University of Southern California Dissertations and Theses

Self-Concept In Adult Women: A Multidimensional Approach

(USC Thesis Other)

Self-Concept In Adult Women: A Multidimensional Approach

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This mam i script has been reproduced from the microfilm master. UM1 films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is depea dent upon the quality o f the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UM3 a complete manuscript and there are mitring pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note win indicate the deletion. Oversize materials (e.g^ maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs Included in the original manuscript have been reproduced xerographicaUy in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. A Ben & Howen information Company 300 North Zeet> Roao Ann Arbor. M l 48106-1346 USA 313 761-4700 800 521-0600 SELF-CONCEPT IN ADULT WOMEN: A MULTIDIMENSIONAL APPROACH by Julia Marie Gearhart A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Counseling Psychology) December 1994 Copyright 1994 Julia Marie Gearhart ONI Number: 9600980 □HI Hlcrofora 9600980 Copyright 1995, by OMI Coapany. All rights rasarvad. This alcroEoca adition Is pcotactad against unauthoclsad copying undar Tltl* 17, On1tad Staton Coda. UMI 300 North Xaab Road Ann Arbor, HI 48103 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANCELES. CALIFORNIA 90007 This dissertation, written by JuXia...CLeajrAia;:.k................................ under the direction of hex. Dissertation Committee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of re quirements for the degree of DOCTOR OF PHILOSOPHY Otan of Graduate Studies Date 199ft..... DISSERTATION COMMITTEE Julia M. Gearhart Michael Newcomb Self-Concept in Adult Women: A Multidimensional Approach Previous research focusing on self-concept has yielded a variety of contradictory results, in large part due to inadequate tools and mis-specified models. Utilizing a multidimensional model and a promising new instrument, the focus of this exploratory study was to lend greater clarification to the relationships among age, social role, and women's self-concept. Social roles were defined as paid worker, student, parent, and marital status, while age was treated both continuously and categorically (18-30, 31-45, 46-65, and 66+). A stratified sample (N-275) of women ages 18-91 was administered the Self-Descriptive Questionnaire III, a tool designed to measure nine facets of self-concept. As the SDQ III has not been normed for older adults, one goal of this investigation was to establish its validity and reliability for use with a mature pop ulation. The results of this study indicate that all nine dimensions of self-concept are reliable for this older population and, furthermore, that the dimensions are relatively distinct. Coefficient alphas varied from .70 to .94 (median - .89) and average correlations among the dimensions was modest (median r * .20). Multiple regression analyses indicated that role occupancy per se was unrelated to self-concept with two exceptions: full-time employment and single status significantly predicted some aspects of self-concept for middle-aged women. In contrast, age was the primary predictor for eight dimensions of self-concept. Specifically, self-concepts for physical appearance, physical ability, and opposite sex relationships were negatively correlated with age, while self-concepts for religion/spirituality, honesty/reliability, and emotional stability were positively correlated. Self- concepts for cognitive ability and same sex relation ships were also related to age, but in a non-linear fashion. Only the dimension of general-esteem was not correlated with age in any manner. These findings were interpreted in terms of both role and maturation theory— with the strongest support for the latter— suggesting that self-concept is not a fixed, stable construct over time, but rather shifts with development throughout the life cycle. (Copies available exclusively from Micrographics Department, Doheny Library, USC, Los Angeles, CA 90089- 0192). Commit Ignature Date ii TABLE OF CONTENTS LIST OF T A B L E S ..................................... iii CHAPTER 1. INTRODUCTION.................................* 1 2. REVIEW OF THE LITERATURE................... * . 8 3. METHODS.........................................30 4. RESULTS.........................................47 5. DISCUSSION...................................68 APPENDIX A. STRUCTURE OF SELF-CONCEPT . ..................95 B. COVER LETTER .................................. 96 C. DEMOGRAPHIC QUESTIONNAIRE . ............... 97 D. SELF DESCRIPTIVE QUESTIONNAIRE I I I ............99 REFERENCES..........................................102 iii LIST OP TABLES 1. Demographic Characteristics of Women (Total Population).........* 3 2 2. Demographic Characteristics of Women by Age . 35 3. Chi-Square and ANOVAS for Demographic Characteristics and Age G r o u p ............... 3 7 4. Chi-Square and ANOVA Difference Tests by Ethnic Group . . 40 5. Reliability Coefficients for the Nine Dimensions of Self-Concept .................... 50 6. Correlations Between the Nine Dimensions of Self-Concept for Total Population (N*265) . . 52 7. Correlations Between the Nine Dimensions of Self-Concept for 18-45 and 46+ Populations of Women.......................................... 53 8. Correlations Between Age and the Nine Dimensions of Self-Concept .................... 55 9. ANOVA Results for the Four Age Groups and Nine Dimensions of Self-Concept ............. 57 10. Multiple Regression Analyses of the Five Predictor Variables with Nine Dimensions of Self-Concept for 18-30 Year-old Women .... 59 11. Multiple Regression Analyses of the Five Predictor Variables with Nine Dimensions of Self-Concept for 31-45 Year-old Women .... 61 12. Multiple Regression Analyses of the Five Predictor Variables with Nine Dimensions of Self-Concept for 46-65 Year-old Women .... 62 13. Multiple Regression Analyses of the Five Predictor Variables with Nine Dimensions of Self-Concept for 66+ Year-old Women ........ 64 14. Multiple Regression Analyses of the Five Predictor Variables with Nine Dimensions of Self-Concept for Total Population, Controlling for A g e ......................................65 1 INTRODUCTION The way people evaluate themselves is widely accepted as important by mental and health care profes sionals. People who have relatively high self-worth tend to be better students (Purkey, 1970), are bothered less by anxiety (Piers & Hams, 1969), are less de pressed (Battle, 1981), demonstrate better physical health (Fitts, 1965), and enjoy better social relation ships (Coopersmith, 1975). In their jobs, they tend to value independence, welcome competition, and expect more success (Rosenberg, 1965). People with high self esteem appear to be adjusted, happy, and competent. One of the goals of the health care professional, therefore, is to facilitate greater self-regard in all people and thus enhance the quality of human life. For the clinician, educator, and researcher alike, the evaluation of self-concept is an initial step toward better understanding human development and motivation (Gilberts, 1983). Most researchers view self-concept as a set of structured self-attitudes that is relatively stable and characteristic of an individual (Demo, 1992). As might be expected, psychologists and sociologists conceptual ize the structure of self-concept somewhat differently, with psychologists typically describing a system of traits and sociologists describing a structure of roles 2 or identities. While numerous studies from both disci plines have examined structural dimensions of self- concept, very few have addressed its contextual as pects, i.e. changes in self-concept from one situation to another, from one relationship to another, or from one year or stage in the life course to another. As a result, little is known about the developmental and social conditions responsible for change and stability in self-concept. A few studies, most designed by developmental psychologists, have followed individuals through ado lescence, and others have examined self-concept during the transition from adolescence to early adulthood. Only indirect attention, however, has been devoted to self-concept at more mature points in the life-cycle— such as mid-life or old age— and much of this research has focused exclusively on men (Costa & McRae, 1980; Erickson, 1970; Levinson, 1977; Vaillant, 1972). Furthermore, those studies that have attempted to employ a temporal perspective have typically examined self-concept in a social vacuum, ignoring relationships between changes in the life course and changes in self- concept . The focus of this study is on the nature of and differences in the self-concept of women over the life course. Utilizing the Self-Descriptive Questionnaire 3 III, a newly developed Instrument designed to measure multiple dimensions of self-concept, this project proposes to investigate the fluctuations in women's self-perceptions across situations and relationships (defined as roles) and life phases (represented by age groups), in hopes of lending greater clarity to the contextual nature of self-concept. Although the self has been a major topic of study for centuries, the evaluation of self-concept is not a simple task. Reviews of self-concept research (Burns, 1979; Demo, 1992; Shavelson, Hubner, and Stanton, 1976; Wells and Harwell, 1976; Wylie, 1974, 1979) emphasize the lack of theoretical basis in most studies, the poor quality of measurement tools used to assess self-con cept, methodological shortcomings, and general lack of consistency in the findings. Harsh and Shavelson (1985) note that self-concept research typically utilizes three different definitions of general self-concept. The first, and most question able, involves a conglomerate score across an assort ment of self-report items. The Piers-Harris and Ten nessee Self-Concept instruments are examples of this usage. This approach attempts to measure a diverse set of facets, but facets which often lack empirical veri fication and weighting. Responses are simply summed to form a total score that is assumed to measure general- 4 self. Marsh and Shavelson suggest that many of the contradictory findings existent in self-concept re search are the result of such poorly defined constructs and tools. The second use of general self-concept refers to an inferred construct, a higher-order factor, that is not directly measurable and is often found in tools utilizing a multifaceted and hierarchical model of self-concept, such as the Self-Descriptive Question naire (SDQ) used in this study. The third— and most common--use of general self-concept refers to a sepa rate, identifiable facet which is viewed as a super ordinate dimension and is often addressed as self esteem. Ninety percent of all studies on the self are actually studies of this single factor {McGuire, 1981). Rosenberg's Self-Esteem Scale (1965) and the SDQ's dimension of general-self are typical of this applica tion. Although the construct of a general self-concept is both theoretically and practically important, recent work indicates that the self is best conceived as having a hierarchical and multifaceted structure (Flem ing k Courtney, 1984; Fleming k Watts, 1980; Harler, 1982; Marsh, Parker, k Smith, 1983; Marsh, Relich, k Smith, 1983; Marsh, Smith, k Barnes, 198 3; Marsh 4 Shavelson, 1985; Shavelson k Bolus, 1982; Wylie, 1979). There is less agreement, however, on th$ specific 5 facets and how they are organized. As children ap proach adulthood, for example, levels of self-concept become more distinct, and their hierarchical structure becomes weaker. A particular problem specific to the area of self perception research is that theorists and researchers alike have used the terms self-concept and self-esteem interchangeably. Self-concept is typically defined as the perception one holds of oneself, totally and with regard to several dimensions— such as personal attrib utes and assumed roles (Shavelson et al., 1976). Self- concept is also influenced by environmental interac tion, such that individuals fulfilling specific roles receive and internalize feedback about the nature and quality of their role performance. Thus a woman may perceive herself as a wife or daughter, as well as a "successful1 1 student or "competent" mother. Self esteem, on the other hand, is the valuative assessment individuals make concerning personal satisfaction with roles and/or the quality of performance— e.g., "I feel good about myself as a mother." Perceptions involving self-esteem are made on the basis of personal values and are often referred to as the affective component of self-perception (Hamachek, 1978). In brief, self- concept refers to the descriptive perceptions of self while self-esteem refers to the valuative assessment of 6 those descriptions (Beare & Lipka, 1980). While self esteem is an important element of the self-system, it must be remembered that it is based on self-concept— i.e., "how I feel about how I see myself"— and thus is, at least theoretically, subordinate to self-concept. Shavelson, Hubner, and Stanton (1976) propose a multifaceted, hierarchical model of self-concept that is utilized in this study (one possible representation is seen in Appendix A). In this model, self-concept is defined as self-perceptions that are formed through experience with and interpretations of one's environ ment, influenced especially by evaluations by signifi cant others, reinforcements, and attributions for one's own behavior (Shavelson & Marsh, 1985). It is further posited that self-concept is characterized by seven major features: (a) it is organized and structured in that people organize information they have about them selves and relate these categories to each other; (b) it is multifaceted, and the specific facets reflect the category system adopted by a person or shared by a group; (c) it is hierarchically organized, with specif ic perceptions of behavior at the base moving to infer ences about the self in subareas, and then to the self in general at the apex; (d) it is stable at the apex of the hierarchy, but becomes situationally specific as one descends the hierarchy, and thus less stable; (e) 7 it is better differentiated for adolescents and adults, with facets becoming more distinct with age; (f) it has both descriptive and evaluative dimensions; (g) and it is differentiable from other constructs (Marsh, Barnes, & Hocevar, 1986). One of the few self-perception instruments to incorporate this multifaceted model of self-concept is the Self-Descriptive Questionnaire III (SDQ III), a tool designed to measure 13 dimensions of academic and non-academic self-concept in young adults (Marsh, 1989). Unlike previous self-concept instruments, the structure of the SDQ III is based on rigorous factor- analytic procedures and its construct validity is substantiated by extensive multitrait-multimethod analyses (Marsh, Barnes, 6 Hocevar, 1985; Marsh f c Byrne, 1989; Marsh, Richards, & Barnes, 1986). As this tool has not been normed for older adults, however, one of the primary goals of this study is to establish its validity and reliability for use with a mature popula tion. Stutaaairfc of the Prnhloa The problem that this study addresses is the possible influence of age group and social role on women's self-concept, defined multidimensionally. This problem may be further clarified by asking a number of questions. 8 (1) Is the SDQ III a valid and reliable instrument for measuring self-concept in adult women? (2) If acceptable validity and reliability can be established, what is the relationship between age and self-concept in adult women? (3) What is the relationship between social roles and self-concept in women? (4) Do age and role together impact on self-con cept in women? This study is important for a number of reasons. First, because there are few instruments that adequate ly measure self-concept, this study will provide cru cial normative data on the adult population for the SDQ III, a promising new tool for measuring self-concept multidimensionally. Secondly, this investigation will add to the sparse literature on the contextual varia tions of self-concept; that is, how self-concept varies by age, situations, and relationships. Third, whereas previous self-perception research has focused over whelmingly on adolescents and college students, this study will involve a diverse population of adults from throughout the life course, specifically women, who have also been neglected in developmental research. at Lifcarafcura As noted earlier, self-perception research has been plagued by methodological problems, imprecise 9 construct definitions, and poor measurement tools, and the studies reviewed here are no exception. In fact, the majority of these studies have not addressed self* concept at all, but investigate a presumably related construct, that of well-being. Others focus on self esteem, an important component of the self-system, but only one dimension of self-concept. Very few investigations have focused specifically on self-concept, perhaps in part due to the lack of adequate instrumentation. The relatively recent vali dation of the self-concept's multifaceted nature has presented the additional challenge of operationalizng this multidimensional model. As applied research has yet to catch up with these developments, much of the current review— by necessity— summarizes studies on self-esteem and well-being, variables with theoretical support for inclusion, but which have not yet been empirically linked with self-concept. Many of the studies analyzed here also suffer from methodological shortcomings. For example, except where noted, most of the investigations have utilized rela tively small samples of subjects from a homogenous populace. Even in the larger studies, women have typically been white, middle or upper middle-class, and drawn from university populations and environs, thus significantly restricting generalizability. In addi- 10 tion, while most self-perception research, including that reported here, tends to be cross-sectional in design, the influence of possible cohort and genera tional effects has been virtually ignored. For the purposes of this review, literature on self-concept, self-esteem, and well-being will be presented for each of the predictor variables of this study: age, parenthood status, marital status, and vocational status. Two major theoretical perspectives on aging are also presented. Age. Age is a primary identifier that has a pro found effect on the roles one occupies and the way one is perceived by others. A plethora of studies have attempted to investigate the relationships between age and self-esteem, life satisfaction, and self-concept, yielding a variety of often contradictory results. Several studies have suggested that women manifest poorer self-image and loss of self-esteem during the mid-years of the forties and fifties. Powell (1977), in a comparison of homemakers and employed women who had entered the empty-nest years, found that homemakers reported significantly greater numbers of emotional and physical problems. In another study, Birnbaum (1975) compared homemakers with career women, most of whom still had children living at home; these homemakers reported significantly fewer positive feelings about 11 themselves even in domestic and child care areas. Lowenthal, Thurnher, and Chiriboga (1975) found middle* aged women, about half of whom were employed, to have poorer self-concepts in comparison to both younger and older subjects. In contrast to these findings, Neugarten (1986) reports that middle-aged women expressed feelings of increased self-confidence and viewed mid-life as a time of greater freedom to develop one's capabilities. Other researchers (Monge, 1975; Puglisi & Jackson, 1981) have found that the highest levels of self-esteem for women occurred during the mid-life decades. Thus, while there is research which strongly suggests that women— particularly those who have invested in tradi tional roles— may experience a loss of self-esteem and hold less positive self-images during the mid-life years, other studies indicate that women in their forties and fifties have more positive self-concepts than younger women and also, in some cases, than older women. Research focusing on older people in their sixties and beyond has produced an equally confusing picture. Some studies report little difference between older and younger women in self-confidence (Wallach f c Kogan, 1961), negative self-perceptions (Gunn et al., 1960), or self-esteem (Erdwins, Mellinger, & Tyer, 1981). 12 Other investigations, however, have reported less favorable self-perceptions in older subjects (Brenneis, 1975; Monge, 1975; Puglisi 6 Jackson, 1980). A number of recent studies offer compelling evidence that self- evaluation— usually measured as global self-esteem— increases incrementally with age group, with older adults reporting the highest self-esteem (Gove, Ortega, and Style, 1989; Hong, Bianca, & Bollington, 1993; Jaquish & Ribble, 1981; Morganti, Nehrke, Hulicka, and Cataldo, 1988; Nehrke, Hulica, & Morganti, 1980). This ”paradox of happiness” (Baltes & Baltes, 1990) is also reflected in related research in which the elderly demonstrated high levels of life satisfaction (Laborda & Powers, 1985), morale (Lawton, 1984), subjective health (Ferraro, 1980; Mossey t Shapiro, 1982), and well-being (Diener, 1984; Larson, 1978). One major weakness associated with the studies reviewed thus far concerns the use of cross-sectional designs to measure, or imply, age-related changes over the 1ife-course. Although highly notable instances of longitudinal age-span research exist— for example, the Berkely and Oakland studies (Block, 1971), The Fels study (Kagan I Moss, 1972), and the Terman study of the gifted (Oden, 1968)— these studies have tended to focus on such global phenomena as achievement and adjustment through life, and not self-perception per se. As a 13 result, there is an absence of relevant data collected on the same population across broad periods of the life course, for different facets of the self-system. Two investigations have attempted to longitudinal ly measure changes in self-concept over the life span— unfortunately, with contradictory results. Lowenthal et al. (1980), studying four waves of subjects at different developmental transitions over a 12-year period, determined that self-concept becomes more negative with age, although life satisfaction generally increases. Helson and Hoane (1987), on the other hand, studying one group of women at three different periods in a 25-year span, found that self-regard became more favorable with age. The most obvious explanation for these discrepant findings is that Lowenthal's population included a much wider age span (17-70 years) than Helson and Hoane's (21-45 years), although the letter's data base had the advantage of being derived from only one group of subjects. Further research with Helson and Hoane's population will undoubtedly continue to yield addition al age-related data. Another major problem with these two studies is that both utilized non-standardized, unidimensional measures of self-concept, reflecting the accepted model and methodology of the times (1960's). Subsequent research has clearly specified the multidi 14 mensional nature of self-concept, however, demonstrat ing one of the integral risks in longitudinal research, i.e. the possibility that significant changes in the construct or its instrumentation might occur over the course of the study. rq!« frhftnr-y and aging. in recent years, sociolo gists have primarily viewed aging as a process involv ing role acquisition and role loss. According to role theory, individuals develop awareness of social posi tion through interaction with others, and these identi ties are developed and sustained in role relationships. Qualitative theorists (Mead, 1934; Rosenberg, 1979; Stryker f c Serpe, 1982) propose that societal roles, such as those associated with age and gender, are closely tied to the norms and values of society. Thus, assuming that age confers and withholds status in all societies, age-related changes in self-concept are largely attributed to characteristics of the social system. In applying role theory to the life experience of the elderly, gerontologists focus on the lack of formal structures and statuses available to the aged in main stream society. Role losses such as retirement, loss of good health, and loss of one's spouse are perceived as resulting in diminished self-concept and a decline in self-esteem, life satisfaction, and morale (Blau, 15 1973; Larson, 1978; Lemon, Bengston, & Peterson, 1972; Pilisuk, 1982; Roscow, 1974). From a role perspective, there are at least three reasons to anticipate self-concept being most positive in early adulthood, then declining with age. First, aging is initially associated with role acquisition, by and large a positive experience. This is followed by role transition, reflecting various degrees of stress, and finally with role loss, seen as frequently problem atic. Secondly, western cultures typically place a high value on the attributes of the young— such as strength, energy, and beauty--and devalues characteris tics of older persons, viewing them as old-fashioned, out of touch, and relatively unattractive. Third, in the rapidly changing milieu of contemporary society, middle-aged and older adults would be expected to experience increasing difficulty keeping pace with these changes, ultimately functioning less and less effectively. Maturation theory and aging. The concept of adult psychological maturation is derived from theories of human development descended from Freud's theory of psychosexual development. According to Freud, individ uals are born with a psychological predisposition to certain characteristics, which will develop naturally if growth is not interrupted. Human development theo- 16 rles of today continue to emphasize stages of develop ment and developmental tasks over the life course, with the presumed outcomes differing substantially from those of a role perspective. Erickson (1963; 1982) describes eight "ages of man,1 * each defined by the duality of possible gains and losses that occur at each stage. With age, Erickson postulates a widening social sphere that moves from the initial mother-child bond to family, school, and peer relationships, and eventually the work place and commu nity. Jung (1958), Buhler (1935), Neugarten (1977), Levinson (1978), Gould (1978), and others have suggest ed a similar progression. These theorists propose that early and middle adulthood is associated with a concern with self that is manifested by a focus on activity, achievement, power, and control. As persons age, howev er, their ego needs gradually decline and they become more reflective, contemplative, and accepting of self and others. This adult maturation approach to aging suggests that aging is a relatively positive process* From this perspective, one does not anticipate a de cline in life satisfaction, self-esteem, or self-con cept over the life course, and--if change does occur— it is expected to be in the direction of an improve ment. 17 Paranthonri. Three areas of research are pertinent to the issue of how parenthood influences self-concept: the research on happiness and quality of life, the research on psychological distress, and studies of adults' perceptions of parenthood. In the first two areas respondents' well-being is measured directly, and comparisons made to determine whether parents are psychologically better or worse off than non-parents. In the studies on perceptions of parenthood, respon dents are asked about their overall evaluation of parenthood and the specific costs and benefits of children. The research on quality of life typically has focused on variations in parents' well-being over the life course and in response to specific transitions, such as birth of the first child. A number of re searchers have commented on the U-shaped curve in marital well-being that is associated with the life cycle state of parenthood (Aldous, 1978; Figley, 197 3; Rollins f c Cannon, 1974), and this variation appears to be particularly strong for wives (Rollins 6 Feldman, 1970). More recent research has found that parents with children at home are psychologically less well off than non-parents on a wide range of indicators (Alwin, 1984; Glen & McLanahan, 1981; McLanahan & Adams, 1984, 1985). A few studies have reported no differences le between parents and non-parents on some indicators of well-being (Andrews & Withey, 1976; Marini, 1980), but no one has found that parents are actually better of than non-parents on any of these measures. Not sur prisingly, women with young and school age children are consistently the least happy and least satisfied of all groups studied (Campbell, 1976; Glen 6 Weaver, 1978; Glen & McLanahan, 1982). Furthermore, the overall presence of children appears to be a more important factor than either number, spacing, or gender of child (Polit, 1982). The two most commonly used constructs in the psychological distress research are depression and anxiety. Most early studies on parenthood and mental health found that parents experienced more depression and anxiety than adults without children (Gove & Gearken, 1977; Pearlin, 1974; Radloff, 1975). However, methodological problems (i.e. couples with children at home were compared to other married couples, not dis tinguishing between those in the empty-nest stage and those who had never had children) rendered these re sults inconclusive, or at best, not very useful. As with the quality of life research, later work on distress encompasses more sophisticated methodolo gies and more complex models. In the last decade researchers have also expanded their focus to include 19 single parents and non-marrled couples with children as well as married couples. Recent research is considerably less conclusive than early studies regarding the effect of children on psychological distress, and especially regarding de pression. Gove and Mangione (1983) found that having children at home increased the anxiety of married women, but had no impact on depression. Cleary and Mechanic (1983) also found that parenthood was not related to depression, except among employed mothers. Furthermore, at least two researchers have reported that parents are better off than childless adults. Kandel et al. (1985) found that parenthood reduced the incidence of depression in women, and Aneshensel et al. (1981) reported similar results for men as well as women. However, both of these studies defined parent hood as having ever had children rather than as living with children, which undoubtedly contributed to these contradictory findings. The most consistent finding in this area of re search is that single mothers are much more distressed than any other group. Pearlin first noted this in the early seventies, and it has since been substantiated by numerous other researchers (Alwin et al., 1984; Brown f r Harris, 1978; Aneshenshel et al., 1981; Guttentag et al., 1980). 20 Another important aspect of the research on dis tress is that it has extended our knowledge about the interaction between employment and parental status. Studies increasingly indicate that employment has positive effects for women without children and, under certain conditions, for women with children* At least two studies have shown that employment has a positive effect for married mothers if their husbands help with child care and housework (Kessler & McRae, 1982; Ross et al*, 1983). However, Ross et al. found found that the benefits of work depended on the couple's prefer ence about whether the wife should work. If they preferred a "traditional" division of labor, working had no psychological benefits for the wife. Studies of adults' perceptions of parenthood have often yielded inconsistent results* Veroff et al. (1981) found that younger couples were less positive and viewed parenthood as more restrictive than couples over 50, and also that mothers were less positive than fathers on their evaluations of children. These re sults were confirmed by McLanahan and Adams (1985). However, Hoffman and Manis (1978) found that mothers reported more parental satisfaction than fathers, and Campbell et al. (1976) found no gender differences at all in parental satisfaction. Goetting (1986) suggests that such inconsistent findings reflect the multidimen- 21 sional and gender-specific nature of the construct— that motherhood brings both greater role fulfillment and more restrictions than fatherhood. Women's perceptions also vary as a function of whether or not they work outside the home and whether they view parenthood from the perspective of tradition al gender roles. Blake & del Pinal (1981) found that women who were employed, part or full-time, were twice as likely as homemakers to say that the costs of chil dren outweighed the benefits. Moreover, women who believed that men were unlikely to share in housework or child care were much more likely to emphasize the direct costs of children. Those findings are consis tent with what is known about the joint effect of employment and motherhood. Research of perceptions of children indicates that women who do not finish high school are more likely to report that the benefits of children outweigh the costs than are women with a high school or college degree. Both Blake & del Pinal (1981) and Veroff et al. (1961) found that women with limited schooling were more likely to view children as a social and financial investment, and were more likely that other groups to say that parenthood gave them a great deal of fulfill ment. 22 In summary, a number of generalizations can be drawn from the literature reviewed thus far. First, parenthood in itself does not increase psychological well-being or satisfaction for women, and often de creases it. Second, single mothers are clearly worse off than other groups. Third, working outside the home improves the psychological well-being of women without children and also improves the well-being of mothers if husband help with child care and if both parties accept the woman's dual role. And fourth, the opportunity costs of children are lower for women with limited education. Vocafcione1 status. The effect of employment on women's self-perception has been largely derived from research focusing on the relationship between multiple role involvement and well-being. The roles of full time homemaker and full-time paid worker are typically addressed, often with the inclusion of motherhood as well. In most of these studies, three dimensions of well-being are analyzed: self-esteem, happi ness/satisfaction with life, and psychiatric symptom atology (depression or anxiety). Two major hypotheses have been postulated concern ing the relation of role involvement to well-being. The "scarcity" hypothesis (Narks, 1977), put forth by Goode (I960) and others, contends that people have 23 limited energy and resources and may become overbur dened by too many role relationships. The competing demands of different social tasks are thus believed to produce increased role strain and conflict and, ulti mately, psychological distress. In contrast, the "enhancement" hypothesis (Harks, 1977; Sieber, 1974) suggests that individuals may actually profit from multiple role involvement, gaining status, privileges, and increased self-esteem, as well as the ability to trade off undesirable aspects of roles. Recent studies have generally substantiated the "enhancement” hypothe sis, encouraging a flurry of research directed at specific life domains. The relationship between voca tional status and women's well-being in particular has been targeted. Host current studies have reported a positive relationship between full-time employment and self esteem in women. Baruch and Barnett (1986), utilizing a large sample of midlife women, found that being a paid worker was consistently associated with higher self-esteem. Pietromonaco, Hanis, and Frohardt-Lane (1986) determined that holding more roles (including full-time employment) was associated with both higher self-esteem and greater job satisfaction. In another study, Coleman and Antonucci (1983) concluded that working is the only significant predictor of self- 24 esteem in midlife women; neither income, education, marital status, or family life-cycle stage were ade quate predictors. Pietromonaco, Nanis, and Marcus (1985), in contrast to these results, however, found that the benefits of full-time employment are attenuat ed for women who are not career-oriented. Although these findings are relatively isolated, they do suggest that some women are perhaps self-selected into employ ment because of higher levels of self-esteem; that higher self-esteem might be the "cause” of employment and not vice versa. Studies by Verbrugge (in press) and Stein, Newcomb, and Bentler (1990) indicate that such selection factors account for only a small (or no) proportion of the findings, however. Research on the risks and benefits of multiple role involvement indicates that women who hold more social roles experience less psychological distress and mental illness (Brown & Harris, 1978; Coleman & Antonucci, 1983; Thoits, 1983), and fewer physical ailments (Northcott, 1981; Verbrugge, 1982). In regard to the specific role of employment, Cowan and Cowan (1985) found that working women were significantly less symptomatic on the CES-D depression scale than homemak ers, and Merikangas (1985) concluded that lack of employment is a risk factor for depression in women. Repetti, Matthews, and Waldron (1989), in a review of 25 the effects of employment on women's mental and physi cal health, determined that working did not have a negative effect on health, and Thomas (1988) found that employment was actually highly correlated with good health. When happiness or satisfaction are used as indices of well-being, the effect of full-time employment is less clear. Barnett and Baruch (1981) found that women with jobs outside the home expressed greater life satisfaction than homemakers, and Manis (1982) reported similar results. However, numerous other studies have found no significant differences in happiness or satis faction between employed and nonemployed women (Camp bell, Converse, & Rogers, 1976; Seriin, 1980; Warr & Pearry, 1982; Wright, 1978). Although research generally supports the positive effects of employment for women without children, the same benefits cannot be claimed for employed women with children. It appears that motherhood and paid employ ment do not mix well for women, insofar as occupying both roles offers no subjective benefits over occupying one or the the other (Ryder, 1979). Two different studies have concluded that employment has positive effects for married mothers only when husbands share domestic and child care responsibilities (Kessler 6 McRae, 1982; Ross k Huber, 1983). Ross and Huber also 26 found that the benefits of employment depended on the couple's preferences about whether the wife (with children) should work. When a "traditional" division of labor was preferred, there were no psychological benefits for the wife. Marital utatue. There is very limited data avail able on the relationship between marital status and a women's self-concept. A considerable amount of re search has focused on a related variable, however, that of subjective well-being. Studies of well-being done in the 1970's consis tently reflect that married women are happier and more satisfied than those who are not, whether the latter are never-married, divorced, or widowed (Campbell et al., 1976; Depner, 1979; Glenn, 1975; Ward, 1979). However, a more recent meta-analytic review of 58 empirical sources reports only weak associations be tween marital status and well-being (Haring-Hidove et al., 1985). Specifically, a mean correlation of .14 was found across all studies, indicating a positive and significant correlation between marriage and well being, but of a magnitude considerably less than would be expected based on previous research. In one of the few investigations specifically addressing the relationship between marital status and self-concept, Gigy (1980) noted that even though mar- 27 ried women appear to be happier and more satisfied, ratings on self-esteem and self-concept are equal to or less than those for single women. Thus while being married may mildly enhance a woman's happiness and sense of satisfaction, it apparently does nothing to boost self-concept or self-esteem. Tn summary, as noted earlier, very few investiga tions have focused specifically on self-concept, and the majority of studies analyzed here have measured a presumably related construct, that of well-being. Others measured self-esteem, an important component of the self-system, but only one dimension of self-con cept. Research addressing the relationships between age and self-concept, self-esteem, and well-being has yielded a plethora of inconsistent findings, primarily due to imprecise construct definitions and methodologi cal problems. A number of recent cross-sectional investigations have demonstrated that self-esteem increases with age group, but this pattern has not been replicated by the two existent longitudinal studies that have measured self-concept (i.e., one found that self-concept increases with age, the other that it decreases). Clearly no definitive relationships be tween self-concept and age or age group have yet been established. 28 Studies investigating the relationship between parenthood and well-being have consistently noted that motherhood does not increase well-being for women, and often decreases it. Research has determined that mothers with children at home are psychologically less well off than non-mothers on a wide range of indica tors; although some studies have reported no differenc es between parents and non-parents, none have shown parents to be better off than non-parents. A considerable amount of research has focused on the relationship between marital status and subjective well-being. In general, studies on well-being have consistently reflected that married women are happier and more satisfied than those who are not. More recent work, however, while continuing to confirm this rela tionship, has noted that the strength of the correla tion has weakened significantly. The effect of employment on women's self-percep tion has been largely derived from research focusing on the relationship between multiple role involvement and well-being, particularly the dimension of self esteem. Most current studies have reported a positive relationship between full-time employment and self esteem in women. These benefits do not extend to employed women with children, however, except when childcare and domestic responsibilities are shared. 29 In summary, then, most past studies— utilizing various measures of well-being— have found positive relationships between well-being and full-time employ ment or marriage, negative or no relationships between well-being and motherhood, and inconsistent findings for the relationships between age and self-concept, self-esteem, or well-being. Research Quest:inns The focus of this study is the nature and degree of change in the self-concepts of four age groups of women. Utilizing the Self-Descriptive Questionnaire III, a newly developed instrument designed to measure multiple dimensions of self-concept in young adults, this study proposes to establish the reliability and validity of this tool for use with a mature population. If adequate reliability and validity can be achieved, this study proposes to investigate the relationships between these dimensions of self-concept and the roles of marital status, vocational status, and parenthood for four age groups of women. As the study is largely exploratory, it is also hoped that the research find ings will generate additional testable hypotheses for this multifaceted model of self-concept. This study proposes that: (1) Adequate reliability and validity criteria will be established for the SDQ III for older adults. 30 (2) Different dimensions of self-concept may be higher for different age groups than others. (3) Different dimensions of self-concept may be higher for different roles (at different ages) than others. As self-concept has seldom been addressed multi- dimensionally, it is difficult to hypothesize specific differences in dimensions. Based on what developmental literature says about aging, however, hypotheses are presented on the dimensions of self-concept for physi cal appearance, self-concept for physical ability, and self-concept for religion/spirituality. It is hypothe sized that: (1) Self-concept for physical appearance will decrease with age in women. (2) Self-concept for physical ability will de crease with age in women. (3) Self-concept for religion/spirituality will increase with age in women. METHODS This chapter will present the methods and proce dures of this study, including a description of the subjects, description of the research instrumentation, measures of the study, and description of the proce dures . 31 naanri ptinn of Suhjants The sample for this study! described in Table 1, consisted of 265 women from the ages of 18 to 91 who reside in San Bernardino and Riverside Counties* Stratified sampling was utilized to ensure equal dis tribution of age and to assure ethnic representation according to county demographics. All participation was voluntary and all subjects were fully informed as to the research topic. Although the majority of this population were Caucasian, approximately 25% were minority and reflect ed the counties' ethnic distributions as per the 1990 U.S. census survey. Approximately 60% of the subjects were employed— the rest being retired or housewives— and most earned less than $30,000 a year. Although 25% of the participants were full-time students, most were also employed in some capacity. This population was generally well educated, with nearly half of them listing some college or vocational/business school training. Marital status was fairly equally distribut ed between married, single, and divorced/widowed, and the majority of women had children either in or out of the home. 32 Table 1 Demographic Characteristics of Women (Total Population) VARIABLE N 16 - 30 84 34 31 - 45 63 24 46 - 65 62 23 66+ 56 21 BfchnlrHfcy Hispanic 32 12 African-American 20 8 Asian 10 4 Pacific Islander 2 1 White 193 72 Other 6 2 Finally Tnefliit Under $15,00 79 30 $15,000 - $30,000 65 25 $30,000 - $45,000 35 13 $45,000 - $60,000 34 13 Over $60,000 47 18 * Combined for couplea; where a woman is the head of household, this represents one income only. For students living at home, this represents parents' income. Wfliir!e+1 on Junior High or Less 4 2 Some High School 19 7 High School Graduate 30 11 Vocational/Trade 20 8 Some College 60 23 2-Year College Degree 43 16 4-Year College Degree 31 12 Graduate Work/Prof 55 21 33 Table 1 (continued) Demographic Characterimtica of Woman _______VARIABLE_________________ N_____________________%__________ Vnff«»toni1 Bfr.atrna Professional* 52 20 Technical 15 6 Manager/Adain 19 7 Clerical 31 12 Sales 8 3 Service 30 11 Other 1 .4 Not Eaployed 46 17 Retired 38 14 Missinq 25 9 * Any job classification typically requiring a college degree. May! fra 1 single 78 30 Harried 97 37 Cohabiting 6 2 Divorced/Separated 48 18 Widowed 34 13 PiranthnnH Children in hoae 73 28 Children out of hose 97 37 Ho Children 93 35 Full-tiae 61 23 Part-tise 27 10 Not Student 177 67 34 Chi-square testis and ANOVAs were used to determine whether these demographic characteristics varied across the four age groups of women (categories were collapsed to maintain adequate cell sizes for each group). As can be seen in Tables 2 and 3, there were a number of significant differences between the age groups, al though most can be explained by developmental and/or socioeconomic trends associated with age. For example, as might be predicted from typical earning patterns, income was lowest for the youngest and oldest (i.e., retired) age groups, with the middle age groups demon strating peak incomes. Similarly, the youngest and oldest age groups had the least amount of formal educa tion— reflecting, respectively, current student status and cohort differences of accessibility to higher education— while the middle age groups exhibited the greatest amount of education. Not surprisingly, the percentage of married women increased with age group unitil 66+, when widowhood became the most prominent feature. Contrariwise, single status was the greatest for the youngest age group and significantly lower for each age group thereafter. As might be expected, the percentage of women who were mothers increased with age group— varying from 37% to 91%— reflecting the influ ence of both developmental cycle and cohort societal patterns. On the variable of vocational status, at 35 Table 2 Dmmogrmphlc Characteristic* of Wommn toy Age p T C t n t of S i i p l i by Ag» Group __________Variable____________ 18-30 31-45 46-65_____ 66+ F+.tiwicifcy Hispanic 22 6 13 4 African-Aseriean 13 8 5 2 Asian 7 6 Pacific Islander 1 2 Mhits 55 71 81 95 Othsr 1 6 2 - - TnrnM Lass than $15,000 48 14 10 45 $15,000 - $30,000 23 21 25 34 $30,000 - $45,000 8 17 15 15 $45,000 - $60,000 5 17 27 6 Graatar than $60,000 17 30 23 — Junior high or lsss l — 2 4 Sosa high school 12 3 3 9 High school graduate 7 6 14 20 Vocational/trado school 6 3 6 16 floss collogo, no dograo 30 21 19 18 2-yoar collags degree 21 10 19 14 4-year college degree 14 23 2 7 Graduate work/prof 7 34 34 13 36 Table 2 (continued) Dmmographic ChmrmctmrlMtlcg of hfommn by Age Percent of Sasple by Age Group Variable 18-30 31-45 46-65 66 + Single Harried Cohabiting Divorced/aeparated Widowed Professional Technical Hanager/Adsin Clerical Sales Service Other Mot employed Petired Children in hose Children out of hose No children Pull-tise Part-tise Hot student 20 67 8 12 7 6 18 6 12 2 38 37 63 49 11 40 52 24 21 2 2 38 5 13 13 5 5 22 56 1 42 22 14 16 61 5 24 8 2 26 5 10 11 31 16 11 76 13 9 14 76 16 7 24 53 2 4 9 79 91 9 100 37 Table 3 X• and ANOVAS tor Demographic Chmr»ctoristic* and Ag« Group _____Var labia__________________IB-30 31-45 46-65 66+ Ethnicity (%) Caucasian 55 71 81 95 All others 45 29 19 5 X*{ 3) - 29.13*** Vocational gtattia (%) Prof ass ional/aanageaent 18 51 35 2 Blue/white collar 44 27 18 11 Unemployed 38 22 31 9 Retired — — 16 79 X*(9) » 183.15*** Married/cohabiting 25 54 63 16 Single 67 24 5 7 Divorced/widowed 8 22 32 86 Xa(6) - 137.3**** Children 37 57 87 91 No children 63 43 13 9 X*{3) - 60.81*** Tnnn— (X income) 2.20 3.38 3.28 1.81 F( 3,256) - 19.01*** Education fff yaara) 3.93 4.7? 4.43 3.83 F(3,25B) * 9.16*** * p < .05; ** p < .01; •** p < .001 38 least 50% of each non-retired age group was employed, although the 31-45 age group had the greatest rate of employment. The 18-30 age group demonstrated the lowest rate of employment, presumably due to mothers at home with young children and young adults with full time student status. Another significant pattern in the demographic characteristics occurred with the 31-45 age group. By and large, this group was the best educated, most employed, and most highly paid of all the age groups. They also held the greatest percentage of professional level jobs. As this is the population that would have benefitted most from socioeconomic changes precipitated by Affirmative Action and the women's movement, these findings are not unexpected. However, even this group experienced typical patterns of marriage, divorce, and parenthood, suggesting that women of this era— the ones in this study, in any case— increased the number of social roles held, rather than opting out of others. The most striking differences between age groups occurred on the variable of ethnicity. Minorities were over-represented in the 18-30 age group (52%), probably due to the utilization of an inner-city junior college and DPSS training program as sites for data collection. The two middle age groups— with 28% and 19% minority populations, respectively— more closely approximated 39 actual county demographics. In the 66+ age group, with only 5% of the total, minority participation was clear ly under-represented. Although adequate numbers of older minority women were recruited to participate in this study, many declined— either verbally to this investigator or by default (i.e., not returning the questionnaire)— suggesting a cultural pattern of reti cence in this population. To determine whether demographic characteristics varied across age groups for minority and Caucasian subjects, chi-square and ANOVAs were again utilized (age groups were collapsed to maintain adequate cell sizes). As illustrated in Table 4, there were signifi cant differences between minority and Caucasian sub jects on most of the demographic variables for the 18- 45 age group. Contrariwise, there were no significant demographic differences between Caucasian and minority participants in the 46-66+ age group. Paralleling national figures (U.S. Department of Labor, 1992), for the 16-45 age group twice as many minorities were unemployed as Caucasians, and those who were employed were under-represented in white collar professions, similarly, income and years of education were signifi cantly lower for these minority participants than for Caucasians. On the variable of marital status, twice 40 Table 4 X* end ANOVA Diffmrmnc* Tmmtm by Ethnic Group ___________________________________31 ~ 45___________46 - 66+ Variable Cau- All Cau- All ______________________________ caalan Othara caalan Othara i mil 1 (%) Profeeaional/managament 36 22 16 40 Blue/white collar 40 33 16 — Unemployed 22 45 19 27 Retired — — 48 33 X*(2)-9.16** X*(3)-7.10 Married/cohabitng 49 20 37 76 Single 36 64 6 7 Divorced/Widowed 13 16 57 24 X*(2)-12.52** X*( 2 )*5. 02 P«r«w»hnnH (%) Children 40 56 SB 93 Mo Children 60 44 12 7 X*(1)-3.6B X*(l)-.33 (x income) i nn (x year a) 2.87 2.30 2.60 2.58 F( 1,143)—4.43* P(1,111)—.003 4.50 3.99 4.13 4.26 F(l,143)-4.43+ Pfl,111)-.003 * p < .05; ** p < .01; *** p < .001 41 as many Caucasians were married as minorities, although no differences were found in the divorce rate. In addition, no significant differences were found on the variable of parenthood. Due to the age span of these subjects, recruitment occurred through a variety of sources. A number of questionnaires were distributed by this researcher and associates at various work sites and social organiza tions and later returned by mail. These sites were widely representative— including primary and secondary schools, hospitals, libraries, church groups, agencies, fitness centers, and professional womens' organiza tions— and ultimately made up 50% of the sample. The remainder of the population was recruited from two colleges, a Department of Public Social Services training program, and two senior citizen organizations. Younger women (approximately 18-30 years) were solicit ed from social science classes and special vocational programs at a local junior college and state college. Both younger and middle-aged women were recruited from the job training program, older women were obtained from two senior citizen organizations which provide peer counseling for other seniors. Participants from these identified sites provided the final 50% of the sample. 42 naggripfcinn of Tngtirument.iit.iQn As seen in Appendix C, a questionnaire soliciting demographic data and information about the relevant variables was developed. Data were collected on age, education, ethnicity, income, employment, age and number and residence of children, and marital status, in addition, each subject filled out the Self-Descrip tion Questionnaire III (Karsh, in press). A copy of this instrument may be found in Appendix D. fiplf-naaciMpti nn Quant-Annnni ra ttt . Because it mea sures different aspects of the self-concept, the Self- Description Questionnaire III (SDQ III) was chosen to measure attitudes toward the self. The SDQ III is designed to measure 13 factors of self-concept: mathe matics, verbal, general academic, problem solving/creative thinking, physical abilities, physical appearance, same sex relationships, opposite sex rela tionships, relations with parents, reli gion/spirituality, honesty/reliability, emotional stability, and general self. Each of the 13 SQD III scales is represented by 10 or 12 items, approximately half of which are negatively worded. Subjects respond on an 8-point response scale where categories range from definitely false (1) to dafinitaly true (6). Due to the focus of this study, only nine dimensions of the SQD III were utilized: problem solving/creative think 43 ing, physical abilities, physical appearance, relation ships with sane sex peers, relationships with opposite sex peers, religion/spirituality, honesty/reliability, emotional stability, and general self (hearafter re ferred to as general-esteem). The construct validity of the SDQ III is solidly established and is based on both within-network and between-network investigations. Early SDQ research, primarily within-network studies, focused on the inter nal characteristics of self-concept, particularly its facets and organization. Numerous conventional and confirmatory factor analyses of self-ratings and re sponses of significant others have identified the 13 dimensions of self-concept that the SDQ III is designed to measure (Harsh, Byrne, and Shavelson, 1986; Harsh, Relich, and Smith, 1983; Marsh and Shavelson, 1985; Marsh, Smith, and Barnes, 1985). Furthermore, correla tions among the facets has been surprisingly small, with average correlations of only .2. Additional multitrait-multimethod (MTMM) analyses, based on Camp bell and Fiske criteria, have strongly supported both the convergent and divergent validity of the SDQ III factors (Marsh, Barnes, and Hocevar, 1985; Marsh and Byrne, 1989; Marsh, Richards, and Barnes, 1986). According to the SDQ III test manual (Marsh 1989), the normative archive of responses used to estimate reli- 44 ability for the 13 SDQ III factors obtained alpha coefficients from .76 to .95 (median * .89). Other reliability studies (Harsh and O'Neil, 1984; Harsh and Shavelson, 1985) have yielded coefficients between .88 and .90, suggesting that the SDQ III is a highly reli able instrument. More recent SDQ research— between-network stud ies— has examined the relationships between dimensions of self-concept and a wide variety of other constructs, including sex (Marsh, 1985), academic performance (Marsh and O'Neil, 1984), responses to other self- concept measures (Marsh and Richards, 1988), self- concept inferred by significant others (Marsh, Smith, and Barnes, 1985), and attributions for success and failure (Marsh, Richards, and Barnes, 1986). SDQ responses have been systematically related to these criteria in a way that is consistent with the theory, providing additional support for the construct validity of the SDQ ill. social utilizing data from the questionnaire, social roles were defined as paid worker (hours em ployed outside the home or 0), student (none, part, or full time), parent (age of youngest child in the home or 0), and marital status (single, married/cohabiting, divorced/widowed). These particular social roles were 45 chosen because they were deemed most representative of those held by women today. salf-Concept. For the purposes of this study, self- concept was defined and measured multi-dimensionally. Utilizing the Self-Descriptive Questionnaire III (Marsh, in press), nine dimensions of self-concept were assessed: problem solving/creative thinking, physical ability, physical appearance, relationships with same sex peers, relationships with opposite sex peers, reli gion/spirituality, honesty/reliability, emotional stability, and general-esteem. Prnrftduras This investigation was an exploratory field study involving survey research at numerous community loca tions. Contact with the directors and administrators of utilized sites was initiated a full year before actual data collection began, as permission was re quired from a number of bureaucratic systems. Those questionnaires obtained from identified sources (i.e., colleges, senior organizations, and job training pro gram) were distributed and collected on-site by this investigator, following a brief description of the study. Other questionnaires were selectively distrib uted by the researcher and associates at various work settings and organizations, along with a stamped and addressed envelope to facilitate return (approximately 46 80% were returned). All questionnaires— both mail-ins and those personally collected— included a cover letter (see Appendix B) which explained the purpose of the research and gave instructions for completion. Al though the questionnaires were coded for anonymity, participants were invited to leave an address with the researcher or on their envelope to receive a summary of the findings. 47 RESULTS Chapter Overview This chapter of the dissertation will deal with the presentation of the results. The chapter has been divided into five sections which correspond with the treatment of the data, reliability figures, intercor relation figures, and the two research questions of the study. Trmttnant af the Data To assess the relationship between age and self- concept, Pearson's correlations were first run between age as a continuous variable and the nine dimensions of self-concept. To test for non-linear relationships between age and self-concept, four new age groups were created (18-30, 31-45, 46-65, and 66+) and a series of one-way analyses of variance (ANOVAs) conducted between the age groups and each dimension of self-concept. In addition to the analyses of variance, Scheffd tests were used to determine further significance of differ ences between the means. The confidence level was established at .05. To investigate the relationship between different dimensions of self-concept and role, four predictor variables ware used to define role: marital status (single, married/cohabiting, divorced/widowed), parent- 48 hood ++{age of youngest child or 0), vocational status (hours employed outside the home or 0), and student status (none, part time, or full time). The parent* hood, vocational, and student variables were all treat ed as continuous data. Because the variable of marital status was defined categorically, however, it required dummy coding and some initial statistical procedures prior to inclusion in the study. Because it was felt that the divorced/widowed factor of marital status might be confounded with age, zero-order correlations were run between the nine dimensions of self-concept and the three marital cate gories for two newly created age groups (18-45 and 46+). As significant correlations were found for each level of marital status in the 46+ group, correlations were rerun, partialing out the effects of age. This time no significant correlations were found between the divorced/widowed factor and age, and the decision was made to eliminate this category of marital status from the study. To explore the relationship between self-concept and role, standard multiple regressions were performed between each dimension of self-concept (criterion variable) and vocational status, parenthood, student status, marriage, and single status (predictor vari ables) for four age groups of women (18-30, 31-45, 46- 49 65, 66+). Interactions of the predictor variables (as a block) were also conducted. Reliability. According to the unpublished test manual, the Self-Descriptive Questionnaire III (SDQ III) is intend ed for use by late adolescents, generally in the age range of 16-25 (Harsh, 1969). Although the author reports that the SDQ III may also be appropriate for older respondents, no normative data is available for this population. Hence, the data resulting from this study are of particular interest. Coefficient alpha estimates of reliability, based on the internal consistency of responses, have been presented in the SDQ III manual. Coefficient alphas for the 9 dimensions of self-concept utilized in this study vary from .74 to .95 (median ■ .90), and only the coefficient alpha of the honesty/reliability factor is less than .84. These alphas were computed from 2,436 sets of responses, 2/3 of which were students, 37% of whom were women, and only 7% over the age of 26. For this study, coefficient alpha estimates of reliability were computed for the total sample, a group 18-45, and 46+ group. In Table 5 you will find relia bilities for each of the nine scales for the total sample, the two age groupings, and the normed popula tion from the SDQ III manual. For the total sample, bO Table 5 Reliability Coefficients tor tha Nine Dimensions of Self-Concept SDQ Seal* Horsed Population Total Population 18-45 46+ Cognitive Ability .84 .79 .79 .78 Physical Appearance .90 .89 .87 .90 Save Sax Ralationshipa .87 .78 .79 .76 Physical Ability .94 .94 .94 .93 Religion/Spirituality .95 .93 .94 .92 Ganaral Eataas .93 .93 .93 .94 Honasty/Rallability .74 .70 .76 .58 Opposite Sax Relationships .92 .88 .86 .90 Esotional Stability .89 .90 .89 .92 51 coefficient samples ranged from .70 to .94 (median * = .89); with the 18*45 group, alphas varied from .76 to .94 (median = .86); the 46+ group had a very wide distribution of alpha coefficients from .58 to .94 (median - .90), but only the factor of hones ty/reliability was less than .76. Considering the differences between this sample and the normed popula tion for age, sex, occupation, and— presumably— life experiences, there was remarkably little diversity in the reliability figures. Tntflrcorrnlations of +ha Nine DimflnaionB In order to examine the degree of association among the nine dimensions of self-concept, correlation matrices were run between all nine dimensions for the total population. As illustrated in Table 6, the highest correlation (r - .62) occurred between self- concept for emotional stability and self-concept for general esteem. Correlations between the remaining dimensions were modest (median r = .20). To test for the possibility that degree of associ ation might differ as a function of age, correlations were also run between the nine dimensions for the 18-45 group and then for the 46+ group (see Table 7). Uti lizing Fisher R to z transformations to test for dif ferences between these pairs of correlations, signifi cant differences between age groups were found only on 52 Table 6 Correlations Between the Nine Dimensions of Self-Concept for Total Population (N - 265) U) fit (1 ) (* ) (1 ) (1 ) r Gopitifi Ability 1 .0 0 Apical tfpMTttCt .3 0 * 1 .0 0 s m Sn Kalttioalipa .IT * .1 1 * 1.0 0 Apical Ability .»* .3 4 * ,1J* 1 .0 0 h lifi* / Spirltality .0 7 -.07 .07 -.17* 1.0 0 Omni b tM .3 4 * .U* .4 1 * .1 0 * .0 1 1.0 0 iBMrtf/ k l lability .1 7 * .10 .1 1 * .0] .2 1 * .2 2 * 1.0 0 O p p o a ita Sn StltUonbifi .3 4 * ,1 7 * .3 1 * .]]* -.10 .4 3 * -.10 bOtiflMl .1 1 * .»* ,M * .0 1 .0 3 .1 3 * .1 4 * Strtiilitj *p< M 53 Tabic 7 Corral ations Batwaan the Nina Dimmnmions of Salf-Concapt tor 18-45 and 46+ Populations of Woman (1 ) (2) (3 ) Ul til (* ) («) HI O ognitiva Ability 1 ,0 0 .3 1 * .»* . w .15 .1 4 * ,2V .2 0 * .3 0 * Apical llfttru o t .1 7 * 1 ,0 0 .0 7 . » * -.01 .4 5 * .2 5 * .4 3 * .1 4 * Sat sat Itlatioethipt . 1 2 * .3 3 * 1 .0 0 .2 2 * .0 5 .3 1 * .0 1 .1 0 * .3 1 * Apical Ability .2 0 * .3 5 * .1 1 1 .0 0 -.02 .1 1 .2 0 * .2 0 * .2 1 * kligion/ sp iritalitj -.01 .0 7 .0 3 -.10* 1 .0 0 .1 0 .2 7 * -.01 .0 1 Carnal btaa .it* .4 3 * .St* .2 0 * - . O f 1.0 0 .3 3 * .3 5 * .1 3 * O aatf/ AliAility . 1 0 * .0 2 .1 4 -.02 .2 4 * .1 1 1 .0 0 .0 2 .1 4 OffMita la Itlitin a tip .3 1 * .4 7 * .S I* .2 0 * -.00 .5 0 * .0 3 1.00 .3 4 * kotiooal Stability .3 5 * .»* .IS * .1 0 -.03 .IS * .1 0 .4 0 * 1 .0 0 » f < .OS loti: Comlttiw aoq IMS ( D ■ 1 4 7 } an 4 m tfe t diapml ad oorralitiea for tte 4 * + poop (o ■ Ul) « b a lo a tit diafoaal. b4 the paired correlations for sane sex relation ships/opposite sex relationships and physical appear ance/same sex relationships, indicating minimal inter correlation between dimensions as a function of age. Salf-Concapt and Age Research question 1 asked whether different dimen sions of self-concept vary as a function of age. Corollary hypotheses proposed that: (1) self-concept for physical appearance will decrease with age, (2) self-concept for physical ability will decrease with age, and (3) self-concept for religion/spirituality will increase with age. Pearson's correlations were run between age and all nine dimensions of self-concept, revealing a linear relationship to age in six of the dimensions. These are presented in Table 8. From these correlations it can be seen that self-concepts for physical appearance, physical ability, and opposite sex relationships are negatively correlated with age (p < .001), indicating that self-regard in these areas is significantly lower for older women than for younger women. On the other hand, self-concepts for religion/spirituality (p < .001), honesty/reliability (p < .001), and emotional stability (p < .01) were found to be positively corre lated with age, suggesting that greater self-percep tions of honesty, spirituality, and stability are 55 Tabla 6 Correlations Between Age and the Bine Dimensions of Self-Concept SDQ Seal* Age Co^nitiva Ability - .10 p - .111 Physical Appaaranca - .24 p “ .000 Saaa sax Ralationships .10 p - .103 Physical Ability - .37 p " .000 Raligion/Spirituality .29 p “ .000 Canaral Estaaa - .02 p - .70 Honas ty/Ral iabi 1 i ty .22 p " .000 Opposlts Sax Ralationships - .31 p " .000 Bnotional Stability .16 p “ .008 56 associated with older age. No significant linear relationships were found between age and the self- concept dimensions of cognitive ability, same-sex relationships, or general-esteem. A series of one-way analyses of variance (ANOVAs) were also performed to investigate the possibility of discontinuous relationships between dimensions of self- concept and age. For this procedure, four new age groups were created: 16-30, 31-45, 46-65, and 66+. As seen in Table 9, ANOVAs run between the age groups and each of the nine dimensions revealed significant dif ferences on the dimension of cognitive ability— with the 66+ group reflecting the weakest scores, but the 31-45 and 46-65 groups reflecting the strongest— indi cating that, for this population, middle-aged women view themselves as most intellectually competent and capable. Significant differences were also found between age groups and the dimension of same sex rela tionships. The 18-30 group experienced the weakest scores and the 46-65 group the strongest, suggesting that these late middle-aged women greatly value and participate in relationships with other women. Of further note, while self-concepts for physical appear ance and physical ability were earlier linked to age in a linear fashion, ANOVA revealed that there were no 57 Table 9 ANOVA Results for tha Four Age Croups and Hina Dimanmiona of Saif Concapt SDQ Scales 18 - 30 31 - 45 46 - 65 66+ F Value Cognitive Ability 5.56 5.80 5.73 5.31 3.04* Physical Appearance 5.58** 5.40 4.99* 4.88* 5.02** Saee Sex Relationships 5.68* 6.01 6.30 5.89* 4.72** Physical Ability 5.40** 5.59** 4.44** 3.98" 15.59*** Religion/ Spirituality 5.36“* 6.08* 6.21* 6.51* 6.86*** General Esteee 6.31 6.40 6.18 6.28 .34 Honesty/ Reliability 6.25*** 6.69* 6.77* 6.64* 8.80*** Opposite Sex Relationships 5.85*** 5.63* 5.63* 4.81* 9.24*** Sectional Stability 5.05 5.12 5.39 5.59 2.15 * p < .05; ** p < .01; *** p < .001 58 significant differences between the two youngest groups on both of these dimensions, indicating that deterio- rating self-worth in the areas of appearance and physi cal ability is first associated with late middle age. Only one dimension of self-concept— that of gener al esteem— was not assoociated linearly or discontinu- ously with age, probably due to its relatively high correlations with the dimensions of emotional stability (r - .62), opposite sex relationships (r * .43), physi cal appearance (r = .44), and same sex relationships (r - .45). Sfllf-goncapt and Role Research question 2 asked whether different dimen sions of self-concept vary as a function of roles (at different ages). Measures of role included single status, marital/cohabiting status, parenthood, voca tional status, and student status. The relationship with age was investigated using the four previously created age groups: 18-30, 31-45, 46-65, and 66+. A series of multiple regressions were conducted to determine the importance of these roles in predicting self-concept (defined multidimensionally) for different age groups of women. As illustrated in Table 10, for the 18-30 age group the multiple correlation for self- concept for honesty was significant, although none of the main effects were. This aberration appears to be 59 T a b l* 10 Multiple Regression Analyses of the Five Predictor Variables with Mine Dimensions of Self-Concept for 18-30 Year-old Women O D ^ D i- tin Utility ftjsial ifpMIUCt * 1 lit nrtiai Utility klifito/ Spiritu ality G M M T tl b tm *m tT/ niiibility flppotit* S a t Mt- tunbipt totisnal Stability Prattbood -.04 .03 .10 -.13 .07 -.05 .10 -.14 -.30 ftploynnt .02 -.11 .12 -.03 .10 .13 .12 -.06 . 0 9 S c fto o l -.02 .10 .10 .15 .04 .20 .20 .11 .17 ■uriad .27 -.10 -.45* .00 .20 -.16 .07 -.03 .14 Suqli . » .03 -.43* .10 .07 -.10 -.22 -.02 .14 1 .10 .22 .30 .17 .24 .21 .40 .11 .23 I2 .03 .O S .09 .0) .06 .04 .16 .03 .05 r .42 .70 1.53 .41 .17 .70 2.7* .54 .It » p < . O S ; • • p < . 0 1 ; * * * p < . 0 0 1 60 the result of the relatively strong correlations (and thus shared variance) between self-concept for honesty and the variables of employment, marriage, single status, and student status (r - .22 - .27). In addi tion, the role variables of marriage and single status were both found to predict lower self-concept for same sex relationships in this age group; as few of this group are divorced, this would seem to indicate that most of the 18-30 year-old women in this study attach relatively little importance to relationships with other women. Table 11 demonstrates the impact of employment on self-concept for women in the 31-45 age group. Employ ment contributed significantly to self-concepts for same sex relationships, opposite sex relationships, and emotional stability, indicating that employed women in this age bracket experience greater self-worth in both male and female relationships than do unemployed women, as well as a greater sense of emotional stability. Self-concept for opposite sex relationships was also significantly influenced by student status. In the 46-65 age group (see Table 12), the multi ple correlation for general esteem was significant, with the predictor variables of employment and single status significantly influencing this dimension of self-concept. The results indicate that single and/or 61 T a b l e 11 Multiple Regression Analyses of the Five Predictor variables with Nine Dimensions of Self-Concept for 31-45 Year-old Nomen Cnpi- tie Ability Fbysial Sat Su klitioo- tbip Apical Ability k lifia / SpiritB- ility Garni bt*a ta rty / laliibility tpntitt So k it- timbipa kotinal Stability tU B ttel .17 .06 .is -.03 -.22 -.01 .00 .05 .01 MoTBMt .11 .03 .41** .23 .17 .14 .05 .27* .34* school -.06 -.10 .17 .09 -.10 .05 -.13 .34* .09 krriid -.06 .04 .05 .01 -.16 .12 -.15 .30 .22 Siajlt .07 -.10 .01 -.10 -.13 -.06 -.10 .07 .03 1 .31 .17 .41 .26 .30 .20 .11 .31 .36 I1 .10 .03 .23 .07 .09 .04 .03 .14 .13 T 1.21 .11 3.46** .12 1.15 .41 - 1.92 1.73 * p < . 0 5 ; * * p < . 0 1 ; • * * p < . 0 0 1 62 Tabitt 12 Multiple Regression Analyses of the Five Predictor Variables w i t h Mine Dimensions of Self-Concept for 46-65 Year-old Women O pi- tiw Uiilitj h p in l IppM T W C t Pm 3a klation- thlp Partial ibilitj Itllfico/ spirita- ilitj hi tM ftr/ k liftilitr O p iitt S a k l f tioahipi botiooii Stability Pirthnrtt -.07 -.32 .11 -.15 -.13 .02 -.02 -.14 .06 .16 ■ 0 2 .25 .17 -.21 .26* -.03 .12 .24 5t*ol 1 s .04 .11 .21 .02 .10 -.03 .14 -.00 ftrriid -.12 -.10 .06 -.07 .10 -.02 -.00 .09 -.01 Siaglt -.06 - . » -.17 -.11 -.02 -.3*** -.16 -.21 -.17 1 .23 .11 .35 .37 .26 .47 .11 .37 .29 I* .05 .17 .12 .14 .06 .22 .03 .14 .01 r .65 2.24 1.59 1.14 .10 3.20* .37 1.76 1.04 • p < . 05; ** p < ,o i; *•* p < .001 63 employed women in this age group enjoy lesser and greater self-esteem, respectively, than married or unemployed women. As seen in Table 13, none of the equations or the predictors were significant for the 66+ age group. Interactions of the predictor variables were also tested as a block for each age group, but no block entries were found to be significant. As an additional analysis, after controlling for the linear effects of age, a multiple regression for the total sample was run to see whether any of the role variables had a significant effect on the nine dimen sions of self-concept. Table 14 reflects a number of effects corresponding with earlier analyses, e.g., the positive impact of employment on self-concepts for same sex relationships, general-esteem, and honesty. Addi tional new findings were also found: the predictor variables of parenthood and employment significantly influenced self-concept for physical ability (although in different directions), suggesting that women in this study with children at home experience a diminished sense of physical ability, but that employment seems to mitigate this effect, enhancing perceptions of physical competence. 64 Table 13 Multiple Regression Analyses of the Five Predictor variables with Nine Dimensions of Self-Concept for 66+ year-old Nomen copi- tin Ability Mqrtiol ifporuct S tt SB k iltie - ibipi 1 n p ia l Ability kliqiaa/ ipirito- ality i i iMMtf/ fclieility Gppttit* S a h lt- tioadiipi hfitiml Stability _ __ _Lt. - -J nrtraoM fcplC fM B t -.17 -.10 .11 .09 .11 -.03 .05 .03 -.14 Sfeol t fcrud -.01 -.06 -.03 -.05 -.01 -.17 -.07 .12 -.19 Siiplt -.16 .11 -.05 -.06 .07 -.21 -.24 -.04 -.24 1 .13 .16 .14 .13 .14 .25 .25 .13 .30 I* .05 .01 .03 .03 .0} .06 .06 .11 .01 1 .95 .44 .11 .31 .M 1.74 1.11 .21 1.66 ■ p < .05; « p < .01; *« p < .001 65 Table 14 M u l t i p l e P e y r e s s i o j i Analyses of the Five Predictor variables with Nine Dimensions of Self-Concept for Total Population, Controlling for Age C n g n i* tin Ability Apical A p p M K B C t S n i t x blititr- itipe Apical Ability taliqioe/ Spirita- ality Gwnl b u n ftanty/ kliAility oppoaiu S e t kli- tlOM bipf botioail Stability bp -.12 -.19* .10 -2.5** .36*** .02 .17 -.26** .11 hrntbood .04 -.01 .12 -.16* -.05 -.04 .01 -.04 -.02 bplaywrt .11 .01 .23*** .13* .11 .15* .15* .04 .12 S c h o o l -.05 .05 .04 .12 .06 .14 .09 .13 .03 tarried -.01 -.06 -.01 -.05 .01 -.07 .04 .11 -.02 Sibflt -.02 -.01 -.10 -.05 .00 -.14 -.22* -.06 -.07 1 .15 .26 .10 .42 .32 .20 . » ,37 .21 l ' .02 .07 .09 .17 .10 .04 .13 .14 .04 r 1.14 3.1*** 4.22*** 9.13*** 4.93*** l.tl 6.39*** 6.10*** 2.00 • p < . 0 5 ; * * p < . 0 1 ; • * * p < . 0 0 1 66 Sum m ary nf RggulHs The results of this study indicate that, for this population, age is the most relevant variable related to several dimensions of self-concept; that is, self- concepts for religion/spirituality, honesty/reliabili ty, and emotional stability were found to be positively correlated with age, while self-concepts for physical appearance, physical ability, and opposite sex rela tionships were negatively correlated. Self-concepts for cognitive ability and opposite sex relationships were also related to age (or, in this case, age group), but in a discontinuous fashion. Although the data were analyzed a number of different ways, role was found to have relatively little influence on self-concepts. Full-time employment was the only predictor for any aspects of self-concept, and then only for the middle age groups. In summary, the 18-30 age group demonstrated the lowest self-concepts of this population for reli gion/spirituality, honesty/reliability, same sex rela tionships, and emotional stability. However, they exhibited the highest self-concepts of this study for physical appearance and opposite sex relationships. Role--marital status, employment, or parenthood— was not a predictor for any aspects of self-concept with this group. In contrast to these results, while the 66+ age group reflected the lowest self-concepts of the sample for physical appearance, physical ability, and opposite sex relationships, they also experienced the highest self-concepts for religion/spirituality, emotional stability, and honesty/reliability. Role was not a predictor for any dimensions of self-concept with this age group. The two middle groups, 31-45 and 46-65, presented the highest self-concepts for this study for cognitive ability and same sex relationships. They were also the only age groups to be influenced by role, notably employment. For the 31-45 age group, employment was associated with greater self-concepts for same sex relationships, opposite sex relationships, and emotion al stability, while employed women in the 46-65 group demonstrated higher general-esteem. Parenthood was not a predictor for any facets of self-concept, but being single was associated with lower general-esteem for the 46-65 age group. 68 DISCUSSION Review nf PurpnRB and nhjerd-i ves The purpose of this study was to examine the relationships between multiple dimensions of self- concept and the roles of marital status, vocational status, and parenthood for four age groups of women, utilizing the SDQ III. Specific objectives included: 1. To determine whether the SDQ III is a valid and reliable instrument for measuring self- concept in adult women. 2. To determine whether different dimensions of self-concept vary as a function of age. 3. To determine whether different dimensions of self-concept vary as a function of different roles (at different ages). It was hypothesized that: 1. Self-concept for physical appearance will decrease with age in women. 2. Self-concept for physical ability will de crease with age in women. 3. Self-concept for religion/spirituality will increase with age in women. Findings Raaad an Rflaoorrh Questions The analysis of the data collected relative to the principle objectives of the study indicated that, for 69 this sample of women, age is the primary predictor for eight dimensions of self-concept. More specifically, self-concepts for physical appearance, physical abili ty, and opposite sex relationships were negatively correlated with age, while self-concepts for reli gion/spirituality, honesty/reliability, and emotional stability were positively correlated. Self-concepts for cognitive ability and same sex relationships were also related to age, but in a non-linear fashion. Only the dimension of general-esteem was not correlated with age in any fashion. Although role— primarily full-time employment and single status— was found to influence some dimensions of self-concept in middle-aged women, it was not a predictor for other dimensions of self- concept at other ages. Major developmental theorists such as Levinson (1978), Gilligan (1980), and Neugarten (1976) have all addressed the seemingly paradoxical gains and deficits associated with aging. Thus the relationships found in this study between age and multiple dimensions of self- concept are not particularly surprising. Intuitively, it makes sense that self-concepts for physical appear ance, physical ability, and opposite sex relationships might decrease with age, due to a pervasive youth culture which associates strengths in these areas with young adulthood. Contrariwise, self-concepts for 70 spirituality, honesty, and emotional stability would be expected to increase. It also seems reasonable that middle-aged women would experience the greatest self- concepts for cognitive ability and same sex relation ships. However, as discussed in Chapter 1, previous self-concept research has failed to empirically identi fy these relationships, or even consistently demon strate others. Although Gould's (1972) early research did not focus on self-concept per se, his findings did suggest that positive feelings about one's self might depend on different aspects of one's life at different times. Two decades later, utilizing a multidimensional model of self-concept, it has been hypothesized that the basis for positive or negative perceptions about oneself shifts with development from younger to midlife to older years, and the results of this study provide considerable support for this view. Much has been written about the positive benefits of employment for women, with current research empha sizing the impact of employment on women's self-esteem (Baruch and Barnett, 1986; Coleman and Antonucci, 1983; Pietromonaco et al., 1986). Paralleling these studies, this investigation effectively demonstrated the posi tive relationships between employment and self-concepts for middle-aged women. Employed women in the 31-45 year-old group were found to experience greater self 71 worth in both male and female relationships than unem ployed women, as well as a greater sense of emotional stability. Even full-time students in this age group— arguably the vocational equivalent of full-time employ ment— reflected greater self-concept for opposite sex relationships. Employment was also a significant predictor of self-concept for 46-65 year-old women, with employed women in this age group exhibiting great er general-esteem than unemployed women. Although research generally supports the positive effects of employment for women without children, this relationship has not been established for women with children. Both Kessler and McRae (1962) and Ross and Huber (1983) concluded that the benefits of employment for married mothers are significantly attenuated by domestic and childcare responsibilities, and other investigators have noted similar trends. In contrast with these findings, this study found no significant differences between the self-concepts of employed mothers and employed childless women. In fact— contrary to extensive studies (see Chapter l) examining the relationships between well-being and parenthood— for this population, only the dimension of physical ability was affected by the presence of children in the home. Although the impact of marital status on self- concept was largely insignificant in this study, of 72 some interest was the finding that single (i.e., never married) women in the 46-65 age group demonstrate significantly lower self-esteem than married women. This is not particularly surprising when one considers the fact that most of these women came of age at a time when marriage and motherhood were the primary socially sanctioned roles for women, and thus an integral part of their identity. Failure to participate in both of these roles would presumably result in the lower self esteem found in this age group of women. Much of the literature on midlife in women charac terizes it as a time of questioning and reevaluation; a period in which women are seen as experiencing in creased psychological anxiety and lower self-esteem. These data, however, suggest that being employed or in school may significantly mediate the distress often associated with midlife. The higher self-esteem, healthy social networks, and greater emotional stabili ty reflected in the middle-aged employed/student popu lation of this study lends support to the view that working women may, in fact, be more psychologically capable of coping with midlife than other women. 73 Other Findings Although the SDQ III is well established as a reliable and valid instrument for use with young adults— and reportedly is appropriate for older respon dents also--no normative data exists for an older population. Thus the data resulting from this investi gation were of particular interest. The results of this study indicate that all nine dimensions of self-concept are reliable for this older population and, furthermore, that the dimensions are relatively distinct. For this population, coefficient alphas varied from .70 to .94 (median ~ .89), contrast ing favorably with the normative sample coefficients ranging from .74 to .95 (median * .90). Although there was greater variability in the intercorrelations of the dimensions for the older versus normative population, average correlations among the dimensions for both populations was modest (median r - .20). The relative ly little diversity between reliability and intercor relation figures for young adult and older populations suggests that the SDQ III may also be a reliable tool for the general population. fnr Th«nr»y The most striking finding of this study is that— for this sample of women— age rather than role is the most important predictor of self-concept in women. 74 Although marital status and employment affected some aspects of self-concept for middle-aged women, and parenthood was associated with self-concept for physi cal ability, role was not a predictor for most dimen sions of self-concept at most age groups. Age, on the other hand, was strongly and consistently associated with eight of the nine dimensions of self-concept. An important implication of the findings reported here is that they are inconsistent with the assumptions underlying much of the developmental research on women today. Current studies typically select a specific life cycle phase or age group of women, then address the relationships between various roles and psychologi cal we11-being--the premise being that multiple or specific combinations of roles facilitate the develop ment of a richer, more complex view of self, which in turn enhances mental health. The results of this study, however, suggest that the developmental phase or age group itself may be the strongest predictor of a healthy self-system, and not role per se. This is of particular significance because it implies that self- concept is not a fixed, stable construct over time, but rather shifts with development throughout the life cycle. One possible explanation for these conflicting findings regarding the importance of role may lie in 75 the construct definition of psychological well-being. Using factor analytic techniques to examine the under lying structure of well-being for women, Brian and Veroff (1982) found empirical support for three dimen sions of well-being; (1) happiness/unhappiness, (2) strain (psychiatric symptomatology), and (3) personal adequacy (self-esteem). Most subsequent role-related research has utilized some variation of this construct as its criterion variable, apparently assuming that psychological well-being is positively correlated with a healthy self-concept. While personal adequacy (i.e., self-esteem) would almost certainly correlate with some aspects of self-concept, there is insufficient evidence to make similar arguments for the other two factors. Thus, while previous role-related studies have undoubt edly measured important components of well-being, they have not measured self-concept, and any comparisons of the effects of role between this investigation and the others must be made with caution. In our society, older age is typically viewed as a period of diminishing returns, and while this has not been substantiated by recent research (Baruch and Brooks-Gunn, 1984), neither has it been disconfirmed, primarily due to an abundance of inconsistent data. In utilizing a multi-dimensional model of self-concept, this study has been able to track the ebb and flow of 76 different dimensions of self-concept over the life course for four age groups of women— in particular, aged women. And while older women do indeed demon strate low self-concept in some areas of their lives, they also reflect the highest self-concepts of this research population in other areas— notably spirituali ty, emotional stability, and honesty— thus redefining the cycle of older age as a time of losses and gains. The overall findings of this study have interest ing ramifications for adult developmental theory, particularly the maturational theory of aging. As presented earlier, this perspective emphasizes stages of development and developmental tasks, suggesting that aging is a relatively positive process, and the results reported here generally support this view. That is, six of the nine dimensions of self-concept investigated were either unchanged or greater for older age groups than for younger groups. That three dimensions of self-concept were lower for the older age groups is not necessarily contraindicated with this approach, but rather lends support to the developmental and multidi mensional nature of self-concept over the life course, i.e. that self-perceptions may depend on different aspects of one's life at different times. It is impor tant to note, however, that there is nothing in the maturational perspective to suggest that loss of roles 77 is not difficult or problematic. Furthermore, as role theory would predict, in this investigation employment and student status were positively correlated with some dimensions of self-concept for the middle age groups of women. Thus, while this study provides strongest support for a maturational approach to aging, it in no way denigrates the relevance of role theory. Alterna tively, it is proposed that aging must be treated as a dynamic process, one in which role theory by itself cannot adequately explain the relationship between age and self-concept. The results of this study also have important implications for further theoretical work in self- concept. Strong evidence was presented for the mul tidimensional ity of self-concept for a mature female population, as well as the relative distinctness of these dimensions. As with previous SDQ III research (Marsh, Barnes, and Hocevar, 1985), however, the gener- al-esteem scale was significantly correlated with the dimensions of emotional stability, same sex relation ships, opposite sex relationships, and physical appear ance. For the normative young adult population, the small size of these correlations has still suggested that general-esteem has little effect on self-concepts in other areas. In this investigation, however, sub stantially greater correlations were found between 78 these dimensions, indicating that— for older adults and/or mature women— general-esteem may have more superordinate qualities. Harsh refers to general-esteem (roughly the equiv alent of self-esteem) as an "experimental scale" and conceptualizes it as a relatively unidimensional con struct that is not specific to any particular dimension of self-concept but could be applied to each. Although he stresses that the other dimensions of self-concept— rather than general-esteem— should be the primary basis for interpretation of the SDQ III, further research is clearly needed to determine the relation between gener- al-esteem and self-concepts in other areas, at least for an older population. Practical Tmpl<catinna Traditional role theory suggests that the compet ing demands of different social tasks produces role strain or conflict in women (Goode, 1960; Merton, 1957; Sarbin and Allen, 1968; Slater, 1963). In contrast, more recent theories propose that individuals may profit from enacting multiple roles (Barnett and Baruch, 1981; Harks, 1977; Sieber, 1977; Thoits, 1983). Alternatively, the results of this study indicate that role contributes relatively little to women's self perceptions. From a mental health perspective, it is important to know whether holding multiple roles in 79 creases or decreases psychological risks. For the practitioner, understanding the risks and benefits attached to specific roles is integral to planning treatment interventions, developing programs, and formulating public policies. Yet in spite of extensive empirical and qualitative research, the nature of the relationship between role and self-concept or self esteem is as ambiguous as ever. In their roles as workers, spouses, and parents, women experience both suffering and gratification. Comtemporary role theorists— focusing on the relation ships between role and well-being— have highlighted gender-specific expectations and experiences in various social roles, suggesting that it may be qualitative rather than quantitative aspects of women's involvement in these roles that contribute most significantly to the development of self-concept or self-esteem (Barnett & Baruch, 1985; Baruch £ Barnett, 1986; Rosenfield, 1985). Furthermore, women who have greater self-esteem or better developed self-concepts may be more likely to acquire or preserve a large number of roles. One challenge for future research will be to clarify the qualitative dimensions and directionality of this relationship. The enhancement of self-esteem/self-concept has long been an important goal for academic and mental 80 health settings. Thus, the measurement of self-con cept— and its effect on other constructs— is recognized as a valuable outcome. Unfortunately, intervention studies have typically failed to demonstrate any mea surable effects on self-concept, possibly due to the use of instruments not designed to measure specific components of the self-system. Utilizing the SDQ I and II with adolescents, systematic interventions have reliably predicted and measured changes in both academ ic and non-academic dimensions of self-concept, and similar studies are currently underway with the SDQ III. At this time, however, no instrument has been developed that reliably measures intervention-based changes in self-concept. Although changes in self-concept cannot yet be evaluated with any surety, this does not preclude the use of the SDQ III as an investigatory tool, particu larly in counseling and mental health settings. The results of this study indicate that the SDQ III might reliably be used with clients to explore their relative strengths and weaknesses in the area of self-concept. And although more research must be done to obtain normative data on mature adults— males as well as females— eventually clinicians would be able to compare different aspects of their clients' self-concepts with standardized norms for their age group or life cycle. 81 The relationships between self-concept/self-esteem and other clinical conditions, such as depression or physi cal illness, have also recently generated much inter est, and could readily be investigated with the SDQ III. Helping professionals have only recently begun to explore the gap between adult development theory and practice. Currently there is increased interest in incorporating developmental perspectives in clinical practice that emphasize psychological growth and pro cesses of change, and the findings of this study have particular ramifications for such an approach. During young adulthood strong societal norms push individuals toward the development of commitments. Lowenthal (1977) suggests that three areas of commit ment— moral, interpersonal, and mastery or competence— are particularly significant during this period, and the results of this investigation bear this out. For this study, the 18-30 age group experienced the lowest self-concepts of the population for spirituality/reli gion, honesty/reliability, and emotional stability, suggesting concerns or deficits in the area of moral development, as well as anxiety related to personal competency. During this period, young adults are typically at their peak physically and are also engaged in the most numerous relationships of their life spans; 82 these strengths are reflected in the highest self- concepts of the sample for the 18-30 group in physical appearance, physical ability, and opposite sex rela tionships. Young adulthood is a time of transition and com mitment, whether to another person, to a job, or to a moral principle. Counseling issues with this popula tion typically focus on the conflicts and anxieties associated with intimate relationships (e.g., separa tion from family, dating, courtship) and major life choices (e.g., career, marriage, parenthood). A prima ry goal for the mental health practitioner would in clude helping the individual work through these devel opmental tasks in a healthy and empowering fashion. Midlife research has only emerged in the past two decades; consequently our understanding of middle adulthood is limited and conflicting images abound. Some views of middle-age emphasize its stability; the life course is seen as a predictable and orderly set of transitions. Other views emphasize the changes which occur during this period, perceiving it as a time of great personal crisis which then precipitates adapta tion or growth. The findings of this investigation lend support to a theory of midlife (and life course) change, at least insofar as self-concept is concerned. That is, the 83 midiife women in this study demonstrated less self- concept for oppostite sex relationships, physical appearance, and physical ability than the youngest age group, but more than the oldest group. Alternatively, they showed greater self-concept for honesty, religion, and emotional stability than the youngest group, but less than the oldest group. They reflected the highest self-concepts of the population for cognitive ability and same sex relationships. The multidirectionality of these self-concepts implies that midlife is a mixture of both negative and positive changes, of peak perfor mance as well as challenging transitions. Nidlife, then, represents a period of both stabil ity and change for many women. On the one hand, most midlife women continue to enjoy good physical health and cognitive functioning, numerous meaningful rela tionships, and a well-developed sense of personal competence. On the other hand, middle-age is a time of significant role transitions: children leave home, divorces occur, there may be a return to college or employment. Thus, from this perspective, role becomes an important mediator of midlife development. The results of this study strongly demonstrate the positive effect of employment for midlife women. Employed women in the 31-45 group had greater self- concepts for opposite sex relationships, same sex 84 relationships, and emotional stability, and even full time students reflected higher self-concept for oppo site sex relationships; employed women in the 46-65 group exhibited greater general-esteem than unemployed women. Surprisingly, parenthood was not a critical issue for the midlife women of this project, insofar as it was not a predictor for any aspects of self-concept with this sample. Being single, on the other hand, was related to lower general-esteem in late middle-age women (46-65), suggesting that being Moff-track" devel- opmentally may impact significantly on self-concept. A popular myth surrounding the mental health of middle-aged women focuses on what has come to be called "the midlife crisis." This crisis is usually attrib uted to losses of physical health and life opportuni ties, as well as a growing sense of one's own mortali ty. Although this study was not designed to specifi cally measure pathology, a positive self-concept is an integral component of good mental health; thus rela tionships found between age or role and lower self- concepts might be viewed as possible indicators of a "midlife crisis." Except perhaps for the relationship between single status and lower general-esteem, howev er, there was no evidence of any such "crisis" with this population, other researchers have also failed to find indications of a midlife crisis (Haan, 1981; 85 Cooper & Gutmann, 1987; Kristal f c chiriboga, 1979). Perhaps the focus needs to be not so much on whether midiife is a crisis, but rather, under what circum stances does change create a crisis. There are several distinct tasks associated with middle adulthood: the redefining of self outside of family and work roles, recognition of biological limi tations, increased awareness of one's mortality, and reassessment of primary relationships. Many women also experience significant changes in roles or the content of continuing roles at midlife. Furthermore, for those women "off-track" for normative life events, middle-age may have entirely different ramifications than for their peers. Clearly there are many opportunities for change in midlife, which may or may not precipitate a crisis. In any case, the successful resolution of these transitions stands out as the principle challenge for middle adulthood. The results of this study hold a number of impor tant implications for the mental health practitioner. First, middle age in women does appear to be associated with developmental change in the self-system, with relative strengths occuring in the areas of cognitive ability and same sex relationships. Thus, no matter what the impact of the initial presenting problem, counseling with midlife women will inevitably be influ 8 6 enced by these developmental transitions. Secondly, employment appears to play a significant mediating role in midlife development. Certainly the employed middle- aged women of this investigation had notably healthier self-concepts than the unemployed women, as did the students. Whatever the initial stresses and anxieties associated with a return to the work force, the bene fits of employment for midlife women continue to out weigh the risks. Third, although midlife does provide many opportunities for change, "change" does not neces sarily equate with "crisis.” This study found no more losses (i.e., lower self-concepts in specific dimen sions of the self-system) in the middle-aged groups than in the youngest and oldest groups. Helping pro fessionals need to stay mindful that "midlife crises” presented in therapy are more likely to stem from longstanding problems— possibly exacerbated by midlife changes— than from a current life stage. Finally, women who are "off-track" for culturally defined se quences of life events may be at increased risk for anxiety, depression, or low self-worth, as reflected in the lower general-esteem of the single (i.e., never married) late middle-aged women of this study (46-65). Even for women who consciously value their non-tradi- tional choices and lifestyles, departure from known patterns can precipitate much painful soul searching 87 during the middle years, often raising concerns about essential femininity and sexuality. It is crucial that mental health professionals be knowledgeable about and sensitive to those developmental issues and realities that are unique to women. The therapist alert to the centrality of these issues can help separate them from the context in which they are raised, thus providing a healthy framework for the exploration of the choices, risks, and responsibilities involved. Developmental research over the life span has emphasized the different challenges which are confront ed at various ages (Ryff, 1989), but nowhere are these more evident than with the older adult. Changes in physical appearance and functioning— however gradual— are difficult to ignore, as are the development of chronic medical problems. More significant is the shift in social interaction patterns that occur due to retirement and deaths of peers, particularly males. These patterns were readily reflected in the findings of this study, with the oldest group demonstrating the lowest self-concepts of the population for opposite sex relationships, physical appearance, and physical abili ty. The research also indicates that— despite these declines— older adults generally view themselves as more satisfied, happy, and competent in life skills than younger adults; the very high self-concepts for 8 8 honesty/reliability, religion/spirituality, and emo tional stability for the oldest group of this study parallelled these findings* One implication for mental health practice is that how older women view their health status may be as important as their actual condition in determining well-being and a positive self-concept. A primary task for all counselors is instilling hope, but this process is particularly relevant for the older client. Con trary to societal attitudes about aging/ dramatic changes "for the worse" are not typical and are likely to be due to specific problems, rather than aging per se. A key role for the mental health practitioner, then, involves helping the older adult adapt to current levels of ability, as well as exploring normative changes in physical functioning, intimacy patterns, and anxieties related to other life spheres (e.g., cogni tive functioning and sexuality). Prnhlftaa wnri limitations There were a number of problems and limitations encountered in the process of implementing this study which should be considered when interpreting the data. First, as with most field studies, subjects participat ed voluntarily and--as such— may represent a special group in and of themselves. Volunteers may have dif ferent attitudes, personalities, and motivational needs 89 than those who choose not to volunteer. Older adults (65+) available to research, in particular, tend to be self-selected and more likely to be postively biased in regard to such factors as intelligence and health (Lachman, 1984). Thus the self-concepts of female volunteers in this study may not be representative of women in general. To enhance the representativeness of women used in the sample, respondents in this study were not recruit ed from any single source, and every effort was made to obtain a well stratified and demographically diverse population. Nonetheless, minority participants were over-represented in the youngest age group and under represented in the oldest age group. Furthermore, chi- square analyses revealed that minority women in the first two age groups (18-30 and 31-45) had significant ly different incomes, education levels, and employment rates than non-minorities. As both income and educa tion have consistently been positively associated with most measures of well-being (Larsen, 1978; Campbell, 1981), it is possible that these findings for self- concept are negatively biased for the generally lower educated and lower incomed minority women of these two groups. Another limitation concerns the inability to control variables which may have influenced the study's 90 outcome. Every self-report measure is susceptible to social desirability influences, and it is possible that all subjects did not answer their questionnaires hon estly or with the best of intentions. Also, there is no definitive way of confirming to what degree the questionnaire was correctly understood and interpreted, or whether the method of data collection influenced the results. There was some evidence, for example, that the ability to correctly complete the questionnaire varied with age, particularly with the older partici pants, who presumably had less exposure to standardized testing in their lifetime. Although all questionnaires included the same written instructions and cover let ter, half of the questionnaires were mailed in, while the other half were completed with the on-site assis tance of this investigator; this might have differen tially affected the way subjects responded to the questions. A significant delimitation of this study involves the choice of variables selected for investigation. Even under rigorous research conditions, self-concept remains a relatively elusive component of the self- system, and its probable relationships with innumerable traits and/or states inevitably necessitates the estab lishment of very firm research perimeters. As the focus of this study was the nature and differences in 91 the self-concept of women over the life course, its scope was deliberately restricted to those variables most often associated with normative life span develop ment: age group, parenthood, vocational status, and marital status. While there are undoubtedly many factors that can and do influence self-concept— e.g., ethnicity, androgeny, academic proficiency— time and economic constraints preclude their inclusion in this project. A major limitation of this investigation was its cross-sectional nature. Although methodological prob lems preclude a longitudinal approach to the age span investigated here, in a cross-sectional design, the directionality of effects cannot be assessed. It is possible, for example, that women with higher self- concepts self-select themselves into employment or college (versus other roles). Furthermore, cohort factors beyond the breadth of this study— such as exposure to social change and major historical events— may have also influenced self-concepts of specific age groups. Thus, while these findings may reflect devel opmental change in self-concept, they may also reflect cohort differences. The issue of cohort differences is particularly relevant and bears additional examination. MCohortM refers to a group of persons born in the same year or 92 decade, and "cohort: effect" means those events or social changes that can be assumed to have had an impact on those in the same cohort (Grambs, 1984). A major weakness associated with the cross-sectional method is that it provides no direct information about intra-individual (age) changes; it can estimate change only through the observation of group (age-cohort) differences. Thus, in this design age and cohort are confounded— groups differ not only with respect to age but also with respect to cohort, which in this study are one and the same. The research on women is partic ularly sensitive to cohort effect since women, far more than men, have been involved in a revolution of life opportunities and potential. Women now in their for ties, for example, were the first cohort to be able to exert absolute fertility control, to have equal access to education and professional opportunities, as well as exposure to civil rights legislation and the women's movement. Such events would be expected to precipitate a quite different process of social and psychological aging than that experienced by previous cohorts. The cross-sectional method has been criticized for providing information about age group differences as opposed to age-related differences. This is not to preclude its usefulness in research, however. Longitu dinal studies have their own methodological problems, 93 such as selective attrition and statistical regression toward the mean, suggesting that design decisions ultimately become a matter of priorities and trade offs. The strength of the cross-sectional design is its parsimony in regards to time— researchers cannot always wait for their subjects to grow older in order to study adult development. Although cross-sectional findings must be interpreted carefully, they are none theless valuable in identifying developmental and psychological trends, which can be further explored in subsequent studies. A problem of some concern in this investigation was the absence of a family relationships scale for the SDQ III. Although the instrument includes a self- concept dimension for relations with parents when used with young adults, the age span addressed here— with its multigenerational family relationships— lended the original scale unusable. As developmental research has stressed the importance of relationships to womens' self-systems, the absence of a family relations scale is a conspicuous deficit that requires remedial mea sures. The development of a multigenerational family relations dimension for the SDQ III will be a major tasX for future investigators. Although the results of this investigation were significant and meaningful for adult women, the find- 94 ings may not be generalizable to a mature male popula tion. Even with young adults, previous SDQ III re search (Marsh, 1965) has identified differences in the strengths of different dimensions of self-concept between males and females, and it is assumed that differences would also exist for an older population. Additional studies are needed to explore the strength and direction of such differences before the SDQ III can be utilized in any applied sense for an older, mixed, population. 95 Appendix A. Structure of Self-Concept (Shavelson, Hubner, and Stanton, 1976) i f c e e i bm ( •«t i * i ■ i it «< • Cw^lMMlI H <» I * I I I II S' t £ £ - M is nisi S iii d d d o 96 Appendix B SCHOOL O f BDUCATION DfVM ON O f C O U N H U N C AND ■DUCATIONAL PSYCHOLOGY - WFH S O S You have been asked to participate in a study for my dissertation in the PhD program in Counseling Psychology at the University of Southern California (USC). The purpose of this study is to explore how women of different ages and different lifestyles think and feel about themselves. You have been invited to participate because you are representative of many women today, Juggling multiple roles and redefining who you are in the process. You w ill be asked to complete two questionnaires. In the first one, self-descriptive statements are made relating to different areas of your life. Using the scale at the top of the page, you are to mark how strongly each statement describes you, from definitely false (1) to definitely true (8). For some other statements, you w ill be asked to rank how important that area of your life is to you, with one (1) being most important, and higher numbers being less important In the second questionnaire you are asked to provide demographic information about yourself and your family. You may decline to answer any of the questions that you wish, but it is most beneficial to the study If you respond to all the Items. Although the questions in this study are personal, they are not sensitive, and as ail questionnaires are anonymous, there is no risk of identification. In addition, all of the research wil be supervised by Dr. Michael Newcomb, a psychology professor at the University of Southern California. Past participants have reported that they enjoyed filling out the questionnaire because it encouraged them to think about aspects of their U fa that they had not thought about before. If you have any questions or information that you wish to provide about the questionnaires or study, please feel free to include it in the Comments section. If you wish to know the results of the study, you may leave your address with or contact the researcher. Thank you very much for your time and cooperation. Julia Gearhart /) / • U L jl . 224 So. Sycamore j/{jm JutM/WU Rialto, CA 92376 / 97 Appendix C. Demographies Questionnaire Please fill in the appropriate answer or circle the number that best describes you. 1. What is your marital status? (1) married (2) single, never married (3) separated or divorced (4) widowed (5) living with mate 2. What is your age?__________ 3. Do you have children?__________ If vas. how manv? What are their aoes? What is your total annual family income per year? (1) Less than *15,000 (2) *15,000**30,000 (3) *30,000- *45,000 (4) *45,000* *60,000 (5) Greater than *60,000 5. Are you employed outside the home?__________ I f yes, how many hours each week?__________ What is your specific employment (type of Job/position)? 6. if you are employed outside the home, what is y o u r annual income (if different from #4)? (1) Less than *16,000 (2) $16,000* *30,000 (3) *30,000- *45,000 (4) *45,000* *60,000 (5) Greater than *60,000 Are you presently attending school?___________ It yes. are you a full-time or part-time student?_______________________ Are you retired?___________ How many hours per week were you employed outside the home prior to retiring? What was your specific employment (type of job/position) prior to retiring? What is your ethnic group? (1) Hispanic (2) African-American (3) Asian (4) Pacific Islander (5) White (6) ___________________________ How much education have you completed (circle all that apply)? (1) Junior high or less (2) Some high school but did not graduate (3) High school diploma (4) Vocational or hade school/business school (5) Some college but no degree (6) Coliege with 2-year degree (7) College with 4-year degree (8) Graduate work or professional degree 99 Appendix O. Self Descriptive Questionnaire Itl (Mersh, in press) M M M r u m I | M m F m m M m T im MaaSy O ain saiy PWSa P alaa PaM a S e n Ti m M an P atna T /va Ti m Ti m I I I I I I I----- i i i 4 i • 7 a 1 . Overall, I have ■ tot of rsapsct tor myself. 2. I often M M iimH lisa to avoid errtoarraeaing sRuattons. 3. I g s a lot ol attention from monsters of the oppoH o mx 4. I am usually pretty calm and relaxed. S. I am never aMa to think up answers to problems that haven't already been figured out. 6. I have a physically anracbue body. 7. I have few friends of the same sex that I can ready count on a I am a good aihlata. 9. I am a spkttuaifreltgioue parson. 10. Overall, I |Kfc self conidenco. 1 1 People can always roly on me 12. I find a dffflcuR to meet members of the oppoeRe sex whom I H ke. 13. I worry a lot. 14. I am good as conMRng Idaaa m ways that others have not tried. IB. I am ugly. IS. f am comfortable talking lo m entors of the same m 17. I am awkward and poorty oootdinsHd at moat sports and physical activates. IS. SpkRuai/rekgtous bakaM have M o te to do wtth my life phMosophy. 19. Overall. I am pretty accepting of myeaff. 20. Being honest is ogt particularly important to ms 21. I have lots of friends of the oppoeRe sex. 22. I am happy moat of the time. 23. I wish I had more knagmadon and artgkiaMy. 24. l have a good body buNd. 2B . I don't gat along w o N with other members of the same sea. 26. I have good endurance and stamina mspona and physical actMUea. 27. Spimualhafigloua baliafe make my M f e batter and make me a happier parson. 28. Overall, l don't have much respect tor myself 29. I nearly always tea the truth. 30. Moat of my Mends are more comfortable wkh members of the opposes sax than l am. 31. I am arudoue much of the tkne. 32. i anfoy wortdng out new ways of solving problems. 33. There are Iota of things about the way l look that I would Hke so change. 34. i make Mends easily wkh manRisra of the earns sax. 100 M MMr P aw n Pahs I----- 1 1 3S. i hM aporta and physical actMtiw. 36. M y apMtuatfraligioua ballott provide the guMahnw by which I conduct my U fa . 37. Overall. I have a lot oi ssftconfldance. » . taomaMmw take thinga » w do not belong to ma. 39. I am comtortablo talking to members of M ie ftppflijf# M X . 40. I hardly ewer leal depraeeed. 41. I'm not much good at problem solving. 42. M y body weight la about right (noiMier too ftinor too Mdnny). 43. Olhar manMers of the aame aw find ma boring. 44. I have a Ngh energy level In aporta and phyaleal actMMw. 45. Cononuoua apMtuaMehgloua y w v to me. 46. Overall. I hew a vary good aettooncapL 47. I never chear 46. I am quae ahy wen mambera of the oppooeo aw . 49. I land to ba Mnh auiam lanaa. and raattaaa. “ a * ■ w iw w r iw i wrw awi^Vi W W 90. I have a lot of intea actual curfoalty. 51 l dlaNka the wav I look. v ■ ■ ■ ^ w w a i y i i w m i 52 I share teat of actM M w wah members of die same aw. _ 5 2 I'm not very good Many actMMaa M W require phyaleal abiity and oootdinoMon. i w w y x w tm w p &n a v n t w ipwm i M M M n O f f O M f lt O U i D f O V O f . W r w v w v W V y n v i y w ■ MamlXM HMOy IMMM| M w n Ma* T im T im Turn I 1 I I-------- a a 7 a 55. Overall, nothing that l do la vary important 66. Being dlehanaet la often the leaeer of two evtta. 57. I make friends aaaM y w N h iiw tt w s of the opposftt oox. 56 . I do not apand a lot of tune worrying about -■ ■ V M fipi. 59. I am not vary original In my Idow, thoughta. and acUone. 60. I hava nice facial faaturaa . 61. Not many people of the aame aw Hke ma 6 2 I H ke to w arclaa vigorously at aporta andfor phyaleal actMMw. 62 I am a boner paraon aa a conaaquanoa of my ap M M u al, raNgMua, boNolh. 64. O varaM , l hava prany poaMva faahnga about fn^NB. 66. I am a vary honaat person. 66. I have had lota of faaNnga of madoquacy about ralMng to mambera of the oppose a aw . 67. I am oflan dapraaaad. 62 I am an ImaginaMw person. 66. I wtah that I were physic ahy more Mtracttve. 70. I am papular with other mambera of the "n™ ** aw . 71. I am poor at moot aporta and physical actMMw. 72 I am baalcaly an aMisfet. and I behave that there la no being hi^ior than men. 73. OvaraM . I have a vary poor self concept. Overall, l am not wary accepting < n myself Few H any ot my friends ara wary spiritual or raagioua. 76. I am inclined toward being an optimist. 77. I would have no interest in being an inwantor. 7B Moat ol my friends are better iooMng than I am. 79. Moat people haw more friends ol the sam e sex than I do. *0. I snfoy sports and physical activities. 91. I beHew that there w M be some form of comfnuatian of my spell or soul after my death. 92. Overift, I hawe pretty negative feelings about myself. 93. i value integrity above aN other rttu ee 94. I newer seem to haw much In common with Be m h e e e ed ibe m V rv v l w m i OppOMI M X , 98. I tend to be a wary nervous person. 96. lean often see batter ways of doing routine tasks. 97. I am good IooMng. 99. I haw lots of friends of the same sax 99. I am a sedentary type who avoids strenuous activity 90. Overall. I do Mia of things that are important. 91 l am not a very rsNabie parson. 92. dpjrauajfreeglous beieM hew Idle to do with the type of person l want to be. 74 I would feet OK Mxxa cheating on a test as long _ _ 94. as I did not gat caught. 96. 79. I am comfonable about being affectionate with members of the opposite sex. 1 0 2 REFERENCES Aldous, J. (1978). Family careers:__Developmental change in families. New York: Wiley. Alwin, D. F., Converse, P. E., £ Martin, S. S. (1984). T,lying arrangementsf social— integration and pfiy- chQlogir.nl well-being. presented at the annual meeting of the Midwest Sociological Association, Chicago, 1L. Andrews, F. M., & Withey, S. B. (1976). Social indica ters of wel 1 —being» Americans' pftrcaptlons of lifa guaii ty. New York: Plenum. Aneshensel, C. S., Frerichs, R. R., & Clark, V. A. (1981). Family roles and sex differences in de pression. .lotima 1 of Health and Social Behaviorr 22, 379-393. Baltes, P. B., 6 Baltes, M. M. (1990). Psychological persepctives on successful aging: A model of se lective optimization with competition. In P. B. Baltes & M. M. Baltes (Eds.), Successful aging: perspectives from the behavioral sciences (pp. 1- 34). Cambridge, England: Cambridge University Press. Barnett, G. C., & Baruch, G. K. (1985). Women's in volvement in multiple roles, role strain, and psychological distress. Journal of Personality and Social Psychology, 22, 135-145. Barnett, R. C., & Baruch, G. K. (1981). Multiple role strain, number nf roles, and psychologica1 well- being (Working Paper No. 73). Wellesley College, Wellesley, MA. Bart, P. B. (1971). Depression in middle-aged women. In V. Gornick and B. K. Moran (Eds.), women in sexist society. New York: Basic Books. Baruch, G., £ Brooks-Gunn, J. (1984). Women in mid- life. New York: Plenum Press. Baruch, G., Barnett, R., £ Rivers, C. (1985). Lifeprints: Mew patterns of love and work for today *s women. New York: Signet Books. 103 Baruck, G. K., & Barnett, R. C. (1980). A new start for women at midlife. New York Times Magizina- Dec. 7. Battle, J. (1981). rulture-frea self-esteem Invento ries for children and adults. Seattle: Special Child Pub. Berner, A. (1981). Women in transition: The self- concept of women at four phases of the adult life cycle. Oissertatinn Ahstracts International, 42 (6-a)r 2496. Birnbaum, J. A. (1975). Life patterns and self-esteem in gifted family oriented and career committed women. In N. T. S. Mednic, S. S. Tangri, & L. W. Hoffman (Eds.), Women and ach i flvamant» Snr.lal and nntivatinnai analysis. Washington: Hemisphere. Blake, J., & del Pinal, J. H. (1981). The childless option: Recent American views of nonparenthood. In G. E. Hendershot & P. J. Placek (Eds.), Pre dicting fartiHty: Demographic studies of hi rth expectations (pp. 235-261). Lexington, HA: Lexington. Blau, Z. S. (1973). Old age in a changing society. New York: Watts. Block, J. (1971). T.iues through time. Berkeley: Bancroft Books. Brown, G. W., & Harris, T. (1978). Social origins of depression. New York: Free Press. Buhler, C. (1935). The curve of live as studied in biographies. .Journal of Applied Psycho1ogyr ±£, 405-409. Campbell, A. (1981). The sense of well-haing in Ameri ca! Recent patterns and trends. New York: McGraw-Hill. Campbell, A., Converse, P. E., & Rogers, W. L. (1976). The situation of women. In A. Campbell, P. E. Converse, £ W. L. Rogers (Eds.), The quality of American life (pp. 395-442). New York: Russell Sage. Campbell, A., converse, P. E., £ Rogers, W. L. (Eds.). (1976). The quality of American life. New York: Russell Sage. 104 Cleary P., 6 Mechanic, D. (1983). Sex differences in psychological distress among married people. Journal of Health and Social Behavior. 2At 111- 121. Coleman, L. M., & Antonucci, T. C. (1982, winter). Women's well-being at midlife. tsr Hawslattar. (Available from Institute for Social Research, University of Michigan, Ann Arbor). Cooper, K., & Gutmann, D. (1987). Gender identify f i t ego mastery style in middle-aged, pre- and post- empty nest women. Gerontologlstf 27(3), 347-352. Coopersmith, S. (1975). Building self-esteem in the classroom. In S. Coopersmith (Ed.), navaloping motivation in young children. Palo Alto, CA: Consulting Psychologists Press. Costa, P., McCrae, R., Zonderman, A., Barbano, H., Lebowitz, B., & Larson, D. (1986). Cross-section al studies and personality in a national sample: Stability in neuroticism, Extroversion and open ness. Psychology and Agingr X, 144-149. Demo, D. (1992). The self-concept over time: Research issues and directions. Annual Review of Sociolo- gy, 1&, 303-326. Depner, C. (1979, August). The parental role and psy chological well-being. Paper presented at the meeting of the American Psychological Association, New York. Diener, E. (1984). Subjective well-being. Psychologi cal Bulletin, 542-575. Erdwins, C. (1981). A comparison of different aspects of self-concept for young, middle-aged, and older women. Journal of Cl inlcal Psychology, 37(n31 f 484-490. Erdwins, C., Mellinger, J., f c Tyer, Z. (1981). A com parison of different aspects of self-concept for young, middle-aged, and older women. Journal of Clinical PsychologyT XZ(3), 484-490. Erickson, E. H. (1970). Identity through the life cycle (2nd ed.). New York: W. W. Norton. 105 Ferraro, K. F. (1980). Self-ratings of health among the old and the old-old. Journal of HMlfch and Social Behaviorr 21, 377-383. Figley, C. R. (1973). Child density and the marital relationship. Journal nf Marriage and the Familyr IS, 272-282. Fitts, W. W. (1965). Tannaaaaa aelf-ooncBpt scale. Nashville, TN: Counselor Recordings and Tests. Fleming, J. s., & Courtney, B. E. (1984). The dimen sionality of self-esteem: II: Hierarchical facet model for revised measurement scales. Journal of Personality and Social Psychology. AS, 404-421. Fleming, J. s., & Watts, W. A. (1980). The dimension ality of self-esteem: Results for a college sam ple. Journal of Personality and Social Psycholo gy/ 22, 921-929. Gigy, L. L. (1960). Self concept of single women. Psychology of Women f>uarterlyr £, 321-340. Gilberts, R. (1983). The evaluation of self-esteem. Family anti Community Health, fi, 29-49. Gilligan, C. (1977). In a different voice. Women's conceptions Of self and morality. Harvard Educa tional Review, A2, 481-517. Gilligan, C. (1982). Tn a different voice. Cambridge, MA: Harvard University Press. Gilligan, C. (1982a). Adult development and women's development: Arrangements for a marriage. In J . Z. Giele (Ed.), Women in the middle years. New York: Wiley. Gilligan, C. (1982b). In a different voice; Psycho logical theory and women's development. Boston, MA: Cambridge University Press. Glenn, N. D. (1975). The contribution of marriage to the psychological well-being of males and females. Journal nf Marriage and the Family, 22, 594-601. Glenn, N. D., & McLanahan, S. (1981). The effects of children on the psychological well-being of older adults. Journal nf Marriage and the Familyr AJ, 409-421. 106 Glenn, N. D., & McLanahan, s. (1982). Children and marital happiness: A further specification of the relationship. .Tnurnal nf Marriage and the Family- AA, 63-72. Glenn, N. D., & Weaver, C. N. (1978). A multivariate, multisurvey study of marital happiness. Journal nf Marriage and fhe Family. AO, 269-281. Goetting, A. (1986). Parental satisfaction: A review of research. Journal of Family Issues, 2, 83-109. Goode, W. J. (1960). A theory of role strain. Ameri - can Sociological Ravieu. 25, 483-496. Gore, S., f c Mangione, T. (1983). Social roles, sex roles, and psychological distress: Additive and interactive models of sex differences. Journal of Health and Social Rflhavior 2A, 300-313. Gould, R. (1978). Tranafomations. New York: Simon & Schuster. Gould, R. L. (1972). The phases of adult life: A study in developmental psychology. Awariran Journal of Psych i atry. 129, 33-43. Gove, w. R., & Geerken, M. R. (1977). The effect of children and employment on the mental health of married men and women, social Force, ££, 66-76. Gove, w. R., & Tudor, J. (1973). Adult sex roles and mental illness. American Journal of Psychologyr 2B, 812-835. Gove, W. R., Ortega, S. T., & Style, C. B. (1989). The maturational and role perspectives on aging and self through the adult years: An empirical evalu ation. American Journal of Sociology, 9A, 1117- 1145. Grambs, J. (1989). Women over forty:__Visions and reaHtias. New York: Springer. Grambs, J. (1989). women over forty. New York: Springer. Guttentag, M., Salasin, S., & Belle D. (1980). The manual health nf women. New York: Academic Press. Haan, N. (1981). Common dimensions of personality development: Early adolescence to middle life. 107 In D. Eichorn, J. Clausen, N. Haan, P. Honzik, and D. Mussen (Eds.), Present and past in Middle life. New York: Academic Press. Hamachek, D. E. (1978). Rnnnuntara with the self (2nd ed.). New York: Holt, Rinehart, & Winston. Haring-Hidore, M., Stock, W. A., Okin, M. A., 6 Witter, R. A. (1985). Marital status and well-being: A subjective synthesis. .Tnurna] nf Marriage and the Family, A2, 947-954. Harter, S. (1982). The Perceived competence Scale for Children. Ch M d Pawl opment r 5l1, 87-97. Helson, R., & Moane G. (1987). Personality change in women from college to midlife. Journal of Person ality and Social Pnychnlogy 176-186. Hoffman, L. W., & Manis, J. B. (1978). Influences of children on marital interaction and parental sat isfactions and dissatisfactions. In R. M. Lerner f i t G. B. Spanier (Eds.), Child Influences nn mari tal and fanny inf-araotion (pp. 165-214). New York: Academic Press. Hong, S., Bianco, M. R., & Bollington, J. (1983). Self-esteem: The effects of life-satisfaction, sex, and age. Psychological Reportsf 22, 95-101. Hunter, S., & Sundel, M. (1989). Midlife myths. Newberry Park, CA: Sage. Jaquish, G. A., 6 Ripple, R. E. (1981). Cognitive creative abilities and self-esteem across the adult life span. Human Development, 2A, 110-119. Jung, C. (1958). Psyche and symhoi. New York: An chor. Kagan, J., f i t Moss, H. A. (1962). Birth to maturity. New York: Wiley. Kandel D. B., Davies, M., & Ravels, V. H. (1985). The stressfulness of daily social roles for women. Journal of Health and Social Behavior, 2£, 64-78. Kessler, R., 6 McRae, J. Jr. (1982). The effects of wives' employment on the mental health of married men and women. American Sociological Review, A2, 216-227. 108 Kessler, R. C., f c McRae, J. A. (1981). Trends in the relationship between sex and psychological dis tresses: 1957-1976. Ameriean Sociological Review. 443-452. Krystal, S., 6 Chriboga, D. (1979). The empty nest process in midlife women and men. Hauritas, X, 215-222. Laborda, J. M., 6 Powers, M. J. (1985). Life satisfac tion, health-control orientation, and illness- related factors in persons with osteoarthritis. Raftftflrrh in Nursing and Health, fl, 183-190. Larson, R. (1978). Thirty years of research on the subjective well-being of older Americans. Journal of Gerontology, JU, 109-125. Lawton, M. P. (1984). The varieties of well-being. In C. Z. Malatesta 6 C. G. I2ard (Eds.), Emotion in adult development (pp. 67-83). Beverly Hills, CA: Sage. Lemon, B., Bengston, V., & Peterson, J. (1972). An exploration of the activity theory: Activity types and life satisfaction among in-movers to a retirement community. Journal of Gerontology, 27(4), 511-523. Levinson, D., Darrow, c., Klein, E., Levinson, M., & McKee, B. (1978). The flflasnne nf a man's life. New York: Knopf. Levinson, P. J. (1977). The mid-life transition: A period in adult psychosocial development. Payohi- atry, AH, 99-112. Lowenthal, M. (1977). Toward a sociopsychological theory of change in adulthood and old age. In J. Birien 6 K. Schaie (Eds.), Handbook of the psy chology nf aging. New York: Van Nostrand- Reinhold. Lowenthal, M. F., Thurnher, M., 6 Chiriboga, 0. (1975). A comparative aturiy nf man and woman faoing tran sitions. San Francisco: Jossey-Bass. Manis, J. (1982, March). Relationships among career orientation, employment status, self-esteem, and life satisfaction of women: An analysis of age differences. In D. Hendel (Chair), Life stages and their role in programs for the non-tinning 109 education of unman. Symposium conducted at the meeting of the American Educational Research Asso ciation, New York. Marini, M. (1980). Effects of the number and spacing of children on marital and parental satisfaction. Demography, 12, 225-242. Marks, S. (1977). Multiple roles and role strain: Some notes on human energy, time, and commitment. American Sociological Review, A2, 921-936. Marsh, H., & Shavelson, R. (1985). Self-concept: Its multi-faceted, hierarchical structure. Education al Psychologist:, 2013) , 107-123. Marsh, H. W., & O'Neill, R. (1984). Self-Description Questionnaire III (SDQ III): The construct valid ity of multidimensional self-concept ratings by late-adolescents. Journal Of Educational Measure ment, 21, 153-174. Marsh, H. W., Parker, J. W., & Smith, I. D. (1983). Preadolescent self-concept; Its relation to self- concept inferred by teachers and to academic abil ity. British Journal of Educational Psychologyr 51, 60-78. Marsh, H. W., Relich, J. D., k Smith, I.D. (1983). Self-concept: The construct validity of interpre tations based upon the SDQ. Journal of Personali ty and Social Psychology, 15, 173-187. Marsh, H. W., Smith, I. D., Barnes, J., & Butler, S. (1983). Self-concept: Reliability, dimensionali ty, validity, and the measurement of change. Journal of Educational Psychology, 25, 772-790. Marsh, H. W., Smith, I.D., 6 Barnes, J. (1983). Multi- trait -multimethod analyses of the Self-Description Questionnaire: Student-teacher agreement on mul tidimensional ratings of student self-concept. American PdnoaMonal Research Journal f 25, 333- 357. McGuire, W. J., Mcguire, C. V. (1981). The spontaneous self-concept as affected by personal distinctive ness. In M.D. Lynch, A. A. Norem-Hebeisen, k K. J. Gergen (Eds.), Self-concept: Advances in Theo- ry t Research (pp. 147-171). Cambridge, MA: Ballinger. 110 McLanahan, S., 6 Adams, J. (1984). Explaining the decline in paranta* psychological well-being:__The rola nf employmentr marital disruption and social integratlon (Working Paper 85-25). University of Wisconsin, Madison Center for Demography and Ecol ogy. McLanahan, S., 6 Adams, J. (1984). The changing affect nf children on subjective wall-being (Working Paper 84-16). University of Wisconsin, Madison Center for Demography and Ecology. Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press. Merikangas, K. (1985, May). Sax differences in depres sion . Paper presented at Murray Center (Radcliffe College) Conference: Mental Health in social Context, Cambridge, MA. Merton, R. K. (1957). Social thanry and eocial Btruc- ture. New York: Free Press. Monge, R. H. (1975). Structure of the self-concept from adolescence through old age. Experimental Aging Researchf 1, 281-291. Morganti, J. B., Nehrke, M. F., Hulicka, I. M., & Cataldo, J. F. (1988). Life span differences in life satisfaction, self-concept, and locus of control. International Journal nf Aging and Human Developmentf 26, 45-56. Mortimer, J. T., Finch, 6 M. D., Kumka, D. (1982). Persistence and change in development: The multi dimensional self-concept. In P. B. Baltes 6 0. G. Brim, Jr. (Eds.), r. if e-span development- and behav ior f Vol. a f (pp. 263-313). New York: Academic Press. Mossey, J. M., 6 Shapiro, E. (1982). Self-rated health: A predictor of mortality among the elder ly. American -Tnurnal of Puhl ic Health, 22, 800- 808. Nehrke, M. F., Hulicka, I. M., 6 Morganti, J. B. (1980). Age differences in life satisfaction, focus of control, and self-concept. International Journal of Aging and Human heve1opmentr J_L, 25-33. Neugarten, B. L. (1976). Adaptation of the life cycle. The CounselIng Psychologiatf 6, 16-20. Ill Neugarten, B. L. (1977). Personality and aging. In I. E. Birren & K. W. Schaie (Eds.), Handbook of the payoho1ogy of aging. New York: Van Nostrand Reinhold. Neugarten, B. L. (1986a). Adult personality: Toward a psychology of life cycle. In B. Neugarten (Ed.), Middle age and aging:__A reader ill social psychol ogy. Chicago: University of Chicago Press. Neugarten, B. L. (1986b). The awareness of middle-age. In B. Neugarten (Ed.), Middle age and aging:_ _ _A reader in social psychology. Chicago: University of Chicago Press. Oden, M. H. (1968). The fulfillment of promise: 40 year follow-up of the Terman gifted group. Genet ic Psychology Monographs, 22, 3-93. Pack, T. (1986). Women's self-definition in adulthood: From a different model? Psychology of W om en Quar terly, 1013), 274-284. Pearlin, L. I. (1974). Sex roles and depression. Piers, E., & Harris, D. (1969). M anual for the Pi arfi- H am s children's self-concept scale. Nashville, TN: Counselor Records and Tests. Pietromonaco, P., Manis, J., & Harkus, H. (1985). The relationship of w orking to s e l f - percept!on and well-being in w om en* A cognitive analysis. Manu script submitted for publication. Pietromonaco, R., Manis, J., 6 Frohardt-Lane, K. (1984, August). Psychological consequences of multiple social roles, paper presented a the meeting of the American Psychological Association, Toronto, Canada. Pilisuk, M. (1982). Delivery of social support: The social inoculation. Journal of Orthopsychiatryf 52, 20-30. Polit, D. F. (1982). Effects of family si as* A c r ltj- cal review nf literature since 1971. Washington, DC: American Institute for Research. Powell, V. (1977). The empty nest, employment, and psychiatric symptoms in college-educated women. Psychology of W om en Quarts!yr 2, 35-43. 1 1 2 Puglisi, J. T., & Jackson, D. W. (1981). Sex role identity and self-esteem in adulthood, interna tion al Journal of Aging and Human Development, in press» Pur key, W. W. (1970). fial f-r.oncapt and school achieve ment. Englewood Cliffs, NJ: Prentice-Hall. Radloff, L. S. (1975). Sex differences in depression: The effects of occupation and marital status, sev Rolesr 1, 249-265. Rollins, B., & Feldman, H. (1970). Marital satisfac tion over the family life cycle. Journal of M ar riage and the Family, 12, 20-28. Rollins, B., & Galligan, R. (1978). The developing child and marital satisfaction of parents, in R. M. Lerner & G. W. Spanier (Eds.), Child influences on marital and family interaction (pp. 71-106). New York: Academic Press. Rollins, B. C., 6 Cannon, K. (1974). Marital satisfac tion over the family life cycle: A reevaluation. Journal of Marriage and the Family, 271-282. ROSCOW, I. (1974). Socialination to old age. Berke ley: University of California Press. Rosenberg, M. (1965). S o cieties and the adolescent self-image. Princeton, NJ: Princeton University Press. Rosenfield, S. (1989). The effects of women's employ ment: Personal control and sex differences in mental health. Journal of Health and Social Be havior f 2£L, 77-91. Ross, C., Mirowsky, J., k Huber, J. (1983). Marriage patterns and depression. American Sociological Review, AIL, 809-823. Ryder, N. B. (1979). The future of American fertility. Snci a 1 Problems, 2£, 359-369. Ryff, C. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well being. Journal of Personality and Social Psychol- ogy r 52, 1069-1081. Sarbin, T. R., & Allen, V. L. (1968). Role Theory. In G. Lindsey 6 E. Aronson (Eds.), The handbook of 113 social psychology (Vol. 1, pp. 488-567). Reading, HA: Add!son-Wesley. Schofield, L. (1971). Self-concepts of mature and young women students. Journal of Col 1 age Student Personnelf J_2, 297-301. Serlin, E. (1980). Emptying the nest: Women in the launching stage, in D. HcGuigan (Ed.), Women * a lives: Maw theoryf research, and policy. Ann Arbor: University of Michigan Press. Shavelson, R., Hubner, J., 6 Stanton, J* (1976). Self- concept: The interplay of theory and methods. Journal nf Educational Psychologyr 2A, 3-17. Shavelson, R. J,, 6 Bolus, R. (1982). Self-concept: The interplay of theory and methods. Journal of Educational Psychology r 2A, 3-17. Shavelson, R. J., Hubner, J. J., and Stanton, G. C. (1976). Validation of construct interpretations. Review nf Educational Research. ££, 407-441. Sieber, S. D. (1974). Toward a theory of role accumu lation. American Sociological Reviewr 22, 567- 578. Slater, P. (1963). On social regression. American Sociological Review, 2£, 339-364. Stryker, S., k Serpe, R. (1982). Commitment, identity, salience and role behavior. In W. inkles & E. Knowles (Eds.), Personalltyf roles r and social behavior (pp. 199-218). Thoits, P. A. (1983). Multiple identities and psycho logical well-being: A reformulation and test of the social isolation hypothesis. American Socio logical Review,. AS, 174-187. Vaillant, G. E. (1977). Adaptation to life. Boston: Little & Brown. Vaillant, G. E., k McArthur, C. C. (1972). Natural history of male psychological health: The adult cycle from 18-50. Seminars inpsychiatry, A, 415- 427. Van Hoose, W. (1985). Midlife myths and realities. Atlanta: Humantics Limited. 114 Verbrugge, L. M. (1982). Women's roles and health. In P. Berman & E. Raney (Eds.), Women; A developmen tal perspective (KIH Publication No. 82-2298). Washington, DC: U. S. Government Printing Office. Veroff, J., Douvan, E., & Kulka, R. (1981). Mental health in America; Patterns Qf he]p-SflCKinq from iQS7-1976. New York: Basic Books. Ward, R. (1979). The never-married in later life. Journal of Gerontology,. 2A, 861-869. Wells, L. E., 6 Marwell, G. (1976). Self-esteem, its conceptual!nation and measurement. Beverly Hills, CA: Sage. Wylie, R. C. (1974). The s e lf-ooneept (Vol. 1, rev. ed.). Lincoln: university of Nebraska Press. Wylie, R. C. (1979). The self-concept (Vol. 2). Lin coln: University of Nebraska Press.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A cross-cultural comparison of marital power and dyadic adjustment among American, Indo-American, and East Indian dual-career and single-career couples

PDF

Concordant and discordant drug use in intimate relationships: A longitudinal study

PDF

Robust Principal Components Analysis: Improving Inference About Largest Eigenvalues In Multivariate Data Analysis Using Resistant Statistical Methods And Bootstrap Resampling Techniques

PDF

Counselor Trainees' Perceptions Of And Attitudes Toward Sexual Harassment In Relation To Feminist Beliefs

PDF

A Comparison Of Female Inmates With And Without Histories Of Prostitution On Selected Psychosocial Variables

PDF

Examining the differences in family functioning and meaning in life between Reform, Conservative, and Orthodox Jewish families

PDF

Impact of language and culture on a neuropsychological screening battery for Hispanics

PDF

Emancipated foster youth: A study of outcomes of emancipated foster youth in a transitional housing program

PDF

Moral development and enviro-ethics: an examination of the theories of Kohlberg and Gilligan

PDF

Gender differences in motivation for sexual intercourse: Implications for risky sexual behavior and substance use in a university and community sample

PDF

Effects of parenting style and ethnic identity on European American and Asian Indian adolescents' academic competence and self esteem

PDF

Cognitive and non-cognitive factors as predictors of retention among academically at-risk college students: A structural equation modelling approach

PDF

Neuropsychological And Psychological Correlates Of Marital Violence In A Clinical Sample

PDF

The imperial family domesticity and nationalism in the Victorian novel

PDF

Cognitive predictors of intention -behavior consistency in safer sex practices

PDF

Money Makes The World Go 'Round: Mother-Custody Families And Downward Mobility

PDF

Criminal justice response to domestic violence: A closer look at Los Angeles County, 1995--1998

PDF

A two-dimensional model of cognitive empathy: An empirical study

PDF

Behavioral outcomes of a change from a traditional office to a virtual office: A longitudinal analysis

PDF

Married Japanese women with children: The relationship between perceived spousal support, childrearing, outside employment, and levels of depression, anxiety, self-esteem, and life satisfaction

Asset Metadata

Creator Gearhart, Julia Marie (author)

Core Title Self-Concept In Adult Women: A Multidimensional Approach

Degree Doctor of Philosophy

Degree Program Counseling Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, general,sociology, individual and family studies,women's studies

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Newcomb, Michael (committee chair), Stoner, Madeleine R. (committee member), Thompson, Chalmer E. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c20-571523

Unique identifier UC11226550

Identifier 9600980.pdf (filename),usctheses-c20-571523 (legacy record id)

Legacy Identifier 9600980.pdf

Dmrecord 571523

Document Type Dissertation

Rights Gearhart, Julia Marie

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA