Domain-specific gender comparisons in identity development among college youth: ideology and relationships.

Pastorino, Ellen ; Dunham, Richard M. ; Kidwell, Jeannie 等

Research on identity formation typically focuses on identity as a developmental process. However, it also is important as an "environmental process" that includes influences on personal choices and decision-making from the social contexts provided by friendship, family, religion, occupation, and politics. Erikson (1968) has emphasized the dual nature of identity, as seen in the identity development of a given individual and in the identity of a given culture or subculture: "Identity is 'all-pervasive' . . . for . . . we deal with a process 'located' in the core of the individual and yet also in the core of his communal culture" (p. 22).

The social context in which identity develops is often referred to in the identity literature as one of several domains, and is typically separated into interpersonal and ideological realms. Interpersonal domains refer to the individual's relational orientation in the context of his or her family and other intimate relationships. The ideological domains reflect the individual's approach to the seemingly more public contexts of religion, politics, and occupation.

Identity formation may be operationally defined in terms of exploration and commitment (Marcia, 1968). Exploration refers to a period of decision-making when previous choices, beliefs, and identifications are questioned by the individual and information or experiences related to alternatives are sought; commitment refers to the choice of a relatively stable set of roles and ideals (Marcia, 1968). Based on the occurrence of exploration and commitment, Marcia (1966, 1967) has proposed a process of identity development consisting of four identity statuses: Diffusion, which is characterized by a lack of exploration and commitment to identity issues; Moratorium, by active exploration of goals, values, and beliefs but without commitment; Foreclosure, by premature commitment in the absence of adequate exploration; and Achievement, by stable commitment based on sufficient exploration.

The empirical question of whether gender differences exist in identity development is complex, involving issues of identity status, the relative importance of commitment and exploration, variations in development across domains, and relationships of identity development with other measures. In addition, there are variations in age, cohort, and geographic subculture, different measures of identity development, and different explanations for similar findings.

Research during the 1970s emphasized several gender differences in identity development (Patterson, Sochting, & Marcia, 1992; Waterman, 1982). Males were more likely to achieve an identity during the college years, whereas females were somewhat more likely to be foreclosed, especially in the ideological areas of occupation, religion, and politics. From these findings it was inferred that exploration in identity issues appeared to be more adaptive for men, whereas for women, commitment seemed to be related to adaptive outcomes (Bernard, 1981). Finally, women's identity development was tied to interpersonal issues and to establishing connections with others, whereas men's development was associated more strongly with intrapersonal issues and assertions of independence and autonomy (Bernard, 1981; Fannin, 1979; Hodgson & Fischer, 1979; Schenkel, 1974).

During the 1980s research findings seemed to point to more similarities than differences between men and women in their identity development (Archer, 1992; Josselson, 1982; Steinberg, 1989; Waterman, 1982). For this decade, men and women were found to be similarly distributed across the four identity statuses, and identity exploration was generally related to adjustment for both female and male college students. For some scholars of gender and identity, the lack of gender differences in identity development remains the predominant theme that summarizes the current state of the field (Archer, 1992).

At the same time, several types of gender differences persist in the adolescent literature (Adams & Jones, 1983; Cella, DeWolfe, & Fitzgibbon, 1987; Gilligan, 1982; Harter, 1990; Kimmel & Weiner, 1985; Patterson, Sochting, & Marcia, 1992; Waterman, 1982). As in the previous decade, this literature suggests that women's identity development relates more strongly to interpersonal issues whereas for men it remains dominated by ideological issues.

In addition, gender differences have been noted in the associations of identity status with other measures. For example, occupational identity achievement for men has been related to a competitive attitude about work and striving for materialistic goals, whereas for women, achievement of an occupational identity has been shown to relate to gaining acceptance and approval from others (Grotevant & Thornbecke, 1982). Gender differences also have been emphasized in the associations of identity development with variables thought to be precursors to identity development. For example, gender differences are evident in relations between family process and identity development (e.g., Adams & Jones, 1983; Cooper & Grotevant, 1987; Jackson, Dunham, & Kidwell, 1987).

Lastly, there are differences in the timing and sequencing of identity development for men and women. According to Patterson et al. (1992), young men commonly achieve an identity during late adolescence. However, it has been suggested that women may achieve an identity during late adolescence, or delay identity achievement until after childbearing has been completed. Consequently, males may be more likely to resolve Erikson's stages as theorized, establishing an identity and then focusing on issues of intimacy, whereas for women, these two tasks may coincide rather than proceed in a serial fashion (Patterson et al., 1992; Kacerquis & Adams, 1980; Shanken, 1984).

The earlier interviews for studying identity development (Marcia, 1964, 1966), which focused on the traditional male domains of identity (occupation, religion, and politics), have been extended to include interpersonal domains (sex roles, dating, and friendship) (Grotevant & Cooper, 1981). The extended interviews arose partly in response to the recognition that interpersonal issues may be more salient in the identity development of women.

Some researchers have recommended a de-emphasis of the examination of gender differences based on scores on the identity interview (Archer, 1989a; Cooper & Grotevant, 1987). Instead, they suggest that emphasis should be placed on relations between identity development and other measures. However, an examination of a group of studies published during the 1980s indicates that in many cases, (1) studies report global identity scores without providing information about domain-specific scores (Cella et al., 1987; Cooper, Grotevant & Condon, 1983; Protinsky & Wilkerson, 1986; Shieldel & Marcia, 1985; Wagner, 1987); (2) studies focus on only one or two domains (Cooper & Grotevant, 1987; Grotevant & Thornbecke, 1982; Werrbach, Grotevant, & Cooper, 1990); (3) the complete interview covers only the original ideological domains of occupation, religion, and politics (Adams & Fitch, 1982; Cella et al., 1987); or (4) the studies provide data on only one gender (Adams, 1985; Adams & Jones, 1983; Archer, 1985). Other studies do not report domain-specific gender comparisons, possibly because these studies focus on defining the psychometric properties of a measure or on demonstrating connections between identity development and family functioning or other aspects of development (Craig-Bray, Adams, & Dobson, 1988; Grotevant & Adams, 1984; Grotevant, Thornbecke, & Meyer, 1982; Jackson et al., 1987; Read, Adams, & Dobson, 1984; Streitmatter, 1988; Streitmatter & Pate, 1989).

While all of these variants take their place in the development of our understanding of a complex and important topic, their pattern makes clear the gap to which the present work is addressed. As noted by Archer (1992), sensitive information may be masked by combining different domain scores: "An important finding in studies conducted, typically using between three and seven of these domains, is that a majority of individuals do not use the same decision-making identity mode in all domains. . . . Finding a most characteristic or global mode can limit our understanding of the task of forming one's identity if it is the only approach used. Examining identity activity separately by domain allows us to document the areas of salience to individuals differing in sex, age, ethnicity, class and so on" (p. 34).

The loss of information that occurs when domains are combined or analyses are limited to status assignments may affect not only gender comparisons but may distort the overall picture of identity development in American youth. Archer (1989a, 1989b) summarized the identity development of youth in this country as primarily showing a lack of identity exploration. Even in studies of college-age youth, she concluded that most studies present a portrait of youth who are predominately foreclosed or diffused on identity issues. Archer has speculated that foreclosure and diffusion may represent the normative developmental condition of most youth in our country. Alternatively, she has suggested that this portrait of generally nonexploring youth may be exaggerated by the tendency for researchers to focus on global scores.

Only a few studies have reported domain-specific gender comparisons for a variety of both interpersonal and ideological domains. These studies have analyzed gender differences using status assignments only. Archer (1989a) has reexamined the findings of three studies of gender differences. In two studies of sixth through tenth graders, domain-specific gender comparisons revealed no differences in the domains of occupation, religion, politics, and sex roles. However, for family roles, males were found to be more often diffused, and females more prevalent in the other three statuses. In a third study of high school juniors who were interviewed as seniors and again during their first year following graduation, Archer (1989a) found no gender differences in the areas of occupation, religion, and sex roles. However, in the political arena, males were more likely to be foreclosed, whereas women were more often diffused. Archer has concluded that gender differences are minimal.

Bilsker, Schiedel, & Marcia (1988) evaluated the identity development of college students in the areas of occupation, ideology (politics and religion), and sexual-interpersonal (sex roles and sexual behavior). They did not find gender differences in statuses for any of these areas nor for a global identity score. Findings for the individual domains of politics, religion, sex roles, and sexual behavior were not provided. They reported that identity development in the area of ideology was more predictive of overall global identity for men, whereas the sexual-interpersonal domain was more predictive for women. The authors concluded that these findings support the position that interpersonal issues are more prominent in the identity development of women, whereas ideological issues are more salient in the identity development of men-an inference consistent with research from the 1970s.

The major focus of the present study was to examine gender differences in domain-specific areas of identity development in a group of college youth. Males and females were compared on both global and domain-specific identity statuses for six domains covering both interpersonal and ideological areas. In addition, separate scores on exploration and commitment were examined. Most studies focus on nominal measures (status assignments), rather than on providing separate interval scores on exploration and commitment, an approach which may be more conducive to inferential statistics. This study also explored gender comparisons of relationships among domains, accomplished through an examination of the relative frequency of the four status assignments across domains, and through the intercorrelations of exploration and commitment scores. Finally, the study provides a domain-specific description of the identity process of a group of American youth. By comparing both global and domain-specific scores, we can determine if the use of global scores obscures meaningful findings.

METHOD

Subjects

Participants in this study were part of an ongoing research project at Florida State University. A total of 210 students, evenly distributed by gender, participated, and all participants were recruited from introductory psychology classes and were between the ages of 18 and 21. Cross-sectional and longitudinal research has suggested that the greatest variability in identity statuses can be found among this age group (Meilman, 1977; Offer, Marcus, & Offer, 1970). All students were from nondivorced, two-parent families. Eighty-seven percent of the participants were white with the remainder black (7.6%), Hispanic (4.3%), and Asian (1%). They were predominantly middle to upper-middle class, as judged by the Hollingshead Four Factor Index of Social Position (1965). Four samples, balanced by gender, were collected successfully between 1988 and 1989. There were no differences among the samples in age, family socioeconomic level, or ethnicity; therefore, the four samples were combined.

Measures

The Ego-Identity Interview (Grotevant & Cooper, 1981) is a semistructured interview that assesses adolescent identity formation in six domains: politics, occupation, religion, dating, friendship, and sex roles. The interview provides domain-specific commitment and exploration scores for each of the six domains.

Procedure

Students were told that they would be participating in a study examining variability among American college students and their families. They completed an inventory of background information that included demographic data and participated in the identity interview. A training manual developed by Grotevant and Cooper (1981) was used to train interviewers and coders according to standardized procedures. The interviews, which were audiotaped, were conducted by trained graduate and undergraduate students.

Scoring Procedures

Undergraduate psychology majors were trained to code the interviews using Grotevant and Cooper's (1981) training and rating procedures. The undergraduates independently coded the interviews using transcriptions of the taped interviews. Following the criteria in Grotevant and Cooper's manual, coders rated each subject's responses on a four-point scale for exploration and commitment in each of the six domains: I corresponds to an absence of exploration or commitment; 4 corresponds to strong exploration or commitment. An identity status assignment for each domain was based on the exploration and commitment ratings according to the procedure described by Grotevant and Cooper. In addition, exploration and commitment scores were obtained by summing scores on the four-point scale for each domain.

The combined ideological commitment and exploration scores were calculated by summing the scores from the occupational, religious, and political domains. Similarly, the combined interpersonal exploration and commitment scores were calculated by summing the scores from the dating, sex roles, and friendship domains. The combined overall exploration and commitment scores were calculated by summing across all six domains.

The combined ideological and interpersonal statuses were obtained by selecting the most common status across the relevant domains as the representative status for that individual. In cases where the statuses for the three relevant domains were each different, that person was excluded from the analysis. The combined overall identity status was based on all six domains and was obtained by independent judgments by the coders and researchers based on their individual global impressions of the levels of exploration and commitment across the six domains that were apparent in the interview protocol. Criteria for the overall global judgments are contained in Grotevant and Cooper's scoring manual.

Reliability

During the data-gathering, reliability checks were performed at random intervals to assess and maintain accurate scoring of exploration, commitment, and identity status for each domain. Reliabilities were based on the number of agreements between two coders divided by the number of judgments. Checks on 30 participants resulted in a mean agreement of .83 for exploration ratings, .84 for commitment ratings, and .75 for identity status judgments. Research using similar interviews has reported comparable reliabilities (Archer, 1989b). After reliabilities were calculated, differences in ratings were reviewed by the researchers and the status most closely related to the rating criteria was selected as the final judgement.

RESULTS

Gender differences in scores on exploration and commitment were compared using t-tests that were studentized to control for the number of comparisons. Analyses were completed for scores on each domain separately; for the three ideological domains combined; for the three interpersonal domains combined; and for all six domains combined.

Mean scores and standard deviations for the exploration and commitment scores may be seen in Tables I and 2, respectively. T-tests for independent sample means detected significant gender differences in identity explorations for the areas of politics (t(1, 207) = 2.44, p [less than] .01) and sex roles (t(1,207) = 2.32, p [less than] .02). For identity commitment, there were significant gender differences in the areas of religion (t(1, 207) = 2.48, p [less than] .01); politics (t(1, 207) = 5.47, p [less than] .02); and dating (t(1, 207) = 3.42, p [less than] .01). These results indicated that, in the area of politics, males explored more than did females. However, in sex roles, females were more active explorers. They were more committed in the areas of religion and dating, whereas males were more committed in politics. Tables 1 and 2 also show the results for the combined interpersonal, ideological, and overall commitment and exploration scores. In general, these gender comparisons did not produce statistically significant differences by gender. However, for combined interpersonal commitment, females scored higher (t(1, 207) = 3.34, p [less than] .01).

The demographic composition of the sample and personal data were analyzed to assess sampling artifacts that might account for the obtained gender differences. Year of college entrance (short-term cohort effects), mother's work status (employed vs. housewife), race, and family socioeconomic level showed no significant contribution to the gender differences described above.

Intercorrelations among the exploration scores and commitment scores for each domain also were examined [ILLUSTRATION FOR FIGURE 1 OMITTED]. In general, for both males and females, exploration in one domain tended to be related to exploration in other domains. For both males and females there were intercorrelations between sex role exploration and exploration in occupation and religion, between dating exploration and political exploration, and between exploration in occupation and religion. In general, exploration in friendship did not relate with exploration in any other area for either gender. For females political exploration was associated with exploration in occupation, whereas for males political exploration was associated with sex role and religious exploration. Dating exploration was associated with occupational exploration for males and with sex role exploration for females.

Table 1

Mean scores and standard deviations for identity exploration by
gender

 Males Females

 M sd M sd Range

Exploration

Individual Domains

Occupation 2.59 .63 2.64 .70 1 - 4
Religion 2.19 .74 2.21 .70 1 - 4
Politics(**) 2.09 .69 1.86 .68 1 - 4
Friendship 2.51 .52 2.54 .64 1 - 4
Dating 2.30 .92 2.37 .58 1 - 4
Sex Roles(*) 2.29 .57 2.48 .61 1 - 4

Combined Domains

Ideological 6.87 1.41 6.71 1.48 1 - 16
Interpersonal 7.09 .97 7.39 1.26 1 - 16
Overall 13.97 2.00 14.10 2.20 1 - 24

* p .05
** p .01

For males only, there tended to be a relationship between exploration and commitment. In fact, for males, exploration in politics and in friendship were both related to the overall commitment score (r = .36, p [less than] .01 and r = .26, p [less than] .01, respectively). Also, for males, dating commitment related to exploration in friendship, religion, and politics. Dating commitment also correlated significantly with the overall exploration score for males (r = .38, p [less than] .01).

The distribution of identity statuses by domain for males and females was examined using chi-square analyses. A series of 2 x 4 (gender by identity status) chi-square analyses indicated significant gender differences in identity status for the domains of politics ([[Chi].sup.2] (3, 205) = 8.59 p [less than] .04); dating ([[Chi].sup.2] (3,205) = 11.29, p [less than] .02); and sex roles ([[Chi].sup.2] (3, 205) = 8.80, p [less than] .04). Each significant chi-square analysis was followed with a standard test for binomial proportions. These analyses indicated that in politics, fewer males were in the diffused status; however, for dating and sex roles, there were fewer females in the diffused status. Status by gender chi-square analyses also were completed on the combined ideological, interpersonal, and overall scores; no significant gender differences were found. The frequency distributions of identity status for the individual and combined domains are presented in Tables 3 and 4.

Table 2

Mean scores and standard deviations for identity commitment by
gender

 Males Females

 M sd m sd Range

Commitment

Individual Domains

Occupation 2.64 .98 2.59 .93 1 - 4
Religion(**) 2.62 .68 2.87 .77 1 - 4
Politics(*) 2.47 .75 2.23 .73 1 - 4
Friendship 2.98 .50 3.08 .48 1 - 4
Dating(***) 2.72 .56 2.96 .47 1 - 4
Sex Roles 2.77 .56 2.87 .54 1 - 4

Combined Domains

Ideological 7.72 1.68 7.69 1.46 1 - 16
Interpersonal(**) 8.44 1.00 8.91 1.02 1 - 16
Overall 16.20 2.20 16.60 1.90 1 - 24

* p .05
** p .01
*** p .001

Finally, a frequency count was conducted separately for males and females to determine the extent to which an individual who was in a given status in one domain was in that status in other domains. Table 5 shows the number of times subjects fell in a given status over several domains. Only one subject, out of 208 was in Moratorium in four or more domains. Only nine subjects were in Moratorium in three domains; 31 subjects were in Moratorium in two domains, with the great majority (80%) in Moratorium in only one or none of the domains. Thirty-nine percent were not in Moratorium in any area.

Similarly, the students were not likely to be broadly diffused across domains. Ninety-two percent were diffused in 0, 1, or 2 domains. On the other hand, they were more likely to be broadly foreclosed or achieved, spanning across four domains: for Foreclosure (86%); Achievement, (76%). Chi-square analyses (gender by person-domain) indicated no gender differences.

DISCUSSION

Gender differences in identity development were examined in a group of college-age youth for six identity domains. Analyses were based on the use of exploration and commitment scores as well as individual status assignments. Identity development was viewed both as an individual process and as a reflection of the larger cultural milieu.

Table 3

Frequency distributions of identity statuses by gender for six
domains

 Identity Status

Gender Achieved Moratorium Foreclosed Diffused

Occupation

Males 30 36 27 16
Females 27 39 22 13
Total 57 (27%) 75 (36%) 49 (23%) 29 (14%)

Religion

Males 18 20 48 23
Females 21 12 53 15
Total 39 (19%) 32 (15%) 101 (48%) 38 (18%)

Politics(a)

Males 13 16 41 39
Females 09 08 28 56
Total 22 (11%) 24 (22%) 69 (33%) 95 (45%)

Friendship

Males 52 04 42 11
Females 53 03 42 03
Total 105 (50%) 07 (03%) 84 (40%) 14 (07%)

Dating(a)

Males 26 07 48 28
Females 35 05 52 09
Total 62 (29%) 12 (06%) 100 (48%) 37 (18%)

Sex Roles(a)

Males 20 16 58 14
Females 37 13 42 09
Total 57 (27%) 29 (13%} 100 (48%} 23 (11%)

a Chi-Square .05
Table 4

Frequency distributions of identity statuses by gender for combined
domains

 Identity Status

Gender Achieved Moratorium Foreclosed Diffused

Ideological

Males 14 15 34 14
Females 13 08 27 16
Total 27 (19%) 23 (16%) 61 (43%) 32 (22%)

Interpersonal

Males 25 02 45 09
Females 36 02 44 04
Total 61 (34%) 04 (02%) 89 (50%) 13 (07%)

Overall

Males 37 10 48 14
Females 39 03 50 09
Total 76 (36%) 13 (06%} 98 (47%) 23 (11%)
Table 5

Number of domains per identity status for males (N=107)
and females (N=101)

 Number of Domains

 0 1 2 3 4 5 6

Females

Achieved 16 35 18 19 11 2 0
Foreclosed 10 20 23 26 17 3 2
Moratorium 42 41 15 3 0 0 0
Diffusion 28 47 22 2 2 0 0

Males

Achieved 30 33 24 11 8 1 0
Foreclosed 4 24 34 22 13 7 3
Moratorium 40 44 16 6 1 0 0
Diffusion 33 35 26 7 3 2 0

In general, gender differences were present when domains were examined separately but not when they were combined. Males were more likely to explore and commit in sex roles and to commit in religion and dating. When identity statuses were examined, males were less likely to be diffused in politics, and females were less likely to be diffused in dating and sex roles.

These results suggest that gender differences are evident in the identity development process, and that they align along the traditional division between interpersonal and ideological development. While both of these areas are important in the lives of both men and women, some aspects of interpersonal identity development occur later for a larger proportion of men, and some aspects of ideological identity development, in this case politics, occur later for a larger proportion of women.

It is also interesting to note that while commitment in general did not consistently relate to exploration or to commitment, dating commitment was an exception. Commitment in dating was related to overall exploration for males who, when committed in dating also were exploring in politics, religion, and friendships. Perhaps a sense of being "settled" in dating frees persons, especially males, to engage in the more "unsettling" business of exploring one's social and political environments. A foundation of intimacy, as reflected by dating commitment, therefore, may be just as relevant for male identity development as for female identity development, as earlier research has suggested.

It also is interesting that sex role exploration resembled an ideological area in that it correlated with exploration in the other ideological areas, whereas on the whole, friendship and dating did not. Because the section of the interview in sex role identity probes the respondents' ideology concerning power in relationships, it may be useful to think of sex roles as a kind of "political" domain rather than a strictly interpersonal, or relationship domain. Future research might examine this further.

The results of the present study suggest that global analyses hide some meaningful gender differences. Whereas some differences were apparent when specific domains were considered individually, they generally were not discernible from the scores that summarized across domains. If we were to confine our findings to analyses which combine domains, we would conclude that there are no gender differences in identity development. We would further erroneously speculate that society has changed over the last few decades to eliminate differential levels of exploration in major social context areas for males and females. Although it is common for studies completed during the 1980s to use a range of content areas, a surprising number of researchers reported an overall status score, or collapsed across the interpersonal and ideological domains to provide two overall scores. Frequently, gender differences in identity status were not found.

The lack of gender differences in the areas of occupation and in friendship is not surprising. One explanation is, of course, that this is a college sample, and much of the college experience is oriented toward these areas for both males and females. In addition, in the occupational realm, females have increasingly experienced more role models such as their mothers and other female mentors who are invested in careers. More American women than ever before are working outside the home. Whether due to economic necessity or to personal satisfaction and interest, this is a trend that is likely to continue (England & McCreary, 1987). Approximately 60% of women now work for a living and 95% of young women expect to work outside the home during adulthood (Betz & Fitzgerald, 1987).

Although women are increasingly entering the work force, housework and family care remain largely the responsibility of women (Feree, 1987; Schooler, Miller, Miller, & Richtand, 1984; Nyquist, Slivken, Spence, & Helmreich, 1985). This may explain the lag in sex role exploration among males. Because women expect to combine the dual roles of homemaker and career, they may invest more of themselves in an effort to cope with potential role conflicts or overload. Males, on the other hand, still expect to fulfill their roles as provider, which in many cases, translates into increased career efforts (Archer, 1992). Archer (1985) found evidence among high school students of different expectations by males and females concerning the combining of career and household work. She found that girls were much more likely to have considered the potential conflicts between career and family; males were more likely to be diffused in this area. If the division of labor in the home becomes more equitable, dual sex role expectations for males may become more salient.

The domain-specific analyses also direct attention to key areas of significance for the youth of our society, regardless of gender. Archer (1989b) has suggested that youth in this country are typically nonexploring in their identity status and suggests that the use of global scores may hide nuances in identity development within domain. The results of the present study support Archer's suggestion. The present results permit a conclusion that the use of combined status assignments presents an incomplete picture of identity development. According to our results, the college-aged participants were generally nonexploring only in some domains, and especially in politics. Over 75% of the students were foreclosed or diffused in politics, and about 60% were in one of these two statuses for dating, religion, and sex roles. However, in the occupational and friendship domains, only 35 to 45% were foreclosed or diffused. In the domain of friendship, half of the students were achieved. Across all domains taken together, however, slightly over half of the subjects (57%) were either foreclosed or diffused.

The intercorrelations among the exploration scores for both males and females tend to suggest a somewhat generalized exploration across domains. However, the more detailed case-by-case analysis showing the number of domains in each status suggests that college youth tend not to be in Moratorium generally across domains at any given period of time. These two analyses present somewhat different pictures of identity exploration. The importance of this is to demonstrate the complexity of research in identity development. Conclusions based on any single analysis alone may not present a complete picture of the process.

The finding that the majority of college students were in Moratorium in only one or none of the domains at a given time suggests the need for a cautious approach to exploration among college youth. Inasmuch as Moratorium represents identity exploration, we might look to Erikson's assertion that exploration is at the heart of the identity crisis (Erikson, 1959). One might wonder whether exploring in more than one domain may be inherently unstable or disruptive, involving dysphoria, confusion, and social unpredictability, as Erikson has suggested (Erikson, 1959). Further research is needed to pursue this question.

The findings on political identity development deserve particular attention. Almost half of the youth participating in this study, both males and females, were diffused in the area of politics, indicating that they have neither explored nor committed in politics. Youth may suffer from a lack of role models for engaging in exploration and commitment in this domain. Our current society may include a large portion of adults who are also apathetic about politics. This may reflect a general decline in belief in political institutions and the individual's capacity to make a difference. Additionally, adults may be less likely to involve youth in active debates that would promote a sense of political identity because of changes in the institutions of family and schools. Finally, the complexity and fast-changing pace of society may press youth to focus most of their energy on occupational and interpersonal identity development. Related to this, concerns with economic security and gaining status in the work force may also steer youth toward occupational identity and less toward politics.

There are obvious limitations to the generalizability of the conclusions drawn from this study. The sample is primarily white and middle class college youth from two-parent, nondivorced families. Nevertheless, these findings are consistent with other studies that include differing samples of youth.

Our intent in this study was to encourage other researchers to examine identity development across a range of domains. We emphasized the strategy of defining identity development based on specific domains rather than focusing solely on global scores or status assignments. In addition, more research attention should be directed to different modes of analysis. The use of domains, statuses, and exploration and commitment scores presents a more complex picture of the identity development of American youth, and of our society.

REFERENCES

Adams, G. R. (1985). Family correlates of female adolescents' ego-identity development. Journal of Adolescence, 8, 69-82.

Adams, G. R., & Fitch, S. A. (1982). Ego stage and identity status development: A cross-sequential analysis. Journal of personality and social psychology, 43, 574-583.

Adams, G. R., & Jones, R. M. (1983). Female adolescents' identity development: Age comparisons and perceived child-rearing experience. Developmental Psychology, 19, 249-256.

Archer, S. W. (1985a). Career and/or family: The identity process for adolescent girls. Youth and Society, 16, 289-314.

Archer, S. L. (1985b). Identity and the choice of social roles. In A. S. Waterman (Ed.), Identity in adolescence: Processes and context (pp. 79-100). San Francisco: Jossey-Bass.

Archer, S. L. (1989a). Gender differences in identity development: Issues of process, domain, and timing. Journal of Adolescence, 12, 117-138.

Archer, S. L. (1989b). The status of identity: Reflections on the need for intervention. Journal of Adolescence, 12, 345-359.

Archer, S. (1992). A feminist's approach to identity research. In G. Adams, T. Gullotta, & R. Montemayor (Eds.), Adolescent identity formation (pp. 25-49). Newbury Park, CA: Sage.

Bernard, H. S. (1981). Identity formation during late adolescence: A review of some empirical findings. Adolescence, 16, 349-358.

Betz, N. E., & Fitzgerald, L. E., (1987). The career psychology of women. New York: Academic Press.

Bilsker, D., Schiedel, D., & Marcia, J. (1988). Sex differences in identity status. Sex Roles, 18(2/4), 231-236.

Cella, D. F., DeWolfe, A. S., & Fitzgibbon, M. (1987). Ego identity status, identification, and decision-making style in late adolescents. Adolescence, 22, 849-861.

Cooper, C. R., & Grotevant, H. D. (1987). Gender issues in the interface of family experience and adolescents' friendship and dating identity. Journal of Youth and Adolescence, 16, 247-264.

Cooper, C. R., Grotevant, H. D., & Condon, S. M. (1983). Individuality and connectedness in the family as a context for adolescent identity formation and role-taking skill. In H. D. Grotevant, & C. R. Cooper (Eds.), Adolescent development in the family: New directions for child development (p. 22). San Francisco: Jossey-Bass.

Craig-Bray, L., Adams, G. R., & Dobson, W. R. (1988). Identity formation and social relations during late adolescence. Journal of Youth and Adolescence, 17, 173-187.

England, P., & McCreary, L. (1987). Gender inequity in paid employment. In B. B. Hess, & M. M. Feree (Eds.), Analyzing gender: A handbook of social science research. (pp. 286-320). Newbury Park, CA: Sage.

Erikson, E. H. (1959). Late adolescence. In D. H. Funkerstein (Ed.), The Student and mental health. Cambridge: Riverside Press.

Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.

Fannin, P. N. (1979). The relation between ego identity status and sex role attitude, work role salience, typicality of major and self-esteem in college women. Journal of Vocational Behavior, 14, 12-22.

Feree, M. M. (1987). She works hard for a living: Gender and class on the job. In B. B. Hess, & M. M. Feree (Eds.), Analyzing gender: A handbook of social science research (pp. 322-347). Newbury Park, CA: Sage.

Gilligan, C. (1982). In a different voice: Psychological theory and women's development. Cambridge, MA: Harvard University Press.

Grotevant, H. D., & Adams, G. R. (1984). Development of an objective measure to assess ego-identity in adolescence: Validation and replication. Journal of Youth and Adolescence, 3, 419-438.

Grotevant, H. D., & Cooper, C. R. (1981). Assessing adolescent identity in the areas of occupation, religion, politics, friendships, dating, and sex roles: Manual for administration and coding of the interview. (Ms. No. 2295.) Journal Supplement Abstract Service Catalog of Selected Documents in Psychology, 11, 52.

Grotevant, H. D., & Thornbecke, W. L. (1982). Sex differences in styles of occupational identity formation in late adolescence. Developmental Psychology, 3, 396-405.

Grotevant, H. D., Thornbecke, W. L., & Meyer, M. L. (1982). An extension of Marcia's identity status interview into the interpersonal domain. Journal of Youth and Adolescence, 11, 33-47.

Harter, S. (1990). Self and identity development. In S. Feldman, & G. Elliott (Eds.), At the threshold: The developing adolescent (pp. 352-387). Cambridge, MA: Harvard University Press.

Hodgson, J. W., & Fischer, J. L. (1979). Sex differences in identity and intimacy development in college youth. Journal of Youth and Adolescence, 8, 37-50.

Hollingshead, A. B. (1965). Four Factor Index of Social Status New Haven, CT: Yale University Press.

Jackson, E. P., Dunham, R. M., & Kidwell, J. S. (1987). The effects of gender and of family cohesion and adaptability on identity status. Journal of Adolescent Research, 5, 161-174.

Josselson, R. (1982). Personality structure and identity status in women as viewed through early memories. Journal of Youth and Adolescence, 11, 293-299.

Kacerquis, M. S., & Adams, G. R. (1980). Erikson stage resolution: The relationship between identity and intimacy. Journal of Youth and Adolescence, 9, 117-126.

Kimmel, D. C., & Weiner, I. B. (1985). Adolescence: A developmental transition. Hillsdale, NJ: Erlbaum.

Marcia, J. E. (1964). Determination and construct validity of ego identity status. Unpublished doctoral dissertation, Ohio State University.

Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personality and Social Psychology, 21, 551-559.

Marcia, J. E. (1967). Ego identity status: Relationship to change in self-esteem, "general maladjustment," and authoritarianism. Journal of Personality, 35, 119-133.

Marcia, J. E. (1968). The case history of a construct: Ego identity status. In E. Vinache (Ed.), Readings in general psychology (pp. 325-332). New York: American Book Co.

Meilman, P. W. (1977). Crisis and commitment in adolescence: A developmental study of ego identity status. Dissertation Abstracts International, 38, 2870B.

Nyquist, L., Slivken, J., Spence, J. T., & Helmlrich, R. L. (1985). Household responsibility in middle-class couples: The contribution of demographic and personality variables. Sex Roles, 12, 15-34.

Offer, D., Marcus, D., & Offer, J., L. (1970). A longitudinal study of normal adolescent boys. American Journal of Psychiatry, 126, 917-924.

Patterson, S., Sochting, I., & Marcia, J. (1992). The inner space and beyond: Women and identity. In G. Adams, T. Gullotta, & R. Montemayor (Eds.), Adolescent identity formation (pp. 9-24). Newbury Park, CA: Sage.

Protinsky, H., & Wilkerson, J. (1986). Ego identity, egocentrism, and formal operations. Adolescence, 82, 149-160.

Read, D., Adams, G. R., & Dobson, W. R. (1984). Ego-identity status, personality, and social-influence style. Journal of Personality and Social Psychology, 46, 169-177.

Scheidel, D. G., & Marcia, J. E. (1985). Ego identity, intimacy, sex role orientation, and gender. Developmental Psychology, 18, 149-160.

Schooler, C., Miller, J., Miller, K., & Richtand, C. N. (1984). Work for the household: Its nature and consequences for husbands and wives. American Journal of Sociology, 90, 97-124.

Schenkel, S. (1974). Relationship among ego identity status and field independence in traditional femininity. Journal of Youth and Adolescence, 4, 73-82.

Shanken, L. C. (1984). Identity and intimacy formation in young adults. Unpublished doctoral dissertation, Columbia University.

Steinberg, L. (1989). Adolescence (p. 262). New York: McGraw-Hill.

Streitmatter, J. L., (1988). Ethnicity as a mediating variable of early adolescent development. Journal of Adolescence, 11, 335-346.

Streitmatter, J. L., & Pate, G. (1989). Identity status development and cognitive prejudice in early adolescents. Journal of Early Adolescence, 9, 142-152.

Wagner, J. A. (1987). Formal operations and ego identity in adolescence. Adolescence, 22, 23-35.

Waterman, A. (1982). Identity development from adolescence to adulthood: An extension of theory and a review of research. Developmental Psychology, 18, 341-358.

Werrbach, G. B., Grotevant, H. D., & Cooper, C. R. (1990). Gender differences in adolescents' identity development in the domain of sex role concepts. Sex Roles, 23, 349-362.

Ellen Pastorino, Ph.D., Associate Professor, Gainesville College, Gainesville, Georgia.

Richard M. Dunham, Ph.D., Professor, Jeannie Kidwell, Ph.D., Adjunct, and Roderick Bacho, M.S., Florida State University.

Susie D. Lamborn, Ph.D., Assistant Professor, University of West Florida.