Ego Identity Research Paper

Ego Identity Status, Identity Style, and Personal Expressiveness: An Empirical Investigation of
Three Convergent Constructs
Seth J. Schwartz
Florida International University
Ronald L. Mullis
Florida State University
Alan S. Waterman
The College of New Jersey
Richard M. Dunham
Florida State University
This study represents an investigation of relationships among three sets of identity con- structs, including the ego identity statuses, the identity styles, and personal expressive- ness. The Ego Identity Process Questionnaire was used to assess identity status as con- ceptualized by J. E. Marcia. The Identity Style Inventory was used to assess identity style as conceptualized by M. D. Berzonsky. The Personally Expressive Activities Question-
naire was used to measure feelings of personal expressiveness as described by A. S. Waterman. Data were collected from two samples of students at two universities. The measures were considered in pairs to examine the convergence among the constructs. Results revealed that these three measures, and perhaps the underlying constructs, are convergent. Associations between measures are discussed in relation to previous research and theory.
Erikson (1968) posited ego identity as the personality component under- going a time of special ascendancy during the adolescent years. He described ego identity as serving a variety of functions including sameness over time, inner coherence, the synthesis of successive identifications, and protection against experiences of sudden discontinuities occasioned by biological development or changes in life circumstances. Erikson placed identity as one
Correspondence and reprint requests should be addressed to Seth J. Schwartz, Department of Psychology, DM 256, Florida International University, University Park Campus, Miami, FL 33199; e-mail: [e-mail].
Journal of Adolescent Research, Vol. 15 No. 4, July 2000 504-521 504

Schwartz et al. / THREE IDENTITY CONSTRUCTS 505 pole on a dimension pertaining to self-knowledge, which extended to identity confusion as the opposite pole.
Whereas Erikson wrote on identity in primary theoretical and clinical terms, Marcia’s (1966) ego identity status paradigm was the first approach to the operationalization of identity to gain acceptance for purposes of the sys- tematic study of identity under circumstances of normative socialization. Marcia (1993) estimated that in the first 25 years of identity research, more than 300 studies had been conducted. There is an extensive body of research literature demonstrating the discriminant validity of the identity statuses (Marcia, Waterman, Matteson, Archer, & Orlofsky, 1993; Meuss, 1996).
Within the past decade, however, there have been numerous efforts to broaden or reformulate the conceptualization of identity formation. Most of these efforts have built on the identity status paradigm, although they have differed in the extent to which the reconceptualizations are compatible with the original paradigm. These approaches include the work of Berzonsky (1989), Côté (1996), Kurtines (1999), Meuss (1996), and Waterman (1992). The present study was undertaken to evaluate the extent of association between measures derived from Marcia’s (1966, 1993) ego identity status paradigm, Berzonsky’s (1989, 1993) identity style paradigm, and Waterman’s (1990, 1992) conceptualization of personally expressive identity functioning.
Marcia: The Ego Identity Status Paradigm
Marcia (1966), working from Erikson’s ego-analytic writings, identified two dimensions in the process of identity formation: exploration and com- mitment. Exploration involves the active consideration of alternative possi- ble identity elements in a quest for a more complete sense of self, whereas commitment represents a decision to adhere to a specific set of goals, values, and beliefs, whether self-initiated or adapted from others.
Based on these two dimensions, Marcia (1966) derived four identity sta- tuses, each of which represents a juxtaposition of levels of exploration and commitment. Someone who commits to a given career, ideology, or interper- sonal style following a period of relatively intense exploration is categorized as Identity Achieved. A person who is still negotiating the exploration pro- cess, and who therefore is uncommitted, is said to be in Moratorium. An indi- vidual who has committed to a set of particular identity elements without having explored alternatives is in the Foreclosure status. Often, Foreclosure commitments are based on identification with parental or other authority fig- ures or are based on the preferences of such individuals. Finally, a person who is without identity commitments and who is not engaged in exploration is said to be Identity Diffused.
506 JOURNAL OF ADOLESCENT RESEARCH / July 2000
Resolutions of the task of identity formation involving the combination of exploration and commitment, that is, Identity Achievement outcomes, are generally viewed as preferable. Exploration is seen as increasing the likeli- hood that the element eventually selected will be one that has a reasonable chance of satisfying the needs of the individual and of receiving support from the society within which the person functions (Grotevant, 1987). Further- more, commitment to selected identity elements increases the likelihood of successfully working through difficulties that might arise in the implementa- tion of specific identity choices (Bosma, 1992).
Berzonsky: The Identity Style Paradigm
Berzonsky (1989, 1990) has developed a process-oriented paradigm of identity formation based on a constructivist theoretical perspective. “Indi- viduals construct both a sense of who they think they are and the ‘reality’ within which they act” (Berzonsky, 1990, p. 156). Berzonsky describes three distinct cognitive orientations, called identity styles, by which self-relevant information is evaluated, structured, revised, and utilized. An Informational style entails a willingness to investigate multiple solutions to a given problem and to explore several options before committing to any one. The Normative style represents growth by conforming to social and familial expectations and a high degree of commitment to authority and to the exercise of judg- ment. The Diffuse/Avoidant style is marked by the tendency to procrastinate and to make decisions on a situation-by-situation basis (Berzonsky, 1993). From the constructivist perspective, the Informational style may be seen as the preferable course for identity formation because it will likely entail a greater openness to experience, action on the basis of deliberate reasoning, pursuit of decision-relevant information, and the exercise of greater ego con- trol (Berzonsky, 1990).
Waterman: Personally Expressive Identity Functioning
Waterman (1990, 1992, 1993a) has developed a eudaimonist perspective on identity functioning. With its origins in classic Hellenic philosophy, most notably in The Nichomachean Ethics (Aristotle, 1985), eudaimonism is a the- ory of ethics, calling on people to recognize and live in accordance with their daimon, or true self, by fulfilling their greatest potentials. Within psychol- ogy, the eudaimonist perspective can be found in the work of Horney (1950), Maslow (1968), and May (1969). With respect to the task of identity forma- tion, “a person’s sense of identity is an approximation of the daimon, a set of goals, values, and beliefs that to a greater or lesser extent correspond to the
Schwartz et al. / THREE IDENTITY CONSTRUCTS 507 actual potentials of the individual” (Waterman, 1992, p. 58). Within this per- spective, feelings of personal expressiveness associated with an activity or identity alternative are interpreted as indicative of a meshing of the identity element and the individual’s existing potentialities.
The discovery or recognition of personally expressive identity choices is viewed as involving the use of both intuitive and rational levels of decision making (Epstein, 1991). Individuals differ not only in the extent to which such personally expressive identity choices have been identified but also in whether such choices are seen as the objectives of the identity formation pro- cess. Some adolescents, in the process of making identity choices, are look- ing to establish instrumental commitments, that is, commitments that will pay off in terms of finding a way of functioning within the community that is deemed successful and that meets with general acceptance of others. The dis- tinction between personally expressive and instrumental identity choices corresponds with that between intrinsically and extrinsically motivated activities (Waterman, 1999), with the former generally recognized as a pref- erable manner of functioning (Csikszentmihalyi, 1990; Deci & Ryan, 1985).
Theoretical and Empirical Linkages Among the Three Conceptualizations of Identity Formation
Ego identity status and identity style. Berzonsky (1989) has linked the identity statuses and identity styles as follows: (a) the Identity Achievement and Moratorium statuses are associated with the Informational style in that each involves a process of active exploration in the process of making iden- tity-related decisions; (b) the Foreclosed status and the Normative style both involve the forming of identity commitments without the consideration of alternative possibilities; and (c) the Identity Diffuse status and the Dif- fuse/Avoidant identity style both entail the absence of identity-related com- mitments and the lack of concern with their development.
In previous studies, most of the hypothesized linkages have been docu- mented. Berzonsky (1990), using the Extended Objective Measures of Ego Identity Status (EOM-EIS) (Bennion & Adams, 1986) and the Identity Style Inventory (ISI) (Berzonsky, 1989), found significant correlations between measures of (a) Identity Achievement and Informational style, (b) Foreclo- sure and Normative style, and (c) Identity Diffusion and Diffuse/Avoidant style. The correlation of measures of the Moratorium status and the Informa- tional style was nonsignificant unless the effect of commitment was statisti- cally controlled (Berzonsky & Neimeyer, 1994). In a replication study, Streitmatter (1993) obtained comparable results.
508 JOURNAL OF ADOLESCENT RESEARCH / July 2000
When Berzonsky and Neimeyer (1994) statistically controlled the effect of commitment, the Informational scores of Moratoriums were similar to those of Identity Achieved participants and were significantly higher than those individuals classified as Diffused or Foreclosed. Likewise, with com- mitment held constant, Normative scores of Identity Achievers were found to be significantly lower than those of Foreclosed individuals but equivalent to those with a Moratorium or Diffuse identity status.
Ego identity status and personal expressiveness. Waterman (1992) pro- posed personal expressiveness versus instrumentality as a third defining dimension of identity, supplementing the dimensions of exploration and commitment developed by Marcia (1966). On a conceptual level, the expres- siveness dimension is considered to be largely independent of the other two. Both Identity Achievement and Foreclosure commitments may be either expressive or instrumental, and a person in the Moratorium status may be seeking either an expressive or an instrumental resolution to the identity search. Only with respect to the Identity Diffusion status was the assumption made that any identity activity present would be instrumental rather than expressive.
The fact that both Identity Achievement and Foreclosure commitments may be either expressive or instrumental does not carry an implication that the probabilities are equivalent for the two identity statuses. To the contrary, a stronger association of expressiveness with Identity Achievement can be anticipated because the more options that are actively explored, the greater the likelihood will be that one or more of the alternatives considered will be consistent with personal potentials and hence be experienced as expressive (Waterman, Lopez, Gruenfeld, & Jessee, 2000). Conversely, the greater the role played by identification in the process of forming Foreclosure commit- ments, the greater the likelihood that a commitment will be based on instru- mental considerations relating to the fulfillment of the expectations of signif- icant others. With respect to the Moratorium status, the absence of commitments can be taken as an indication that personally expressive iden- tity alternatives have not yet been identified, even when this is the objective of the identity search. Hence, no relationship between measures of the Mora- torium status and personal expressiveness would be anticipated.
Relationships between measures of identity status and experiences of per- sonal expressiveness have been investigated in several unpublished studies (Gordon, 1989; Waterman et al., 2000; Waterman & Mannion, 1994). In these studies, significant positive correlations have been found between the Iden- tity Achievement status scale of the EOM-EIS (Bennion & Adams, 1986) and the personal expressiveness scale of the Personally Expressive Activities
Schwartz et al. / THREE IDENTITY CONSTRUCTS 509
Questionnaire (PEAQ) (Waterman, 1993b) but not with the Identity Achievement scale of the Dellas Identity Status Inventory (DISI) (Dellas & Jernigan, 1987). As expected, the Identity Diffusion scales of the EOM-EIS and DISI have been negatively correlated with personal expressiveness. In previous studies, correlations between personal expressiveness and mea- sures of the Moratorium and Foreclosed statuses have been consistently nonsignificant.
Identity style and personal expressiveness. The identity style paradigm and the eudaimonist construct of personal expressiveness are based on con- trasting philosophical assumptions with regard to the utility of a concept of a true self. Using a constructivist orientation, Berzonsky (1990) is explicit in rejecting the concept of a true self, whereas the existence of such potentiali- ties has been central in Waterman’s theoretical writings (1990, 1993a, 1999). It might be concluded, therefore, that there should be no empirical associa- tion between measures derived from these two perspectives. However, it can be seen from the sections above that both the Informational style and the per- sonal expressiveness are linked with Identity Achievement and that both the Diffuse/Avoidant style and the absence of personal expressiveness are linked with the Identity Diffusion status. It is logically plausible that two sets of con- structs that are systematically related to a third conceptual scheme may be systematically related to each other.
This study was undertaken to explore the linkage among these three sets of identity constructs. With respect to the association of identity status and iden- tity style constructs and the association of identity status and personal expres- siveness constructs, this research represents the replication of previous stud- ies. This research represents the first evaluation of the linkage of identity style and personal expressiveness constructs.
METHOD
Participants
The research was conducted with two samples drawn from two southeast- ern universities differing in ethnic composition. Drawing the sample from two diverse student populations was done to assess the replicability and generalizability of the aforementioned theoretical linkages.
Sample 1. One hundred thirteen university undergraduates (13 males, 100 females) from an introductory psychology course at the Florida State
510 JOURNAL OF ADOLESCENT RESEARCH / July 2000
University participated in this study. The ethnic composition of the student body at Florida State University is approximately 75% Caucasian, 10% Afri- can American, 10% Hispanic, and 5% Other. Although ethnicity data were not gathered from this sample, the composition of the sample approximated that of the university. (Responses from two additional students were not included in the analyses because of missing data.)
Sample 2. Participants in Sample 2 were 196 university undergraduates (54 males, 155 females, 19 not reporting gender) from a variety of psychol- ogy courses at Florida International University. Participants were given extra credit in their courses for completing the questionnaire packets. This sample consisted of 33 Caucasians, 9 African Americans, 131 Hispanics, 3 Others, and 52 participants who did not report their ethnicity. The ethnic composition of the student body at Florida International University is approximately 20% Caucasian, 10% African American, 60% Hispanic, and 10% Other. Once again, the composition of the sample approximated that of the university.
Measures
Identity status. The Ego Identity Process Questionnaire (EIPQ) (Balistreri, Busch-Rossnagel, & Geisinger, 1995) is a 32-item scale assess- ing ego identity in four ideological domains (occupational choice, political preference, religious beliefs, and personal values) and in four interpersonal domains (friendships, dating, sex roles, and family). Sixteen of the EIPQ items assess exploration and 16 assess commitment. There are two explora- tion and two commitment items per domain. Each item is rated on a 6-point Likert-type scale ranging from strongly agree to strongly disagree. Informa- tion on the psychometric properties of the EIPQ exploration and commitment measures is provided by Balistreri et al. (1995).
Status assignments were performed according to the median split tech- nique used by Balistreri et al. (1995). For both exploration and commitment, scores falling on or above the median were classified as high, the remainder, low. Identity status categories were assigned on the basis of these median splits. A person who was high on both the exploration and commitment scales was considered to be in the Identity Achievement status. Someone who was high on exploration and low on commitment was judged to be in the Moratorium status. A participant with the reverse pattern was considered to be Foreclosed. Someone low on both scales was placed in the Identity Diffu- sion category.
Schwartz et al. / THREE IDENTITY CONSTRUCTS 511
Identity style. The Identity Style Inventory (ISI) (Berzonsky, 1997) is a 40- item measure with 11 items assessing the Informational style, 9 items for the Normative style, and 10 items tapping the Diffuse/Avoidant style. Items are responded to on 5-point Likert-type scales. Given the unequal number of items in each scale, comparable measures were generated by dividing the scale sum for each style by the number of items in the scale. (A commitment scale, which was not analyzed for this study, is also included in the measure.) Berzonsky has provided a description of the psychometric properties of the ISI.
In this study, identity style was used as a continuous measure in its rela- tionship to identity status but as a categorical variable in its relationship to personal expressiveness. To obtain continuous measures of identity style, it is necessary to sum the responses to the items on each of the three scales and divide each total score by the number of items on the respective scale. Cate- gorical measures of style are obtained by standardizing the three style scores and using the style with the highest standard score (see Berzonsky, 1989).
Personal expressiveness. The Personally Expressive Activities Question- naire (PEAQ) (Waterman, 1993b) requires participants to list five activities of importance to themselves that they would use to describe themselves to another person. Each activity is then rated with respect to a variety of scales including the extent to which it is experienced as personally expressive. The scale for personal expressiveness consists of six statements, each rated on a 7-point scale with endpoints labeled strongly agree and strongly disagree. Scores on these six items, summed across the five activities, were used as a continuous measure of the extent to which individuals engaged in personally expressive activities. Waterman has provided a description of the psychometric properties of the PEAQ.
Procedure
Sample 1. Each participant completed the questionnaires during a 1-hour testing session. After signing a consent form, the participants completed the instruments in the following order: the PEAQ, EIPQ, and ISI. Three addi- tional measures followed to satisfy the requirements of a larger study. The usual completion time for all measures was 45 minutes.
512 JOURNAL OF ADOLESCENT RESEARCH / July 2000
Sample 2. All participants completed the questionnaire packets at home and returned them the following week. The order of measures was again the PEAQ, EIPQ, and ISI. Four other measures, which were not analyzed for this study, were included after the ISI. The average completion time for all mea- sures, as reported by participants, was 90 minutes.
RESUL TS
Data Analytic Strategy
Because the hypotheses put forth in this study were pairwise comparisons among identity status, identity style, and personal expressiveness, one-way analyses of variance were the most direct tests of these hypotheses. However, there was a need to determine the effects of two demographic variables, gen- der and sample, on the results. Main effects of gender or sample are presented separately in a demographic variables section. The presence of gender or sample differences in themselves do not bear directly on the hypotheses. However, any interactions of gender or sample with identity status or style would contraindicate the use of one-way ANOVAs to test the hypotheses.
The number of males in Sample 1 was insufficient to generate minimally required sample sizes in all cells of a three-way ANOVA. For this reason, separate series of two-way ANOV As were conducted for gender and sample with identity status or identity style to assess possible two-way interactions. Because no significant two-way interactions for gender or sample were observed, the results of one-way ANOVAs will be reported as tests of the specified hypotheses.
Gender Differences
There was a notable discrepancy in the ratio of female to male participants in the two samples. Comparisons of the mean scores of males and females, across samples, on each of the measures employed in this study did not reveal any statistically significant differences. Furthermore, when gender was included as an independent variable in the analyses of variance used to test the hypotheses on which the study was based, it did not yield any significant main effects or interactions.
Sample Differences
Schwartz et al. / THREE IDENTITY CONSTRUCTS 513
Significant main effects of sample emerged in all two-way ANOV As that were conducted. With respect to the continuous identity style measures, there was a significant main effect of sample on the Informational, F(1, 313) = 15.03, p < .001; Normative, F(1, 313) = 16.45, p < .001; and Diffuse/ Avoidant, F(1, 316) = 18.21, p < .001, styles. The mean Informational style score was greater in Sample 2 (M = 3.57, SD = 1.69) than in Sample 1 (M = 3.40, SD = 1.04). The mean Normative style score was greater in Sample 1 (M = 3.38, SD = 1.66) than in Sample 2 (M = 3.21, SD = 1.95). The mean Diffuse/ Avoidant style score was greater in Sample 1 (M = 2.73, SD = 1.88) than in Sample 2 (M = 2.43, SD = 1.97). There was also a significant main effect of sample on personal expressiveness scores, F(1, 313) = 4.83, p < .03. The mean personal expressiveness score was greater in Sample 1 (M = 146.20, SD = 24.23) than in Sample 2 (M = 139.57, SD = 27.59).
Identity status categorizations also differed by sample, χ2(3, N = 327) = 9.29, p < .03. Specifically, there was a greater percentage of participants in the Foreclosed status (41% to 26%) and a smaller percentage of participants in the Moratorium status (16% to 27%) in Sample 2 than in Sample 1. The percentages of participants classified into the Identity Diffused and Identity Achieved statuses did not differ appreciably by sample (i.e., the differences were both less than 5%). Identity style categorizations did not differ by sample.
There were no interactions between sample and either identity status or style. Separate one-way ANOV As were run for Sample 1 and Sample 2 because there were main effects of sample on all of the dependent variables used in these analyses.
Frequency Distributions
Because identity status and style are used as categorical variables in anal- yses reported here, frequency distributions are provided for both. The iden- tity status distribution for Sample 1 was 32 Diffused (28.3%), 29 Foreclosed (25.7%), 30 Moratorium (26.5%), and 22 Achieved (19.5%). For Sample 2, the identity status distribution was: 59 Diffused (27.8%), 87 Foreclosed (41.0%), 34 Moratorium (16.0%), and 32 Achieved (15.1%).
The identity style distribution for Sample 1 was: 42 Informational (37.2%), 40 Normative (35.4%), and 31 Diffuse/Avoidant (27.4%). For Sample 2, the identity style distribution was: 80 Informational (38.8%), 59 Normative (28.4%), and 67 Diffuse/Avoidant (31.6%).
514 JOURNAL OF ADOLESCENT RESEARCH / July 2000
TABLE 1: Pairwise Comparisons for Identity Style Variables by Ego Identity Status
Style
Informational Sample 1
Sample 2
Normative Sample 1
Sample 2
Diffuse/Avoidant Sample 1
Sample 2
Ego Identity Status Group
Diff For Mor Ach abbb 3.09 3.39 3.60 3.61
F Ratio
8.62** 19.67**
8.34** 12.50**
3.09* 8.95**
(1.61) (1.15) (1.74) (1.37) abcc
3.25 3.55 3.75 3.97 (1.42) (1.62) (1.40) (1.38)
2.99 (1.72)
3.26 (1.25)
2.74 3.26 a (1.56) b (1.70)
(1.37) (1.30) c ac
2.78
3.15
2.54 2.77 (1.76) (1.94) abcb abbb 3.32 2.90 2.92 2.93
(1.86) (1.97) (1.92) (1.98) abbb
3.08 2.58 2.58 2.50 (2.01) (1.67) (1.98) (2.53)
NOTE: Within each row, means with the same superscript are not significantly different from one another. Statuses: Diff = Diffusion; For = Foreclosure; Mor = Moratorium; Ach = Achievement.
*p < .05. **p < .001.
Interrelationships of Identity Status and Identity Style
Sample 1. For Sample 1, participants in the four identity statuses were found to differ significantly on each of the three identity style measures: (a) Infor- mational style, F(3, 109) = 8.69, p < .001; (b) Normative style, F(3, 109) = 8.34, p < .001; and (c) Diffuse/Avoidant style, F(3, 109) = 3.09, p < .03.
Sample 2. For Sample 2, participants in the four identity statuses were again found to differ significantly on each of the three identity style mea- sures: (a) Informational style, F(3, 204) = 19.67, p < .001; (b) Normative style, F(3, 204) = 12.50, p < .001; and (c) Diffuse/Avoidant style, F(3, 207) = 8.95, p < .001 (see Table 1).
Post-hoc comparisons. Tukey’s Least Significant Difference (LSD) pairwise comparisons were conducted on both data sets (see Table 1).
TABLE 2:
Sample Sample 1 Sample 2
Personal Expressiveness by Identity Status Category
Schwartz et al. / THREE IDENTITY CONSTRUCTS 515
Diff
137.22
For
147.07
Mor
150.73
Ach
151.95
F Ratio 2.32** 4.84*
(20.05)
132.20 143.39 135.39 153.31
(23.50)
(25.50) (30.02) (29.27) (20.14)
(28.75)
a bc ab c
(20.62)
NOTE: Within each row, means with the same superscript do not differ significantly from one another. Statuses: Diff = Diffusion; For = Foreclosure; Mor = Moratorium; Ach = Achievement.
*p < .005. **p < .08.
The results were consistent with theoretical expectations. With regard to the Informational style, Identity Achievers scored highest and Identity Diffu- sions scored lowest. For both samples, the Identity Diffusion status was sig- nificantly different from each of the other statuses. For Sample 1, the other statuses did not differ among themselves, whereas for Sample 2, the Identity Achievement and Moratorium statuses scored significantly higher than did the Foreclosed status. For the Normative style scores, the participants in the Foreclosed status scored highest for both samples, with participants in the Moratorium status scoring lowest. Foreclosures were significantly different from all other statuses except with respect to the Identity Achieved status in Sample 1. For the Diffuse/Avoidant style, in both samples, participants in the Diffused status scored highest and were significantly different from partici- pants in all of the other statuses.
Interrelationships of Identity Status and Personal Expressiveness
Sample 1. For Sample 1, the relationship between identity status and per- sonal expressiveness approached significance, F(3, 109) = 2.32, p < .08.
Sample 2. For Sample 2, participants in the four identity statuses were found to differ significantly on the measure of personal expressiveness, F(3, 198) = 4.84, p < .005 (see Table 2).
Post-hoc comparisons. The pattern of means for Sample 2 revealed that participants in the Identity Achieved status scored highest on personal expressiveness, whereas those in the Identity Diffused status scored lowest. Tukey’s LSD pairwise comparisons for Sample 2 indicated that participants in the Identity Diffused and Moratorium statuses were significantly lower on this measure than were those in the Foreclosed and Identity Achieved
516 JOURNAL OF ADOLESCENT RESEARCH / July 2000
TABLE 3:
Sample Sample 1 Sample 2
Personal Expressiveness by Identity Style Category
Diff/Av Norm Info a ab b 139.40 147.33 153.97
F Ratio 3.43* 5.08**
(26.25) (24.52) (18.45) abb
132.18 142.64 145.69 (29.81) (25.91) (24.45)
NOTE: Within each row, means with the same superscript do not differ significantly from one another. Styles: Diff/Av = Diffuse/Avoidant; Norm = Normative; Info = Informational. *p < .05. **p < .01. statuses, with the exception that the Moratorium and Foreclosed statuses did not differ significantly.
Interrelationships of Identity Style and Personal Expressiveness
Sample 1. For Sample 1, participants in the three identity style groups were found to differ significantly on the measure of personal expressiveness, F(2, 110) = 3.43, p < .04.
Sample 2. For Sample 2, participants in the three identity style groups were again found to differ significantly on the measure of personal expres- siveness, F(2, 203) = 5.08, p < .01 (see Table 3).
Post-hoc comparisons. As expected, the patterns of means for the two samples indicated that participants in the Informational style category were highest on personal expressiveness, whereas those in the Diffuse/Avoidant style category were lowest. Tukey’s LSD pairwise comparison tests revealed that, in both Sample 1 and Sample 2, scores for the Informational style cate- gory were significantly higher than were those for the Diffuse/Avoidant style category. For Sample 2 only, the mean personal expressiveness score for the Normative style category was significantly higher than that for the Dif- fuse/Avoidant style category. The Informational and Normative groups did not differ significantly from one another.
DISCUSSION
The findings from this study provide support for the hypothesized links among ego identity status, identity style, and personal expressiveness. Simi-
Schwartz et al. / THREE IDENTITY CONSTRUCTS 517 lar results were obtained with two independent samples drawn from different universities having different ethnic distributions.
The Identity Achieved status was found to be associated with high use of the Informational style, low use of the Diffuse/Avoidant style, and with scores indicating the highest amount of involvement in personally expressive activities (although not significantly higher than those of Foreclosed individ- uals). Participants in the Moratorium status were the lowest in use of the Nor- mative style, generally low in use of the Diffuse/Avoidant style, and fairly low in involvement in personally expressive activities. Foreclosures demon- strated the strongest linkage to use of the Normative style, were low in use of the Diffuse/Avoidant style, and were intermediate with respect to personally expressive activities. Participants in the Identity Diffused status showed the reverse pattern to that of Identity Achievers. They had the highest scores for the Diffuse/Avoidant style, the lowest scores for the Informational style, and the lowest reported involvement in personally expressive activities. The link- ages found for the identity style and personal expressiveness variables indi- cated the greatest involvement in personally expressive activities on the part of participants using a predominantly Informational style, and the least involvement by those using a predominantly Diffuse/Avoidant style.
The main effects of sample also pose an interesting set of questions. Par- ticipants in Sample 2 reported greater use of the Informational style than did those in Sample 1, whereas participants in Sample 1 reported greater use of the Normative and Diffuse/Avoidant styles, as well as greater levels of per- sonal expressiveness, than did those in Sample 2. Proportionally, there were more participants assigned to the Foreclosed status, and fewer assigned to the Moratorium status, in Sample 2 than in Sample 1.
There are a number of important differences between the samples, as well as between the student bodies of the universities from which the samples were drawn. The most notable difference is ethnicity, but this is not the only difference. Sample 1 was drawn from a university in a “college town” where the majority of students live on campus or in off-campus apartments, whereas Sample 2 was drawn from a university in an urban setting where the majority of students live at home with their parents.
The results obtained here with respect to the linkages between identity sta- tus and identity style reaffirm findings previously generated by Berzonsky (1989), Berzonsky and Neimeyer (1994), and Streitmatter (1993). Individ- uals in the various identity statuses demonstrated different approaches to pro- cessing information pertinent to the task of developing personally relevant goals, values, and beliefs.
The differences between the participants in various identity statuses with respect to the extent of their involvement in personally expressive activities
518 JOURNAL OF ADOLESCENT RESEARCH / July 2000 support the linkage previously found in unpublished studies by Gordon (1989), Waterman et al. (2000), and Waterman and Mannion (1994). As in these earlier studies, Identity Diffusions were lowest in personal expressive- ness. Findings related to an association between the Identity Achieved status and personal expressiveness have been inconsistent in the prior studies, so the findings here constitute increased support for this hypothesized linkage. Supportive evidence for the relationship between personal expressiveness and Identity Achievement has now been obtained with both the EOM-EIS and the EIPQ, but not with the DISI.
The present study has been the first to investigate the potential linkage between identity style and personal expressiveness. Although these two con- structs were drawn from theoretical perspectives with contrasting assump- tions about the nature of the self, the results with both samples demonstrated that individuals differing in their processing of identity-relevant information also differ in the degree to which they experience self-relevant activities as personally expressive.
Waterman (1984) put forth two metaphors for the nature of identity for- mation, namely discovery and creation. The discovery metaphor, finding something that was present but previously not known, corresponds to a eudaimonist approach to establishing an identity (Waterman, 1990). The cre- ation metaphor, bringing into being something that did not previously exist, corresponds to the constructivist approach to establishing an identity (Berzonsky, 1990). Waterman (1984) speculated that adolescents in the pro- cess of forming their sense of identity might use the language of one or the other of these metaphors and that such a difference might relate to the even- tual identity outcomes. The results of the present study suggest that those establishing identity commitments by way of active exploration, that is, Iden- tity Achievers, may be making use of both types of metaphorical processes.
A number of significant sample differences emerged in the findings from this study. Despite the differences between the samples in mean scores on various measures, the interrelationships among the scales were parallel in virtually every instance.
Although the focus of the current study was on the substantive relation- ships among three views of identity, not on these demographic differences, the differences between the samples in mean scores nevertheless are poten- tially of considerable interest and warrant further study.
In summary, the findings reported here demonstrate the empirical conver- gence of the three theoretical schemes from which the constructs of ego iden- tity status, identity style, and personal expressiveness were drawn. Whereas the Erikson/Marcia theory of identity, defined in terms of exploration and commitment, is conceptually compatible with both Berzonsky’s
Schwartz et al. / THREE IDENTITY CONSTRUCTS 519 constructivist theory and Waterman’s eudaimonist identity perspective, the latter two theories are mutually incompatible, at least with respect to the con- cept of a true self. The clear associations among measures of the three iden- tity constructs pose a challenge to develop an integrative theoretical perspec- tive that incorporates these results. It is beyond the scope of this report, however, to put forth theoretical speculations toward that objective.
REFERENCES
Aristotle. (1985). The Nichomachean Ethics (T. Irwin, Trans.). Indianapolis, IN: Hackett. Balistreri, E., Busch-Rossnagel, N. A., & Geisinger, K. F. (1995). Development and preliminary validation of the Ego Identity Process Questionnaire. Journal of Adolescence, 18, 179-190. Bennion, L. D., & Adams, G. R. (1986). A revision of the extended version of the Objective Measure of Ego Identity Status: An identity instrument for use with late adolescents. Journal of Adolescent Research, 1, 183-198.
Berzonsky, M. D. (1989). Identity style: Conceptualization and measurement. Journal of Ado- lescent Research, 4, 267-281.
Berzonsky, M. D. (1990). Self-construction over the lifespan: A process perspective on identity formation. In G. J. Neimeyer & R. A. Neimeyer (Eds.), Advances in personal construct the- ory (Vol. 1, pp. 155-186). Greenwich, CT: JAI.
Berzonsky, M. D. (1993). A constructivist view of identity formation: People as postpositivist
self-theorists.
In J. Kroger (Ed.), Discussions on ego identity (pp. 169-203). Hillsdale, NJ:
Lawrence Erlbaum.
Berzonsky, M. D. (1997). Identity Style Inventory, Version 3. Unpublished measure, State Uni- versity of New York, Cortland.
Berzonsky, M. D., & Neimeyer, G. J. (1994). Ego identity status and identity processing orienta- tion: The mediating role of commitment. Journal of Research in Personality, 28, 425-435. Bosma, H. A. (1992). Identity in adolescence: Managing commitments. In G. R. Adams, T. P. Gullotta, & R. Montemayor (Eds.), Adolescent identity formation: Advances in adolescent development (pp. 91-121). Newbury Park, CA: Sage.
Côté, J. E. (1996). Sociological perspectives on identity formation: The culture-identity link and identity capital. Journal of Adolescence, 19, 419-430.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Basic
Books.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behav- ior. New York: Plenum.
Dellas, M., & Jernigan, L. P. (1987). Occupational identity status development, gender compari- sons, and internal-external locus of control in first-year Air Force cadets. Journal of
Youth and Adolescence, 16, 587-600.
Epstein, S. (1991). Cognitive-experiential self theory: Implications for developmental psychol- ogy. In M. R. Gunnar & L. A. Sroufe (Eds.), Self processes and development: The Minnesota symposia on child psychology (Vol. 23, pp. 79-123). Hillsdale, NJ: Lawrence Erlbaum. Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.
Gordon, L. L. (1989). Identity development and personal expressiveness. Unpublished master’s thesis, University of Arizona, Tucson.
520 JOURNAL OF ADOLESCENT RESEARCH / July 2000
Grotevant, H. D. (1987). Toward a process model of identity formation. Journal of Adolescent Research, 2, 203-222.
Horney, K. (1950). New ways in psychoanalysis. New York: Norton.
Kurtines, W. M. (1999). A co-constructivist perspective on human behavior and development.
Unpublished manuscript, Florida International University, Miami.
Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personality and Social Psychology, 5, 551-558.
Marcia, J. E. (1993). The ego identity status approach to ego identity. In J. E. Marcia, A. S.
Waterman, D. R. Matteson, S. L. Archer, & J. L. Orlofsky (Eds.), Ego identity: A handbook for psychosocial research (pp. 1-21). New York: Springer-Verlag.
Marcia, J. E., Waterman, A. S., Matteson, D. R., Archer, S. L., & Orlofsky, J. L. (1993). Ego identity: A handbook for psychosocial research. New York: Springer-Verlag.
Maslow, A. H. (1968). The farther reaches of human nature. New York: Penguin.
May, R. (1969). Love and will. New York: Norton.
Meuss, W. (1996). Studies on identity development in adolescence: An overview of research and some new data. Journal of Youth and Adolescence, 25, 569-598.
Streitmatter, J. L. (1993). Identity status and identity style: A replication study. Journal of Ado- lescence, 16, 211-215.
Waterman, A. S. (1984). Identity formation: Discovery or creation? Journal of Early Adoles- cence, 4, 329-341.
Waterman, A. S. (1990). Personal expressiveness: Philosophical and psychological founda- tions. Journal of Mind and Behavior, 11, 47-74.
Waterman, A. S. (1992). Identity as an aspect of optimal psychological functioning. In G. R.
Adams, T. P. Gullotta, & R. Montemayor (Eds.), Adolescent identity formation: Advances in adolescent development (pp. 50-72). Newbury Park, CA: Sage.
Waterman, A. S. (1993a). Finding something to do or someone to be: A eudaimonist perspective on identity formation. In J. Kroger (Ed.), Discussions on ego identity (pp. 147-168).
Hillsdale, NJ: Lawrence Erlbaum.
Waterman, A. S. (1993b). Psychometric properties of the Personally Expressive Activities
Questionnaire. Unpublished manuscript, The College of New Jersey, Ewing.
Waterman, A. S. (1999). The problem of selectivity in intrinsic motivation and identity forma-
tion.
Manuscript submitted for publication.
Waterman, A. S., Lopez, L. M., Gruenfeld, K. E., & Jessee, C. L. (2000). The relationship of per- sonal expressiveness to intrinsic motivation, purpose in life, identity achievement, and prin- cipled morality. Manuscript in preparation, The College of New Jersey, Ewing. Waterman, A. S., & Mannion, K. (1994, June). On the relationship of optimal experience to identity formation, psychosocial identity, and self-actualization. Paper presented at the bien- nial meeting of the International Society for the Study of Behavioral Development, Amster- dam, the Netherlands.
Seth J. Schwartz was born in New York City and received his bachelor’ s degree in psy- chology from Florida State University. He has a master’ s degree in family and child sci- ences from Florida State University and is currently a doctorate in life span develop- mental psychology from Florida International University. His interests lie in the area of identity development, particularly in designing interventions to facilitate identity in ado- lescents and young adults. His dissertation was a comparison of identity interventions drawn from self-construction and self-discovery approaches and, along with
several
Schwartz et al. / THREE IDENTITY CONSTRUCTS 521 senior colleagues, he has submitted a grant proposal to implement the interventions with a marginalized high school population.
Ronald L. Mullis has been a professor of family and child sciences at Florida State Uni- versity since 1989. His areas of teaching and research include adolescent identity for- mation, adolescent at risk, and parent-child relations. He earned his Ph.D. in child de- velopment from Iowa State University and has served on the faculty at North Dakota State University and California State University–Chico. He is codirector of the FSU Family Institute Policy and Research Program and is a charter member of the Florida Inter- University Consortium for Child, Family, and Community Studies.
Alan S. Waterman is a professor in the Department of Psychology at The College of New Jersey. He received his Ph.D. from the State University of New York at Buffalo. His prin- cipal research interests are focused on positive psychological functioning including identity development, intrinsic motivation, and self-realization and on service-learning educational techniques. He is a fellow in the American Psychological Association and American Psychological Society and currently serves as the North American editor of the Journal of Adolescence.
Richard M. Dunham is retired from the Florida State University where he was professor of psychology and directed the Identity Development Research Laboratory. Previously, he served on the faculty at the University of Houston, with NASA in the Astronaut Training Office of the Manned Spacecraft Center, and as chief psychologist at Dorothea Dix Hospital in Raleigh. His doctorate is from Duke University. His research has fo- cused on the familial and societal supports of the development of mental illnesses and other psychological qualities.