Return to Commission IX Clearinghouse Index
The Erwin Identity Scale
T. Dary Erwin and Albert B. Hood
The establishment of identity is the core developmental vector in Chickering's schema, "...that solid sense of self that assumes form as the developmental tasks for competence, emotions, and autonomy are undertaken with some success, and which, as it becomes more firm, provides a framework for interpersonal relationships, purposes, and integrity" (Chickering, l969, p. 80). According to Chickering, itentity then is a primary element around which other vectors revolve. Identity is generally referred to as the answer to the question of: who am I? That is, identity refers to the type of person the individual thinks he or she is; the sameness, the unity, or the individual comprehensiveness of the person.
For Chickering, the quest for personal identity in self- understanding is a major life time developmental task. The search for identity reaches its highest point, however, during late adolescence and earlier adulthood (ages l8-25). At that time when body changes are profound, genital maturity is developing, and other approaching issues of adulthood are eminent, one's focus turns to developing a sense of self. Chickering directly specified two aspects of identity-- conceptions concerning body and appearance, and clarification of sexual identification. For Chickering, the changing, growing, physical body and self-probing questions about self become partially focused on one's presentation of self. Appearance is both superficial and fundamental. The superficiality is exhibited through style of dress and external appearances. More fundamental issues include feelings, questioning self-esteem, such as unattractiveness and worthlessness. In the establishment of identity, one's presentation of self is not overly influenced by current styles or by the dictates of peers or persons in authority. It is the integration of influences--of others, of situations, and of personal appearances that affirms an established identity.
The second component of identity that Chickering specifically mentioned is sexual identification. He suggested the need for clarification of a male or female role and the need for clarification between heterosexuality and homosexuality.
Erwin (l978) suggested a third component of identity-- personal confidence. He has maintained that although Chickering did not directly mention it when he quoted from Erickson about "intercapital" and "accrued confidence," he implied that self- assurance was a necessary component of identity.
The Construction of the Erwin Identity Scale
It was around these three areas that the Erwin Identity Scale was constructed. It is a 58-item instrument with three sub-scales: confidence, sexual identity, and conceptions about body and appearance.
Confidence Scale
Confidence is an assuredness in one's self and in one's capabilities. Confidence includes a conscious self-reliance while recognizing the necessary dependence on outside sources. This recognition is an awareness and faith in one's own capabilities yet a realization that there are limits to these processes. The confident person has some understanding of his or her limitations. A self-confident individual feels comfortable about expressing beliefs, making decisions, and behaving competently, even though action may not be taken in these areas. A sample item on this subscale is "I am as sure of myself as most other people seem to be sure of themselves."
Sexual Identity
The Sexual Identity Scale attempts to measure the clarification, understanding, and acceptance of one's sexual feelings. The person with a high degree of sexual identity recognizes his or her sexual feelings as natural and normal. There is an absence of guilt because of their presence. Sexual identity includes not only a positive acceptance of one's sexual feelings but also a control of one's sexual feelings. For instance, a person's sexual feelings are not overwhelming and do not interfere in interactions with other people. Moreover, sexual feelings are accepted as a normal part of close love relationships. This recognition and acceptance of sexual feelings does not imply sexual activity or lack of it. A sample item is, "I believe that I understand the sexual functions of men and women."
Conceptions about Body and Appearance
Identity also includes an accurate self-perception and acceptance of one's body and one's appearance. It is an issue of presentation of self. What do I think of my body? How do I conceive of myself and my appearance? An increasing acceptance of one's body particularly in relation to other people is a necessary component. In addition, one's appearance and dress are resolved issues representing a varied balancing of personal preferences, the desires of others, and situational expectations. A person with a high degree of identity exhibits a personal dress style governed by individual tastes rather than the dictates or expectations of other people, such as peers or persons in authority. A sample item is, "I would not change my style of clothes just because my boss indicated that I should dress more like him or her."
First Draft of the EIS
A preliminary form of the EIS with 78 items was administered to a group of undergraduates at The University of Iowa in the late spring of l977. A stratified random sample of 80 students was drawn balanced by sex and freshmen and senior classes. The sample of freshmen and seniors was selected since Chickering hypothesized that seniors would have a more established identity than freshmen. Two additional scales--Lack of Anxiety and Personal Integration from the Omnibus Personality Inventory (Heist & Yonge, l968) were administered in conjunction with the EIS to examine construct validity. According to Chickering, the establishment of identity is accompanied by a lessening anxiety and a greater degree of personal integration and stability. These were also the same two scales Chickering reported using.
The freshmen were tested near the end of their second semester in college and the seniors tested near their graduation. Sixty one of the 80 students (76%) completed the inventory. Reliability estimates using Cronbach's coefficient of internal consistency (l95l) yielded coefficients in the vicinity of .70 for each of the sub-scales. These coefficients were determined to be satisfactory for the initial form of the instrument. Correlations between the EIS sub-scales and between the OPI scales and the EIS sub-scales ranged in the vicinity of .5 to .7. As Chickering theorized, and was expected, a moderate degree of relationship was shown between the EIS measures and the lack of anxiety and personal integration measures.
Several 2 x 2 (sex x class) ANOVA tests were performed on the EIS sub-scales. Scores on the EIS were tested for mean differences across groups of men and women, and freshmen and seniors. No statistical differences among the means of the four groups were found. One possible reason for this is that changes in identity may occur before the end of the freshmen year or after the senior year contrary to Chickering's hypotheses. Another possibility is that the EIS is not validly measuring identity which is purported by Chickering to differ between freshmen and seniors. Correlations between the EIS and college grade point averages ranged from -.04 to .l6 suggesting little or no relationship between grades and identity as measured by the EIS.
Since no differences occurred between the freshmen and seniors in the undergraduate sample and because the testing was done near the end of the academic year (the freshmen were almost sophomores) it was decided to sample students from lower and higher educational levels. Small samples of high school students and graduate students were administered the EIS and the results suggested that a change in identity as measured by the EIS may come between high school and late in the freshman year.
Longitudinal Studies
The EIS was administered to a random sample of l69 students who participated in a summer orientation program as entering freshmen at The University of Iowa in l977. The majority of these students were white, single, and under l9 years of age. Half of these students (those who Social Security Numbers ended in an odd digit) were asked to return and have the EIS readministered to them during the middle of their first semester--the fall semester of l977. Fifty five students participated in this re-test. The remaining students--the 84 with even numbered Social Security Numbers--were asked to take the EIS during the middle of their second semester--the Spring semester of l978. Seventy four re-took the EIS at that time. Descriptive statistics and reliability estimates for the total freshmen orientation testing and the first and second freshmen re-tests are given in Table D-l.
Reliability estimates fell in the range of .7 to .8 and remained stable over the freshman year. For the 55 students re- tested during the first semester of their freshman year, a significant increase was found on the confidence sub-scale and a significant decrease on the conceptions about body and appearance sub-scale (Table D-2). Those students with even numbered Social Security Numbers followed up in the Spring during the middle of their second semester did not show any significant changes in their mean scores on either the confidence or the sexual identity sub-scale. They also showed a significant decrease on the conceputions about body and appearance sub-scale.
One hundred thirty four students in the freshmen orientation sample had ACT scores and granted permission for their use in this study. The ACT scores included five measures: English, mathematics, social science, natural science, and composite. In addition to ACT scores, self reported high school grade point average and vocational interest scores were also obtained from the ACT Student Profile Section. The Vocational Interest Inventory is based on Holland's theory of vocational choice and has six l5-item scales measuring the following interests: realistic or technical, investigative or science, artistic or creative arts, social or social science, enterprising or business contact, and convention or business detail. Correlation coefficients between EIS scores and ACT scores ranged from -.087 to +.l95. Between EIS sub-scales and reported high school grade point averages the range was from -.0l0 to +.l69 and between the different sub-scales and the various vocational interest scores the range was from -.l48 to +.l44. Thus there does not appear to be any relationship between these cognitive and vocational interest measures and the identity measures of the EIS.
Senior Year Follow-Up Study
Of the l69 entering freshmen who took the EIS, ll8 still remained on the campus four years later during what would ordinarily be the second semester of their senior year. These students were contacted a number of times asking them to re-take the EIS at a number of convenient locations established for this purpose around the campus. Eventually, completed forms were obtained from 82 of ll8 students representing 69% of this group. Mean scores for each of the sub-scales of the EIS for the 82 students tested at the end of their senior year are shown in Table D-3. The mean scores on all three sub-scales--Confidence, Sexual Identity, and Concepts about Body and Appearance-- increased from freshmen orientation to senior year. Scores on the Confidence sub-scale which had increased slightly during the freshmen year continued to show a further increase by the end of the senior year. The Sexual Identity sub-scale scores which had remained the same during the freshmen year also increased by the end of the senior year. The scale which showed a decrease during the freshmen year--Conceptions about Body and Appearance--showed a smaller but still statistically significant increase from the original summer orientation testing and a larger increase from the freshmen year re-tests.
Of the 83 students who participated in the fourth year follow-up, 33 were males and 50 were females. Mean scores of the sexes were compared on all three sub-scales of the EIS to determine if there were any sex related differences. The results which appear in Table D-4 revealed that men had significantly higher scores on the Confidence sub-scale in their freshmen year. In the fourth year follow-up, men scored significantly higher on the Confidence sub-scale as well as on the Conceptions about Body and Appearnace sub-scale. Students of both sexes showed substantial gains on all sub-scales over the four year period with smallest change occurring among the females on the Conceptions about Body and Appearance sub-scale.
High School Sample
One hundred seventy-one students in a parochial high school were administered the EIS and several other instruments. Descriptive statistics for this group are shown in Table D-5. The internal consistency coefficients were somewhat lower for the high school students and several reasons may account for this difference. The testing was done over a two day period with only 25 minutes allowed for a single day of testing and interruption proved disruptive. Secondly, a number of instruments contained obviously invalid results as several students did not follow the correct directions in responding to the items. In addition, the seriousness of the high school students during the testing periods was not as great as was the diligence observed for the college students.
To examine several aspects of construct validity for the EIS, sub-groups of the high school students were administered several other instruments. Thirty seven of them completed the Simmons' Identity Achievement Scale (l973)--a measure of Marcia's and Erikson conceptulation of identity. This instrument contains 24 items designed to measure identity achievement and Simmons has reported a test re-test liability coefficient of .76. The correlation of .46 was obtained between the EIS and the confidence sub-scale and correlations of .29 obtained for the sexual identity sub-scale and .l8 with the conceptions about body and appearance sub-scale. The Rotter Internal-External Scale (l965) was administered to 35 of the high school students. This scale is a measure of the generalized expectancy or belief in internal control versus external control of reinforcement as a psychological variable. Internal control is the degree to which an individual perceives that reinforcement "follows from or is contingent upon his/her own behavior or attributes." The opposite determinant or external control is the degree to which an individual "feels the reward or reinforcement is controlled by forces outside of himself/herself and may occur independently of his/her own reactions." Rotter reports test re-test reliability coefficients ranging from .49 to .83 and has cited evidence for construct validity from a variety of studies. A correlation of -.36 was found between the EIS sexual identity sub-scale and the internal-external scale. For the confidence sub-scale the correlation was .2l and for the conceptions about body and appearances sub-scale the correlation was -.0l. These results suggest that high school students who score higher on the confidence and sexual identity sub-scales tend to have a more internal control of reinforcement.
Many other researchers studying identity have found that it is a complex concept. The problems encountered in the construction of the Erwin Identity Scale also showed that identity or the sense of self is a difficult concept to explain and therefore measure. Given these problems, the three subscales on the Erwin Identity Scale do offer further information on the general development of students and provided an objectively scored instrument for this vague but widely discussed concept.
References
Chickering, A. W. (1969). Education and identity. San Francisco: Jossey-Bass.
Cronbach, T. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297-334.
Erwin, T. D. (1979). The validation of the Erwin Identity Scale. (Doctoral dissertation, The University of Iowa, 1978). Dissertation Abstracts International, 39, 4818A.
Heist, P., & Yonge, G. (1968). Omnibus Personality Inventory Manual. New York: The Psychological Corporation.
Rotter, J. B. (1965). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80, 1-28.
Simmons, D. D. (1973). Development of an objective measure of identity achievement status. Journal of Projective Techniques and Personality Assessment, 34, 241-244.
**************************************************
Table D-1
EIS Means, Standard Deviations and Reliabilities
Orientation First Retest Second Retest
Subscale Mean S.D. Rel. Mean S.D. Rel. Mean S.D. Rel.
Confidence 63.3 10.73 .81 61.6 11.2 .82 65.1 11.0 .83
Sexual 78.8 10.88 .75 76.9 10.4 .71 80.9 10.9 .77
Identity
Conceptions
about Body 58.7 10.36 .79 57.5 9.7 .75 57.7 11.8 .86
and
Appearance
N = 169 55 74
**************************************************
Table D-2
Means on EIS Sub-scales for Original and
First and Second Retests
________________________________________________________
Subscale Summer Summer Fall Summer Spring
Testing Testing Retest Testing Retest
(All)
Confidence 63.3 6l.8 64.4* 64.7 65.1
Sexual 78.8 77.6 78.6 80.4 80.9
Identity
Conceptions
about Body 58.67 62.0 58.0* 6l.8 57.7*
and
Appearance
N = l69 55 55 74 74
________________________________________________________
*Significant change from Summer testing at .05 level.
**************************************************
Table D-3
Freshmen vs. Senior Means on E.I.S.
________________________________________________
Summer Orientation Senior Year
Subscale Mean S.D. Mean S.D. t
Confidence 63.9 11.1 68.1 10.8 2.85**
Sexual Identity 74.0 11.2 84.1 10.2 2.99**
Conceptions
about Body 59.2 10.1 62.8 9.9 2.27*
and Appearance
N = 82 82
__________________________________________________
*Significant change from Summer testing at .05 level
**Significant change from Summer testing at .01 level
**************************************************
Table D-4
Male and Female Scores on E.I.S.
_____________________________________________________
1977 Means 1981 Means
Subscale Male Female t Male Female t
Confidence 67.8 61.7 2.57* 73.7 65.8 3.58**
Sexual Identity 76.0 80.7 1.80 84.0 84.1 0.04
Conceptions
about Body 60.9 58.1 1.22 65.3 61.0 2.02*
and Appearance
N = 32 50 33 49
_____________________________________________________
*Significant at .05 level
**Significant at .01 level
**************************************************
Table D-5
High School Students - Descriptive Statistics
____________________________________________________
Standard Internal Consistency
EIS Subscale Mean Deviation Coefficient
Confidence 57.5 9.2 .71
Sexual Identity 72.8 9.8 .70
Conceptions
about Body 52.6 9.5 .74
and Appearance
N = 171
Return to Iowa Index