IDENTITY STATUS FORMULAE: GENERATING CONTINUOUS MEASURES OF THE IDENTITY STATUSES FROM MEASURES OF EXPLORATION AND COMMITMENT.

Schwartz, Seth J. ; Dunham, Richard M.

ABSTRACT

Mathematical formulae were devised for the purpose of generating continuous measures of the four identity statuses from measures of exploration and commitment. The formulae were consistent with the conceptual definitions of the statuses. They were found to be effective both in terms of generating continuous measures of the statuses and in terms of deriving status assignments.

During the last thirty years, identity formation has been a worthy focus of considerable theoretical and empirical research. In 1993, Marcia estimated that more than 300 studies had been published on identity. Most of this research drew on the formulation that Marcia (1966) derived from the work of Erikson (1950). This formulation is quite parsimonious and has been shown to possess adequate construct validity (Waterman, 1988).

Marcia extracted two principal dimensions from Erikson's work, exploration and commitment. Exploration signifies the search for a more complete sense of self. Commitment is the act of deciding on a particular set of goals, values, and beliefs to which one will adhere.

Traditionally, exploration and commitment have each been bifurcated into high and low levels, with the median score on each serving as the dividing point. Marcia derived four identity statuses by juxtaposing each level of exploration with each level of commitment. These statuses are achievement, moratorium, foreclosure, and diffusion.

Achievement represents high levels of both exploration and commitment. It is the state of having committed to specific identity elements following a period of exploration. Achieved individuals tend to be balanced thinkers (Marcia, 1980) and to be willing to explore further if the situation requires it (Stephen, Fraser, & Marcia, 1992).

Moratorium is the state of active exploration without much commitment. In developmental terms, it usually precedes the achieved status (Waterman, 1988). Individuals in moratorium tend to be anxious and depressed (Kidwell, Dunham, Bacho, Pastorino, & Portes, 1995), and they have been shown to display more creative thought than do those in the other statuses (Berman, Schwartz, Kurtines, & Berman, 2000).

Foreclosure represents a commitment formed without adequate prior exploration. This commitment often stems from the expectations of parents or other significant figures in the individual's life. Foreclosure may be associated with smug self-satisfaction (Marcia, 1980) and authoritarianism (Marcia, 1967).

Diffusion is the relative absence of exploration and commitment. Diffused individuals are often apathetic and disinterested (Marcia, 1980). They may also be especially prone to academic difficulties (Berzonsky, 1985).

Formulae are provided here for calculating continuous measures of the four basic identity statuses from continuous measures of exploration and commitment. Tests of the effectiveness of the derived identity measures were accomplished by: (a) comparing two relationships, namely that between those derived continuous measures and an independent measure with that between existing continuous measures of identity status and the same independent measure; and (b) comparing categorization data based on the use of the formulae with data using more usual methods of categorization, on a single instrument.

Classification of Instruments that Assess Identity Status

There are two general forms of identity measures, called here direct (designed to measure identity status directly) and derived (identity status is derived from measures of exploration and commitment). In addition, there are two types of identity status scaling, categorical and continuous. Table 1 presents several instruments classified by mode of assessment and type of scaling. There have been no derived continuous identity measures.

Categorical measures of identity status are commonly used. They lend themselves to statistical analyses that are less powerful than those that are possible with continuous measures. The validity of categorical measures derived from continuous measures, such as measures of exploration and commitment, may be diminished due to the inherent uncertainty and ambiguity of setting thresholds and interpreting values near those thresholds. Instruments that assign status directly, through the subjective judgment of a human rater or coder, may also face the reliability problems that characterize subjective judgments of complex conditions.

In addition to allowing statistical analyses with potentially greater power, continuous measures of identity status can be correlated with variables from other constructs. Further, scores may indicate not only identity status, but also how much of each status is manifested in a given individual. Thus, continuous measures may facilitate the detection of low-magnitude differences in identity status (Adams, Bennion, & Huh, 1989; Bennion & Adams, 1986).

The measurement of identity status has most commonly involved the structured interview. Structured interviews provide for the direct categorical assignment of an overall identity status as well as status within specific content areas (such as politics and occupation), which are called domains (e.g., Grotevant & Cooper, 1981; Marcia & Archer, 1993). As incidental steps in their use, structured interviews may yield continuous measures of overall or domain-specific exploration or commitment (Jackson, Dunham, & Kidwell, 1990), such as when coders are instructed to assign each participant an exploration score and a commitment score (both usually between 1 and 4) for each domain, as a method of deriving status categorizations. The continuous measures have, at times, also been used to assign identity status by domain. Thus, derived categorical functioning may be superimposed on an instrument designed for direct categorical measurement.

The Formulae

In light of the distinctions among identity status instruments, the need for derived continuous measures is clear. Such measures would build upon available continuous measures of exploration and commitment, but would respect past theory, methodology, and findings with regard to identity statuses. Therefore, the measures would have to be demonstrably: (a) based on theoretically grounded definitions of the statuses, and (b) plausible in terms of the outcomes of their usage (the outcomes would need to be reasonable, in comparison with outcomes from preexisting approaches, when conceived as continuous measures and also when reduced to categorical status assignments).

What follows is a presentation of four formulae, one for each of the basic identity statuses, together with a brief discussion of the theoretical justification for the formulae. Two sets of calculations, from two samples of identity data, are then presented to assess the plausibility of the formulae in comparison with established approaches. The scores generated by the formulae are compared with those obtained from a direct continuous measure in terms of their relationships to a third identity instrument, the Identity Style Inventory (Berzonsky, 1997), which has been shown to be related to both derived categorical and direct continuous identity measures (Berzonsky, 1989; Schwartz, 1996).

Marcia presented exploration and commitment in a four-cell diagram to account for the four identity statuses. Table 2 shows a similar formulation, but the terms "manifest" and "less manifest" have been used. This reflects the fact that questions concerning how much exploration or commitment is necessary to cross the threshold between identity statuses have not been systematically addressed in the empirical literature. Each derived categorical instrument has its own rules for status assignment. Implicitly, the direct categorical instruments do also, in that they identify exploration and commitment levels by domain, and then invite a coder or interviewer to judge or assign the status (Jackson, Dunham, & Kidwell, 1990). Presumably, exploration and commitment, or any given status, may exist to some degree or intensity within the individual and will be reflected in relevant ideation, perception, and behavior (e.g., cognition, stances or styles, self-ascriptions, information-seeking and decision-making behavior, a nd imitation or conformity). When the strength of exploration or commitment is low, the individual is taken to fall into a less mature status. Low strength may be judged in relative terms. It may fall below some threshold, perhaps a median, in the distribution of exploration or commitment scores for a given sample or population.

Much the same reasoning applies to the assignment of statuses. Although individuals may differ in status from domain to domain, overall assignment proceeds on the assumption that the behavior of an individual is best reflected in a single status. The behaviors characteristic of the other statuses are presumed to be less observable or less manifest. When they are less manifest, it may be because they do not exist in the repertory of the individual, because they cannot compete successfully for expression with others that are stronger, or because of some other suppressor process. Since developmental expectations would almost always lead to the belief that the component behaviors of all four statuses are established in the repertories for the ages in question--and because they are frequently observed to be present in analyses by domain--it may be assumed that less-manifest statuses have been suppressed or inhibited. They have lost out to other, momentarily stronger tendencies, or been influenced by the inhibitor y social prescriptions of authoritative models.

The classes of instruments referred to in Table 1 either directly assess the signs of the presence of a status or assess the signs of the presence of exploration and commitment. Even the direct continuous measures were created, and are used, with the assumption that exploration and commitment are underlying orthogonal processes. As shown in Table 3, the information in Table 2 may be recast to clarify the underlying logical processes affecting definitions of the four identity statuses.

In order to construct a set of equations to express those conditions and relationships, three representations must be contrived. The first is the representation of exploration and commitment in quantitative terms. That, of course, is provided by the scores of any derived continuous instrument. The second is the representation of "and" as reflecting the relationships between exploration and commitment for the four statuses. Third, the terms "manifest" and "less manifest" must be represented.

Syntax provides "and" as the expression of the logical operator "conjunction." That logical operator is expressed in mathematical terms as addition or multiplication. In the present research, the operator "and" was treated as reflecting a process of multiplication rather than merely one of addition. The choice of a multiplicative process rests on the assumption that, whereas exploration and commitment may be independent and mathematically orthogonal processes, they are interactive, and with regard to the identity achievement status, they may be synergistic (e.g., Jessee, Gruenfeld, & Waterman, 1995). Thus, because identity achievement represents exploration and commitment, the identity achievement score (A) may be construed as the product of the exploration (E) and commitment (C) scores: A = E X C.

The third representation is of the terms "manifest" and "less manifest." In the rationale for the formula for achievement, the manifestations of exploration and commitment are represented as whole, continuously distributed numbers. "Less manifest" is, of course, some degree of negation of "manifest." Following Freud (1905/1950) and others, negation is the logical maneuver underlying suppression and repression of an idea. In the mathematized psychology of Herbart (1866/1961), repression is represented mathematically as inversion. As such, "less exploration" or "less commitment" becomes the inverse, or reciprocal, of exploration or commitment, that is 1/E or 1/C.

Putting the previous representations together yields the additional necessary formulae. Moratorium symbolizes exploration in the relative absence of commitment, with the moratorium score (M) being the product of the exploration score and the inverse of the commitment score: M = E/C. Similarly, foreclosure embodies the concept of commitment without exploration, with the foreclosure score (F) equaling the commitment score multiplied by the inverse of the exploration score: F = C/E. Finally, identity diffusion represents the relative absence of exploration and commitment; therefore, the diffusion score (D) would be calculated as the product of the inverses of the exploration and commitment scores: D = (1/E) x (1/C).

Having sketched the logic and theoretical base of the formulae, a test of the comparability of the continuous status measures produced by the formulae with other well-established procedures was conducted. Berzonsky (1989) has provided findings on the relationship of a direct continuous measure and his Identity Style Inventory, a measure that is theoretically and empirically related to ego identity. The present research investigated the relationship of a derived categorical measure, which has been afforded derived continuous properties by the formulae, and the same independent measure. The task was to determine whether continuous measures based on exploration and commitment scores from a derived categorical measure relate in theoretically and methodologically sensible ways to the independent measure, in light of previous findings with a direct continuous measure.

In addition, status assignments using the formulae should be comparable to status assignments made with the same sample and measure (the Ego Identity Process Questionnaire) using the conventional median split technique offered by Balistreri, Busch-Rossnagel, and Geisinger (1995). The effectiveness of the formulae was tested in this way.

Two studies were conducted in order to test the effectiveness of the formulae in producing continuous identity status scores. Study 1 tested the formulae in the two ways outlined above. Study 2 replicated Study 1 with a larger sample.

STUDY 1: METHOD

Participants

The participants for Study 1 were 113 Florida State University undergraduates (13 males, 100 females), between 18 and 22 years of age, who were enrolled in introductory psychology courses. Participation in the study partially fulfilled the research requirement for those courses.

Measures

Three instruments were used: the Extended Objective Measure of Ego Identity Status (EOM-EIS; Bennion & Adams, 1986), the Ego Identity Process Questionnaire (EIPQ; Balistreri, Busch-Rossnagel, & Geisinger, 1995), and the Identity Style Inventory (Berzonsky, 1997). The EIPQ is a derived categorical measure; the EOM-EIS is a direct continuous measure; and the Identity Style Inventory is a prominent independent measure within the identity literature. The relationship of the EIPQ and Identity Style Inventory was estimated on the basis of data gathered for this report. The relationship of the EOM-EIS and Identity Style Inventory was estimated from a study by Berzonsky (1989).

Ego identity. The EIPQ is an objective instrument that generates scores for exploration and commitment, sums those scores across eight domains, and uses median splits to categorize individuals into identity statuses. It does not, however, generate continuous scores for the four statuses.

The EIPQ contains 32 items, 16 assessing exploration and 16 targeting commitment. The EIPQ assesses ego identity in four ideological domains (occupational choice, political preference, religious affiliation, and personal values) and in four interpersonal domains (friendships, dating, sex roles, and family). A six-point Likert-type scale, ranging from "strongly disagree" to "strongly agree," is used to rate each item. There are two exploration and two commitment items per domain.

Cronbach's alpha for the EIPQ was .76 for exploration and .75 for commitment. Test-retest reliability coefficients were .90 for exploration and .76 for commitment.

Following recommendations by Balistreri et al. (1995), median splits were conducted on the exploration and commitment scores. Scores falling at or above the median were classified as "high," and the remainder as "low." Identity status categories were assigned on the basis of those median splits. High scores on both exploration and commitment placed the respondent in the identity achievement category. One who had a high score on exploration and a low score on commitment was judged to be in moratorium, whereas someone with the reverse pattern was considered to be foreclosed. Low scores on both scales placed the individual in the identity diffusion category.

For the formulae, a status assignment technique was created based on the classification method for the Identity Style Inventory and the EOM-EIS. All four scores for each participant were converted to standard scores, and the participant was assigned to the status that had the highest standard score. However, following recommendations by Bennion and Adams (1986), participants whose four status scores were all within one standard deviation of their respective means were assigned to a fifth status, low profile moratorium.

Identity style. Building on Kelly's (1955) personal construct theory, Berzonsky (1989) designed an objective measure that also uses continuous scores to obtain a categorical placement. Berzonsky has not purported to measure identity status; rather, he has claimed to measure identity style. The Identity Style Inventory proceeds according to the assumption that all individuals form and maintain their identities through the use of any combination of three distinct processing styles. The identity style paradigm differs from, and builds on, identity status theory in that it is intended to represent a dynamic problem--solving process. One may show a predilection for an extended period of time, although one may not remain within a single identity status for very long (see Waterman & Waterman, 1971). Therefore, identity style is taken to represent the general process through which an individual makes identity-related decisions. Berzonsky correlated identity status scores from the EOM-EIS with the Identity Style Inve ntory scales and found compatibility between the identity status and identity style constructs.

The first and most cognitively sophisticated of the three identity styles is the informational approach. When informational individuals are presented with a problem, they tend to seek advice and/or facts from various sources (e.g., friends, books). They often avoid committing to any specific solution until they have investigated multiple courses of action. Once decided, those individuals remain open to new ideas and facts and are willing to change course.

The normative identity style is characterized by adherence to familial, social, societal, or other group standards. When faced with a predicament, normative individuals are apt to seek counsel from people who serve as authority figures. The result is closed-mindedness toward non-authoritative input and a rigid decision-making style. Not surprisingly, normative individuals tend to be highly committed (Berzonsky & Neimeyer, 1994).

The third identity style, diffuse / avoidant, symbolizes the inability or unwillingness to make decisions at all. Diffuse/avoidant individuals tend to procrastinate and avoid thinking about prospective or imminent situations (Berzonsky, 1992). Consequently, their levels of commitment are low (Berzonsky & Neimeyer, 1994).

The 40-item Identity Style Inventory (Berzonsky, 1997) measures the use of the three identity styles, as well as level of ideological and interpersonal commitment (the version of the Identity Style Inventory used in the two studies reported here is Version 3; Berzonsky used the original 20-item version in his 1989 study). Responses to each item are made on a five-point Likert scale, ranging from "not at all like me" to "very much like me." Eleven items assess the informational style, 9 target the normative style, 10 are aimed at the diffuse/avoidant style, and 10 measure commitment (not analyzed here). Of the three styles, the one assigned the highest standard score becomes the participant's classification.

Internal reliability coefficients (Cronbach's alpha) for the Identity Style Inventory, as reported by Berzonsky (1997), were: informational style, .70; normative style, .64; and diffuse/avoidant style, .76. Test-retest reliabilities over a two-week interval were: informational style, .87; normative style, .87; and diffuse/avoidant style, .83.

Procedure

Approval for the current study was obtained from the Florida State University Human Subjects Committee. Each participant signed a consent form before completing either of the measures. Participants were identifiable only by a code number and were therefore anonymous.

The EIPQ was presented first, followed by the Identity Style Inventory. Both instruments were presented in their paper-and-pencil forms.

STUDY 1: RESULTS

Exploration and commitment scores from the EIPQ were transformed into identity status scores using the formulae introduced in this article. The continuous status scores were then correlated with the Identity Style Inventory variables, and the correlation coefficients were compared with those from the study by Berzonsky (1989). (See Table 4.)

In order to further compare the transformation formulae with the existing continuous measures of identity status, the correlation coefficients from both methods were converted to percentages of explained variance. Explained variance obtained using the EOM-EIS was subtracted from that obtained using the EIPQ. The results of those calculations are displayed in Figure 1.

In 8 of the 12 comparisons between the correlation coefficients, the differences were not statistically significant. The formulae produced relatively strong and theoretically predictable relationships between the Identity Style Inventory informational style measure and the derived identity status scales. With respect to the diffuse/avoidant style scale, the derived continuous scores and the existing continuous scores performed in an equivalent manner, with the exception that the existing diffusion measure appeared to relate more strongly. The two sets of identity measures were similar in their relationships to the Identity Style Inventory normative measure, with the exception that the existing measure of achievement portrayed a stronger relationship to the normative style than did the derived continuous measure of achievement.

For the Identity Style Inventory informational style scale, two of the four differences in correlation coefficients were significant at the .05 level. For both achievement and diffusion, the continuous scores derived from the formulae were more strongly correlated with the informational style than were the existing continuous scores. A third difference, in which the continuous measure of foreclosure correlated more strongly with the informational style scale than did the existing continuous measure of foreclosure, approached statistical significance.

For the Identity Style Inventory normative style scale, one difference in correlation coefficients was significant at the .05 level. The existing continuous measure of achievement was more strongly correlated with the normative style than was the achievement measure derived from the formulae.

For the Identity Style Inventory diffuse/avoidant style scale, onedifference in correlation coefficients was significant at the .05 level. The existing continuous measure of diffusion was more strongly correlated with the diffuse/avoidant style than was the diffusion measure derived from the formulae.

The formulae have been offered to provide the power of continuous measures in statistical analysis. In order to assess the possibility that they also provide adequate categorizations by status, status assignments were made both by the median split technique and the standard score technique, and those sets of assignments were compared with one another. After the 55 participants assigned to low profile moratorium were eliminated from the analysis, the two techniques produced the same categorizations in 55 of 58 cases (94.8%). Among the three discrepancies, two participants classified as foreclosed using the median splits were classified as achieved using the formulae, and one participant assigned to moratorium using the median splits was classified as achieved using the formulae.

STUDY 2: METHOD

Participants

The participants in Study 2 were 325 Florida International University students (83 males, 219 females, 23 did not report gender), between 18 and 27 years of age, who were enrolled in psychology courses. These students received course credit for their participation.

Measures

The EIPQ and Identity Style Inventory were used.

Procedure

Approval for this study was obtained from the Research Review Committee of Florida International University. Each participant signed a consent form and was assigned a code number to ensure anonymity. All participants completed the EIPQ and Identity Style Inventory in a research laboratory on the university campus.

STUDY 2: RESULTS

As in Study 1, exploration and commitment scores from the EIPQ were transformed into identity status scores using the formulae. These status scores were then correlated with the Identity Style Inventory style measures, and the resulting correlations were compared with those from Study 1. No correlations were significantly different between Study 1 and Study 2, and all of the differences that were significant between Study 1 and the Berzonsky (1989) study were also significant between Study 2 and the Berzonsky study.

As in Study 1, status assignments were made using the formulae and were compared with those obstained using the median split technique. After the 142 participants assigned to low profile moratorium using the Bennion and Adams (1986) criteria were eliminated from the analysis, the two techniques produced the same classifications for 176 of the remaining 183 participants (96.2%); [[chi].sup.2](9, 183) = 678.17, p [less than] .00001. Of the seven participants who were not placed into the same status using both techniques, three were classified as diffused using the formulae and as foreclosed using the median splits, two were classified as diffused using the formulae and assigned to moratorium using the median splits, and two were assigned to moratorium using the formulae and classified as achieved using the median splits.

DISCUSSION

The present report has offered simple formulae for deriving continuous measures of ego identity status from continuous measures of exploration and commitment. The effectiveness of the formulae was tested by means of comparing (a) the relationship of continuous measures of status on the EIPQ to Identity Style Inventory style measures with (b) the relationship of continuous measures of status on the EOM-EIS with Identity Style Inventory style measures.

The advantages of generating continuous status scores are threefold. First, continuous measures of identity status empower the investigation of correlates to each status. Second, continuous status scores allow for status assignment by means of standardization procedures rather than simply through median splits. Finally, and perhaps most importantly, continuous status scores may offer a more sensitive measure of the relative strength of each status for a given individual.

The continuous scores derived using the formulae did not differ from the direct continuous measures in terms of 8 of the 12 relationships between identity status and identity style. In all 12 cases, the formulae produced theoretically consistent relationships between identity status and identity style. Among the four differences, two were found for the informational style, one for the normative style, and one for the diffuse/avoidant style.

First, the correlation between the informational style and the achievement status, which was theoretically expected to be high and positive (Berzonsky, 1990), was significantly greater in the current study than in the results by Berzonsky (1989). The larger correlation between achievement and the informational style obtained using the formulae may, to some extent, be a result of the presence of both exploration and commitment, each of which has been shown to be significantly and positively related to the informational style (Schwartz, 1996), in the numerator of the achievement formula.

Second, the diffusion measure created by the formulae was negatively correlated with the Identity Style Inventory informational style scale to a significantly greater extent than was the EOM-EIS diffusion scale. This effect may be the inverse expression of that which produced the higher correlation between the EIPQ achievement measure and the Identity Style Inventory informational style scale. That is, it may be caused by the presence of both exploration and commitment in the denominator of the diffusion formula.

Third, Berzonsky (1989) found a substantial correlation between achievement and the normative style, which was not theoretically expected, whereas the current study produced a much more modest finding. Inasmuch as the normative style is negatively correlated with exploration (Schwartz, 1996), the presence of the exploration term in the numerator of the achievement formula may have served to lower the relationship of the resulting achievement scale with the normative style.

Finally, Berzonsky reported a higher correlation between diffusion and the diffuse/avoidant style, whereas the relationship obtained with the formulae was much lower. The unusually high correlation reported by Berzonsky may be, to some extent, an artifact of overlapping item content between the EOM-EIS diffusion measure and the Identity Style Inventory diffuse/avoidant scale. An inspection of the two scales has revealed some degree of item overlap.

It is presumed that the formulae can be applied to any identity instrument, direct or derived, that produces continuous measures of exploration and commitment. Jackson et al. (1990) derived status assignments from continuous measures of exploration and commitment, on the Grotevant and Cooper (1981) Ego Identity Interview, by setting rational thresholds. Their method did not produce continuous measures of status, however.

The multiplicative formulae put forth produced nearly identical results as those for similarly constructed additive formulae, which were calculated but not reported here. The choice of multiplicative formulae over additive formulae rests on the logic of Freud and Herbart, which appears to support a multiplication algorithm. However, it is for further research to introduce alternative formulae and to test their effectiveness.

The present investigation was not designed to compare the EIPQ and the EOM-EIS; additional research will be needed to accomplish that. It was conducted simply to determine whether a scoring method performs in a way that is intelligible given the standards of the field. On the whole, it appears that the transformed status scores perform acceptably.

The two analyses provide some initial evidence that the formulae: (a) provide meaningful continuous measures of identity status, and (b) may reliably be used for categorizations. It is perhaps obvious that the level of match in categorizations is mathematically ordained. Central tendency, either the median or the mean, z = 0, is the dividing line in each method, and the continuous values of E and C are basic to the categorization in each method. However, whenever there was a discrepancy between the status assigned using the formulae and that assigned using the median splits, the formulae consistently assigned the more mature status.

The conceptual difference between the formulae and median split status assignment techniques is only that the formulae take the additional step of making clear that the statuses may be thought of as continuously distributed resultants of the simultaneous consideration of exploration and commitment measures. This additional step is not only compatible with the common definitions of the concepts in the source theories and mathematically obvious, but it may also anticipate a better treatment of amount or level of a status in future research. It may anticipate a clear recognition that a psychological resultant is not merely an amalgam of its antecedent components, but has additional meaning as an emergent phenomenon. The argument here is not that , for example, achievement is more than the simultaneous occurrence of exploration and commitment, but that research with the formulae is more likely to clarify whether it is more.

Although it may prove fruitful to test the formulae introduced here on other identity measures, a review and evaluation of the range of measures within the identity formation literature was beyond the scope of the present study. It will remain for other studies to clarify the limits of application for the formulae.

It should be noted that the foreclosure and moratorium equations are the mathematical inverses of one another, as are the diffusion and achievement equations. Thus, those two pairs of mathematical inverses yield scores that are reciprocals of each other. As an artifact of the mathematical inverse procedure, two orthogonal identity status axes have been created. The two axes are independent of one another; their measures are not correlated. Yet, the axes created by the formulae may reflect some sort of psychological reality. One of the axes connects the diffusion and achievement statuses, whereas the other axis connects the foreclosure and moratorium statuses. The Cartesian plane comprised of the two axes represents a 45-degree rotation of the exploration and commitment axis matrix offered by Marcia. Further, it is possible that the achievement-diffusion axis represents the bipolar dimension proposed by Erikson, ranging from identity synthesis to identity confusion (see Cote, 1984).

The equations introduced here are intended to reflect the fundamental epistemological processes that enable four combinations of exploration and commitment to yield four identity statuses. In the context of a specific methodological contribution, it is not possible to devote the space that would be required to treat the historical, philosophical, and psychometric roots of the four equations fully. Their rationale has merely been outlined.

A number of historical, philosophical, mathematical, and psychological points remain to be explored or elaborated. Among those are the debts to the logic of Herbart and Freud and an extended discussion of the exact psychological conditions to be reflected in the formulae. One item within that discussion might be whether the iteration of additive effects of simultaneous or immediately sequential psychological processes yields a multiplicative overall effect. For now, however, it seems sufficient to say that a base has been laid for further empirical work to ascertain the conditions under which the formulae may be useful or to test the comparative value of other formulae that might be offered.

Richard M. Dunham, Department of Psychology, Florida State University, Tallahassee, Florida.

REFERENCES

Adams, G. R., Bennian, L. D., & Huh, K. (1989). Objective Measure of Ego Identity Status: A reference manual. Unpublished manuscript, Utah State University, Logan.

Balistreri, E., Busch-Rossnagel, N. A., & Geisinger, K. F. (1995). Development and validation of the Ego Identity Process Questionnaire. Journal of Adolescence, 18, 179-192.

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.

Berman, A. M., Schwartz, S. J., Kurtines, W. M., & Berman, S. L. (2000). The role of style and competence in identity formation. Manuscript submitted for publication.

Berzonsky, M. D. (1985). Diffusion within Marcia's identity status paradigm:n Does it foreshadow academic problems? Journal of Youth and Adolescence, 14, 527-538.

Berzonsky, M. D. (1989). Identity style: Conceptualization and measurement. Journal of Adolescent Research, 5, 268-282.

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 theory: Volume 1 (pp. 155-186). Greenwich, CT: JAI Press.

Berzonsky, M. D. (1992). Identity style and coping strategies. Journal of Personality, 60, 771-788.

Berzonsky, M. D. (1997). Psychometric properties of the Identity Style Inventory, Version 3. Unpublished manuscript, State University of New York at Cortland.

Berzonsky, M. D., & Neimeyer, G. J. (1994). Ego identity status and identity processing orientation: The mediating role of commitment. Journal of Research in Personality, 28, 425-435.

C[hat{o}]tacute{e} J. E. (1984). The identity crisis: A formulation and empirical test of Erikson's theory of ego identity formation. Unpublished doctoral dissertation, York University, Ontario, Canada.

Erikson, E. H. (1950). Childhood and society. New York: Norton.

Freud, 5. (1950). Negation. In D. Rapaport (Ed.), Organization and pathology of thought: Selected sources (pp. 75-92) New York: Columbia University Press (Original work published 1905).

Grotevant, H. D., & Cooper, C. R. (1981). Assessing adolescent identity in the areas of occupation, religion, politics, friendships, dating, and sex roles: Manual for the administration and coding of the interview. Journal Supplement Abstract Service Catalog of Selected Documents in Psychology, 11, 52-53 (Ms. No. 2295).

Herbart, J. F. (1961). Psychology as a science, newly founded upon experience, metaphysics, and mathematics. In T. Shipley (Ed.), Classics in psychology (pp. 22-50). New York: Philosophical Library (Original work published 1866).

Jackson, E. P., Dunham, R. M., & Kidwell, J. S. (1990). The effects of gender and family cohesion and adaptability on identity status. Journal of Adolescent Research, 5, 161-174.

Jessee, C. L., Gruenfeld, K. E., & Waterman, A. S. (1995, February). The relationship of identity status to optimal psychological functioning and decision making. Paper presented at the second biennial meeting of the Society for Research on Identity Formation, Dog Island, FL.

Kelly, G. A. (1955). The psychology of personal constructs. New York: Norton.

Kidwell, J. S., Dunham, R. M., Bacho, R. A., Pastorino, E., & Portes, P. R. (1995). Adolescent identity exploration: A test of Erikson's theory of transitional crisis. Adolescence, 30, 785-793.

Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personality and Social Psychology, 5, 551-558.

Marcia, J. E. (1967). Ego identity status: Relationship to change in self-esteem, "general maladjustment," and authoritarianism. Journal of Personality, 35, 118-133.

Marcia, J. E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handbook of adolescent psychology (pp. 159-187). New York: Wiley.

Marcia, J. E. (1988). Common processes underlying ego identity, cognitive/moral development, and individuation. In D. K. Lapsley & F. C. Power (Eds.), Self, ego, and identity: Integrative approaches (pp. 211-225). New York: Springer-Verlag.

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., & Archer, S. L. (1993). The Identity Status Interview: Late Adolescent College Form. 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. 318-333). New York: Springer-Verlag.

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.

Schwartz, S. J. (1996). Ego identity, personal expressiveness, and identity style. Unpublished master's thesis, Florida State University, Tallahassee.

Stephen, J., Fraser, E., & Marcia, J. E. (1992). Moratorium-achievement (MAMA) cycles in lifespan identity development: Value orientations and reasoning system correlates. Journal of Adolescence, 15, 283-300.

Waterman, A. S. (1988). Identity status theory and Erikson's theory: Commonalities and differences. Developmental Review, 8, 185-208.

Waterman, A. S., & Waterman, C. K. (1971). A longitudinal study of changes in ego identity status during the freshman year at college. Developmental Psychology, 5, 167-173.

 Some Identity Status Instruments by Mode of Status
 Assessment and by Type of Status Scaling
Assessment Categorical Scaling Continuous Scaling
Direct Identity Status Interview Extended Objective Measure
 (Marcia & Archer, 1993) of Ego Identity Status
 Ego Identity Interview (Bennion & Adams, 1986)
 (Grotevant & Cooper, 1981)
Derived [*] Ego Identity Process Questionnaire None [**]
 (Balistreri, Busch-Rossnagel, &
 Geisinger, 1995)
(*.)calculated from measures of exploration and commitment
(**.)see Jackson, Dunham, & Kidwell, 1990.
 Identity Statuses, Reflecting Manifest and
 Less Manifest Exploration and Commitment
 Commitment
 MANIFEST LESS MANIFEST
Exploration MANIFEST Achievement Moratorium
 LESS MANIFEST Foreclosure Diffusion
 Logical Recasting of Interactions of Exploration and Commitment
Status Exploration Logical Relationship Commitment
Achievement manifest AND manifest
Moratorium manifest AND less manifest
Foreclosure less manifest AND manifest
Diffusion less manifest AND less manifest
 Comparison between EOM-EIS and EIPQ
 Correlations with the Identity Style Inventory
 Identity Style
 Informational Normative Diffuse/Avoidant
Achievement
EOM-EIS .25 .52 -.40
EIPQ .53 .20 -.36
Difference (z) 2.27 [*] 2.61 [**] 0.28
Moratorium
EOM-EIS .06 -.29 .29
EIPQ .15 -.50 .02
Difference (z) 0.62 1.72 1.93
Foreclosure
EOM-EIS -.02 .47 -.06
EIPQ -.14 .48 -.10
Difference (z) 0.83 0.07 0.28
Diffusion
EOM-EIS -.28 -.18 .62
EIPQ -.54 -.28 .31
Difference (z) 2.13 [*] 0.76 2.82 [**]
Note: The z test determined whether the correlation
coefficients were significantly different from each other.
(*.)p [less than] .05
(**.)p [less than] .01