fenigstein et al., 1975), a life events scale, and a recent ilnes questionnaire. Thre weks later, Time 2, the students filed out the life events and ilnes measure. Mullen and Suls found that for students high in self- consciousnes the occurrence of stresful life events did not increase their risk of ilnes. Students low in self-consciousnes who experienced uncontrollable negative life events were more likely to report ilneses. The authors contend that participants high in self- consciousnes can more efectively manage stres and ilnes-inducing negative life events, as compared to those low in self-consciousnes. They suggest that people with high self-consciousnes are beter able to take action to more efectively cope with their Sojourner Adjustment 28 situation because they are more aware of their psychological and physical reactions to stresful life events. Frable, Plat, and Hoey (1998) found an opposite patern of results. They conducted a study focused on the experience of 18 university students with concealable culturaly defined stigmas (i.e., gay students, people with bulimia, and adult children of economicaly disadvantaged parents). These 18 university students, along with students in thre other groups (i.e., those with visible stigmas, concealable valued conditions, and visible valued conditions), were asked to wear stopwatches that periodicaly reminded them to fil out self-reports of their anxiety, depresion, self-regard, social confidence, and felings about physical appearance and abilities. Additionaly, they were asked to make an indication as to whether they were in the company of similar others (e.g., a gay student with another gay student) or disimilar others (e.g., an economicaly disadvantaged student with a group of economicaly advantaged students). Across al situations those with concealable stigmas reported significantly lower psychological functioning. They reported higher situational anxiety, more felings of depresion, lower self-estem, lower self-regard, lower social confidence, lower perceived physical atractivenes, and lower perceived physical ability in comparison with the other groups. However, when in the company of similar others those with concealable stigmas generaly felt beter about themselves. They experienced les anxiety and les depresion when with similar others, as compared to times when they were surrounded by disimilar others. This patern of results was only present for students with concealable stigmas, suggesting that something about the proces of living with a concealable stigma results in Sojourner Adjustment 29 negative outcomes. When these results are interpreted in light of Frable et al.?s (1990) earlier work suggesting that those with concealable stigmas spend more time engaged in self-awarenes than other groups, it may indicate that higher self-awarenes is related to poorer psychological outcomes. The improved psychological functioning around similar others also sems to suggest that when degres of distinctivenes are lesened, self- awarenes (and its deleterious outcomes) lesen. This may point to the dynamic shifts in self-perception that may occur as self-awarenes is manipulated. Intercultural Adjustment Self-Eficacy as Moderator of the Relationship betwen Self- Awarenes and Sojourner Adjustment In light of studies that found conflicting outcomes of self-focused atention, it is dificult to speculate about the nature of the relationship betwen self-awarenes and sojourner adjustment. In some instances being highly aware of one?s self appears to lead to beter interpersonal functioning and health promoting behaviors, but in other instances it leads to more atention paid to one?s failings (Davis & Franzoi, 1999). To acount for these diferent outcomes, it is possible that another variable moderates the relationship betwen self-awarenes and its outcomes. In the current context, a sojourner?s confidence in his or her ability to handle intercultural adjustment may change the relationship betwen self-awarenes and adjustment. Specificaly, one might find a positive relationship betwen self-awarenes and sojourner adjustment when intercultural adjustment self-eficacy beliefs are high, and find a negative (or significantly les positive) relationship betwen self-awarenes and adjustment when intercultural adjustment self-eficacy is low. A study in the domestic literature makes this sem like a plausible hypothesis. Sojourner Adjustment 30 Alden, Bieling, and Walace (1994) studied the efect of self-focused atention on socialy anxious individual?s perceptions of others? expectations of them and their level of self-eficacy to live up to those expectations. They utilized a sample of 90 female undergraduates who where divided into anxious and non-anxious groups based on their responses to the Social Avoidance and Distres Scale (D. Watson & Friend, 1969). They administered the Multidimensional Perfectionism Scale (Hewit & Flet, 1991), which includes a subscale of perceptions of others? unrealistic expectations of social behavior, and a self-eficacy measure asking them to evaluate their ability to perform social behaviors. In addition, they were asked to rate how often (from seldom to frequently) they evaluate their behavior during social interactions, as a measure of self-awarenes. In this study, anxious individuals, as compared to non-anxious individuals, believed that others expected them to demonstrate excedingly high levels of social competence, and yet they felt ineficacious in their ability to perform social behaviors competently (Alden et al., 1994). Additionaly, perceptions of others? expectations for excedingly high social behavior was significantly correlated with high self-awarenes. Taken together, these results suggest that high self-awarenes and low self-eficacy may be related to problematic perceptions of others? expectations, which appears to be a component of high social anxiety. This is not necesarily evidence that self-eficacy moderates the relationship betwen self-awarenes and adjustment, but it does give further credence to the theoretical position that self-eficacy beliefs are an important determinant of the valence (e.g., positive self-evaluation, negative self-evaluation) of the outcome of self-focused atention (Bandura, 1986). Sojourner Adjustment 31 Statement of the Problem A clear understanding of the proces and progresion of intercultural adjustment is lacking in the intercultural transition literature. Many prognostications have been made about the trajectory of sojourner adjustment, and many models of adjustment have been put forth based on those ideas (Zapf, 1991). Most models tend to rely on the asumption that sojourner adjustment follows a U-curve patern. However, few studies have found this theorized change in sojourner adjustment. Moreover, most studies have not found any evidence of any kind of systematic change. This has not stopped intercultural researchers from searching for the supposed change. Overal, it appears that cross-sectional atempts to explore changes in sojourner adjustment do not uncover consistent changes in adjustment (e.g., Searle & Ward, 1990; Ward & Kennedy, 1992; Ward & Searle, 1991). Change over time research designs, however, have had more succes in finding temporal changes in adjustment (e.g., Kealey, 1989). People change in ways other than level of adjustment as they progres through life in a foreign country. Self-eficacy beliefs are beginning to be recognized as an important variable in intercultural transitions (Harison et al., 1996; Jerusalem & Mitag, 1997; Mak & Tran, 2001), but it is not understood how they change over the course of a sojourn. They may approximate the same growth curves as adjustment, or they may show a distinct patern; further exploration is needed. Self-eficacy beliefs have only been recently included in studies examining their relationship to sojourner adjustment. This general lack of inclusion in studies of sojourner adjustment may be due to the fact that self-eficacy measures have only recently been tailored for use in this context. Though generalized scales of self-eficacy Sojourner Adjustment 32 have yielded encouraging results (e.g., Harison et al., 1996; Jerusalem & Mitag, 1997), research with context-specific measures of intercultural adjustment self-eficacy is needed (Brenner, 2001). Not only may intercultural adjustment self-eficacy and sojourner adjustment be related, dynamic changes in intercultural adjustment self- eficacy beliefs may lead to changing rates of sojourner adjustment. Levels of self-awarenes have received no exploration in the intercultural literature. Only two studies were identified that studied the similar construct of self- monitoring (Harison et al., 1996; Kealey, 1989). Though sojourners theoreticaly experience higher levels of self-awarenes as a result of their experience (Adler, 1975), these changes have not been documented. Change over time studies of self-awarenes over the course of a sojourn are, therefore, needed. Though intercultural transition theorists have long suspected that self-awarenes is an important correlate of changing sojourner adjustment (Adler, 1975), it is a construct that has never been studied in intercultural contexts. It is dificult to ascertain how it may afect sojourner adjustment, however. Studies of self-awarenes caried out in domestic setings have found that it shows contradictory paterns (Davis & Franzoi, 1999). At times it is linked to positive results, and at other times it is correlated with negative outcomes. To explain these diferent outcomes, it is necesary to look to other theories of the self for possible answers. Bandura?s (1977; 1986; 1989; 1997) self-eficacy theory is a logical place to start. An important mechanism involved in self-awarenes is the proces of making self-evaluations (Duval & Wicklund, 1972; Silvia & Duval, 2001). Bandura (1986) suggested that a critical component of the self-appraisal proces (e.g., appraisals of one?s psychological or sociocultural adjustment) is an individual?s beliefs Sojourner Adjustment 33 about his or her ability to cope with a situation, conceived as self-eficacy beliefs. Changing levels of sojourner adjustment, driven by self-awarenes, may be dependent on one?s intercultural adjustment self-eficacy beliefs. Study of this possibility is needed. Hypotheses and Research Questions Research Question 1a: Does sojourner psychological adjustment change in a systematic manner? Research Question 1b: Does sojourner sociocultural adjustment change in a systematic manner? Change over time studies of sojourner adjustment have found that sojourners? adjustment levels fluctuate as a function of time in a foreign country (Kealey, 1989). Moreover, other methodologies, like critical incidents qualitative methodology, also have found evidence of changing adjustment (Pedersen, 1995). Though often postulated to change in an initial negative curvilinear manner (e.g., U-curve hypothesis), empirical evidence does not support any particular trajectory (e.g., positive linear change, negative linear change, positive curvilinear change, negative curvilinear change). Research Question 2: Do sojourner intercultural adjustment self-eficacy beliefs change in a systematic manner? Much les speculation has been made about the changing nature of self-eficacy beliefs while overseas. Though it has been found that self-eficacy beliefs do not vary in intercultural contexts with time, those findings are based on measures of generalized self- eficacy, which are much les prone to situational influence (Jerusalem & Mitag, 1997). Context-specific self-eficacy beliefs are more closely tied to specific situational Sojourner Adjustment 34 demands, and, therefore, may fluctuate as a result of time engaged in context-specific activities. Research Question 3: Does sojourner self-awarenes change in a systematic manner? Heightened self-focused atention has long been hypothesized as an outcome of intercultural transition (Adler, 1975). No research to date has examined the supposed changes in self-awarenes. Objective Self-Awarenes theory (Duval & Wicklund, 1972; Silvia & Duval, 2001) and theories and studies of social distinctivenes (Frable et al., 1990; Frable et al, 1998; McGuire et al., 1978) suggest that cues in sojourners? environment should result in heightened self-awarenes. However, specification of the expected growth patern (e.g., positive linear, negative linear, curvilinear) over the course of a sojourn is not waranted as no theory or research guides this supposition. Hypothesis 1a: Higher intercultural adjustment self-eficacy positively affects change in sojourner psychological adjustment. Hypothesis 1b: Higher intercultural adjustment self-eficacy positively affects change in sojourner sociocultural adjustment. Higher levels of positive intercultural adjustment self-eficacy beliefs are hypothesized to result in a positive change in sojourner adjustment. Therefore, higher levels of intercultural adjustment self-eficacy ? that is, higher levels of confidence in one?s ability to manage dificult situations in a foreign country -- is hypothesized to result in increased levels of sojourner adjustment. Previous studies in intercultural transition contexts (Harison et al., 1996; Jerusalem & Mitag, 1997) and in the domestic sphere (Holahan & Holahan, 1987) support the contention that self-eficacy is positively related Sojourner Adjustment 35 to adjustment, but none have examined the hypothesis that intercultural adjustment self- eficacy demonstrates temporal precedence over adjustment for sojourners. Hypothesis 2a: Intercultural adjustment self-eficacy beliefs moderate the relationship betwen self-awarenes and sojourner psychological adjustment, such that for those with high self-eficacy beliefs, the relationship betwen self-awarenes and psychological adjustment wil become more positive as self-awarenes increases. Hypothesis 2b: Intercultural adjustment self-eficacy beliefs moderate the relationship betwen self-awarenes and sojourner sociocultural adjustment, such that for those with high self-eficacy beliefs, the relationship betwen self-awarenes and sociocultural adjustment wil become more positive as self-awarenes increases. High self-awarenes has been found to engender an inherently self-evaluative proces, yet it is not possible to specify that self-awarenes results in negative or positive self-evaluation (Silvia & Duval, 2001), suggesting that a third variable may moderate its outcome (e.g., higher or lower sojourner adjustment). The ameliorative efects of high self-eficacy beliefs are wel supported in domestic contexts (Bandura, 1997). However, no research has been conducted which examines the possibility that self-eficacy beliefs enhance adjustment during an intercultural transition. In addition to afecting adjustment directly (Hypotheses 1a and 1b), it is expected that intercultural adjustment self-eficacy beliefs wil moderate the relationship betwen self-awarenes and sojourner adjustment. Specificaly, it is expected that for sojourners with higher intercultural adjustment self-eficacy, the efects of self-awarenes on sojourner adjustment wil become more positive. Se Figure 1 for a graphical representation of the hypothesized relationship. Sojourner Adjustment 36 Low High Self-Awarenes L ow H i gh A dj us t m e nt High Self-Eficacy Low Self-Eficacy Figure 1: Hypothesized Moderating Efects of Self-Eficacy on the Relationship Betwen Self-Awarenes and Sojourner Adjustment Sojourner Adjustment 37 Chapter 3: Method Participants Participants were U.S. university students embarking on a sponsored semester or year-long study abroad program. A total of 103 pre-departure sojourners were secured as being wiling to participate in the current study. They were the number of pre-departure students who completed the first wave of data collection after they had received an invitation to complete the survey. The invitation was sent on the author?s behalf by study abroad personnel to ensure that students? privacy and anonymity would be maintained if they chose not to take part in the survey. The number of pre-departure students who received the invitation to participate is, thus, unknown. Not al of the first wave participants completed al waves of data collection; therefore, some of the participants? data were ineligible for use in the current analyses. Seventy-five participants completed the survey at Time 2, 64 participants completed Time 3, and 62 completed Time 4. Because participation by some respondents was not uniform during the entire survey time period, the number of students who responded to al four waves of data collection was 52. In acordance with typical enrollment paterns in study abroad programs, the vast majority (82.7%; n=43) of the participants were in their third-year of undergraduate study. Two second-year undergraduates (3.8%), and 7 fourth-year undergraduates (13.5%) also took part in the survey. Participants ranged in age from 19 to 24 years, with the mean age being 20.1 years (SD=.89). Many U.S. universities were represented among the participants. Smith College represented the largest contingent of participants (36%; n=19), followed by the University of California at Santa Barbara (26.9%; n=14), Sojourner Adjustment 38 Middlebury College (17.3%; n=9), the University of California at Davis (7.7%; n=4), the University of Maryland (5.8%; n=3), the University of Delaware (3.8%; n=2), and the University of Akron (1.9%; n=1). The sample was biased toward women, with 49 women (94.2%) and 3 men (5.8%) completing al waves of the survey. Most of the sample identified as being heterosexual (76.9%; n=40). Other participants identified as bisexual (11.5%; n=6), gay/lesbian (5.8%; n=3), unsure of their orientation (3.8%, n=2), and 1 participant labeled her- or himself as queer. Participants were asked to indicate the income level of their parents. The range of parental income for the current sample was from under $10,000 to over $100,000. One participant (1.9%) indicated an income level of $0-9,999, 1 participant (1.9%) indicated an income level of $10,000-19,999, 3 participants (5.8%) indicated an income level of $20,000-29,999, 3 participants (5.8%) indicated an income level of $30,000-39,999, 3 participants (5.8%) indicated an income level of $40,000-49,999, 6 participants (11.5%) indicated an income level of $50,000-59,999, 3 participants (5.8%)indicated an income level of $60,000-69,999, 7 participants (13.5%) indicated an income level of $80,000- 89,999, 4 participants (7.7%) indicated an income level of $90,000-99,999, and 19 participants (36.5%) indicated an income level over $100,000. Two participants did not provide information about income. The current sample was largely White/Caucasian (85.6%; n=45). There were 2 Asian American/Pacific Islanders (3.8%), 1 Hispanic (1.9%), and 2 individuals (3.8%) who identified as biracial who took part in the survey as wel. An additional 2 participants (3.8%) identified as being foreign nationals of Nigerian and Venezuelan/Italian descent. Respondents were asked to indicate their U.S. generation Sojourner Adjustment 39 status, where third generation status was defined as having U.S.-born parents and grandparents. Most indicated that they were third generation U.S. citizens (71.2%; n=37). Four participants (7.7%) indicated that they were second generation, 5 (9.6%) indicated that they were first generation, and 6 (11.5%) indicated that they were foreign born. Participants traveled to 12 countries. Included among the countries of destination were France (7.7% of the sample), Germany (7.7%), Hong Kong (1.9%), Ireland (5.8%), Italy (17.3%), Japan (1.9%), Mexico (1.9%), Russia (1.9%), Spain (15.4%), Switzerland (5.8%), The Netherlands (5.8%), and the United Kingdom (26.9%). Participants also provided general information about their living arangements in the country of destination. Most (59.6%; n=31) lived in residence hals or flats inhabited mostly by local nationals. Eight (15.4%) lived in residence hals or flats inhabited mostly by U.S. nationals. One-quarter of the participants (n=13) lived with families in a homestay arangement. Respondents indicated the amount of time that they had lived in a country other than the U.S. The amount of time reported by participants ranged from 0 to 216 months. Though several participants indicated extensive time spent in a foreign country prior to study abroad, the median amount of time spent in another country was 2 months. Participants gave an indication of the relative amount of time they spent engaged in activities with individuals from the host culture. On a scale from 0% to 100% of the time, the mean amount of time spent with host nationals at Time 2 was 36.7%, at Time 3 was 39.8%, and at Time 4 was 39.2%. Paired sample t-tests found that the amount of time spent with host nationals did not significantly change across the times reported. Sojourner Adjustment 40 Measures An important consideration in the current study is the efect that measurement has on levels of self-awarenes. As Osberg (1985) reported, two considerations are important when considering the use of self-awarenes measures. First, the length of the overal questionnaire in which the self-awarenes measure is embedded should be recognized as a potential confound of the level of self-reported self-awarenes. Osberg found that survey length is correlated with self-awarenes scores, such that longer questionnaires are predictive of higher levels of self-awarenes. Minimizing the overal length of the questionnaire was, therefore, adopted as a guiding principle in an efort to curtail the efects of this potential confound. Moreover, since potential participants were asked to complete the questionnaire a total of four times, utilizing a shorter questionnaire had an additional benefit of minimizing respondent fatigue. The second consideration when using self-awarenes measures is their relative position in the administration sequence. Osberg (1985) found that scores on a self- awarenes measure administered at the end of a questionnaire were significantly higher than scores obtained from the same self-awarenes measure that was presented first. Order of administration, thus, appears to be particularly important when utilizing measures of self-awarenes. In the current study, the measure of self-awarenes was fixed to the first position of the questionnaire for al participants, while al other measures were presented randomly. This was planned to minimize undesired confounding variance in self-awarenes scores. Psychological adjustment. The Mental Health Inventory (Davies, Sherbourne, Peterson, & Ware, 1988) was used to measure sojourners? level of psychological Sojourner Adjustment 41 adjustment. The MHI-5 consists of 5 items that measure psychological adjustment by tapping areas of behavioral dysfunction, felings of psychological distres, and general positive afect. Respondents were asked to indicate the frequency of dysfunction, distres, or positive afect that they experienced in the past month. Item responses were obtained using a 6-point scale from 1 (none of the time) to 6 (al of the time). A sample item reads ?How much of the time during the past month have you felt calm and peaceful?? Once al items were properly reversed coded and averaged, scores on the MHI-5 could potentialy range from 1 to 6, where higher scores indicate a relative absence of distres and presence of wel-being. Initial study of the MHI-5 found a Cronbach?s alpha of .84 (Ware, Snow, Kosinski, & Gandek, 1993), and subsequent studies have found similar levels of internal reliability (McCabe, Thomas, Brazier, & Coleman, 1996). In the current study the MHI- 5 demonstrated adequate reliability with a range of Cronbach?s alphas of .80 to .85 across the 4 waves of data collection. An 8-wek test-retest correlation of .46 with people who were diagnosed with euthymic afect suggests that scores are only moderately stable over time (Leidy, Palmer, Murray, Robb, & Revicki, 1998). This level of variability is important given the current study?s aims to uncover changing psychological adjustment. Validity for the measure is supported by high significant correlations with other measures of psychological adjustment (McCabe et al., 1996), and sensitivity to improved psychological functioning for people with depresion after 8 weks of therapy (Leidy et al., 1998). Moreover, Berwick et al. (1991) found that the MHI-5 was the most eficient measure, compared with 3 other measures of psychological adjustment, in correctly detecting the absence or presence of mental disorders as clasified by the National Sojourner Adjustment 42 Institutes of Mental Health Diagnostic Interview Schedule (DIS, Robins, Helzer, Croughan, & Ratclif, 1981). The MHI-5 outperformed the General Health Questionnaire (GHQ-30, Vieweg & Hedlund, 1983), the Somatic Symptom Inventory (SI-28, Barsky, Wyshak, & Klerman, 1986), and a longer 18-item version of the MHI (MHI-18, Ware, Johnston, & Davies-Avery, 1979). Sociocultural adjustment. The Sociocultural Adaptation Scale (SCAS, Ward & Kennedy, 1999) was used to measure sociocultural adjustment. The SCAS is variable in item length as it is intended to be adaptable for sojourners traveling to many diferent countries and under diferent circumstances. A set of 10 items has been used across al research utilizing the scale. An additional 4 items developed by the original authors that pertain to academic adjustment in a foreign country were selected for the current study. Thus, the current version of the SCAS was comprised of 14 items that ases cultural competency and behavioral adaptability of students in a foreign country. Item responses were obtained using a five-point scale ranging from 1 (no dificulty) to 5 (extreme dificulty). Scores on the SCAS were computed by calculating the mean score of the 14 items. Respondents were asked to rate the amount of dificulty experienced, for instance, in ?making friends? or ?understanding what is required of you at University.? Though item composition varies slightly from study to study, Ward and Kennedy (1999) reported a mean coeficient alpha of .85 across 16 studies. In the current study, the SCAS demonstrated adequate reliability with a range of Cronbach?s alphas of .81 to .87 across the 4 waves of data collection. Validity of the measure is supported by theory-consistent moderate positive correlations with psychological adjustment (Searle & Ward, 1990). Sojourner Adjustment 43 Intercultural adjustment self-eficacy. Participants? confidence in their ability to negotiate intercultural transitions was measured by the Intercultural Adjustment Self- Eficacy Scale (IASE; Brenner, 2001). The IASE consists of 27 items asesing respondents? confidence in their ability to manage eight aspects of an intercultural transition. Item responses were obtained using a 10-point scale ranging from ?not at al confident? (0) to ?completely confident? (9). A sample item reads ?How confident are you in your ability to manage felings of frustration with the overseas experience?? The total scale score previously demonstrated adequate internal consistency (Cronbach?s alpha = .95). In the present study, IASE demonstrated adequate reliability with a range of Cronbach?s alphas of .94 to .97 across the 4 waves of data collection. In terms of validity, the IASE also exhibited social cognitive theory-consistent relationships with interest in intercultural experiences and goal commitment regarding cultural imersion (Brenner, 2001). During measure development, exploratory principal axis factor analysis with an oblique rotation revealed an eight-factor structure (Cronbach?s alphas ranged from .83 to .91). Aculturate, Factor 1, composed of 4 items, aseses sojourner beliefs about their ability to adjust to the culture of the host country. A sample item from that factor asks participants to rate their confidence in their ability to ?get acustomed to the local culture.? Factor 2, Personal Care, composed of 4 items, asks about one?s abilities to deal efectively with food and personal care needs in a foreign country. For instance, one sample item asks about one?s ability to ?prepare meals with available local ingredients.? Logistics of the Country, Factor 3, containing 3 items, is made up of questions related to the use of basic financial, communication, and transportation systems. For this factor Sojourner Adjustment 44 participants are asked to rate their ability to ?convert and use the local currency,? among other items. Factor 4, Emergency Management, is composed of thre items about managing unexpected negative events while overseas. A sample item from this factor asks about one?s confidence to ?deal with the loss of your pasport or other important paperwork.? Interpersonal Abilities, Factor 5, composed of 3 items, taps a sojourner?s confidence about socializing with host nationals. For instance, participants are asked to rate their ability to ?initiate relationships with local people.? Factor 6, Psychological Strengths, is composed of four items about managing the emotional strains of life overseas. A sample item of that factor asks how one wil ?manage felings of social rejection.? Factor 7, Cultural Justification, containing 3 items, aseses a sojourner?s belief in his or her ability to handle situations where his or her status as an American is met with negative reaction. For instance, sojourners are asked to rate their ability to ?manage anti-American sentiment.? Lastly, Educational Adaptation, Factor 8, composed of 3 items, taps a participant?s ability to adjust to the local educational system. One item from this factor asks sojourners to rate their ability to ?interact with local instructors.? Further evidence for the construct validity of the IASE subscales is found in a qualitative study of sojourner concerns that uncovered many of the same themes of adjustment (e.g., personal care, logistics of the country, emergency management, interpersonal abilities, psychological adjustment, and educational adaptation) (Ryan & Twibel, 2000). Although it is conceivable that each subscale may offer unique contributions to the study of intercultural transition, the current study was focused on the contributions of a more general intercultural adjustment self-eficacy, rather than detailed Sojourner Adjustment 45 examinations of the subscales. Therefore, a total scale score was computed by averaging the responses to the 27 items. Self-awarenes. The private and public self-awarenes subscales from a modified Situational Self-Awarenes Scale (Govern & Marsch, 2001) was used to measure self- awarenes. Private self-awarenes is the atention one pays to internal, personal aspects of one?s self (e.g., mood). Public self-awarenes is the atention one gives to aspects of one?s self that are presented to others (e.g., behaviors). The 9-item SSAS consists of 6 items asesing respondents? level of self-awarenes (i.e., private and public) and 3 items asesing awarenes of surroundings. Only the 6 self-awarenes items were used. Item responses were obtained using a 7-point scale ranging from ?strongly disagre? (1) to ?strongly agre (7). The SSAS was modified from its original focus on the imediate present to a focus on the last month. Participants were instructed to indicate their level of agrement with each item. A sample item reads, ?I have been conscious of my inner felings.? The SSAS private and public self-awarenes subscales (3 items each) have previously demonstrated adequate internal reliability (Cronbach?s alphas = .70 and .82, respectively), and a moderate level of intercorrelation (r = .31). In the current study, the SAS demonstrated adequate reliability with a range of Cronbach?s alphas of .73 to .78 across the first 3 waves of data collection, and marginal internal consistency (alpha=.66) in the 4 th wave. The mean of the 6 items, which was computed to yield a general index of self-awarenes, can range from 1 to 7. The validity of the subscales has been supported by their ability to detect theory- consistent experimental manipulations of self-awarenes. Participants? scores on the private self-awarenes subscale were significantly higher than their scores on the public Sojourner Adjustment 46 self-awarenes subscales when they completed the measure in the presence of a smal miror and a memory recal condition. Conversely, participants? scores on the public self-awarenes subscale were significantly higher than their scores on the private self- awarenes subscale when they completed the measure in the presence of a large miror and video camera. These results are consistent with theory suggesting that these manipulations cause private or public self-awarenes, respectively (Buss, 1980). Demographics. Participants were asked to provide information about their age, year in university, parents? income, race/ethnicity, sex, and sexual orientation. They also were asked to indicate in which country they were completing their study abroad, which study abroad office had sponsored their program, and to specify whether they were completing a semester or year-long program. Participants also were asked to indicate the total number of months they had previously traveled or lived overseas. Al of the measures can be viewed in Appendix A. Procedure The questionnaire and al acompanying information (e.g., informed consent, instructions for incentive opportunities) were hosted on a University web server, and were created with web authoring software. Participants? responses to the questionnaire were sent via the Internet as an electronic mail mesage and a data file. Only the primary researcher had aces to the pasword-protected data. While it is conceivable that respondents? data could have been intercepted during transmision betwen their computer terminal and the University web server, that possibility is unlikely. Furthermore, respondents submited no personaly identifying information. Respondents only were asked to identify their responses with a predetermined identification number; Sojourner Adjustment 47 they did not provide information that could be used by others to identify them even if their data had been surreptitiously intercepted. Aces to participants was gained by contacting university-based study abroad sponsoring agencies. Agency involvement was solicited through an email mesage. Once an agency agred to participate, the agency was provided with text of an introduction to the survey and was asked to forward it to their students. They forwarded the introductory text via email 1-2 weks before their students were scheduled to embark on their study abroad program. The introductory text contained a brief overview of the study?s purposes and the incentives available to them if they were to complete the study in its entirety. Participants were notified that they would be asked to complete the questionnaire a total of four times. The complete introduction to the survey is reprinted in Appendix B. Interested potential participants first were directed to an informed consent page. Se Appendix C for the complete text of the informed consent. Once they agred to the informed consent they clicked on an ?I Agre. Take Me to the Survey? link which forwarded them to the questionnaire for wave 1 (Time 1). At the top of the questionnaire participants were asked to type an easily remembered four-digit identification number of their choosing, their first and last initial, and an email addres that could be used to contact them for reminders for subsequent waves of data collection. After typing in that information they were presented with demographic questions, the Situational Self- Awarenes Scale, the Intercultural Adjustment Self-Eficacy scale, the Mental Health Inventory-5, and the Sociocultural Adaptation Scale. The Situational Self-Awarenes Scale was always presented first to al participants. The other measures were presented Sojourner Adjustment 48 to participants in a randomized order. At the bottom of the questionnaire page participants were asked to click a button to submit their responses. After participants had been in their country of destination for four weks they were notified via email of the second wave of surveying. Se Appendix E for the text of the reminder email. If a participant had not responded within 5 days of the reminder, she or he was sent another notification (se Appendix F for text). At that time (Time 2) and al subsequent times (Time 3 and 4), participants completed al questionnaires, minus the demographic questions. For Times 2, 3, and 4, the following question was added to the questionnaire, ?Please indicate the percentage below that best describes the amount of time that you spend engaged in activities (e.g., studying, socializing, etc.) with people from the host country in which you are studying.? Participants had the option of selecting 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%. Participants received an email one month from the second phase of surveying (8 weks into their study abroad program) and one month from the third wave of surveying (12 weks into their study abroad program) notifying them that it was time to complete the questionnaire again. If a participant completed al aspects of the survey, he or she would have responded to the measures a total of four times, except for the demographic questions, which only were presented once. Those who participated in al four waves of data collection had the option to specify whether they wanted to be entered into a drawing for a gift certificate. Ten prizes of $50 gift certificates were given to 10 randomly chosen participants. Sojourner Adjustment 49 Chapter 4: Results Latent growth modeling (LGM) with factor mean estimation was the primary method of analysis. LGM was chosen because of the asumptions that it makes about the nature of growth. Not only can LGM be used to analyze data collected over time, with the addition of procedures used to estimate the mean intercept and mean slope of the function of change (e.g., factor means estimation), researchers can meaningfully describe the rate of growth found in the data (Lawrence & Hancock, 1998). Researchers utilizing this approach can test for change over time and describe the initial factor mean of the variable of interest and specify the change in mean scores that one would expect at any of the measured points in time. Compared to other statistical methods commonly used to examine growth over time, LGM does not asume that al individuals change in the same exact manner. It only asumes that individuals? growth follows the same functional form (e.g., al participants change in some linear fashion), but it does not asume that al individuals change at the same rate over time (e.g., some individuals may grow slower, faster, or even in an opposite direction) (Lawrence & Hancock, 1998). Moreover, since individuals change over time at diferent rates, the interelationship betwen individuals? scores on measured variables consequently wil change, altering the strength of correlations among the variables across time. Though this violation of the sphericity asumption is problematic for other commonly used methods for analyzing data collected at multiple points in time (e.g., repeated measures analysis of variance), LGM is not based on this asumption (Lawrence & Hancock, 1998). Sojourner Adjustment 50 Furthermore, the sample size of the current study is adequate for LGM analysis. Traditional estimates of structural equation modeling (of which LGM is a specific subtype) suggest that a minimum of 5 participants per estimated parameter are needed (Bentler & Chou, 1987). As can been sen in the Results, 9 parameters were estimated in the study?s most complex LGM models, thus, requiring a minimum of 45 participants. Though the sample is adequate as specified by the 5:1 ratio of participants to parameter guideline, concern with smaler sample sizes in LGM were addresed in the current study. A recent study on the efect of sample size on LGM was used in determining that the current sample was adequate for the purposes of investigating the research questions and hypotheses in the current study. Hamilton, Hancock, and Gagn? (2003) studied the efects that sample size has on rates of model convergence and parameter bias. Their results suggest that for samples of 50 participants the average convergence rate for LGM was 96% when data were collected over 4 time points. In other words, with the current sample size (N=52) and 4 waves of data, there was only a 4% chance that the proposed models would fail to converge and provide interpretable information. Secondly, and more importantly, they found that with sample sizes of 50 participants, parameter biases were esentialy meaningles and nonexistent. That is, with the current sample size, provided that the models converge and show good fit to the data, the parameter estimates produced by the models are trustworthy, and can be treated as highly acurate estimates of the theoretical population parameters that the model is atempting to fit. With the relatively smal sample size, it was important to make an informed decision about which fit indexes to utilize when evaluating the models proposed by the Sojourner Adjustment 51 current study. Foremost, a tandem fit index criteria was adopted for the current study. When evaluating the current models, the Comparative Fit Index (CFI) and Standardized Root Mean Squared Residual (SRMR) were used together to minimize the chances of rejecting an appropriate model and retaining an inappropriate model. A CFI equal to or greater than .96 and a SRMR equal to or les than .10 has been suggested to be highly efective in minimizing both Type I and Type II eror rates in samples of under 250 participants (Hu & Bentler, 1999). Therefore, this two-index method served as a guideline for evaluating the present models. The choice of the CFI as one part of the tandem criteria was not arbitrary, even though many diferent fit indexes combinations have been proposed (e.g, Hu & Bentler, 1998; 1999). It is a member of a group of fit indexes, incremental fit indexes, that have been shown to demonstrate relatively smal amounts of sensitivity to bias introduced by sample size (Hu & Bentler, 1998). Specificaly, the CFI has demonstrated very low levels of sensitivity to sample size, even when compared to other incremental fit indices in Monte Carlo studies (Jackson, 2001), and showed almost no mean fluctuation when compared across sample sizes of 20 to 1000 (Tanguma, 2001). In short, the use of tandem criteria and the inclusion of the CFI as one of the fit indexes minimized the undesirable efects on model fit that can occur with relatively smal sample size. Provided that the hypothesized models survived convergence failures (e.g., failure of SEM to converge on solution before reaching maximum alowed iterations) and improper solutions (e.g., negative residual variance), the criteria adopted here alow the latent growth models and their constituent parameter estimates to be viewed with a very high level of confidence for the current sample. Sojourner Adjustment 52 Examination of the proposed research questions and hypotheses was acomplished in a thre-step procedure utilizing the statistical software package EQS 5.8 (Bentler, 1985). First, univariate growth models tested for the existence of systematic growth for each variable of interest. This alowed for an examination of Research Questions 1a, 1b, 2, and 3. In LGM, growth is represented by two factors for each variable of interest. They are represented as a latent intercept factor and a latent slope factor. In the second step, the latent growth factors (two possible intercept factors, two possible slope factors) of psychological adjustment and sociocultural adjustment were regresed on the variable of intercultural adjustment self-eficacy at Time 1, provided that systematic change was found in psychological adjustment and sociocultural adjustment. This alowed for a test of Hypotheses 1a and 1b. In the third step, path modeling, utilizing the mean of repeated measures as variables, was used to examine the viability of a predictive model where intercultural adjustment self-eficacy moderates the relationship of self-awarenes to psychological and sociocultural adjustment. These analyses tested Hypothesis 2a and 2b. An intercorrelation matrix indicating the four variables of interest at each of the four waves is presented in Table 1. Means, standard deviations, and internal consistency estimates of the variables also are presented. 53 Table 1: Intercorrelations, Means, and Standard Deviations of Four Waves of Psychological Adjustment, Sociocultural Adjustment, Intercultural Adjustment Self-Eficacy, and Self-Awarenes. ?MHI-5 T1MHI-5 T2MHI-5 T3MHI-5 T4SCAS T1SCAS T2SCAS T3SCAS T4IASE T1IASE T2IASE T3IASE T4SAS T1SAS T2SAS T3SAS T4 MHI-5 T1 (.80) MHI-5 T2 0.23(.83) MHI-5 T3 0.060.60(.85) MHI-5 T4 0.100.410.74(.82) SCAS T1 0.370.090.020.03(.85) SCAS T2 0.120.540.290.250.26(.87) SCAS T3 -0.0030.440.440.440.110.74(.81) SCAS T4 -0.030.260.350.370.290.710.72(.84) IASE T1 0.08-0.050.020.030.280.240.260.46(.95) IASE T2 0.270.440.310.270.400.690.530.570.50(.94) IASE T3 0.040.260.390.410.260.600.650.730.58-0.76(.95) IASE T4 0.010.220.310.330.180.450.480.570.510.520.76(.97) SAS T1 -0.24-0.290.04-0.02-0.160.100.320.350.330.140.340.22(.79) SAS T2 0.05-0.55-0.18-0.14-0.11-0.21-0.08-0.010.13-0.060.050.030.58(.73) SAS T3 0.14-0.38-0.04-0.130.06-0.080.010.050.150.070.12-0.050.440.73(.76) SAS T4 -0.15-0.45-0.11-0.11-0.08-0.13-0.010.060.10-0.050.120.100.620.760.70(.64) MEANS 4.514.464.574.734.594.074.164.326.406.537.057.345.385.905.845.68 SD .70.76.75.65.42.60.45.441.341.221.161.22.99.71.72.62 Note: Cronbach?s alphas are indicated in parentheses on the diagonal. T1, T2, T3, and T4 represent Time 1, Time 2, Time 3 and Time 4 measurements, respectively. MHI-5 = Mental Health Indicator SCAS = Sociocultural Adjustment Scale IASE = Intercultural Adjustment Self-Eficacy SSAS = Situational Self-Awarenes Scale 54 Research Question 1a: Does sojourner psychological adjustment change in a systematic manner? A univariate growth model was tested for psychological adjustment acording to the method described by Lawrence and Hancock (1998). A two-factor growth model was chosen to consist of an intercept and slope factor for psychological adjustment (se Figure 2). Factor loadings for the four observed measures (i.e., Time 1 through 4 measures of psychological adjustment) were fixed to 1 for the intercept. This was done since it was asumed that sojourners? psychological adjustment at any given time was a function of their own initial value, or intercept. This alowed for the intercept to be ?locked in? to psychological adjustment, permiting the slope to describe change beyond the initial level of psychological adjustment. The factor loadings for the observed measures were fixed to 0 and 1, respectively, for the first two loadings for the slope. This served to create a reference for growth on which to interpret the estimated path coeficients for the third and fourth factor loadings for the slope (factor loadings for Time 3 and Time 4 in Figure 2). Factor loadings for Time 3 and Time 4 were not fixed since it was desired to estimate the growth of psychological adjustment, instead of testing a predetermined trajectory (e.g., linear, curvilinear). The addition of what EQS refers to as ?V999? to the model alowed for factor means and variances to be estimated (Lawrence & Hancock, 1998). The parameters from (a) V999 and the intercept factor and (b) V999 and the slope factor represent the factor mean for the intercept and slope, respectively. Thus, the parameter from V999 to the intercept provides an estimate of sojourners? initial psychological adjustment, and the parameter from V999 to the slope provides an estimate of the mean change in 55 psychological adjustment across time. Additionaly, estimates of the variance around the mean intercept and slope factors were calculated (disturbance, or eror, around the factors). Lastly, the parameter betwen the variances of the mean intercept and slope factors alowed the intercept and slope of psychological adjustment to covary (through the disturbances). This parameter estimated the extent to which changes in sojourners? psychological adjustment is related to their initial psychological adjustment. The proposed model of change in psychological adjustment produced an improper solution (e.g., the variance of one of the eror terms was forced to be zero). These results suggest that no model of change in psychological adjustment is tenable for the current sample. Hence, sojourner psychological adjustment does not appear to change in a systematic manner. None of the resulting fit indices and parameter estimates are reliable, nor are they interpretable. Though not a model with good fit, the described parameters of LGM with latent mean estimation for psychological adjustment can be viewed in Figure 2. The relative stability of psychological adjustment can be viewed in Figure 3, a graphical representation of the sample?s mean psychological adjustment scores at Times 1, 2, 3, and 4. Note these means in Figure 3 are the sample?s means, and are not derived from the LGM solution. 56 Figure 3: Psychological Adjustment Sample Means 3.5 3.7 3.9 4.1 4.3 4.5 4.7 4.9 5.1 5.3 5.5 1 2 3 4 Time Note: PSYC1, PSYC2, PSYC3, and PSYC4 denote psychological adjustment measured at four one-month intervals. Intercept Slope PSYC1 PSYC3 PSYC2 PSYC4 V999 E1 E2 E3 E4 D2 D1 1 1 1 1 0 1 * * * * * Figure 2: Structural Model of Latent Growth of Psychological Adjustment Improper Solution Produced. Fit indices not valid. 57 Research Question 1b: Does sojourner sociocultural adjustment change in a systematic manner? A model of implied initial positive growth (factor loadings from the slope factor to sociocultural adjustment at Time 1 were set to 0, and were set to 1 for sociocultural adjustment at Time 2) was first examined for goodnes of fit, resulting in lack of model convergence. As sen in Figure 4, an alternative model was tested whereby initial negative growth was postulated (e.g., the first two slope factor loadings were set to 1 and 0, respectively). This model of change in sociocultural adjustment demonstrated adequate fit to the data (CFI=.97; SRMR=.07). Examination of the factor loadings for the latent slope factor shows that sociocultural adjustment decreased one month into a sojourn (predetermined estimate of 0), then increased at Time 3 (factor loading of .209) and Time 4 (factor loading of .439) for this sample. These factor loadings are used when computing the latent mean scores at Times 1-4. Examination of the model suggests that the latent mean score was 4.603 just prior to departure. This was calculated by adding .540, the product of 1 (the factor loading from the slope factor to the measure of sociocultural adjustment at Time 1) and .540 (the parameter estimate from V999 to the latent slope factor), to 4.063 (the model?s baseline factor mean ? Time 2 in this model; this is the parameter estimate from V999 to the intercept factor). The latent mean decreased to 4.063 at Time 2. In this model, the latent mean for Time 2 is the same as the model?s baseline latent mean (as sen in the parameter estimate from V999 to the intercept factor). This is so because Time 2?s factor loading for the slope factor to Time 2 sociocultural adjustment was set to 0, indicating that its mean is the model?s baseline mean plus 0 adjustment. To calculate Time 3 and 58 Time 4 latent means the same procedure was used. For Time 3, the baseline mean (4.063) was adjusted upward by .113, resulting in a latent mean of 4.176. This adjustment was calculated by multiplying the factor loading from the slope factor to Time 3 (.209) by .540 (the parameter estimate from V999 to the slope factor). Similarly, for Time 4, the baseline mean (4.063) was adjusted upward by .237, resulting in a latent mean of 4.3. This adjustment was calculated by multiplying the factor loading from the slope factor to Time 4 (.439) by .540 (the parameter estimate from V999 to the slope factor). The latent mean scores are graphed in Figure 5 for easier interpretation. In sum, the results indicate that sojourner sociocultural adjustment changes in a systematic manner. It is worth noting here that, although the model (and al subsequent latent growth models) is presented such that parameter estimates that difer significantly from zero are indicated in bold in acordance with traditional methods of model representation, statistical significance from zero is not necesarily useful and meaningful information with these models. The model?s parameter estimates are proportional to the metric of the model, which was set by the predetermined factor loadings of 1 and 0 (or whatever predetermined factor loadings may be used in subsequent models) for the slope factor?s first 2 factor loadings. That is, the diference betwen 1 and 0 determines the parameter estimates in proportion to the predetermined slope factors as they are optimized by EQS. For the model depicted in Figure 4, the factor loadings from the slope factor to sociocultural adjustment at time 3 and 4 happen to be statisticaly significant from zero, but within the context of the current analyses this information is not meaningful or important to model interpretation. Likewise, the parameter estimates from (a) V999 to 59 the intercept factor (4.063) and (b) V999 to the slope factor (.540) also happen to be diferent from zero. Again, these parameter estimates are tied proportionaly to the metric set forth by the slope factor?s 2 first factor loadings, making their significance from zero interesting, but not necesary for interpretation. Intercept Slope SOC1 SOC3 SOC2 SOC4 V999 E1 E2 E3 E4 .172 .250 1* 1* 1* 1* 1* 0* .209 .439 1 1 1* 1* 1* 1* 4.063 .540 -.197 Figure 4: Structural Model of Latent Growth of Sociocultural Adjustment. Note: SOC1, SOC2, SOC3, and SOC4 denote sociocultural adjustment measured at four one-month intervals. Bold parameter estimates indicate statistical diference from zero. Parameter estimates denoted with an asterisk were predetermined parameter estimates. CFI=.97 SRMR=.07 60 Figure 5: Latent Growth of Sociocultural Adjustment Factor Means 3.5 3.7 3.9 4.1 4.3 4.5 4.7 4.9 5.1 5.3 5.5 1 2 3 4 Time Research Question 2: Do sojourner intercultural adjustment self-eficacy beliefs change in a systematic manner? A model of implied initial positive growth (factor loading from the slope factor to intercultural adjustment self-eficacy at Time 1 was set to 0, and was set to 1 for self- eficacy at Time 2) was examined for goodnes of fit, resulting in indications of good fit (CFI=.96; SRMR=.098). As sen in Figure 6, an examination of the factor loadings for the latent slope factor shows that intercultural adjustment self-eficacy increased one month (Time 2) into the sojourn (predetermined loading of 1), then increased further at Time 3 (factor loading of 5.55) and Time 4 (factor loading of 7.919) for this sample. Examination of the model suggests that the latent mean of intercultural adjustment self-eficacy was 6.41 imediately prior to departure (Time 1). The factor means of intercultural adjustment self-eficacy at Times 2, 3, and 4 were computed by adding the appropriate adjustment for growth as specified by the model. This resulted in Note: The sample SD for sociocultural adjustment is .48. The latent mean change betwen T1 and T2 is -.540, betwen T2 and T3 is .113, and betwen T3 and T4 is .124 61 latent mean scores of 6.527, 7.059, and 7.337, respectively. The latent means for intercultural adjustment self-eficacy are graphed in Figure 7. These result indicate that sojourner self-beliefs change in a systematic manner. Intercept Slope SE1 SE3 SE2 SE4 V999 E1 E2 E3 E4 .009 .957 1* 1* 1* 1* 0* 1* 5.55 7.919 1 1 1* 1* 1* 1* 6.41 .117 .002 Note: SE1, SE2, SE3, and SE4 denote intercultural adjustment self-eficacy measured at four one-month intervals. Bold parameter estimates indicate statistical diference from zero. Parameter estimates denoted with an asterisk were predetermined parameter estimates. Figure 6: Structural Model of Latent Growth of Self-Eficacy. CFI=.97 SRMR=.098 62 Figure 7: Latent Growth of Intercultural Adjustment Self-Eficacy Factor Means 5.8 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 1 2 3 4 Time Research Question 3: Does sojourner self-awarenes change in a systematic manner? As with the previous analysis, a model of implied initial positive growth was examined for goodnes of fit, resulting in indications of good fit (CFI=.997; SRMR=.06). Figure 7 shows that self-awarenes increased one month (Time 2) into the sojourn (predetermined factor loading of 1), then decreased at Time 3 (factor loading of .927) and decreased further at Time 4 (factor loading of .517) for this sample. Examination of the model suggests that the latent mean of self-awarenes was 5.382 imediately prior to departure. The factor means of self-awarenes at Times 2, 3, and 4 were computed by adding the appropriate adjustment for growth as specified by the model. This resulted in latent mean scores of 5.896, 5.858, 5.675, respectively. Figure 8 Note: The sample SD for intercultural adjustment self-eficacy is 1.24. The latent mean change betwen T1 and T2 is .117, betwen T2 and T3 is .532, and betwen T3 and T4 is .277. 63 is a graphical representation of the latent means for self-awarenes. In, sum these findings show that sojourner self-awarenes does change in a systematic manner. Intercept Slope SA1 SA3 SA2 SA4 V999 E1 E2 E3 E4 .137 .422 1* 1* 1* 1* 0* 1* .927 .571 1 1 1* 1* 1* 1* 5.382 .514 -.097 Note: SA1, SA2, SA3, and SA4 denote sojourner self-awarenes measured at four one-month intervals. Bold parameter estimates indicate statistical diference from zero. Parameter estimates denoted with an asterisk were predetermined parameter estimates. Figure 8: Structural Model of Latent Growth of Self-Awarenes. CFI=.997 SRMR=.06 64 Figure 9: Latent Growth of Self-Awarenes Factor Means 4.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9 6.1 6.3 6.5 1 2 3 4 Time Hypothesis 1a: Higher intercultural adjustment self-eficacy positively affects change in sojourner psychological adjustment. This hypothesis was not tested as no model of systematic change in psychological adjustment converged (se Research Question 1a), thus negating the need to investigate whether intercultural adjustment self-eficacy could change its growth. Hypothesis 1b: Higher intercultural adjustment self-eficacy positively affects change in sojourner sociocultural adjustment. To examine whether intercultural adjustment self-eficacy influenced the developmental trajectory of sociocultural adjustment, the sociocultural growth factors (intercept and slope) were regresed on intercultural adjustment self-eficacy at Time 1, as shown in Figure 10. The resulting model provides a test of whether participants? initial level of intercultural adjustment self-eficacy afects the growth of sociocultural Note: The sample SD for self-awarenes is .76. The latent mean change betwen T1 and T2 is .514, betwen T2 and T3 is -.037, and betwen T3 and T4 is -.183 65 adjustment over time. Most importantly, for the purposes of testing Hypotheses1b, the regresion parameters from the intercultural adjustment self-eficacy variable to the intercept and slope factors estimate the efects of initial intercultural adjustment self- eficacy beliefs on (a) the starting sociocultural adjustment values and (b) the rate of change in sociocultural adjustment over time, respectively. Unlike previous models, the statistical significance of these parameters is meaningful, as they measure whether self- eficacy?s relationship to the intercept and slope factors is significantly diferent from zero. For the current model this information is necesary for interpretation. Four starting values for the slope factor were utilized in estimating the model. These factor loadings (1, 0, .21, and .44) were drawn from earlier analysis of the change of sociocultural adjustment. The overal model demonstrated adequate fit (CFI=.97; SRMR=.08). Furthermore, the parameter estimate from intercultural adjustment self- eficacy at Time 1 to the intercept factor was significant (p<.05). This parameter estimate suggests that for each 1-point increase in pre-departure intercultural adjustment self- eficacy, one could expect a .108 increase in initial levels of sociocultural adjustment. This relationship has significance for the overal levels of sociocultural adjustment one would expect to observe over time, as is graphed in Figure 11. Those who start with higher levels of intercultural adjustment self-eficacy ? though they follow the same average trajectory ? do not experience as low a sociocultural score during a sojourn as compared to their les self-eficacious peers. The results, therefore, provide support for Hypothesis 1b. 66 Figure 11: Efects of Self-Eficacy at Time 1 on Latent Growth of Sociocultural Adjustent 1 2 3 4 Time S oc i oc ul t ur a l A dj us t m e nt High Self-Eficacy Average Self-Eficacy Low Self-Eficacy Intercept Slope SOC1 SOC3 SOC2 SOC4 Self-Eficacy at Time 1 E1 E2 E3 E4 .155 .229 1* 1* 1* 1* 1* 0* .21* .44* 1 1 1* 1* 1* 1* .108.012 -.194 Figure 10: Structural Model of Efect of Self-Eficacy at Time 1 on Latent Growth of Sociocultural Adjustment. CFI=.97 SRMR=.08 Note: Bold parameter estimates indicate statistical diference from zero. Parameter estimates denoted with an asterisk were predetermined parameter estimates. 67 Hypothesis 2a: Intercultural adjustment self-eficacy beliefs moderate the relationship betwen self-awarenes and sojourner psychological adjustment, such that for those with high self-eficacy beliefs, the relationship betwen self-awarenes and psychological adjustment wil become more positive as self-awarenes increases. Hypothesis 2a was tested by evaluating a path model where the four waves of data were averaged to produce mean repeated-measure variables of intercultural adjustment self-eficacy, self-awarenes, and psychological adjustment (se Figure 12). A product term of intercultural adjustment self-eficacy by self-awarenes was computed for inclusion in the model. In addition to averaging across the four waves of data and computing a product variable, the variables were mean centered. This was done so that collinearity betwen the product variable and its constituent components would be minimized, providing a beter chance of detecting interaction efects (Ping, 1996). The path model demonstrated a reasonable fit to the data (CFI=.96; SRMR=.07). However, the interaction term did not relate in a statisticaly significant manner to psychological adjustment suggesting that (a) intercultural adjustment self-eficacy and self-awarenes each predict psychological adjustment over and above the interaction term, but (b) intercultural adjustment self-eficacy does not moderate the relationship of self-awarenes to psychological adjustment for this sample. Hypothesis 2a is, therefore, not supported. Unlike the latent growth models where statistical significance is relative to a predetermined aspect of the slope factor, statistical significance in this, and the next path model, is meaningful and necesary for interpretation. 68 Hypothesis 2b: Intercultural adjustment self-eficacy beliefs moderate the relationship betwen self-awarenes and sojourner sociocultural adjustment, such that for those with high self-eficacy beliefs, the relationship betwen self-awarenes and sociocultural adjustment wil become more positive as self-awarenes increases. A similar procedure was used to test the moderating hypothesis for sociocultural adjustment. The path model for sociocultural adjustment (se Figure 13) demonstrated adequate fit to the data (CFI=.99; SRMR=.07). Unlike the previous test of moderation, this model found that intercultural adjustment self-eficacy moderates the relationship betwen self-awarenes and sociocultural adjustment. This is sen in the significant Figure 12: Standardized Path Analysis Model of Moderating Efects of Intercultural Adjustment Self-Eficacy on the Relationship Betwen Self- Awarenes and Psychological Adjustment. .408 .091 -.309 Self-Eficacy/Self- Awarenes Interaction Term Self-Awarenes Psychological Adjustment CFI=.96 SRMR=.07 Self-Eficacy -.083 -.172 E 69 relationship betwen the interaction term and sociocultural adjustment (p<.01). To aid in interpreting the results, a graph was produced with the predicted mean outcomes for 4 conditions. These conditions included low intercultural adjustment self-eficacy/low self- awarenes, low intercultural adjustment self-eficacy/high self-awarenes, high intercultural adjustment self-eficacy/low self-awarenes, and high intercultural adjustment self-eficacy/high self-awarenes. The low condition was defined as 1 standard deviation below the mean intercultural adjustment self-eficacy and self- awarenes scores for this sample. The high condition was defined as 1 standard deviation above the mean intercultural adjustment self-eficacy and self-awarenes scores for this sample. Based on the regresion equation that resulted from the path model and utilizing the low or high scores for self-awarenes and intercultural adjustment self-eficacy, it was possible to derive a predicted score of sociocultural adjustment for the 4 conditions. The predicted sociocultural scores for the four conditions are presented in the graph in Figure 14. Intercultural adjustment self-eficacy appears to moderate the relationship such that higher levels of self-eficacy are predictive of a positive relationship betwen self- awarenes and sociocultural adjustment, whereas relatively lower levels of intercultural adjustment self-eficacy are predictive of a negative relationship betwen self-awarenes and sociocultural adjustment. 70 Figure 14: Moderating Efects of Self-Eficacy on the Relationship Betwen Self-Awarenes and Sociocultural Adjustment 3 3.5 4 4.5 5 5.5 6 Low High Self-Awarenes S oc i oc ul t ur a l A dj us t m e nt High Self-Eficacy Low Self-Eficacy Figure 13: Standardized Mean Centered Path Analysis Model of Moderating Efects of Intercultural Adjustment Self-Eficacy on the Relationship Betwen Self-Awarenes and Sociocultural Adjustment. .780 .332 .-.046 Self-Eficacy/Self- Awarenes Interaction Term Self-Awarenes Sociocultural Adjustment CFI=.99 SRMR=.07 Self-Eficacy -.083 -.172 E 71 Chapter 5: Discussion The findings from this study are the first identified empirical evidence of systematic change in U.S. sojourner adjustment during the course of a study abroad program. The findings represent changes over time of adjustment experienced by individuals as they live and learn in a foreign culture. For sociocultural adjustment, the latent means of participants? scores approximate a growth curve similar to the long sought U-curve hypothesis (Lysgaard, 1955). The results suggest that sociocultural adjustment decreases sharply when sojourners are first imersed in a new cultural milieu, but that upward and steady improvement in sociocultural adjustment is the norm for the current participants. The decrease in sociocultural adjustment reflects sojourners? self-reported initial dificulty adapting to new cultural norms as they relate to management of everyday tasks and academic expectations. Acording to the current data, participants apparently begin to adapt to these diferences and start improving in their ability to cope with them within approximately two months of entry into a new culture. By the third month they improve further, and, if the trajectory were to hold, they would, in time, re-atain a sociocultural adjustment score similar to their pre-departure score. These scores may reflect sojourners? increasing ability to recognize and rely on the resources available to them in their host country. The social and financial resources aforded to university students in foreign countries are significant and may underlie the relatively quick recovery in sociocultural scores sen in the current sample. It may also be related to a stabilization of social identity proceses (Elemers et al., 2002) where continued exposure to a foreign 72 culture alows individuals to regain their ability to utilize contextual fedback to atain a renewed sense of their social position and ability to cope in the environment. The same finding of systematic change, however, cannot be reported of psychological adjustment. No identifiable systematic patern of growth could be found for the current sample utilizing LGM methods. It sems as though, on average, participants remained relatively constant in their self-appraisals of psychological functioning during their sojourns. The mean psychological adjustment with which sojourners entered the country of destination was approximately the same level of mean adjustment they reported thre months into their sojourn. If this acurately reflects the underlying phenomenon, psychological adjustment appears relatively unchangeable during the first thre months of a sojourn. Unchanging psychological adjustment in the sample may be related to the fact that study abroad program participants eagerly volunter for the opportunity to live and study in a foreign country. Knowing that their move to a foreign country is temporary may impact how psychologicaly involved, and therefore, psychologicaly efected they become. If this sample?s self-reported percentage of time involved with host nationals is any indication (approximately 38% of the time), the sample, as a whole, was not wel integrated into the host country. Compared to sociocultural adjustment, it may be that psychological adjustment is les sensitive to imediate change of environment, and that systematic variability in the later might only be found in studies examining a longer period of time, such as the more than 6 year time frame studied by Hsiao-Ying (1995). It could be that longer periods of residence in a foreign country are needed to change in psychological adjustment. Then 73 again, psychological adjustment may not have fluctuated on the whole in the current sample due to the efects of self-selection. Only students who fel as though they are psychologicaly equipped to embark on an emotionaly demanding transition to a foreign country may elect to complete a study abroad program. Moreover, study abroad program advisors select students who they believe in their profesional judgment to be capable of adapting wel to the intercultural and educational demands of studying at a foreign institution. These selection presures may have had some bearing on the current finding that psychological adjustment did not change for the current sample Alternatively, the current measure of psychological adjustment, the MHI-5, may not be sensitive enough to the vagaries related to transitioning to life in a foreign culture. Use of the MHI-5 in this study?s non-clinical population may have contributed to the lack of findings for changes in psychological adjustment. It might be that the measure is more suited for use in detecting the relative presence or absence of psychopathology (McCabe et al., 1996), and is not sensitive to changes in mental functioning experienced by les disturbed or clinicaly-defined ?normal? populations. In other words, it may be that the measure is les sensitive to detecting relatively smal changes in mental functioning for people in the ?normal? population, and is a beter at ascertaining gross distinctions betwen ?normal? populations and clinicaly-defined ?disturbed? populations. Other measures of psychological adjustment should be utilized with data collection over time and LGM to ascertain whether they uncover stable or fluctuating psychological adjustment. Much of the discussion about the lack of systematic change in psychological adjustment should be regarded tentatively. Though LGM was not able to find systematic 74 change in psychological adjustment for the sample, it does not necesarily mean that individuals did not change over time on this variable. In trying to understand the lack of convergence of the growth model for psychological adjustment, the data were plotted so that each participant?s four waves of data were represented on a distinct line. Examination of the data in this way showed that there was an extremely high level of variation of trajectories in the sample. It may be that the growth of psychological adjustment was so variable across individuals that LGM was unable to arive at an aceptable solution. That is, LGM may not have been able to compute a solution within the given statistical parameters so that a functional form (e.g., linear, curvilinear) could be discerned. This clearly requires further investigation and is a highly intriguing finding. Several imediate questions arise from plotting the data like this. One of the most important questions is why is psychological adjustment so highly variable in this sample, and is this similar to a non-sojourning sample. An important finding in the current study is the diference betwen the growth paterns of the two types of adjustment. Identifying changing sociocultural adjustment and stable psychological adjustment underscores the importance of conceptualizing and measuring the two phenomena separately. The current data demonstrate that they likely do not share the same growth trajectory, which further supports the notion that sojourner adjustment is best conceptualized as a multidimensional construct (Ward, 1996). The current study also contributes to understanding of other aspects of how sojourners change as a result of transitioning to a foreign culture. Identifiable paterns of growth were found for both intercultural adjustment self-eficacy and for self-awarenes. The current study was the first examination of the sensitivity of a recently created 75 measure of intercultural adjustment self-eficacy (Brenner, 2001) to register change in in- vivo self-eficacy beliefs. These data sem to suggest that the measure is capable of detecting changes experienced by sojourners. Moreover, the results show that the relatively high latent mean of 6.41for the sample just prior to departure increased steadily and significantly during participants? sojourn. The data do not appear to reflect a concern that the participants over-estimated their intercultural adjustment self-eficacy beliefs prior to departure. This would have been reflected in a downward trend in the latent means after arival in their country of destination. The data also suggest that much growth in confidence to handle the transition to a new culture occurred during the second month of their sojourn. It is notable that this increase in intercultural adjustment self- eficacy occurred at the same time that sociocultural adjustment began to improve for sojourners, and at the same time that their level of self-awarenes begins its downward trend. Self-awarenes in the current sample exhibited a positive curvilinear growth patern, much like an inverted ?U?. Growth of sojourners? levels of self-awarenes was strongest betwen pre-departure and one month into the sojourn. Measurements at the second and third months of life in a foreign culture appear to show that levels of self- awarenes began to taper off. However, the latent means of self-awarenes remain at a comparatively high level at the end of thre months. These results are the first empirical evidence of Adler?s (1975) contention that self-awarenes is raised as a result of transitioning to life in a foreign culture. He postulated that self-awarenes plays a pivotal role in the changes in adjustment experienced by sojourners. These results appear to suggest that sojourners do, in fact, 76 change in the amount of time in which they spend in self-focused atention states. Unlike earlier atempts to study the phenomenon (Harison et al., 1996; Kealey, 1989), which utilized personality-like traits of self-consciousnes and self-monitoring as indicators of self-awarenes, situational self-awarenes sems to be sensitive enough to detect changes in individuals? levels of self-focused atention as a result of an intercultural transition. Prediction of Sojourner Adjustment The data collection methods used in the current study produced intriguing findings related to how sojourners change across time. The results are informative when trying to construct estimates of how sojourners react and adapt to transitions experienced in foreign cultures. The current study also examined the predictive ability of intercultural adjustment self-eficacy and self-awarenes. Specificaly, intercultural adjustment self- eficacy?s ability to predict subsequent levels of adjustment was examined, as was the contention that it moderates the relationship betwen sojourner self-awarenes and adjustment. The results provide encouraging evidence of the role played by intercultural adjustment self-eficacy in building predictive models of sojourner adjustment. Pre- departure intercultural adjustment self-eficacy was examined for the efect that it has on the growth of sociocultural adjustment. As can be sen in Figure 11, intercultural adjustment self-eficacy afects the trajectory of sociocultural adjustment such that higher levels of self-eficacy are predictive of higher levels of adjustment. Those who begin a sojourn with higher levels of confidence in their abilities to navigate an intercultural transition enter with relatively higher levels of sociocultural adjustment and maintain those comparatively higher levels of adjustment across the sojourn. Higher intercultural 77 adjustment self-eficacy beliefs do not appear to change the functional form of the adjustment patern (e.g., those with low, average, and high self-eficacy beliefs al experience the same approximate ?U? curve of adjustment), but they appear to be predictive of the relative levels of adjustment experienced by sojourners. These results are the first empirical evidence of the predictive ability of context- specific self-eficacy beliefs in foreteling the adjustment of sojourners. Unlike cross- sectional research in this area (e.g., Harison et al., 1996), these results provide estimates of the temporal precedence of self-eficacy over sociocultural adjustment. They clearly suggest that higher intercultural adjustment self-eficacy beliefs were important in predicting the relative dificulty experienced by sojourners in adapting to the cultural and educational chalenges of their country of destination. Previous research has almost exclusively relied on measures of generalized self-eficacy (which are stable, personality- like traits) when studying sojourner adjustment (e.g., Jerusalem & Mitag, 1997). Context-specific self-eficacy beliefs are more modifiable, potentialy making them an important intervention target. Beyond this direct impact on sociocultural adjustment, intercultural adjustment self-eficacy beliefs were found to play an important moderating role in the relationship betwen self-awarenes and sociocultural adjustment. Previous research on the relationship of self-awarenes to adjustment is contradictory, with some studies suggesting that heightened self-awarenes has positive adjustment outcomes, and other studies suggesting it has negative adjustment outcomes (Silvia & Duval, 2001). It was proposed here that intercultural adjustment self-eficacy may be a moderating variable that partialy explains these diferences, and the results support this hypothesis, at least 78 for sociocultural adjustment. For the current sample, sojourners with high levels of intercultural adjustment self-eficacy benefited from higher levels of self-awarenes. For them, increasingly higher levels of self-awarenes were related to increasingly higher scores on the sociocultural adjustment measure. These results might be evidence that sojourners with much confidence in their abilities to transition to a new culture were able to benefit from the increased atention they paid to their selves in the foreign context. It may be that they were beter able to succesfully adapt their behaviors to the demands of the situation given their higher levels of confidence and self-focused atention. Self-awarenes proved to be much les helpful for sojourners who reported les confidence in their abilities to handle the intercultural transition. For these sojourners, heightened levels of self-awarenes were related to lower levels of sociocultural adjustment. It appears as though self-focused atention for these people was related to negative appraisals of their adaptation in a foreign country. In the absence of higher levels of intercultural adjustment self-eficacy, the increased self-awarenes may have evoked more atention to one?s failures and inadequacies. Regardles of the exact mechanism by which self-eficacy moderates the relationship betwen self-awarenes and sojourner adjustment, these results are the first known empirical indication that intercultural adjustment self-eficacy beliefs can modify the valence (e.g., positive or negative) of self-reported adjustment appraisals for sojourners. Future Research The methods and analysis techniques utilized in the current study were shown to be useful in detecting systematic changes in sojourner characteristics and uncovering significant relationships among the variables. Although intercultural transition research 79 remains logisticaly dificult, web-based data collection used in the current study facilitated data collection over time. Thus, despite the inherent complexities of this type of research, technical advances in web-based data collection make it possible to pursue further meaningful and important research in this area of inquiry. The intercultural transitions literature has long been criticized for its atheoretical nature (Furnham & Bochner, 1986). The approach used here suggests that it may be time to abandon the old ?shotgun? approach in favor of constructing hypotheses and theoreticaly-based predictive models. Although exploratory studies have their use, it appears as though adherence to theory ? even theory adapted from outside the intercultural transition literature per se ? would alow researchers to beter understand the complex changes in sojourner experiences, and how to go about predicting those changes with an adequate level of precision. Beyond the general benefits of continuing to utilize web-based data collection over time and LGM statistical analysis, several specific areas of inquiry could greatly extend the current findings. Most readily, it appears that extending the time period of examination for change over time would garner a more thorough opportunity to study changing sojourner experiences and outcomes. It would alow researchers to ascertain if and how the growth patern deviates from the ones found currently (e.g., does self- awarenes continue to decrease throughout a sojourn, does intercultural adjustment self- eficacy continue to increase, and at what point does it appear to level off?). Expanded time periods also may alow researchers the opportunity to detect systematic changes in psychological adjustment, which were not found in the 3-month time frame studied here. A longer time frame may be necesary to result in psychological fluctuations created by 80 changes in social identity proceses (Elemers et al., 2002) The re-entry period after a sojourn could be included as wel. Studies with longer time periods also would alow for replication and extension of the current results. Though a problematic predictor in past studies (Church, 1982), acounting for personality variables (e.g., positive or negative afect, optimism, pesimism) within the context of the studied models could result in even greater predictive eficiency. The current data are based on the experience of university students embarking on study abroad programs. Further research is needed on sojourner adjustment in the context of busines and government sectors. In those situations, many of the results found in the current study might generalize, but other factors might influence the growth of adjustment. For instance, sojourners? dedication to furthering the corporate or political mision of their firms may impact their ability to succesfully adjust to life and work in a foreign country. Studies in the busines sector have shown that alegiance to the parent company versus the local company difers among employees and is speculated to impact workplace adjustment, but has not yet been studied for its relationship to emotional adjustment (Van Oudenhoven, Van Der Ze, & Van Kooten, 2001). Other adjustment outcomes also might be important to study with these individuals such as implementation of job-related goals, increased productivity, or furthering international cooperation. Study of periods of transition and adjustment in setings outside of the international sphere may also benefit from the techniques and theory developed here. For instance, first year college students? experience could be improved if it were beter understood how they adjust to university life during their first year of life away from home. Likewise, increased understanding of the systematic ways in which newly 81 employed individuals adjust to life in a new workplace or corporate culture could facilitate the school-to-work transition. The transition of individuals who face particular chalenges of cultural transition could be studied in similar ways. The lives of religious and ethnic/racial minority group members, for example, could be conceptualized using an adjustment model similar to the one proposed here. It would be interesting to examine the lives of African-Americans from the perspective of the transition proces ? including psychological and sociocultural adjustment, self-eficacy, and self-awarenes ? when faced with working or living in a predominantly Euro-American environment. The same could be said of many groups of minorities in the U.S., including sexual orientation identity minorities or diferently-abled individuals (e.g., people with hearing impairments interacting with the hearing world). Extending this adjustment model to those populations would necesitate modifications to acount for the basic diferences in level of personal control (e.g., sojourners choose their minority status by voluntarily moving to a new country, whereas marginalized groups do not), degre of status permanence (e.g., ?sojourner? status is temporary, unlike relatively permanent minority status), and degre of knowledge of new culture (e.g., minority status members often have extensive knowledge of the majority culture, whereas sojourner are generaly unaware of cultural norms of the host country). It also is important that future research on the study abroad experience be conducted with adequate samples of men as wel as women. Though study abroad programs have feminized (Ludwig, 2000), the current study?s participants were largely female, thus limiting the generalizability of the findings to women. Similarly, further study is needed to examine whether the experiences of diferent student groups (e.g., 82 minority status race/ethnicity members, sexual orientation identity minority members, religious minority members) are invariant, in structural terms, as compared to the models and results discussed in the current study. Likewise, further research of the utility and validity of the intercultural adjustment self-eficacy measure (general score) is suggested, as is research relating to the invariance of predictive eficiency of its subscales. Study and design of theory-based interventions for study abroad students also would provide rich data. Studies that examine how pre-departure training influences adjustment would add greatly to the understanding of the cause and efect relationship betwen some of the variables studied here. Beyond that consideration, the service provided to students would likely be positive and aid in their enjoyment of their study abroad experience. Implications for Counseling Interventions The results of the current study have some obvious implications for intervening with sojourners. Pre-departure training programs that delineate the diference betwen psychological and sociocultural adjustment would help sojourners to understand that their upcoming proces of adjustment is multifaceted. Training that focuses pre-departure sojourners on sociocultural adjustment as one of their most imediate post-arival concerns appears waranted. Furthermore, citing the familiar ?U?-curve hypothesis (Lysgaard, 1955), as it applies to sociocultural adjustment, has initial support from the current findings. Educating study abroad students on the sociocultural and academic expectations that wil face them in their country of destination may ease their transition in and of itself, but it also may raise their intercultural adjustment self-eficacy for handling some of the dificulties that may come their way. 83 It also appears that sojourners should have the opportunity to anticipate, learn about, and discuss the rapid increase in self-focused atention that may acompany intercultural transitions. It may prove enlightening for sojourners to grasp that a semingly inherent aspect of a sojourn is an increase in the amount of time they spend paying atention to aspects of the self. If they were to be made aware that many sojourners experience heightened self-awarenes as a result of intercultural transition, they may be beter able to normalize and, perhaps, shorten their experience of self- focused atention. It may be useful for pre-departure sojourners to be educated that the increase in self-focused atention may be the result of extraneous contextual factors (e.g., diferent cultural norms), rather than some deficit on their part. Sojourners may find that heightened self-awarenes ultimately leads to growth and self-discovery, especialy for those who could reframe the heightened self-awarenes as an opportunity to learn about themselves and grow from an examination of their beliefs and values The ability to positively utilize heightened intercultural adjustment self-awarenes appears linked to feling more eficacious about the upcoming transition; therefore, programs that are designed to raise pre-departure intercultural adjustment self-eficacy sem like a potentialy efective means of impacting sojourners? subsequent adjustment proces. Inviting speakers to talk about their succesful, but varied, adjustment proces may help students cultivate their intercultural adjustment self-eficacy beliefs vicariously, or help to persuade them that they already posses many of the skils necesary for a positive international transition. Furthermore, framing the intercultural transition proces in the larger context of life transitions may help pre-departure students to draw upon their 84 existing competencies and previous transition succeses and skils when contemplating their likelihood of having a succesful study abroad program. It also is interesting to speculate how the current findings may bear on the instruction of counselor multicultural competence. In the context of the current study, heightened self-awarenes when combined with heightened intercultural adjustment self- eficacy appears to have a positive efect on participants? experience of life in a foreign culture. Efective multicultural training might be best brought to fruition by marshaling heightened self-focus as a cultured-being in the context of elevated multicultural counseling self-eficacy beliefs. Qualitative studies have shown that beginning counselors tend to be most afected by experiential activities that raise their level of multicultural self-awarenes (Heppner & O'Brien, 1994), which may share similarities to foreign culture imersion of study abroad. It is interesting to note that Heppner and O?Brien reported that trainees appeared most concerned with their eficacy related to applying their new knowledge and atitudes. From the point of view of the current study, aiding in the development of multicultural counseling self-eficacy may be important in the context of heightened self-awarenes engendered by experientialy-based multicultural training. If this type of self-eficacy were raised while self-awarenes is higher than normal, then trainees might be able to beter, and more thoroughly, utilize information that comes from self-examination of cultural biases and asesment of one?s level of multicultural development that influences clinical practice. On the other hand, if heightened multicultural self-awarenes is paired with relatively low levels of multicultural counseling self-eficacy, then trainees may discontinue self-examination. Worse yet, they may eschew multiculturaly competent intervention because the 85 culturaly-stimulated heightened self-awarenes (brought about by the formulation of a culturaly-informed intervention) is experienced as a noxious reminder of one?s feling of ineficacy. Limitations As with any study there are limitations to the current investigation. Studies conducted with structural equation modeling techniques, of which latent growth modeling is an extension, are limited in that they fairly rigidly specify the relationships betwen the latent variables. Although latent modeling capitalizes on shared variance, and minimizes eror asociated with measurement, the techniques may make the results appear to be a perfect reproduction of reality. It is imperative that researchers remind themselves that structural equation modeling can only support (or not support) the plausibility of a model. It does not confirm that a model is the true representation of reality. Additionaly, it is important for researchers to be cognizant that alternate structural models may exist that acount for the observed data just as wel as, if not beter than, one?s proposed model. Latent modeling techniques offer much to researchers, but they are not without limitations. The current findings can be most readily generalized to other study abroad program participants. Several aspects of the current sample should be kept in mind when ascertaining to whom and under what circumstances these findings are most relevant. The participants in the current study mostly were White, around 20 years old, female, heterosexual, from high income families, and most traveled to Western European countries. Beyond these identifiable aspects of the current sample, other aspects of the current sample limit generalizability. U.S. study abroad participants undergo 86 intercultural transition under a unique set of circumstances. Some of the more important aspects of these circumstances are that study abroad participants move temporarily to a new country because they voluntarily chose to do so. They move with relatively smal concern for their financial security, and are working under the asumption that they are, at most, on a year-long limited absence from the U.S. Moreover, most do not have to contend with the chalenge of seking social networks in a foreign country since most study abroad programs and host universities provide built-in opportunities for engaging in social activities. Other limitations of the current study include the fact that the participants self- reported their level of adjustment, which may or may not be an acurate reflection of their true adjustment proces. Finaly, though the numbers of participants in the current study were appropriate for the statistical methods employed, they represent a smal sample of the overal numbers of students embarking on a study abroad program. Even though much confounding variance was minimized by the repeated measures design and analysis, it is important not to generalize the findings beyond what current theory and the sample?s characteristics can support. 87 Appendix A Mental Health Indicator How much of the time during the past month? None of the time A litle of the time Some of the time A good bit of the time Most of the time Al of the time ? have you been a very nervous person? 1 2 3 4 5 6 ? have you felt calm and peaceful? 1 2 3 4 5 6 ? have your felt down-hearted and blue? 1 2 3 4 5 6 ? have you been a happy person? 1 2 3 4 5 6 ? have you felt so down in the dumps that nothing could cheer you up?1 2 3 4 5 6 88 Sociocultural Adjustment Scale The following questions ask about the amount of dificulty you have experienced during the last month with the following activities. There are no right or wrong answers. Please answer honestly and candidly. no dificulty slight dificulty moderate dificulty great dificulty extreme dificulty 1. Making friends 1 2 3 4 5 2. Using the transport system 1 2 3 4 5 3. Making yourself understood 1 2 3 4 5 4. Geting used to the pace of life 1 2 3 4 5 5. Going shopping 1 2 3 4 5 6. Going to social events/gatherings/functions 1 2 3 4 5 7. Worshipping in your usual way 1 2 3 4 5 8. Talking about yourself with others 1 2 3 4 5 9. Understanding jokes and humor 1 2 3 4 5 10. Dealing with someone who is unpleasant/cross/aggresive 1 2 3 4 5 11. Living away from family members overseas/independently from your parents 1 2 3 4 5 12. Understanding what is required of you at university. 1 2 3 4 5 13. Coping with academic work. 1 2 3 4 5 14. Expresing your ideas in clas 1 2 3 4 5 89 Intercultural Adjustment Self-Eficacy The following questions ask about your beliefs in your abilities to do certain things efectively while on a study abroad program in a foreign country. Rate yourself acording to your expected capabilities at the beginning of your time abroad (e.g., in the first month). There are no right or wrong answers. Please answer honestly and candidly. How confident are you in your ability to do the following things efectively? Not Confident Somewhat Completely At Al Confident Confident 1. socialize with the local people. 0123456789 2. convert and use the local currency. 0123456789 3. handle derogatory American jokes. 0123456789 4. use the local public transportation system. 0123456789 5. handle felings of disorientation. 0123456789 6. buy health and hygiene products at a local store. 0123456789 7. act acording to local customs. 0123456789 8. order food at a local restaurant. 0123456789 9. resolve a dispute with an instructor. 0123456789 10. expand your understanding of the foreign country?s political system, society, and culture. 0123456789 11. interact with local instructors. 0123456789 12. deal with the loss of your pasport or other important paperwork. 0123456789 13. use the local phone system. 0123456789 14. discuss political views diferent from your own. 0123456789 15. manage felings of frustration with the overseas experience. 0123456789 16. adapt to diferent educational practices. 0123456789 17. manage felings of social rejection. 0123456789 18. prepare meals with available local ingredients. 0123456789 19. exhibit appropriate social behavior. 0123456789 90 20. interact with local students. 0123456789 21. replace a lost or stolen train ticket. 0123456789 22. get acustomed to the local culture. 0123456789 23. help a friend find medical atention for a serious health concern. 0123456789 24. handle felings of lonelines. 0123456789 25. manage anti-American sentiment. 0123456789 26. initiate relationships with local people. 0123456789 27. shop for food at a local market or grocery. 0123456789 91 Situational Self-Awarenes Scale Please respond to each statement based on how you have felt during the last month. There are no right or wrong answers ? just be honest. Strongly disagre Moderately disagre Disagre Neutral Agre Moderately agre Strongly agre 1. I have been conscious of my inner felings 1 2 3 4 5 6 7 2. I have been concerned about the way I present myself 1 2 3 4 5 6 7 3. I have been self- conscious about the way I look 1 2 3 4 5 6 7 4. I have been reflective about my life 1 2 3 4 5 6 7 5. I have been concerned about what other people think of me 1 2 3 4 5 6 7 6. I have been aware of my innermost thoughts 1 2 3 4 5 6 7 92 Background Questionnaire I am interested in gathering a litle information about your background. Your responses to the following questions wil be held in strict confidence. Please print your answers clearly. 1. Sex: (1) Male (2) Female 2. Age: 3. Which of the following comes closest to describing your race or ethnic group? (1) Black or African-American (2) Latino/a or Hispanic (3) White or European-American (4) Asian/Pacific Islander-American (5) Native American _ (6) Middle Eastern-American (6) Other (please specify) 4. What is your sexual orientation: (1) Heterosexual (2) Homosexual (3) Bisexual (4) Unsure (6) Other 5. Please indicate your year in college: (1) firstyear (2) sophomore (3) junior (4) senior (5) graduate student 6. List the country where you wil be studying. 93 7. Circle the length of your study abroad program. Semester Year 8. List the name of your study abroad office 9. List the name of your U.S. university. 10.In what country were you born? Please list the country in which each of the following were born: Father Mother Paternal Grandfather Maternal Grandfather Paternal Grandmother Maternal Grandmother 11.Indicate your family?s approximate yearly income: $0-9,999 $10,000-19999 $20,000-29,999 $30,000-39,999 $40,000-49,999 $50,000-59,999 $60,000-69,999 $70,000-79,999 $80,000-89,999 $90,000-99,999 over $100,000 12. Indicate the amount of time (in months) that you have previously spent traveling or living in a foreign country. 94 Appendix B Introduction to Study Leter Dear University of [Insert Name Here] Study Abroad Student, Congratulations on your decision to study in a foreign country. The next few months you spend overseas wil likely have an ongoing impact on the rest of your life. I am a counseling psychology doctoral student examining the experiences of U.S. study abroad students. I hope to include your experiences with theirs, and, as such, would like to invite you to become part of a voluntary study. Your responses wil be aggregated with other participants? responses and statisticaly analyzed for my doctoral disertation. Your responses wil not be shared with your study abroad office Your study abroad office is emailing you on my behalf so that I could make you aware of an important research project designed to investigate the experience of study abroad students. Your participation in the current research project wil result in: 1. Augmented knowledge of cultural shock. 2. Increased international understanding. 3. Improved study abroad programs. As you can se, although your participation is completely voluntary, if you chose to become involved with the Culture Shock research project you wil be doing a great service. To become involved in this project you must complete the first wave of a brief survey sometime in the wek prior to your departure. You wil then receive email reminders to complete the same survey thre times again over a thre month period while you are overseas. If you complete the entire research project you wil have the option to enter yourself into a drawing for 1 of 10 $50 gift certificates to Target or Amazon.com To learn more about this research project and to complete the 10-15 minute survey please visit the following website http:/ww.otal.umd.edu/cultureshock I hope you wil find this research meaningful, particularly in light of recent world events. Best wishes on your upcoming overseas journey, Bradley Brenner, MPhil, MA Doctoral Candidate University of Maryland 95 Appendix C Informed Consent Please carefully read the following terms of consent: I am at least 18 years of age. This study includes questionnaires about perceptions of adjustment, confidence in my ability to handle diferent situations while overseas, and levels of self-awarenes. I understand that I wil be asked to complete the questionnaires a total of four times, once prior to departing for my study abroad program and thre more times at one-month intervals after my arival in my country of destination. My participation in this research is voluntary. I am fre to stop participating at any point without penalty. No information wil be submited to the University of Maryland server until I click on the final "Submit My Responses" button at the end of each page. If at any point within one month after I have submited my responses I wish to withdraw my participation in this study, I may contact the researcher at brenner@glue.umd.edu to request deletion of my data. The data gathered in this study wil be treated confidentialy. The data wil be stored with a code number, and wil be pasword protected. In order to asure anonymity as wel as a high quality data set, the Internet Protocol (IP) addres of each computer from which data are submited, along with a time/date stamp of when the data were submited, wil be recorded. As with other Internet transmisions, it is conceivable that the researcher could use the IP addres to gain aces to my identity; however I understand that he wil not use the information for that purpose. I understand that there are no other known risks asociated with participation in this study. The benefits of this study are not intended to help me personaly, but rather to help the investigator learn more about how study abroad students function while they are overseas. I am fre to ask any questions I may have now or at a later time. I may contact the researcher, Bradley Brenner, MPhil, MA at brenner@glue.umd.edu, or his faculty advisor, Dr. Robert Lent at boblent@wam.umd.edu. By clicking "I agre" below, I atest to the fact that I have read, understand, and agre to the above statements and that I voluntarily agre to participate in this study. I Agre. Take me to the survey. 96 Appendix D Wave Notification Email Text Dear Study Abroad Student, Thank you for your involvement in the Culture Shock research project. You may recal that the research project is designed to improve international understanding and to gain beter understanding about the proces of culture shock. To met that goal your experiences as a study abroad student are vitaly needed. Please take this opportunity to complete the third wave of the survey. It should take about 5 minutes of your time. You can find the survey at http:/ww.otal.umd.edu/cultureshock/secondtime.html Your responses to this survey are critical to the research project and ensure that you wil be enrolled for the lottery for 1 of 10 $50 gift certificates. Thank you for your contribution to this important research, Brad Brenner, MPhil, MA University of Maryland 97 Appendix E Wave Notification Reminder Email Text Dear Study Abroad Student, I realize that you're busy with your study abroad program, but would like to ask you to take the latest instalment of Culture Shock survey. It wil likely take you 5 minutes to complete. Please go to http:/ww.otal.umd.edu/cultureshock/secondtime.html to complete the survey. Your responses at al points of data collection are vital to this research project and ensure that you wil be eligible to be enrolled in the drawing for 1 of 10 $50 gift certificates. Thank you, Brad Brenner, MPhil, MA University of Maryland 98 REFERENCES Adler, P. S. (1975). The transitional experience: An alternative view of culture shock. Journal of Humanistic Psychology, 15(4), 13-23. Alden, L. E., Bieling, P. J., & Walace, S. T. (1994). Perfectionism in an interpersonal context: A self-regulation analysis of dysphoria and social anxiety. Cognitive Therapy and Research, 18, 297-316. Bandura, A. (1977). Self-eficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Clifs, NJ: Prentice-Hal. Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44, 1175-1184. Bandura, A. (1997). Self-eficacy: The exercise of control. New York: W. H. Freman and Company. Barsky, A. J., Wyshak, G., & Klerman, G. L. (1986). Medical and psychiatric determinants of outpatient medical utilization. Medical Care, 24, 548-560. Benson, P. G. (1978). Measuring cross-cultural adjustment: The problem of criteria. International Journal of Intercultural Relations, 2, 21-37. Bentler, P. M. (1985). Theory and Implementation of EQS, a Structural Equations Program (Version 5.8). Los Angeles: BMDP Statistical Software. 99 Bentler, P. M., & Chou, C. (1987). Practical isues in structural modeling research. Sociological Methods and Research, 16, 78-117. Berwick, D. M., Murphy, J. M., Goldman, P. A., Ware, J. E., Barsky, A. J., & Weinstein, M. C. (1991). Performance of a five-item mental health screning test. Medical Care, 29, 169-176. Black, J. S., & Stephens, G. K. (1989). The influence of the spouse on American expatriate adjustment and intent to stay in Pacific Rim overseas asignments. Journal of Management, 15, 529-544. Brenner, B. R. (2001). Measuring sojourner experiences in a new culture: Development of an intercultural adjustment self-eficacy scale. Unpublished Masters Thesis, University of Maryland, College Park. Buss, A. H. (1980). A modification of Fenigstein, Scheier, and Buss Self-Conscious Scale. Journal of Personality Asesment, 48, 629-637. Cantril, H. (1965). The pattern of human concerns. New Brunswick, NJ: Rutgers University Pres. Church, A. T. (1982). Sojourner adjustment. Psychological Bulletin, 91, 540-572. Davies, A. R., Sherbourne, C. D., Peterson, J. R., & Ware, J., J. E. (1988). Scoring manual: Adult health status and patient satisfaction measures used in RAND's health insurance experiment. Santa Monica: The RAND Corporation. Davis, M. H., & Franzoi, S. L. (1999). Self-awarenes and self-consciousnes. In V. J. Derlega, B. A. Winstead & W. H. Jones (Eds.), Personality: Contemporary theory and research (pp. 307-338). Chicago: Nelson-Hal, Inc. 100 Duval, T. S., & Wicklund, R. A. (1972). A theory of objective self awarenes. New York: Academic Pres. Elemers, N., Spears, R., & Doosje, B. (2002). Self and social identity. Annual Review of Psychology, 53, 161-168. Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private self- consciousnes: Asesment and theory. Journal of Clinical and Consulting Psychology, 43, 522-527. Frable, D. E., Blackstone, T., & Scherbaum, C. (1990). Marginal and mindful: Deviants in social interactions. Journal of Personality and Social Psychology, 59, 140-149. Frable, D. E., Plat, L., & Hoey, S. (1998). Concealable stigmas and positive self- perceptions: Feling beter around similar others. Journal of Personality and Social Psychology, 74, 909-922. Furnham, A., & Bochner, S. (1986). Culture shock: Psychological reaction to unfamiliar environments. London: Methuen. Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood Clifs, NJ: Prentice-Hal. Govern, J. M., & Marsch, L. A. (2001). Development and validation of the situational self-awarenes scale. Consciousnes and Cognition, 10, 366-378. Hamilton, J., Hancock, G. R., & Gagn?, P. E. (2003). The efect of sample size on latent growth models. Paper presented at the National Council on Measurement in Education, Chicago, IL. 101 Harison, J. K., Chadwick, M., & Scales, M. (1996). The relationship betwen cross- cultural adjustment and the personality variables of self-eficacy and self- monitoring. International Journal of Intercultural Relations, 20, 167-188. Heppner, M. J., & O'Brien, K. M. (1994). Multicultural counselor training: Students' perceptions of helpful and hindering events. Counselor Education and Supervision, 34, 4-18. Hewit, G. L., & Flet, G. L. (1991). Perfectionism in the self and social contexts: Conceptualization, asesment and asociation with psychopathology. Journal of Personality and Social Psychology, 60, 456-470. Hodapp, V., Schwarzer, R., Schwenkmezger, P., Laux, L., & Spielberger, C. D. (1988). State-Trait Personality Inventory (STPI). Unpublished manuscript, D?sseldorf, Germany. Holahan, C. K., & Holahan, C. J. (1987). Self-eficacy, social support, and depresion in aging: A longitudinal analysis. Journal of Gerontology, 42, 65-68. Holahan, C. K., Holahan, C. J., & Belk, S. S. (1984). Adjustment in aging: The roles of life stres, hasles, and self-eficacy. Health Psychology, 3, 5-328. Hormuth, S. E. (1990). The ecology of the self: Relocation and self-concept change. Cambridge: Cambridge University Pres. Hsiao-Ying, T. (1995). Sojourner adjustment: The case of foreigners in Japan. Journal of Cross-Cultural Psychology, 26, 523-536. Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to underparameterized model mispecification. Psychological Methods, 3, 424- 453. 102 Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1-55. Jackson, D. L. (2001). Sample size and number of parameter estimates in maximum likelihood confirmatory factor analysis: A Monte Carlo investigation. Structural Equation Modeling, 8, 205-223. Janssens, M. (1995). Intercultural interaction: A burden on international managers? Journal of Organizational Behavior, 16, 155-167. Jerusalem, M., & Mitag, W. (1997). Self-eficacy in stresful life transitions. In A. Bandura (Ed.), Self-eficacy in changing societies (pp. 177-201). Cambridge: Cambridge University Pres. Jerusalem, M., & Schwarzer, R. (1986). Selbstwirksamkeit [Self-Eficacy Scale]. In R. Schwarzer (Ed.), Skalen zur befindlichkeit und pers?sonlichkeit (pp. 15-28). Berlin: Freie Universit?t Institut f?r Psychologie. Kealey, D. J. (1989). A study of cross-cultural efectivenes: Theoretical isues, practical applications. International Journal of Intercultural Relations, 13, 387-428. Lawrence, F. R., & Hancock, G. R. (1998). Asesing change over time using latent growth modeling. Measurement and Evaluation in Counseling and Development, 30, 211-224. Leidy, N. K., Palmer, C., Murray, M., Robb, J., & Revicki, D. A. (1998). Health-related quality of life asesment in euthymic and depresed patients with bipolar disorder: Psychometric performance of four self-report measures. Journal of Afective Disorders, 48, 207-214. 103 Lent, R. W., Hacket, G., & Brown, S. D. (1998). Extending social cognitive theory to counselor training: Problems and prospects. The Counseling Psychologist, 26, 295-306. Ludwig, M. (2000). The next ten years: Trends afecting study abroad participation for US students. International Educator, 9(4). Lysgaard, S. (1955). Adjustment in a foreign society: Norwegian Fulbright grantes visiting the United States. International Social Science Bulletin, 7, 45-51. Mak, A., & Tran, C. (2001). Big five personality and cultural relocation factors in Vietnamese Australian students' intercultural social self-eficacy. International Journal of Intercultural Relations, 25(2), 181-201. McCabe, C. J., Thomas, K. J., Brazier, J. E., & Coleman, P. (1996). Measuring the mental health status of a population: A comparison of the GHQ-12 and the SF-36 (MHI-5). British Journal of Psychiatry, 169, 517-521. McGuire, W. J., McGuire, C. V., Child, P., & Fujioka, T. (1978). Salience of ethnicity in the spontaneous self-concept as a function of one's ethnic distinctivenes in the social environment. Journal of Personality and Social Psychology, 36, 511-520. McNair, D., Lorr, M., & Droppleman, L. (1971). Profile of mood state. San Diego: Educational and Industrial Testing Services. Mullen, B., & Suls, J. (1982). "Know thyself": Stresful life changes and the ameliorative efect of private self-consciousnes. Journal of Experimental Social Psychology, 18, 43-55. 104 Osberg, T. M. (1985). Order efects in the administration of personality measures: The case of the self-consciousnes scale. Journal of Personality Asesment, 49, 536- 540. Pedersen, P. (1995). The five stages of culture shock: Critical incidents around the world. Westport, CT, US: Grenwood Pres/Grenwood Publishing Group Inc. Ping, R. A. (1996). Latent variable regresion: A technique for estimating interaction and quadratic coeficients. Multivariate Behavioral Research, 31, 95-120. Reves, A. L., Watson, P. J., Ramsey, A., & Morris, R. J. (1995). Private self- consciousnes factors, need for cognition and depresion. Journal of Social Behavior and Personality, 10, 431-444. Robins, L. N., Helzer, J. E., Croughan, J., & Ratclif, K. S. (1981). National Institutes of Mental Health Diagnostic Interview Schedule: Its history, characteristics, and validity. Archives of General Psychiatry, 38, 381-389. Ryan, M. E., & Twibel, R. S. (2000). Concerns, values, stres, coping, health and educational outcomes of college students who studied abroad. International Journal of Intercultural Relations, 24, 409-435. Scheier, M. F., & Carver, C. S. (1985). The Self-Consciousnes Scale: A revised version for use with general populations. Journal of Applied Social Psychology, 15, 687- 699. Searle, W., & Ward, C. (1990). The prediction of psychological and sociocultural adjustment during cross-cultural transitions. International Journal of Intercultural Relations, 14, 449-464. 105 Sherer, M., Maddux, J. E., Mercandante, B., Prentice-Dunn, S., Jacobs, B., & Rogers, R. W. (1982). The Self-eficacy Scale: Construction and validation. Psychological Reports, 51, 663-671. Silvia, P. J., & Duval, T. S. (2001). Objective Self-Awarenes Theory: Recent progres and enduring problems. Personality and Social Psychology Review, 5, 230-241. Smith, M. A., & Leigh, B. (1997). Virtual subjects: Using the internet as an alternative source of subjects and research environment. Behavior Research Methods, Instruments, and Computers, 29, 496-505. Snyder, M. (1974). Self-monitoring of expresive behavior. Journal of Personality and Social Psychology, 30, 526-537. Sussman, N. M. (2000). The dynamic nature of cultural identity throughout cultural transitions: Why home is not so swet. Personality and Social Psychology Review, 4, 355-373. Tanguma, J. (2001). Efects of sample size on the distribution of selected fit indices: A graphical approach. Educational and Psychological Measurement, 61, 759-776. Van Oudenhoven, J. P., Van Der Ze, K. I., & Van Kooten, M. (2001). Succesful adaptation strategies acording to expatriates. International Journal of Intercultural Relations, 25, 467-482. Vieweg, B. W., & Hedlund, J. L. (1983). The General Health Questionnaire (GHQ): A comprehensive review. Journal of Operational Psychiatry, 14, 74-81. Ward, C. (1996). Aculturation. In D. Landis & R. S. Bhagat (Eds.), Handbook of intercultural training (2nd ed., pp. 124-147). Thousand Oaks: Sage. 106 Ward, C., & Kennedy, A. (1992). Locus of control, mood disturbance, and social dificulty during cross-cultural transitions. International Journal of Intercultural Relations, 16, 175-194. Ward, C., & Kennedy, A. (1999). The measurement of sociocultural adaptation. International Journal of Intercultural Relations, 23(4), 659-677. Ward, C., & Searle, W. (1991). The impact of value discrepancies and cultural identity on psychological and sociocultural adjustment of sojourners. International Journal of Intercultural Relations, 15, 209-225. Ware, J. E., Johnston, S. A., & Davies-Avery, A. (1979). Conceptualization and measurement of health for adults in the Health Insurance Experiment, volume II: Mental health. Santa Monica, CA: The Rand Corporation. Ware, J. E., Snow, K. K., Kosinski, M., & Gandek, B. (1993). SF-36 Health Survey manual and the interpretation guide. Boston: New England Medical Center, The Health Institute. Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Clinical and Consulting Psychology, 33, 448-457. Watson, P. J., Hickman, S. E., Morris, R. J., Stutz, N. L., & Whiting, L. (1994). Complexity of self-consciousnes subscales: Correlations of factors with self- estem and dietary restraint. Journal of Social Behavior and Personality, 9, (761- 774). Zapf, M. K. (1991). Cross-cultural transitions and welnes: Dealing with culture shock. International Journal of Intercultural Relations, 14, 105-119.