Printer Friendly

Gender differences in identity development: an examination of longitudinal data.

Identity development, as theorized by Erikson (1968) over twenty years ago, is the focus of psychosocial development during the adolescent years. Of the "eight stages of man," the fifth stage, identity achievement versus role confusion, involves youth from the onset of puberty.

This stage has been an area of particular interest, commencing with attempts to operationalize Erikson's conceptualization of the growing-up process of young people. Marcia (1966, 1980) conceptualized the process of identity development as passage through some or all of four identity statuses. The central issues of crisis and commitment correspond to Erikson's concept of identity achievement. In order to operationalize this interpretation, Marcia developed an interview format as a means of determining an individual's identity status.

Examination of identity development by gender has received increased attention. Two particular veins of this research have surfaced. Studies conducted by Archer(1982, 1989), Adams and Fitch (1982), and Grotevant and Thorbecke (1982) have examined gender differences in identity development within the more traditional Eriksonian framework. In contrast, work by Gilligan and associates (1979, 1988) and Lyons (1983) challenges Erikson's theory of identity development with regard to gender. They argue that Erikson's concepts of psychosocial development are male oriented and do not reflect the female experience. Because the methodologies, analyses, and findings are in stark contrast, a brief review of each side of the issue of whether the process of identity development tends to be gender specific follows.

Archer's (1982) cross-sectional study of early and middle adolescents addressed the issue of gender differences in identity development, in addition to age-related differences. Using Marcia's (1966) ego-identity interview format, a total of 160 subjects were interviewed. Results indicated no significant gender differences according to grade level. Further, when responses were examined according to specific content areas (e.g., vocational, religious, and political), no significant differences in identity status by gender were obtained. Archer concluded that females and males proceed through the identity statuses in like fashion.

A second work by Archer (1989) examined gender differences in identity development according to three dimensions. Does the process of identity formation differ by gender? Do the domains in which females and males define themselves differ? Is the timing of attention to identity development concerns differentiated by gender? Results from the three studies suggested that the genders involved themselves in the identity process similarly, with the exception of the foreclosure status. Males were more often identified as foreclosed than were females. When measuring political ideology, males were more foreclosed and females more diffused. Moratorium and identity achievement were more prevalent for females than for males in the area of family roles. Finally, Archer found no timing differences in the identity development process by gender in two of the three studies. As in the earlier study, Archer used Marcia's (1966) interview format to obtain data.

In light of the considerable research examining differences between the sexes during adolescence (other than identity development), this finding may appear puzzling. For example, it has been substantiated that females mature earlier than males in several developmental aspects; females reach puberty before males (e.g., Blyth, Simmons, & Zakin, 1985) and, although results are somewhat conflicting, some studies indicate gender differences in psychosocial development. Douvan and Adelson (1966) concluded that the route to definition of oneself is gender specific. Self-esteem is an additional area of apparent gender differentiation. A number of studies suggest that females have lower self-esteem than males, particularly during early adolescence (Simmons, Brown, Bush, & Blyth, 1978).

Principally through ethnographic methodology, the work of researchers such as Gilligan (1982, 1988) and Miller (1976) suggests that the path of identity development for females may be different than that for males. Gilligan (1979) argues that Erikson's theory, embodied in the "eight stages of man," is based upon a masculine developmental model and is biased to the point of excluding the developmental process of females. Each of Erikson's stages leading to adolescence, the focus of which is upon identity development, concerns itself with the successful resolution of issues which are most consistent with male socialization. In sequence, psychosocially "healthy" children succeed in individuation, a sense of separateness and agency, a growing sense of autonomy, and (preceding adolescence) a demonstration of competence through a "mastery of the technology of their culture in order to recognize themselves and be recognized as capable of becoming adults" (Gilligan, 1979, p. 437). Gilligan (1979) describes the bias inherent in Erikson's theory: "for men, identity precedes intimacy and generativity in the optimal cycle of human separation and attachment; for women, these tasks seem instead to be fused. Intimacy precedes, or rather goes along with, identity as the female comes to know herself as she is known, through her relationships with others".

The essence of Gilligan's work is the idea that females tend to define themselves through their relationships with others, while males follow "traditional masculine" lines of self-definition--according to their occupational selves. Gilligan examined these ideas within a framework of the conception of self and morality. Similarly, Miller (1976) concluded that women's conception of themselves is associated with their ability to make and maintain relationships.

The findings of Noddings' (1983) cross-sectional study of male and female children, adolescents, and adults supported Gilligan's central assertion of gender differences in psychological development. Within a framework of moral development, she concluded that females more frequently use a connected or relationship-oriented self-definition, while males more often use a separate/objective self-characterization.

Gilligan's work, and that of others whose conclusions are similar, maintains that the identity development process of females is quite different from that of males. Intimacy is a primary issue for females. Early on, females' constructions of themselves focus on building and maintaining relationships. This focus may lessen consideration by females of those issues which are held to be important for males, and the essence of Erikson's notion of successful identity achievement--autonomy and individuation.

In contrast to these findings, quantitative work done within the more traditional Eriksonian framework via Marcia's (1980) Identity Interview has suggested that identity development is essentially the same regardless of gender. If Gilligan's conception of female identity development is correct, would one expect to find discrepancies by gender on Eriksonian measures of identity development?

This study sought to examine two questions about identity development using the Extended Objective Measure of Ego Identity Status (EOM-EIS). The first question was one of timing. During adolescence, do females and males address identity issues at similar or different times? The second question was whether the two genders address identity comparably as measured by the EOM-EIS.



Subjects for this study were drawn from a junior high school in an urban southwestern setting. The entire student population of seventh and eighth graders, approximately 500 students, completed a questionnaire. Three years later, a follow-up was conducted at the two high schools which these students attended based on their residence. All tenth and eleventh graders in the two high schools were administered the same questionnaire. A total of 105 respondents were identified as the cohort.

The attrition rate was not unexpected. The mobility rate in this urban center is quite high. Many schools in the district experience up to 50% turnover in a given year. The breakdown according to gender was 44 (41.9%) male and 61 (58%) female. Because the sample was overwhelmingly white (97 students, or 92%), race/ethnicity was excluded as a variable.


The Extended Objective Measure of Ego Identity Status (Grotevant & Adams, 1984) was used to measure the identity status of the respondents on both assessment occasions. The questionnaire has 64 items which are designed to assess ego identity with regard to religious, occupational, political, philosophical, and social contexts. The instrument is designed around the dimensions of ideological and interpersonal identity development. Two items measure each of the four statuses (diffusion, foreclosure, moratorium, and achievement)for each of the four ideological and four interpersonal content areas.

Previous studies have reported impressive psychometric properties for the EOM-EIS (e.g., Grotevant & Adams, 1984). Alpha coefficients from Jones and Hartmann's (1985) study ranged from .75 (diffusion) to .85 (foreclosure). Estimates of reliability and validity with a sample of early and middle adolescents were acceptable as well (Jones & Streitmatter, 1987). Cronbach alpha coefficients were .52 for the interpersonal diffusion subscale, .80 for total foreclosure, .74 for moratorium, and .77 for achievement, with a median coefficient of .67. Estimates of convergent-divergent validity ranged from .52 to .80.


During initial data collection, questionnaires were administered by social studies teachers to all students in attendance that day. Teachers were allowed to help students with vocabulary problems; otherwise, they completed the instruments on their own. During the second administration (in the high schools), all tenth and eleventh graders in attendance that day completed the questionnaire in their English classes. Again students could be assisted with vocabulary if necessary. None had a primary language other than English, and none were classified as special education students.

On the instruction sheet for both administrations, students provided their telephone numbers. The instruction sheet for the high school administration included an item in which students indicated the junior high school they attended. Following the survey in the high schools, students were matched using the phone numbers. In several cases, students' phone numbers had changed. In those cases, birth dates were matched. In a few instances, parents' occupations also were used to verify that the student was appropriately part of the longitudinal sample.


Three methods of data analyses were employed. Due to the limited sample size, one-way analysis of variance was used to compare identity scores across gender at Time 1 and again at Time 2. Paired t tests were then employed to evaluate any significant change over time in total identity scores separately for each gender. Finally, analysis of covariance was employed to investigate the relationship between identity scores at Time 1 and Time 2. Four identical analyses were conducted, each employing one of the four total identity scales at Time 2 (foreclosure, diffusion, moratorium, and achievement) as dependent measures and their Time 1 counterparts as covariates.

The analysis of variance for identity status development over time (junior high and high school administrations) and across gender provided the following results. When total diffusion was entered as the dependent variable, a significant decrease in scores was evident for both genders. The decrease in total diffusion scores was similar for boys and girls; that is, non-significant gender differences emerged in the analysis.
Time 1
Source SS df MS F
Between Groups 1.56 1 1.56 .03
Within Groups 5108.50 103 49.60
Total 5110.06 104
Time 2
Between Groups .66 1 .66 .01
Within Groups 6198.54 103 60.18
Total 6199.20 104

As with diffusion scores, no significant gender differences in foreclosure scores were present at Time 1 and Time 2. Between early and middle adolescence, the foreclosure scores of both males and females decreased significantly, although no statistically significant differences between genders emerged.

No significant effects by gender were observed when total moratorium was entered as the dependent variable. Once again, scores for males and females were parallel, with an apparent rise in moratorium. Increased activity in dealing with identity issues was apparent for both genders as they moved from early to middle adolescence.
Time 1
Source SS df MS F
Between Groups 155.83 1 155.83 2.99
Within Groups 5366.22 103 52.10
Total 5522.05 104
Time 2
Between Groups 21.72 1 21.72 .52
Within Groups 4297.84 103 41.73
Total 4319.56 104

There were no significant effects by gender when total achievement was entered as the dependent variable. This cohort did not appear to be identity achieved. It is interesting to note that male scores actually decreased over time, while female scores were stable.

Collectively, results from the analysis of variance are consistent with Erikson's theory. Total foreclosure and diffusion scores decreased from Time 1 to Time 2. Total moratorium scores increased with age, indicating that as the male and female respondents grew older, they experienced more intense questioning and exploring of identity. Finally, total achievement scores showed no significant change over time, suggesting that respondents were not yet finished struggling with alternatives in order to arrive at a clear commitment.
Time 1
Source SS df MS F
Between Groups 76.49 1 76.49 1.50
Within Groups 5238.90 103 50.86
Total 5315.39 104
Time 2
Between Groups 6.42 1 6.42 .11
Within Groups 6100.63 103 59.23
Total 6107.05 104

Table 5 presents the mean identity scores for males and females for Time 1 and Time 2. Males demonstrated a significant decrease in total foreclosure and diffusion scores and a significant increase in moratorium scores over time. Consistent with Erikson's theory, as boys leave early adolescence and obtain greater life experience, they increasingly shed imposed values and attitudes prescribed by parents and/or other significant adults. No significant difference was observed in total achievement scores between Time 1 and Time 2.

The findings were similar for girls. Mean scores for total foreclosure and diffusion decreased significantly between Time 1 and Time 2. Total moratorium scores for females increased significantly over time, whereas achievement scores remained stable. Although the levels of significance were somewhat more variable for males as compared to females, the pattern of change was the same when examining total identity scores over time.
Time 1
Source SS df MS F
Between Groups 10.42 1 10.42 .19
Within Groups 5541.42 103 53.80
Total 5551.84 104
Time 2
Between Groups 221.57 1 221.57 3.36
Within Groups 6783.35 103 65.86
Total 7004.92 104


The results of this study, taken as a whole, appear consistent with Erikson's theory of psychosocial development within the fifth stage of identity achievement versus role confusion. As respondents aged over the three-year period, a pattern of decreased foreclosure and diffusion scores was found. Moratorium scores increased significantly while achievement scores remained fairly stable. The individuals in the study were moving from the foreclosure status, where attitudes and values are accepted from parents or other important adults, into that of moratorium, where such attitudes are questioned. Following this, the respondents began to consider, within the framework of exploration (Waterman, 1982) or crisis (Erikson, 1968), a perspective which they felt suited them individually. As indicated by the lack of change in achievement scores, this process of exploration was not yet complete. An achieved identity status was not yet accomplished by the respondents, as measured by the EOM-EIS.
Time 1 Time 2
 x SD x SD t
Males 56.11 7.36 50.44 7.41 4.99(*)
Females 56.36 6.80 50.60 7.93 5.88(*)
Males 59.98 6.52 42.05 6.58 13.75(*)
Females 57.51 7.69 42.97 6.37 12.03(*)
Males 54.80 7.65 57.61 8.93 -2.40(*)
Females 53.07 6.74 58.11 6.67 -5.23(*)
Males 50.59 7.31 48.32 7.73 1.67
Females 51.23 7.35 51.26 8.38 -.03
* p |is less than~ .05

When comparing identity status development by gender, the results of this study support previous work based on Eriksonian theory. As with Archer's work (1982, 1989), there were no statistically significant differences found between boys and girls in their progress through the identity statuses. However, several differences exist between Archer's (1982, 1989) work and the present study. Archer employed Marcia's interview format and used a cross-sectional design. In contrast, this work utilized the EOM-EIS with longitudinal data. The findings of the present study may serve to underscore that, when utilizing either of the primary assessment tools based on Eriksonian theory, the identity development process does not appear to be differentiated by gender.

What, then, of Gilligan's work which suggests that females' identity concerns are focused on issues other than those associated with the path toward identity achievement--individuation and autonomy? In fact, the "female" issues of connectedness and relationship maintenance are quite opposite from "male" identity issues.

The discrepancy between Gilligan's work and the findings of Archer (1982, 1989) as well as the present study may lie in the different methodologies and assessment tools. Eriksonian-based identity development instruments, principally Marcia's (1980) interview format and the EOM-EIS (Grotevant and Adams, 1984), do not examine intimacy or relationship issues during the identity achievement versus diffusion stage. In contrast, Gilligan's ethnographic interviews dwell on these issues. Perhaps the contradictory nature of the results of these two veins of research lies within the different developmental concerns tapped; therefore, each offers only a partial picture of identity development by gender. As this study's findings suggest, females, similar to males, deal with such identity issues as independence. Gilligan suggests they also construct parts of themselves in accordance with caring and relationship nurturance. Females may understand the demands of the male-dominated society which clearly pictures success in traditional "masculine" terms. Therefore, their identity development may be structured according to those demands. Simultaneously, females develop the "feminine" parts of their identity in response to other societal demands.

As Archer (1989) points out, identity development may be more complex for females. The contradiction in the research findings taken as a whole may indicate that only a partial picture has emerged in considering female identity development, with each of the two approaches resulting in separate views. Perhaps what is needed is a more comprehensive method of investigating female identity development, whereby both the "masculine" pathway according to Erikson and the "feminine" pathway according to Gilligan can be made complementary rather than conflicting.

Reprint requests to Janice Streitmatter, Associate Professor, Department of Teaching and Teacher Education, University of Arizona, Tucson, Arizona 85721.


Adams, G.R., & Fitch, S.A. (1982). Ego stage and identity status development: A cross-sectional analysis. Journal of Personality and Social Psychology, 43, 574-583.

Archer, S.L. (1982). The lower age boundaries of identity development. Child Development, 53, 1551-1556.

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

Blyth, C., Simmons, R., & Zakin, D. (1985). Satisfaction with body image for early adolescent females: The impact of pubertal timing within different school environments. Journal of Youth and Adolescence, 14, 227-236.

Douvan, E., & Adelson, J. (1966). The adolescent experience. New York: Wiley.

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

Gilligan, C. (1979). Woman's place in man's life cycle. Harvard Review, 49, 431-446.

Gilligan, C. (1982). In a different voice: Women's conceptions of the self and of morality. Cambridge, MA: Harvard University Press.

Gilligan, C., Ward, J. V., Taylor, J. M., & Bardige, B. (1988). Mapping the moral domain. 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, 13, 419-438.

Grotevant, H. D., & Thorbecke, W. (1982). Sex differences in styles of occupational identity formation in late adolescence. Developmental Psychology, 18, 396-405.

Jones, R. M., & Hartmann, B. R. (1985). Ego identity, self esteem, and substance abuse during adolescence. Tucson: University of Arizona, College of Medicine (ERIC Document Reproduction Service No. ED 263 454).

Jones, R. M., & Streitmatter, J. L. (1987). Validity and reliability of the EOM-EIS for early adolescents. Adolescence, 22, 647-660.

Lyons, N. (1983). Two perspectives: On self, relationships and morality. Harvard Review, 53, 125-145.

Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personality and Social Psychology, 3, 551-558.

Marcia, J. E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handbook of adolescent psychology. New York: Wiley.

Miller, J. B. (1976). Toward a new psychology of women. Boston: Beacon Press.

Noddings, N. (1983). Formal modes of knowing. In E. Eisner (Ed.), Learning and teaching the ways of knowing. Chicago: University of Chicago Press.

Simmons, R., Brown, L., Bush, D., & Blyth, D. (1978). Self esteem and achievement of black and white adolescents. Social Problems, 26, 86-96.

Waterman, A. S. (1982). Identity development from adolescence to adulthood: An extension of theory and a review of the research. Developmental Psychology, 18, 341-358.
COPYRIGHT 1993 Libra Publishers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Streitmatter, Janice
Date:Mar 22, 1993
Previous Article:Predictors of depression in street youth.
Next Article:Premarital sexuality: a five-year follow-up study of attitudes and behavior by dating stage.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters