tllHllIIUIHHWIHIHHWWWIWllHMWIHIWI Thins Date 0-7 639 IIIIIIIIIIIIIIIIIIIIIIIIIIIII WWI!”Ill!llllHlHlllllllllllt'llllillllllllllllllllll 3 1293 00881 0982 This is to certify that the thesis entitled A STUDY OF THE ROLE OF SELF-ESTEEM IN THE STATUS ATTAINMENT PROCESS presented by Ching-Miao Lan has been accepted towards fulfillment of the requirements for M. A. degree in Sociology dfla‘%©«%x Major professor Oct 16, 1991 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY 1 Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or betore date due. DATE DUE DATE DUE DATE DUE __ ll |[:JL__% : [___ ll MSU Is An Affirmative Action/Equal Opportunity Institution ammo-9.1 A STUDY OF THE ROLE OF SELF-ESTEEM IN THE STATUS ATTAINMENT PROCESS By Ching-Miao Lan A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology 1991 635- 6’05 '7 ABSTRACT A STUDY OF THE ROLE OF SELF-ESTEEM IN THE STATUS ATTAINMENT PROCESS By Ching-Miao Lan This research examined the social psychological process through which individuals come to acquire positions in the educational and occupational hierarchies in a society. Utilizing a national representative sample and improved measures of parental socioeconomic status and educational aspiration, this study proposes a social psychological elaboration of the Wisconsin model by incorporating ”self-esteem” to improve the explanatory power. The results indicate that the inclusion of self-esteem does not significantly improve the explained variance of educational and occupational attainment. If self-esteem affects status attainment at all, its effect is indirect by way of its influence on academic performance. However, self—esteem mediates a small portion of the effect of mental ability on status attainment. Hence, the analysis helps to understand the social psychological process of status attainment. In the summary, some possible directions for further work are suggested. ACKNOWLEDGEMENTS I wish to thank Dr. William Faunce, Thomas Conner, and Stan Kaplowitz for serving on my thesis committee. I would especially like to express my appreciation to Dr. Faunce for giving me his valuable advice and time. I would like to acknowledge my friends Tracy Chang, Su-hao Tu, I Wayan Dipta, and Chun-hao Li for their suggestions in methodology and assistance in data analysis. The encouragement and moral support that I received from my friends Zoe Chen, Su-hao Tu, Su-fen Weng, Chia-lin Li, Wei-ssu Chien, and Phillips Gunawan were vital to the accomplishment of this thesis. I really appreciate them. Finally, a special note of thanks goes to my parents and family for their care and support. iii TABLE OF CONTENTS Page LIST OF TABLES ...................... v LIST OF FIGURES ...................... vi Statement of the Problem ................... 1 Theoretical Argument .................... 5 The Model ....................... 5 Literature Review ...................... 9 Status Attainment Model ................ 9 Self-Esteem and Academic Achievement ......... 18 Self-Esteem and Occupational Attainment ........ 21 Social Class of Origin and Self-Esteem ......... 22 Method ............................ 25 Data Source ...................... 25 Subjects ........................ 26 Measures ....................... 26 Data Analysis ..................... 30 Result ............................ 32 The Wisconsin Model of Status Attainment ........ 32 The Elaborated Model of Status Attainment ....... 34 Discussion and Conclusion .................. 47 Discussion ....................... 47 Conclusion ....................... 49 References .......................... 52 iv LIST OF TABLES Table Page 1. Zero-order Correlations, Means, and Standard Deviations of the Nine Variables for White Men ........... 36 2. Direct and Indirect Effects of SES, MA, SE, AP and 801 on Aspirations and Attainments ............. 38 3. Standardized Coefficients for Reduced Form Equations . . . 4O 4. Standardized Path Coefficient for the Baseline and the Elaborated Model on Aspirations and Attainments . . . . 46 LIST OF FIGURES Figure Page 1. The Elaborated Social Psychological Model of Status Attainment ....................... 5 2. The Blau-Duncan Model of Status Attainment ........ 10 3. The Sewell-Haller-Portes Model of Status Attainment . . . . 12 4. The Sewell-Haller-Ohlendorf Model of Status Attainment . . l4 5. Standardized Path Coefficients for the Baseline (SHO) Model ......................... 33 6. Standardized Path Coefficients for the Elaborated Model . . . 37 vi CHAPTER I STATEMENT OF THE PROBLEM This study addresses the social psychological process through which individuals come to acquire positions in the educational and occupational hierarchies in a society. In the past two decades, researchers have undertaken numerous studies to develop a social psychological theory of status attainment. The Wisconsin social psychological model (Sewell, Haller and Portes, 1969; Sewell, Haller, and Ohlendorf, 1970; Sewell and Hauser, 1972, 1975) has been the most prominent life-cycle model of young men’s socioeconomic achievement. This model posits that the effects of parental socioeconomic status and mental ability on an individual’s educational and occupational status attainment are largely mediated by such social psychological variables as academic performance, significant others’ influence, and educational and occupational aspirations. The Wisconsin model of status attainment has successfully accounted for a substantial amount of the variance in status attainment and specified the mechanism for educational and occupational status attainment. The basic Wisconsin model has been widely criticized, replicated, and elaborated. A number of replications of the model have been done by utilizing improved measures (Jencks, Crouse, and Mueser, 1983) and large representative national samples (Alexander, Eckland and Griff'm, 1975; Otto and Haller, 1979) to show the model’s applicability to national samples. Numerous elaborations of the model have sought to establish and interpret the roles of other social psychological variables that were 1 2 not included in the initial formulations. These variables are such as self- assessment of school ability and the evaluation of self-direction and conformity (Wilson and Portes, 1975; Looker and Pineo, 1983). The elaborations have attempted to make the model more powerful, but none has improved the fit of the model. Moreover, some studies criticized that while the model traces correlations between individual workers’ traits and occupational rewards, it fails to address the uneven development of segmented labor markets (Bowles and Gintis, 1975, 1976; Beck et al., 1978). The model’s atheoretical assumption of a unitary labor market led some critics to partition samples corresponding to different labor pools and introduce vectors representing market and class segments (cf. Beck et al., 1978; Wright and Perrone, 1977). Utilizing a national representative sample and improved measures of parental socioeconomic status and educational aspiration, this study proposes a social psychological elaboration of the Wisconsin model by incorporating ”self-esteem”, an achievement-related personality variable, to improve the model’s explanatory power. The Sewell-Haller—Ohlendorf (SHO) model will be treated as a baseline model and compared with an elaborated model that includes self-esteem. Professional psychologists as early as William James emphasized that a person’s beliefs about himself will influence his decisions and actions. James viewed self-concept as arising from the experience of self as object. The forefathers of American social psychology, Cooley and Mead, described the self as a social entity formed by appraisal reflected from other persons. Following Mead and Cooley, symbolic interactionists hypothesized that a positive self-concept will lead to constructive, socially desirable behavior, and conversely that a distorted self-concept will lead to deviant, socially inadequate behaviors. Thus, within the frameworks of several theorists, 3 the role of self-esteem in initiating and guiding behavior has been equated with motivation (e. g., Combs and Snygg, 1959; Gordon, 1968; Rogers, 1951; Sears and Sherman, 1964). Consequently, variations in human behavior spanning a wide range of performance situations have been attributed to individual differences in self-esteem. As early as 1951, Rogers stated that the importance of a favorable or positive attitude toward oneself has been regarded as a major factor in all aspects of adjustment by a number of personality theorists. The psychological studies on self-esteem provide strong evidence that the very nature of one’s self-evaluation is significant to behavior and has a profound effect on the individual’s emotions, desires, values and goals. N orem-Hebeisen’s proposal that enhanced or positive self-concept should function to support organismic functioning seems almost a truism. Thus, people have a tendency to process information and to behave in a way that is most enhancing to self-esteem. Some studies suggest that persons with positive self-esteem seem to be superior in cognitive processing. This enhances their personal effectiveness and competence. Combs, Snygg, and Rogers believed that the maintenance and enhancement of the perceived self is the motive behind all behavior. It follows that there is only one basic kind of motivation, and that is the personal motivation that each human being has when engaged in activities. Given the cognitive and motivational aspects of self-esteem, we might expect persons with positive self-esteem to be in a superior position to exert control over their environment. In addition to influencing mastery over the environment, self- esteem may be positively linked to such achievement behaviors as educational and occupational attainments. That self-esteem may contribute directly to attainment is plausible. Positive self-esteem is 4 likely to make individuals more highly ambitious, that is, they will set relatively high levels of aspiration and make an effort to influence their environment. Consequently their positive views of their own abilities and competence will help them to have substantial control over their environment and to get things accomplished. A number of studies have found that self-esteem has a positive impact on educational and occupational attainment (Coopersmith, 1967; Gergen, 1971; Luck and Heiss, 1972; Purkey, 1970; Rosenberg, 1965; Rosenberg and Simmons, 1971). With respect to social antecedents of self-esteem, some studies suggested that social class and race appear to have meaningful effects on self-esteem. Most of the results show that blacks are more likely to have higher self-esteem than whites, and higher-class individuals have somewhat more positive self-images than those from less advantaged backgrounds. However, other studies of class and self-esteem show no relationship at all. By virtue of the achievement-related characteristics of self-esteem, this study proposes a status attainment model which argues that the degree to which persons in general feel positive about themselves should affect their socioeconomic status attainment. Furthermore, the manner in which the effects of self-esteem operate in the status attainment process is systematic. Self-esteem affects status attainment either directly or indirectly through intervening variables in the model, including academic performance, significant others, and educational and occupational aspirations. The purpose of this study is to examine the role of self- esteem in the socioeconomic status attainment process, to analyze social psychological mechanisms of the socioeconomic status attainment process, and to improve the explanatory power of the social psychological model of socioeconomic status attainment. CHAPTER II THEORETICAL ARGUMENT The Model The social psychological model of socioeconomic status attainment proposed by this study is displayed as a path diagram in Figure 1. Several versions of the Wisconsin model (Sewell, Haller, and Portes, 1969; Sewell, Haller, and Ohlendorf, 1970, and Sewell and Hauser, 1972, 1975) have appeared in the literature; the above one is an elaboration of the Sewell-Haller-Ohlendorf (1970) revised model. In their original paper, Sewell, Haller and Portes (1969) provided a detailed rationale for the incorporation of the social psychological intervening variables. The hypothesized relationships between the nine variables are as follows. SE\ > sor OCAP \ s W > OCAT / $EDAT/ EDAP/ MA SES: Socioeconomic Status EDAP: Educational Aspiration MA: Mental Ability OCAP: Occupational Aspiration SE: Self-Esteem EDAT: Educational Attainmmt SOI: Significant Others’ Influence Figure l. The elaborated social psychological model of status attainment. 6 The Wisconsin model posits that socioeconomic status of family of origin is positively correlated with mental ability. Mental ability is anticipated to have a substantial direct effect on academic performance as well as an indirect effect through academic performance on significant others’ influence. Parental socioeconomic status, mental ability, and academic performance are assumed to have direct effects on significant others’ influence. The higher a person’s socioeconomic status, the higher will be the socioeconomic status of those with whom he interacts and the greater likelihood that one’s significant others will encourage him to achieve high-status oriented goals. Furthermore, significant others base their expectations or encouragement for educational attainment on the youth’s demonstrated abilities, including mental ability and academic performance in high school. Academic performance and significant others’ influence are assumed to affect educational and occupational aspiration and educational attainment. A person develops and adjusts his aspirations according to both the encouragement he receives from his significant others and his own perceived ability on the basis of objective information provided by academic performance. Educational aspiration affects educational attainment while occupational aspiration affects occupational attainment. Both occupational aspiration and educational attainment are assumed to affect occupational attainment. The early versions of the Wisconsin model treated occupational attainment as an ultimate dependent variable. Later, Sewell and Hauser (1972) extended the model to include earnings as a final status attainment variable. However, the present model does not use their finding that educational and occupational attainment are central determinants of earnings. 7 The set of relationships hypothesized above has been well established in previous studies (Sewell, Haller, and Portes, 1970; Sewell and Hauser, 1972). The present model adds seven new paths to or from self-esteem. Parental socioeconomic status and mental ability are assumed to affect a person’s self-esteem. From an interactionist’s perspective, self- esteem emerges in social situations and serves as an intervening variable between social structure and individual behavior; then, it further influences the individual’s behavior. According to Purkey’s argument, although the emotional climate of the family is more important than its economic status, personal success of parents seems likely to contribute to the emotional climate. Thus, self-esteem should be positively related to social class. The correlation between mental ability and self-esteem has been consistently found in the psychological literature on self—esteem. However, the data do not provide clear-cut evidence about the nature or causes of relationship between self-esteem and mental ability. An achievement model proposed by Keith, Pottebaum, and Eberhart (1986) provided evidence that parental socioeconomic status and mental ability affect self-esteem, which in turn affects academic achievement. In keeping with social learning theory, achievement can be viewed as a behavioral as well as a motivational variable. Therefore, in the status attainment model, self-esteem is assumed to affect not only achievement behaviors such as academic performance, educational, and occupational attainment, but also motivational factors such as educational and occupational aspirations. Reitzes and Mutran (1980) found that self- concept variables are directly associated with both academic performance and aspiration, and also are major intervening variables that mediate the effect of family background on academic performance and aspiration. 8 The elaborated model argues that a person who holds a positive self- image will have higher educational and occupational aspirations. Educational attainment is an outcome of the earlier phases of the life cycle. Meanwhile, it is a valuable and long-term human capital investment and the benefits of educational attainment are distant. Because each individual is striving to behave in ways which are consistent with his self-interpretation, those who possess positive self-esteem would be more likely to invest time and make long-term efforts to finish high school and attend college. Consequently, they would acquire a higher social status than would those with negative self-esteem. In addition to having a direct effect on educational attainment, self-esteem would affect educational attainment indirectly by way of academic performance and educational aspiration. The psychological literature provides strong evidence that individuals with positive self-esteem are more competent, more highly motivated, and more influential over their peers and environment. Therefore, when educational attainment is controlled, positives are expected to search more carefully for job opportunities and to utilize the opportunities more efficiently when they arise. As a consequence of diligent preparation, job search efforts, and the full utilization of their potential, positives are expected to have higher occupational attainment than negatives. CHAPTER III LITERATURE REVIEW The Blau-Duncan and Wisconsin models are the two basic theoretical models of status attainment. While the Blau-Duncan model focuses on the social structure of status transmission, the Wisconsin model focuses on social psychological dynamics that mediate the structural impact of family socioeconomic status on youths’ socioeconomic status attainment. During the past two decades, a number of studies have been undertaken to replicate or elaborate the Blau-Duncan and Wisconsin models. Since the present study begins with a social psychological elaboration of the Wisconsin model, the following review will be restricted to studies that attempted to develop a social psychological model of status attainment. Although the Blau-Duncan model of status attainment was not intended to be social psychological, it has made a tremendous contribution to the development of the social psychological model of status attainment. Blau and Duncan’s (1967) analytic framework and research technology helped to reconceptualize conventional mobility research and provoked new work. The traditional concern with problems of mobility was the distance of movement in a social status hierarchy within and between generations. The Blau-Duncan analysis shifted from this traditional conception of mobility to the causal sequences of status transmission and attainment, intergenerationally and intragenerationally, on the basis of structural variables. For example, instead of treating parental status as a 9 10 starting point for mobility, Blau and Duncan treated it as one of the causal factors leading to status attainment. The Blau-Duncan model, as shown in Figure 2, begins with father’s educational and occupational statuses, followed by son’s education, first job, and occupational status in 1962. Basically, this model indicates that parental status has an important effect on son’s occupational attainment, but it is indirect through son’s educational attainment. By using survey data from a 1962 national sample of males aged 20 to 64, the Blau-Duncan model accounted for about 26% of the variance in son’s education, 33% of the variance in first job, and 42% of the variance in son’s occupational status in 1962 (Blau and Duncan, 1967). Blau and Duncan’s most fundamental finding was that educational attainment is an important mechanism that mediates the effect of original socioeconomic status on occupational attainment. The crucial question remaining unanswered was, by what mediating process does parental status affect educational and occupational attainment? $359 .310 ) EDAT FAED K V1,753 .516 .27 440 OCAT .281 .224 FAOC +IOB1 T318 OCAT: Occupational Attainment FAOC: Father’s Occupation 1081: First Job FAED: Father’s Education EDAT: Educational Attainment Figure 2. The Blau-Duncan model of status attainment. 11 The answer to this question suggested by Sewell, Haller, and Portes requires examination of the status attainment processes at a specific social psychological levell . Such an analysis could possibly add to the explanation of variance in the dependent variables. Building upon the influential Blau-Duncan model, they developed a social psychological model of educational and occupational status attainment. The Sewell- Haller-Portes (SHP) model, as shown in Figure 3, argued that socioeconomic status and mental ability are positively correlated, and that mental ability affects academic performance. Meanwhile, socioeconomic status and academic performance affect significant others’ influence. Sewell, Haller and Portes asserted that the higher a person’s socioeconomic status, the higher would be the socioeconomic status of those with whom he interacts and more likely they would encourage him to achieve higher education signifying a high social status. Further, a youth’s demonstrated ability (e. g. high school grades) would serve as a basis on which significant others might form expectations about his educational attainment. Thus, it is implied that mental ability will have a potential indirect effect on significant others’ influence. In turn, significant others’ influence transmits the effects of socioeconomic status, mental ability, and academic performance to educational and occupational aspirations, and then through them, to educational and occupational attainment. It can be seen that significant others’ influence is identified as a central mediating variable in the SHP model of status attainment. Educational aspiration instead of occupational aspiration affects 1 They commented that the weakness of the Blau-Duncan model is the omission of important psychological variables, such as mental ability, and social psychological intervening variables, such as reference groups, significant others, self-concept, behavior expectation, levels of educational and occupational aspiration, and experiences of success or failure in school (Sewell, Haller, and Portes, 1969). 12 educational attainment; likewise, occupational aspiration instead of educational aspiration affects occupational attainment. There is no causal link between educational and occupational aspirations. However, their unmeasured causes (e3 and e4) are known to be highly correlated. Meanwhile, educational attainment is not only an important status variable in its own right, but also a profound determinant of occupational attainment. In addition to these eight theoretically expected paths, two possible paths, though theoretically doubtful, were also postulated in the SHP model. One is from socioeconomic status to academic performance 2, and the second is from significant others’ influence to educational attainment. e5 .84 54 333 '22 > sor‘—'——> OCAP 21 cl \ ~. ~ \ $82 \ 01 .21 \ '\ 9erarmocrrfl>ocnr \ 47 u l... .44 63 >AP 62 TJS e<5 OCAT: Occupational Attainment SOI: Significant Others’ Influence EDAT: Educational Attainment AP: Academic Performance OCAP: Occupational Aspiration SES: Socioeconomic Status EDAP: Educational Aspiration MA: Mental Ability Figure 3. The Sewell-Haller-Portes model of status attainment. 2 There has been conflicting evidence regarding the direct effect of family’s SES on youth’s academic performance (I-Iavighurst and Neugarten, 1957; Wilson, 1959). Sewell, Haller, and Portes argued that in large high schools, often far removed from the youths’ home and neighborhood, it is nevertheless possible that school grades are partly determined by teachers’ desires to please prestigious parents or to reward ”middle- class”behavior; therefore the path from SES to academic performance may not be as direct as it seems (Sewell, Haller, and Portes, 1969). 13 Using longitudinal data for farm boys who graduated from Wisconsin high schools in 1957 and were followed up in 1964, Sewell, Haller and Portes (1969) found substantial coefficients for the paths hypothesized in the model, and that the effects of significant others’ influence were both theoretically and empirically ”far-reaching” in the status attainment process. They pointed out that the effect of significant others’ influence on educational attainment was unexpectedly large, and that the path from socioeconomic background to academic performance was negligible. Although paths from academic performance to level of educational and occupational aspirations were also possible, they did not analyze these paths further. The SHP model accounted for 50% of the variance in educational attainment and 34% of the variance in early occupational attainment. The most encouraging finding of the Sewell-Haller—Portes study was that their model successfully explained the complex process by which social psychological mechanisms mediate the influence of status origins and mental ability on educational and occupational attainments; the model also added a great deal to the explanation of variance in educational attainment. The SHP Wisconsin model was first used to describe data about the subsample in five different residential areas -- farm, village, small city, medium city, and large city. Using the Wisconsin sample, Sewell, Haller and Ohlendorf (1970) reestirnated the SHP model and modified it slightly. In the Sewell-Haller-Ohlendorf (SHO) model, as shown in Figure 4, the path from socioeconomic status to academic performance was deleted because of its irrelevance. The path from significant others’ influence to educational attainment, which seemed suspicious to Sewell, Haller, and l4 e5 $326 3133 '2“ > sor——‘i4-‘——>OCAP 61 .152 .778 .233 .179 EDAT:”; OCAT T654 MA '63 e2 OCAT: Occupational Attainment SOI: Significant Others’ Influence EDAT: Educational Attainment AP: Academic Performance OCAP: Occupational Aspiration SES: Socioeconomic Status EDAP: Educational Aspiration MA: Mental Ability Figure 4. The Sewell-Haller-Ohlendorf model of status attainment. Portes, was retained because of its unexpectedly large path coefficient 3. Three new direct paths were added to the SH0 model. The path from ability to significant others’ influence was inserted because the coefficient was unexpectedly too large to ignore in the urban sample. Sewell, Haller, and Ohlendorf (1970) reasoned that significant others have access to youth’s academic potential apart from their grades in school. In fact, the SH0 model showed a fundamental reassessment of the effects of academic performance, measured by rank in high school class, in the status attainment process. As a result, although confirming the critical role of significant others’ influence, Sewell, Haller, and Ohlendorf (1970) found that academic performance has direct effects on educational and occupational aspirations and indirect effects on educational and 3 A quite arbitrary criterion of standardized coefficient equal or greater than .15 was used for the retention of paths in the revised model (Sewell, Haller, and Ohlendorf, 1970). 15 occupational attainment that are not mediated by significant others’ influence. Consequently, they concluded that ” individuals are not wholly dependent upon their significant others for guidance in the status aspects of their career decisions” (p.1025). The SHO model accounted for 57% of the variance in educational attainment and 40% of the variance in occupational attainment in the male Wisconsin sample. Most of the replications and elaborations of the Wisconsin model are based on the original Wisconsin model (Sewell, Haller, and Portes, 1969; Sewell, Haller, and Ohlendorf, 1970), rather than on the disaggregated model (Sewell and Hauser, 1972, 1975). These replications have often been incomplete because data have been missing for some variables, especially intellectual ability and occupational attainment. In some instances, educational and occupational aspirations were treated as the ultimate dependent variables. For instance, the Wisconsin model has been used to study youths’ aspiration formation process (Haller and Woelfel, 1972, Williams, 1972; Kerckhoff and Huf, 1974; Spencer, 1976; Davis and Kandel, 1980). But in general, the results of the replications -- using various measures, methods of data collection, and national and international samples -- have been remarkably consistent with those of the Wisconsin model (Kerckhoff, 1974; Porter, 1976; Williams, 1972; Nachmias, 1977; Otto, and Haller, 1979; Looker and Pineo, 1983). Nevertheless, one challenging finding has appeared. Wilson and Portes (1975) found that major social psychological intervening variables, such as significant others’ influence, self-assessment of abilities, and aspirations, are not as important as they appeared in the SH0 model; mental ability and socioeconomic background are more important. Educational aspiration still shows a strong effect on educational attainment, but it fails to mediate the direct effects of ability and 16 background variables on educational attainment. Significant others’ influence does not directly affect attainment and emerges as a less important variable in the status attainment process. Socioeconomic background has a significant direct effect on educational attainment, a fact confirming the predictions of the SH0 model. Academic performance and mental ability emerge as major predictor variables. Wilson and Portes (1975) argued that variables such as socioeconomic background, mental ability, and academic performance have ”social psychological” effects as well as ”structural” effects in the status attainment process. On the one hand, when the social psychological function is dominant, the effects of socioeconomic background and mental ability will be largely mediated by social psychological intervening variables; on the other hand, to the extent that the effects are not mediated by the social psychological intervening variables, socioeconomic status and mental ability can be assumed to be exercising their ”structural” effects (Wilson and Portes, 1975). Furthermore, rather than replicating the Wisconsin model, a number of studies have attempted to construct a more powerful model by incorporating other relevant social-psychological variables into it (Porter, 1976; Otto and Alwin, 1977; Looker and Pineo, 1983). Using a five-year longitudinal study of a national sample of white males, Porter (1976) examined the roles of two mental ability domains (creativity and intelligence) and two aspects of personality (conformity and ambition) in the educational and early occupational attainment processes. Porter (1976) found that intellectual ability and significant others’ influence have important indirect effects on occupational attainment. The effect of intelligence on occupational attainment is mediated by educational attainment, while the effect of significant others’ influence on educational l7 and occupational attainment is mediated by personality attributes, especially ambition. In addition, ambition was found to have important effects on both occupational expectation and aspiration. Generally speaking, the roles of creativity and conformity were found to be less important than those of intelligence and ambition in the status attainment process (Porter, 197 6). Otto and Alwin (1977) found that participation in athletics has positive effects on educational and occupational aspirations, educational and occupational attainment. Significant others’ influence, as expected, has a strong mediating effect on aspirations and attainments. However, they found no support for the hypothesis that the perception of acceptance by peers functions as an intervening mechanism in the socioeconomic status attainment process. Looker and Pineo (1983) found that while the values of self- direction and conformity do not appear to have mediating effects on status attainment, the teenager’s self-concept of ability is an important intervening variable, particularly as a predictor of parental aspirations. Parental aspirations are affected in youths’ perceptions. Both self-concept of ability and parental aspirations affect educational and occupational aspirations and through them, educational attainment (Looker and Pineo, 1983). All of these studies presented so far employ path analytic techniques to test the theoretical model of status attainment. Using a more powerful statistical technique--LISREL—-Hauser, Tsai, and Sewell (1983) evaluated the Sewell-Haller-Portes and Sewell-Haller—Ohlendorf specifications by taking measurement error into account. Hauser, Tsai, and Sewell (1983) found that the Sewell-Haller-Portes model substantially underestimated the importance of mental ability and academic 18 performance in the status attainment process, but the Sewell-Haller- Ohlendorf model did not. Also, they found that it is necessary to posit a direct path from socioeconomic background to early occupational attainment. They found no support for the argument that youths’ reports of significant others’ encouragement are only a reflection of their own goals. While the major features of the early version of the Wisconsin model have persisted in their analysis, the Sewell-Haller—Ohlendorf model was found to be more powerful in explaining the process of educational and occupational attainment (Hauser, Tsai, and Sewell, 1983). However, none of the social psychological elaborations of the Wisconsin model of status attainment significantly improve the model in terms of the explained variance or its ability to explain the process of status attainment. Hauser, Tsai and Sewell (1983) suggested an analytic framework (the SHP model) for further model development. One alternative is to extend the model by incorporating other variables. The present study attempts to do this by including self-esteem to improve the model’s explanatory power of status attainment. A substantial number of studies have been conducted to examine the relationships between self-esteem and other important individual psychological states and social behaviors in a wide variety of settings. This section will review studies on the relation between self-esteem and achievement-related behaviors in education and occupation, as well as the social antecedents of self-esteem. Self—Esteem and Academic Achievement Academic performance is one set of achievement-related behaviors affected by self-esteem which has received most extensive concern in the 19 self-esteem literature. Numerous studies have provided persistent support for the positive relation between self-esteem and academic achievement. According to Fitts (1972), the two most common academic achievement criteria were standardized achievement tests and grade point average (GPA). Gay (1966) reported the highest correlations from studies involving the Tennessee Self-Concept Scale (T SCS), a self-administering scale comprised of 100 statements concerning self- concept and achievement tests. Positive relationships were also found between scores on the TSCS and mathematical and reading tests of the California Achievement Test battery for a group of students in a study by Williams and Cole (1968). Studies conducted by Delisle (1953), Sterens (1956), and Fink (1962) revealed a positive correlation between a favorable self-concept and academic success. This finding was supported by Coopersmith (1959) who reported that the child possessing a positive self-concept was more likely to succeed academically than the child with a negative self- concept. The Iowa Test of Basic Skills, an instrument which measures vocabulary, reading, language, work-study, and arithmetic skills for each grade level from three through nine, was utilized as a measure of academic achievement in a study by Caplin ( 1968). In support of Coopersmith’s findings, Caplin’s study revealed that children having more positive self-concepts and higher levels of aspiration had higher I academic achievement. Results of a study by Primavera, Simon and Primavera (1974) investigating the relationship between academic achievement and self-esteem with reference to possible sex differences, utilized standardized achievement scores which the researchers considered to be unbiased measures of academic achievement for both males and 20 females. Results indicated that self-esteem was related significantly to academic achievement. Self-concept was found to relate significantly to achievement when an individual’s own capabilities were considered. Brookover and Gottlieb (1964) stated that individuals develop a self-concept of ability which functionally limits one’s school achievement. Fitts (1972) reviewed this as a logical relationship indicating that an individual’s course grades are more a product of his total self than are standardized test results. In essence, a student’s school performance may have a greater significance for his total self-concept than a score on a nationally standardized test. Studies by J enson, Demetrardes, Haynes, and Parker supported the premise that self-concept correlates more highly with achievement based upon potential than with achievement on an absolute basis. Several recent studies have attempted to explore directly the causal relations between self-esteem and achievement. Bachman and O’Malley (1977) analyzed longitudinal data employing an a priori ”causal” model relating self-esteem and achievement that focused on the influence of self- esteem at high school age on the educational and occupational attainment of young men. High school self-esteem was found to exert no significant causal influence on either educational or occupational attainment; instead the positive zero-order relations between self-esteem and attainment appeared to result from shared antecedents. The views of Clark and Coleman et al. (1966) suggested that improving self-esteem will result in improved academic achievement. That is, they presupposed a positive relation between self-esteem and achievement, and place self-esteem causally prior to, or at least as reciprocally related to, achievement. Further, they suggested that self- 21 esteem mediates the relation between background variables and academic achievement. Having analyzed the results of numerous studies on the relationship between self-concept and academic achievement, Purkey (1970) concluded that the literature indicates a strong reciprocal relationship and gives us reason to assume that enhancing the self-concept is a vital influence in improving academic performance. Self-Esteem and Occupational Attainment It is often assumed that work experience has broad impacts on the personality, influencing the individual’s attitudes, values, and perceptions of self. As a result, successes and failures in work will have important psychological consequences. There is considerable evidence that persons choose their work on the basis of already-formed psychological characteristics (Rosenberg, 1957; Davis, 1965; Holland, 1976). Korman (1966, 1967) found that individuals of high self-esteem were more likely to have chosen occupations where they thought they had high abilities than individuals of low self-esteem. Another consistent finding is Katz’s (1964) findings in his research program relating to achievement of blacks that the performance of blacks is directly related to their expectancy of success in the situation. Hall (1971) hypothesized that high self-esteem and confidence engender information search and risk-taking, activities that facilitate the mastery of occupational tasks. Morrison’s (1977) cross-sectional data support this position; managers who had adapted successfully to changing role demands had higher self-esteem and risk-taking propensity than those who did not. Furthermore, Bachman and his colleagues (Bachman and O’Malley, 1977; Bachman et al., 1978) have reported a small but positive 22 relationship between self-esteem, measured during high school, and occupational status five years following graduation. Luck and Heiss (1972) reported that self-esteem is positively related to a number of job dimensions, including income, prestige, upward mobility, and personal satisfaction with occupational achievement. Wylie’s earlier review of the self-concept literature (Wylie, 1961) was consistent with the positive relationships noted above; however, Wylie’s recent (1979) review of a much larger number of studies has not revealed a clear and strong pattern of positive associations between self-esteem and socioeconomic level. The conventional wisdom regarding work and self-esteem is that the former is a major determinant of the latter. While it is commonly assumed that occupational attainment necessarily affects self-esteem, the research evidence does not support this assumption totally. Faunce (1989) argued that the more often we think about our level of occupational achievement, the greater will be the effect of work-related values on self- esteem. Then, he suggested the assumption that work necessarily affects self-esteem should be rejected and directs attention to variables that may influence this relationship. Thus we must conclude that, although the empirical evidence to date has not been entirely persuasive, there are theoretical bases for expecting a positive association between self-esteem and occupational attainment under specific conditions. Social Class of Origin and Self-Esteem The relationship between parental social class and children’s self— esteem has received less attention. Nevertheless, some studies have found that individuals of higher socioeconomic status do actually reflect the more positive image society holds of them (Rosenberg, 1965; 23 Coopersmith, 1967; Rosenberg and Pearlin, 1978; Demo and Savin- Williams, 1983). However, other studies show little or no relationship. Rosenberg (1967) studied child’s self-esteem and found that there is a weak, nonsignificant relationship between self-esteem and social class. He indicated that children in the upper middle class are more likely to have high esteem and those in the lower middle class low or medium esteem, but the effects of differing social position are not very striking. Discussing Rosenberg’s previous study, Coopersmith suggested that Rosenberg’s study result may be due to the smaller sample size. Two studies examining this relationship are compared in Rosenberg and Pearlin’s study, one, a sample of children aged 8-18; the other, a sample of adults aged 18-65. The results indicate virtually no association for younger children, a modest association for early adolescents, and a considerably stronger relationship among adults. They deduced that one reason social class has little effect on the self-esteem of children is that children are not yet exposed to the class-related occupational conditions that help to shape self-esteem; the other reason they suggested is that the SES of his interpersonal environment is more homogeneous. Demo and Savin-Williams (1983) reported that the ascribed nature of social class among young adolescents makes it a weak determinant of their self- esteem, but that with increasing age socioeconomic position becomes more meaningful and thus more consequential for self-esteem. They found that data reveal a stronger association between social class and self- esteem among eighth grades than among fifth graders. Some investigations appear to challenge the relationship between self-concept and socioeconomic status. Zirkel and Moses (1971) argued that reasons for the inconsistencies seem to be varied and diverse. Some of these may be attributed to differences in definitions, instruments, 24 research designs, age groups, regions, times, and the individuality of human beings which defies categorization. In sum, while self-esteem has been found to be an important psychological state affecting academic achievement in many previous studies, studies undertaken to investigate the relationship between self- esteem and work experience show equivocal findings. Some indicate that self-esteem is positively related to occupational attainment while others show no relationship. Where a relationship has been found, the causality seems to be reciprocal. Since these results are not conclusive, a determination of the impact of self-esteem on occupational attainment awaits further empirical support. CHAPTER IV METHOD Data Source The data utilized for this study come from the National Center for Education Statistics’ High School and Beyond (HSB) six-year longitudinal study. The base year, first, second, and third follow-up surveys were conducted in 1980, 1982, 1984, and 1986, respectively. The High School and Beyond (HSB) survey is a nationally representative, two-stage probability sample of approximately 30,000 high school sophomores and 28,000 seniors in U.S. public and private high schools 4. Only the senior cohort’s base year (1980) and third follow-up (1986) data for white male are used. The base—year survey employed a two-stage stratified sample design with schools as the first stage units and students within schools as the second stage units. There were 1,122 schools selected in relation to their estimated enrollment and other demographic features from a frame of 24,725 schools in the US. Within each school, 36 seniors were randomly selected. In schools with fewer than 36 seniors, all eligible students were selected. Questionnaires and cognitive tests were administered to each senior student in the HSB sample. Most of the questions in the Senior Cohort Base Year Questionnaire focused on students’ behavior and experiences in the secondary school setting. Also included were questions 4 For more information concerning HSB, see High School and Beyond 1980 Senior Cohort Third Follow-up (1986) Volume II: Data File User’s Manual. 25 26 about family background, post secondary educational and occupational aspirations, and personal attitudes and beliefs. There were 28,240 seniors who actually participated in the base year survey. All senior students selected for the base year sample had a known, non-zero chance of being selected for all subsequent follow-up surveys. The third follow-up sample consists of 11,995 students selected from the base year probability sample. In the Senior Cohort Third Follow-Up Questionnaire, respondents were asked to update background information and to provide information about their work experience, employment history, and educational attainment. The total number of participants in the third follow-up survey is 10,536. Subjects There are 2,319 white males who participated in both the base and third follow-up survey. Only those white males who held one full-time job at the time of the third follow-up survey in 1986 were selected as subjects for this study. A full-time job is defined as working thirty-five hours or more per week Thus, those white males who were in military service, unemployed, or part-time employed are excluded from this study. As a result, the sample consists of 1,031 white males. easures The variables used in this study were measured as follows: Socioeconomic Status (SE8): The HSB data are rich in measures involving parental characteristics and family environment. A measure of socioeconomic status (SES) has been already scaled on the HSB base year (1980) data tape. The socioeconomic status composite was constructed from five components: father’s occupation, father’s education, mother’s 27 education, family income, and material possessions in the household. Father and mother’s education include nine categories: ”less than high school;” ”high school grade;” ”less than 2 years vocational school;” ”2 or more years vocational school;” ”less than two years college;” ”two or more years college;” ”four or five years college degree;” ”Master’s degree;” ”Ph.D., M.D., or advanced degree.” Father’s occupation is coded by the HSB occupational categoriesS. Family income was reported in the Base Year Parent Questionnaire. If not available, it was taken from the Base Year Student Questionnaire. It includes seven categories which ranged from ”$6,999 or less” to ”$38,000 or more.” With regard to material possessions in the household, students were asked if they had each of the following home possessions: a place to study, a newspaper, an encyclopedia, a typewriter, a dishwasher, two or more cars, more than fifty books, their own room, and a calculator. The composite score is the simple average of the non-missing components after each component score has been standardized. This composite measures general family status, and, perhaps more importantly, it taps the quality of the home environment available to the respondent. 5 The HSB occupational categories are: (l) CLERICAL: banker teller, bookkeeper, secretary, typist, mail carrier, ticket agent; (2) CRAFTSMAN: baker, automobile mechanic, machinist, painter, plumber, telephone installer, carpenter; (3)FARMER, FARMER MANAGER; (4) HOME MAKER OR HOUSEWIFE ONLY; (5) LABOR: construction worker, car washer, sanitary worker, farm labor; (6) MANAGER, ADMINISTRATOR: sales manager, office manager, school administrator, buyer, restaurant manager, government official; (7) MILITARY; (8) OPERATIVE: meat cutter, assembler, machine operator, welder, taxicab, bus or truck driver; (9) PROFESSIONAL: accountant, artist, registered nurse, engineer, librarian, writer, social worker, actor, actress. athlete, politician, but not including school teacher, (10) PROFESSIONAL: clergyman, dentist, physician, lawyer, scientists, college teacher; (11) PROPRIETOR OR OWNER: owner of a small business, contractor, restaurant owner; (12) PROTECTIVE SERVICE: decisive, police officer or guard, sheriff, fire fighter; (13) SALES: salesperson, advertising or insurance agent, real broker; (14) SCHOOL TEACHER: elementary or secondary; (15) SERVICE: barber, beautician, practical nurse, private household worker, janitor, waiter; (16) TECHNICAL: draftsman, medical or dental technician, computer programmer; (17) NOT WORKING . 28 Mental Ability (MA): Mental ability is a weighted composite score derived from factor analysis of six HSB vocabulary and nonverbal mental tests: Vocabulary Part I, Vocabulary Part II, Mosaic Comparison Part I, Mosaic Comparison Part II, Picture-Number and Visualization in Three Dimensions. The factor score coefficients for the unrotated first principal factor were used to create an ability factor score for each subject in the present analysis. The specific measurement of ability used in this study is developed by Page and Keith (1981) 6. Self-Esteem (SE): A psychological scale of self-esteem was constructed on the HSB base year (1980) data tape. The self-esteem is measured by four items: ”I take a positive attitude toward myself;” ”I feel I am a person of worth, on an equal plane with others;” ” I am able to do things as well as most other people;” on the whole, I am satisfied with myself.” The composite scale is the average of the standardized scores of the four question items. Although the self-esteem composite is quite short, there is evidence to support the reliability and validity of the psychological scale of self-esteem derived from these items. The Cronbach’s alpha value for the self-esteem scale is .85. (Keith, Pottebaum, and Eberhart, 1986). Academic Performance (AP): Academic performance is measured by respondents’ self-reported grade point average received in high school. It was divided into the following eight categories: ”mostly A”; 6 The HSB senior cohort completed a battery of cognitive tests: Vocabulary (I and II), Reading Mathematics (1 and II),Picture Number, Mosaic Comparisons (I and II), and Visualization in Three Dimensions. Using the HSB data, Page and Keith (1981) argued that the best measure of ”general” ability is to use the ability tests without school effects. Vocabulary, Reading, and Mathematics tests are considered as the relatively school-related tests, yet, vocabulary has a special status in intelligence testing. Therefore, two vocabulary tests are incorporated with the other four mental non-Verbal tests as measure of general ability. This measurement of ability has been used in several published studies based on HSB data (Page and Keith, 1981; Roberson, Keith, and Page, 1983; Keith and Page, 1985; and Keith, Pottebaum and Eberhart, 1986). 29 ”half A and B”; ”mostly B”; ”half B and C”; ”mostly C”; ”half C and D”; ”mostly D”; and ”below D”. Those categories are coded from eight to one respectively. Significant Others’ Influence (SOI): Significant others’ influence is an unweighted composite score of four questions from the base year survey: ”what do the following people (father, mother, teacher, and friends) think you ought to do after high school?” It is coded 1 if respondent reported college attendance and 0 for other plans for each question. The score of the composite variable varies from zero to four. Mycational Aspiration (EDAP): The level of educational aspiration is measured by the question from the base year survey; ”what is the lowest schooling you will be satisfied with?” It consists of nine categories: ”less than high school graduation;” ”high school graduation only;” ” less than two years vocational school;” ”two years of vocational II II II II school;” ”less than two years college; two or greater years college; finish college;” ”Master’s degree;” and ”Ph.D. or MD.” These categories are coded from one to nine respectively. Occupational Aspiration (OCAP): The level of occupational aspiration is measured by the question from the base year survey: ”what kind of job or occupation will you expect or plan to have when your are thirty years old?” It was originally coded by the HSB occupational categories. For the purpose of this study, it was recoded by the mean Duncan’s SEI score computed for each HSB occupational category. Educational Attainment (EDAT): Educational attainment is a composite variable created in the third follow-up survey by searching the first, second, and third follow-up data to determine the highest level of education respondent attained. It consists of seven categories: ”less than high school diploma;” ”high school diploma;” ”license or certificate;” 30 ”two-three years vocational program degree;” ”four years BA. degree;” ”Master’s degree;” and ”Ph. D. or M. D.” These categories are coded from one to seven respectively. Occupational Attainment (OCAT): Occupational attainment is measured by Duncan’s SE1 score of the job or occupation the respondent had at the time of the third follow-up survey in 1986 7. Data Analysis Path analysis is the statistical technique used in the analysis. The data are analyzed according to the path model shown in Figure 1, a recursive, over-identified path model with two exogenous variables. One of the most important stages in constructing a causal model is the empirical test of the theoretically postulated relationships. In an over- identified model, testing is done in two steps: (1) testing for the significance of postulated paths and (2) testing for the non-significance of omitted paths. Every direct effect omitted from a model implicitly asserts that the total association between the causally prior variable and the dependent one is entirely mediated by other variables in the system. The procedure of the analysis is: (l) to estimate the path coefficients and coefficients of determination of the baseline model (the Wisconsin model without the inclusion of self-esteem), (2) to estimate the path coefficients and coefficients of determination of the elaborated model (the status model with the incorporation of self-esteem); (3) to calculate the direct and indirect effects of independent variables on dependent variables by path multiplication according to the specification 7 The means and standard deviation of the occupational attainment are 44.508 and 24.042 respectively. Thus, it indicated that, among the young men, there was an adequate variation for this measure. 31 of the model with self-esteem; (4) to estimate the fully specified model. Since the large sample size would result in a higher probability of rejecting the null hypothesis, the level of significance test is set to .01. CHAPTER IV RESULT The zero-order correlation coefficients, means, and standard deviations of the nine variables are presented in Table 1. Path coefficients for the baseline model (SHO) are presented in Figure 5 and those for the elaborated model in Figure 6. By multiplication of the hypothesized paths, direct, indirect, and total causal effects (i.e. the sum of direct and indirect effects) are computed and presented in Table 2. Standardized regression coefficients of reduced-form and full structural equations for the elaborated model are displayed in Table 3. Table 3 provides one basis for estimating the extent to which the effects of socioeconomic background and mental ability on socioeconomic status attainment are mediated by the intervening variables in the model. The results of the present analysis are presented as they relate to the baseline (SHO) and elaborated model. Table 4 compares the path coefficients and variance explained in the two models. The Wisconsin Model of Status Attainment The standardized regression coefficients of the hypothesized paths of the baseline Wisconsin model (SHO) of status attainment are shown in Figure 5. An initial comparison between the present results using HSB data and the result of previous studies (Sewell, Haller, and Portes, 1969; Sewell, Haller and Ohlendorf, 1970; Sewell and Hauser, 1972, 1975) indicates a basic similarity in terms of overall structure. All standardized 32 33 regression coefficients for the causal model of the baseline model (SHO) are significant and similar to those of previous studies. Major trends in the present result are comparable to the results of past studies. SES '2 > $01 399* 90cm .233* 205* 261* 288* - /C()CAT 240* EDAT 373" 4,0. .167, >EDAP%r SES: Socioeconomic Status EDAP: Educational Aspiration MA: Mental Ability OCAP: Occupational Aspiration SOI: Significant Others’ Influence EDAT: Educational Attainment Figure 5. Standardized path coefficients for the Baseline (SHO) model. First, earlier research on status attainment identified significant others’ influence as a pivotal variable. In both the SHP and SHO models and subsequent replications, significant others’ influence mediated sizable portions of the effects of parental socioeconomic status and of mental ability on educational and occupational attainment. In addition, significant others’ influence had the strongest effects on both educational and occupational aspirations and a direct effect on educational attainment (Sewell, Haller, Portes, 1969; Sewell, Haller, and Ohlendorf, 1970). The HSB data support previous findings that significant others’ influence is a crucial factor in the process of aspiration formation and educational attainment and a major mediator of the effects of parental socioeconomic status on educational and occupational aspirations, and educational and occupational attainment. Second, besides confirming the importance of significant others’ influence, the data also support Wilson and Portes’ (1975) finding of the 34 emergent importance of academic performance in the status attainment process. Wilson and Portes (1975) found that neither educational nor occupational aspirations are capable of largely mediating the effect of academic performance on educational attainment. The present results indicate that academic performance is the major mediator of the effects of mental ability on aspirations and attainment variables. Furthermore, academic performance is the most important determinant of educational attainment in terms of its direct effect on educational attainment (Figure 5). Third, with regard to the paths which were not hypothesized in the SHP model, the present findings are consistent with those of the previous studies in showing that socioeconomic status has an independent direct effect on educational attainment. Moreover, one unexpected finding is that academic performance has a significant direct effect on occupational attainment when all other variables are controlled. Fourth, using the HSB data, the baseline model (SHO) accounts for 23% of the variance in educational attainment, and 24% of the variance in occupational attainment, which is similar to that of previous studies (Sewell, Haller, and Portes, 1969; Sewell, Haller, and Ohlendorf, 1970). The Elaborated Model of Status Attainment The purpose of the present study is to scrutinize the impact of self- esteem on the process of status attainment. As in most replications of status attainment models, the model estimated here does not result in any appreciable increase in explained variance. It does enhance our understanding of the social psychological process through which parental socioeconomic status and mental ability affect young white men’s socioeconomic status attainment. The effects of self-esteem on the 35 process of status attainment are presented in detail as they relate to other attainment antecedents, educational attainment, and occupational attainment. Causal Structure Among Attainment Antecedents Table 1 shows that socioeconomic status and mental ability are positively correlated with self-esteem. However, when included in the model of status attainment, neither variable has a significant direct effect on self-esteem. Mental ability has a sizable direct effect on academic performance net of self-esteem (Beta=.445). Self-esteem also has a weak direct effect on academic performance net of mental ability with regard to the paths which are not hypothesized in the SHO model. The data indicate that socioeconomic status has no direct effect on academic performance independent of mental ability and self-esteem (Table 3). Socioeconomic status, mental ability, and academic performance are found to have significant direct effects on significant others’ . encouragement of college attendance. In turn, significant others’ influence has the strongest direct effects on educational and occupational aspirations (Figure 6). Significant others’ influence is the most important determinant of educational and occupational aspirations in terms of its direct effects (Table 2). Self-esteem has a positive direct effect on educational aspirations but a negative direct effect on occupational aspirations. Academic performance has significant direct effects on educational and occupational aspirations. Approximately half of the causal effects of academic performance on aspirations are indirect by way of significant others’ influence. The elaborated model accounts for 47% of the variance of educational aspiration and 23% of the variance of occupational aspiration. 36 cones—EH .8050 “enema—ma ”HOm Eon—ESE Ecocaoeemw macaw Became—om "Mm Seneca. 523880 ”nice be? 852 ”$2 52832 fieozaosem ”awn—m mESm omEoeoooomoom ”mm—m Nvoém mom; 53.3 RNN mwm; men; fined 9.5.x wee. dd womév god ONE woe SQN vmwfi SQ: ecofic 95. .882 ..... «ewe. .«mmm. tmwm. sewn. atom. o3. can. comm. H» c8 mozatg new: 2: .«o mootagoe Became was .338 woos—£830 coecgoN A oBaH 33* 67* )/OCAT / .oso ‘ EDAT '37” .. 243* MA 445* )AP .166. EDAP .181* SES: Socioeconomic Status EDAP: Educational Aspiration MA: Mental Ability OCAP: Occupational Aspiration SE: Self-Esteem EDAT: Educational Attainment SOI: Significant Others’ Influence Figure 6. Standardized path coefficients for the elaborated model. In the elaborated model, about half of the effects of socioeconomic status and of mental ability on educational and occupational aspirations are mediated by self-esteem, academic performance, and significant others’ influence (Table 3). About 45% of the effects of socioeconomic status on educational and occupational aspirations are mediated by significant others’ influence and almost none by self-esteem. About half of the effect of mental ability on educational aspiration is mediated by significant others’ influence, 30% by academic performance, and 1% by self-esteem. In sum, the data indicate that self-esteem has very slight direct effects on educational and occupational aspirations. Note that the direct effect of self-esteem on occupational aspirations tends to be negative, but it is not significant. However, significant others’ influence plays the most important role on both aspirations in this research. Although the addition of self-esteem does not significantly improve the explained variance in aspirations, it mediates the effects of socioeconomic status and mental ability on educational and occupational aspirations. 38 :2. E: :2. .23 E. 82. 9:. 8m «2. 2:. as. m< as. - o8. «3. - mm on. on. :1 <2 m8. “8. E: mmm seem sage Bop seem seas: seem span Eggnog—cm 8... E: 82.. Ba an. 2 _. m2. m< 28. - 28. «no. - mm 8a. SN. --..- <2 8_. 8.. E: «mm seem case 38. seem sees 329025 mam. 3m. 8m 5. E. 82. n< ”8. >8. :c. mm 8m. 8m. 5.. <2 «2. NB. :1 new team. _§6 .58. e85 833 seem .025 83693883 asses/3022.03 assess sa Eocene. 8 8m .5 m< .3 .<2 .32 co use» 82%.: e2 625 .u use. 39 .mcoobo 88:2: 8a 68:. .«o 88 8a 8850 88.? .338 3.80.30 2: .«o 8333:0on 2: o. 8:882. 8058:8838 £8 .3 .8888 8a Somme 80.582 .88 3885093 com «830E000 £8 comm—8888..“ 8a 808% 885. men. :3. men. Dam mam. ll- mmu. .200 on— . on u . 2...... ~Om e2. e3. .2... m< N8. Dov >8. mm 2 ~. 2 _. .i..- <2 m3. m3. It- mmm 88mm $850 3cm. Seam 68:2: «comm 685 23:88:80 8.889 80.80qu 833.8.» «outcome—c5 .8838 .N 2.3 40 Sm. Sn. ”3. 3w". Rm. Eu. mom. 8m. 8m. m8. 2 .8... 8m ta. Eh. «SN. .2 :9 Re. a8. MB. .68. £8. mm «.5. .8“. «an. :8. «SN. .93. «can. 63.. :9. so. <2 $2. :8. .53. «an. .55. «SN. .55. 8.... m8... «8. ma 5 5 AS 5 5 5 5 8v 3 3 .25 Ba .2 mm 85%; 8022.5 8...§> 80—5835 .8289? 83m v8.68 no.“ 953580 vofigaSm .m 033. 25. 25. 52. 8a. 5:. 5:. 5mm. «2. 2. 3:. mm 58. .200 .«Dm— . «mvu . A—(Dm .52. .52. .53. .58. 8m .55. .55. .58. 5mm. .52. 52. .2 55.- :5. - 25.- 58. - 85. - N8. .. m8. - 58.- mm .58. 58. .58. .52. 58. 5mm. .52. 5%. 5mm. .59... <2 52. .52. .52. 52. .52. 5.2. .555. .52. 5:. 5:. 8a a: 6 3 5 6 3 As 6 6 3 55m .28 838.5 .8255 852.3 Because—2: 3.5.80 .5. 2.3. 42 3N. EN. .EN. SN. 55. bag. >3. um .35. Exam *5 a. «v2. m