Citation
Acculturation, family characteristics, gender, and identity development of African American, Caucasian, and Hispanic adolescents

Material Information

Title:
Acculturation, family characteristics, gender, and identity development of African American, Caucasian, and Hispanic adolescents
Creator:
Forbes, Sean Alan
Publication Date:
Language:
English
Physical Description:
vi, 74 leaves : ; 29 cm.

Subjects

Subjects / Keywords:
Acculturation ( jstor )
Adolescents ( jstor )
African Americans ( jstor )
Ethnicity ( jstor )
Foreclosures ( jstor )
High school students ( jstor )
Hispanic Americans ( jstor )
Hispanics ( jstor )
Parenting style ( jstor )
Tax moratoriums ( jstor )
Dissertations, Academic -- Educational Psychology -- UF ( lcsh )
Educational Psychology thesis, Ph. D ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1999.
Bibliography:
Includes bibliographical references (leaves 68-73).
Additional Physical Form:
Also available online.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Sean Alan Forbes.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021557649 ( ALEPH )
43637782 ( OCLC )

Downloads

This item has the following downloads:


Full Text










ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND HISPANIC ADOLESCENTS











By

SEAN ALAN FORBES









A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1999














ACKNOWLEDGEMENTS

Patricia Ashton has guided me since I began graduate school. Her wisdom and patience have not been lost on me. I thank her for her unwavering support and am indebted for the lessons she has taught me. This dissertation is dedicated to her. I would also like to thank the other members of my committee for their assistance. James Algina was a sobering reminder of the diligence needed for any worthwhile pursuit. John Kranzler, who has known me since I was an undergraduate student, has counseled me many a time and his advice has served me well. Barry Guinagh has been the kindest man I have ever met. Greg Neimeyer and Lee Mullally, my outside committee members, have saved my skin on more than one occasion. Collectively, this committee has been outstanding.

I must also recognize Randall Scott Hewitt-the most natural philosopher I hav, known. Graduate school would not have been complete without him. He and his family have opened my eyes to other worlds and I will miss their companionship.
















ii















TABLE OF CONTENTS


ACKNOWLEDGEMENTS ......................................................................... ii

ABSTRACT................................. ...................... ............................................... v

CHAPTERS

1 INTRODUCTION.....................................................................................................1

Statement of the Problem......................................................................................... 3
Purpose of the Study............................................................................................. 9
Ecological Variables Related to Identity Status....................................... .............. 0
Research Questions....................................... .......... ...... ......... ............. .. 27
Significance of Study............................................................................................... 28

2 METHOD .............................................................................................................. 30

Research Participants.............................................................................................30
Instruments............................................... 30
Procedures........................................ ................................................................ 34
Research Questions............................................ 34

3 RESULTS ............................................................................................................... 36

Summary Statistics ................................................................................................. 36
Results of Statistical Analyses ............................................................................ 42

4 DISCUSSION ...................................................................................................... 51

Discussion of Results of Statistical Analyses ....................................... .... .. 57
Limitations of the Study.............................................................. ........................ 62
Implications of the Study................................................................................... 63
Recommendations for Future Research ............................. ..................... .......... 64

APPENDICES

A OVERVIEW OF IDENTITY STUDIES ................................................................. 66



iii









B INFORMED CONSENT ............................ ........................................................... 67

REFER EN CE S ............................ ........... .... .................. .. ........ ........ ............... ......... 68

BIOGRAPHICAL SKETCH ......................................... 74


















































iv














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND HISPANIC ADOLESCENTS

By

Sean Alan Forbes

December 1999



Chair: Patricia Ashton
Major Department: Educational Psychology

Adolescence is a critical period in the development of psychosocial identity. That is, as the transition is made from child to adult, the adolescent must reconcile childhood identifications with the coming expectations of adulthood. The goal is to establish a sense of individuality in the larger society. This goal is attained through the development of an identity. The identity crisis includes four statuses: (a) achievement-wherein an adolescent experiences an identity crisis and makes a commitment to alternatives, (b) moratorium-wherein an adolescent experiences an identity crisis but makes no commitments, (c) foreclosure- wherein an adolescent bypasses an identity crisis by adopting the commitments of parents or other significant figures, and (d) diffusionwherein an adolescent avoids the experience of a crisis and postpones commitments. One of the less considered areas in identity status research has been the comparison of identity



V








development among ethnic groups. In the studies investigating the topic, inconsistent results have been found. Some researchers have reported that African American and Hispanic adolescents were more likely to be foreclosed than their Caucasian peers. In light of the differences among ethnic groups in parenting style and familism reported in some studies, this study was designed to determine whether the inconsistent findings regarding the relation between identity status and ethnic group could be explained by parenting style, familism, SES, gender, and/or family configuration. Also, as adolescents from cultural minorities must negotiate a bicultural existence, it was also of interest to consider the relation between identity development and acculturation for African American and Hispanic adolescents. Three hundred college aged adolescents participated in this study. Of the total, 82 identified themselves as African American and 98 identified themselves as Caucasian. One hundred twenty participants identified themselves as Hispanic. Analysis of variance indicated no significant differences among groups in identity development and no relationship between identity development and familism. Significant relationships were found between parenting style and acculturation and identity development.


















vi















CHAPTER I
INTRODUCTION

According to Erikson (1963), the formation of an identity is the dominant psychosocial crisis of adolescence. Commonly referred to as the identity crisis, this process is characterized by a delay of adult commitments that permits the adolescent to experiment with various roles and ultimately to commit to available identity alternatives in ideological and interpersonal domains (Erikson, 1968, p. 159). As the goal of an identity crisis is the synthesis of the identifications of childhood and the opportunities and expectations of adulthood, identity resolution is the foundation upon which future adult decisions are based (Erikson, 1987; Spencer & Markstrom-Adams, 1990). Moreover, the individual who successfully resolves the crisis becomes a secure, independent adult prepared to establish intimacy with others, whereas those who are unable to establish an identity enter adulthood without a firm sense of how they fit into society (Elkind, 1978; Peretti & Wilson, 1995).

Marcia (1966, 1976), in an extension of Erikson's theory, described four possible outcomes of the crisis: identity achievement, moratorium, foreclosure, and diffusion. Identity achievement is defined by a commitment to identity alternatives after a period of exploration, whereas moratorium is defined as a period of exploration of alternative roles during which commitment to an identity is delayed. Foreclosure is defined by identity commitment without prior exploration, whereas diffusion is defined by the unwillingness to commit to an identity with or without exploration.


I





2


The achievement and moratorium statuses, by involving exploration, are more progressive identity statuses than foreclosure and diffusion, which do not involve exploration (Waterman, 1993). That is, adolescents must play an active role in the resolution of childhood identifications with adult opportunities and expectations; they must experiment with the identity alternatives available if a positive resolution of the crisis is to be achieved.

Taking an active part in resolving their identity crisis often leaves the adolescent open for problems. For example, adolescents classified in the moratorium status are more prone to anxiety than those in foreclosure because of the lack of commitment in their lives (Marcia, 1976; Sterling & Van Horn, 1989). Yet, the overall outcome of being identity achieved or in moratorium is positive. That is, by taking an active role in their identity crises, adolescents in the moratorium or achievement status are psychologically healthier than those in diffusion and foreclosure. For example, those in moratorium and achievement are less likely than those in foreclosure to be authoritarian or to use stereotypical thinking (Marcia & Friedman, 1970; Streitmatter & Pate, 1989). Those in achievement and moratorium are more likely to possess an internal locus of control (Dellas & Jernigan, 1987), and to be more autonomous than those in foreclosure and diffusion (Andrews, 1973). In the areas of cognitive performance and cognitive style, those in the achieved and moratorium statuses often perform at a higher level than those in foreclosure and diffusion. For example, adolescents in the achievement status have been found to have better study habits (Waterman & Waterman, 1975) and higher grade point averages than those in other statuses (Cross & Allen, 1970, Raphael, 1977). Waterman and Waterman (1975) found that individuals in achievement and moratorium





3


were more reflective and less impulsive than those in foreclosure and diffusion, and Adams, Abraham, and Markstrom (1987) found that individuals in the higher statuses were less self-conscious than those in foreclosure and diffusion.

Statement of the Problem

As individuals move through adolescence, most tend toward the progressive

statuses of achievement and moratorium rather than the regressive statuses of diffusion and foreclosure. Several studies, however, have suggested that progressive identity development is more common for Caucasian adolescents than for their African American and Hispanic peers.

In a longitudinal study of ideological identity in 22 low socioeconomic status

(SES) African American and Caucasian high school males, Hauser (1972) administered a Q-sort annually for 3 years and obtained correlations within each annual Q-sort and between each annual Q-sort to estimate the consistency within aspects of the students' self-image and change over time in their self-image. From these correlations, Hauser concluded that, compared to the Caucasian students, the African American students were more foreclosed than the Caucasian students.

Several problems in the Hauser (1972) study raise questions about the validity of his findings. First, the study consisted of a convenience sample of 22 matched African American and Caucasian students. The small size of the sample and the inability to determine the adequacy of the matching contribute to doubts about the validity of the findings. Second, the author used a Q-sort of undetermined reliability rather than the traditional interview or questionnaire measures of identity. Third, the reliability of the differences between the correlations in the African American and Caucasian groups is









unclear. Given these weaknesses in Hauser's study, the validity of his conclusion that the African American students were more foreclosed than their Caucasian counterparts must be questioned. Furthermore, the study was conducted over 30 years ago (between 1962 and 1967). Hauser referred to students' comments during the interview that suggested that racial restrictions on the African American students' daily lives (e.g., restrictions in jobs, social life, recreation, and housing), limitations in work opportunities, and the lack of role models who had succeeded in the majority culture may have contributed to their sense of foreclosure. The greater social and economic opportunities available to African Americans today may have reduced that sense of foreclosure.

Despite the weaknesses in Hauser's study, a more recent study by Abraham

(1986) supported Hauser's findings regarding the relationship between ethnic-minority status and identity foreclosure. Abraham investigated the ideological and interpersonal identity of 841 Mexican American and Caucasian students in grades 9-12 at a high school in the southwest United States. Respondents completed the Extended Objective Measure of Ego Identity Status (EOM-EIS) (Grotevant & Adams, 1984), a 64-item measure of identity in four areas of ideological identity (occupation, religion, politics, and lifestyle) and interpersonal identity (gender roles, dating, friendship, and recreation). In a multivariate analysis of covariance, Abraham used mothers' and fathers' education as covariates to control for differences in SES in the two groups. The results indicated Caucasians and Mexican Americans differed in ideological identity. A greater percentage of Mexican American respondents were foreclosed than their Caucasian counterparts. Abraham concluded, therefore, that the Mexican American adolescents seemed to be more receptive to their parents' views and ideals on politics, religion, occupation, and






5


philosophical lifestyle. No differences between ethnic groups were found on interpersonal identity.

Abraham (1986) offered two alternative explanations for her findings. First, she suggested that Hauser's belief that the students' ethnic minority status may have limited the range of their exposure to ideological options. Second, she noted that differences in parenting style might account for the results, though this variable was not included in the analysis. Abraham concluded that the limitations in options and parenting style may have contributed to the greater number of the Mexican American students in the foreclosure status. Selection of participants from a single school raises questions, however, about the generalizability of Abraham's findings.

Administering the EOM-EIS to 367 students in grades 7 and 8 in a junior high school in the Southwest, Streitmatter (1988) compared the identity development of Caucasians and an ethnic minority group composed of Hispanic Americans, African Americans, Native Americans, and Asian Americans. Like Hauser (1972) and Abraham (1986), Streitmatter found that the Caucasian adolescents were less foreclosed than the minority respondents. Unlike the previous studies, however, differences were found in the interpersonal domain as well as the ideological. Streitmatter suggested that the greater incidence of foreclosure in the minority groups was the result of minority adolescents' feeling more uncomfortable in the majority culture than their Caucasian peers. She proposed that this discomfort led them to accept parental values and reduced their willingness to experiment with ideas that conflicted with their parents' expectations. Further, she pointed out that the minority students were bussed into the school, which might have added to the students' discomfort and inhibited their exploration. She referred








to research on parenting style that has suggested that a warm, supportive, and controlling parenting style is related to the foreclosure status and proposed that the Mexican American parents may be more likely to use such a style with their children. In light of her findings, Streitmatter suggested that studies must be done to identify mediators of the relationships between ethnicity and identity development.

Like Hauser (1972) and Abraham (1986), the Streitmatter study is seriously flawed. She combined the four ethnic groups into one group in her analysis. Citing problems such as poor sampling techniques and a lack of control for SES in previous studies, Markstrom-Adams and Adams (1995) designed a study comparing the identity development of African American, American Indian, Mexican American, and Caucasian adolescents to address these design flaws. Their respondent pool roughly approximated the racial composition of the area from which the respondents were sampled, and SES was controlled by obtaining the respondents' parents' educational level. MarkstromAdams and Adams also investigated the role that age and gender roles play in the development of identity.

For the Markstrom-Adams and Adams (1995) study, 123 adolescents in grades 10-12 in a high school in the rural Southwest completed the EOM-EIS (Grotevant & Adams, 1984), the Personal Attributes Questionnaire (Spence, Helmreich, & Stapp, 1974), a gender role measure, the Nowicki-Strickland Locus of Control Scale for children (Nowicki & Strickland, 1973), and questions about parents' level of education. Like the previous researchers, Markstrom-Adams and Adams found that the minority adolescents were more foreclosed than their Caucasian peers. In addition, they found a nonsignificant trend suggesting that the minority adolescents were also more likely to be classified as






7


diffused than their Caucasian counterparts. Also, relationships between the contextual variables and identity status emerged. Higher parent education was related to lower foreclosure rates for ideological identity. In concluding, Markstrom-Adams and Adams (1995) agreed with the earlier researchers who argued that the greater foreclosure in nonWhite adolescents might be because the minority groups are more accepting of foreclosure than Caucasians, who may require experimentation and autonomous choice in achieving an identity.

Two other studies, however, did not report differences in foreclosure in minority and Caucasian adolescents (Abraham, 1983; Rotheram-Borus, 1989). Abraham (1983) investigated the identity development of 223 Caucasian and Mexican American students from a rural area in the southwest with the Objective Measure of Ego Identity Status (OM-EIS), a 24-item measure of ideological identity, including the areas of occupation, politics, and religion (Adams, Shea, & Fitch, 1979). She found that more Caucasian students than Mexican American students were classified in diffusion. Although more Mexican American students than Caucasian students were classified as foreclosed (18.8% compared to 9.6%) and more Caucasian students were classified as identity achieved (10.1% compared to 16.4%), these differences were not significantly different.

In her discussion, Abraham (1983) pointed out that SES level has been identified as a variable related to identity development. A post hoc analysis of the parental educational level of the students in her study revealed that the parents of the Mexican American students were lower in educational level than the parents of the Caucasian students. Because a higher SES level would be expected to be associated with lower levels of diffusion, she proposed that parental socialization style might be the influential





8


variable in her study. She pointed out that Adams and Jones (1983) found that a maternal style characterized by control, regulation, and strong encouragement of independence and a paternal style characterized by praise, approval, and less fair discipline were related to diffusion, parental styles that might have been more common among the parents of the Caucasian students compared to the Mexican American students.

Using the OM-EIS (Adams et al., 1979), Rotheram-Borus (1989), in an investigation of the identity status of 330 African American, Caucasian, Filipino, Hispanic, and Native American high school students, found that there was little relation between identity status and ethnicity. Her only significant finding was that Caucasian students in the 1 Ith and 12th grades were more likely than the other students to have high moratorium scores on the OM-EIS. The students, however, were not classified in the moratorium status at a rate different from others. Rotheram-Borus argued that because of the school's high degree of integration, ethnic differences in the stages of identity did not exist.

In summary, Hauser (1972), Abraham (1986), Streitmatter (1988), and

Markstrom-Adams and Adams (1995) reported that the minority students in their studies were more foreclosed that their Caucasian peers. Abraham (1983) and Rotherham-Borus (1989) reported no significant differences in foreclosure (see Appendix A for a comparison of these studies). The inconsistency of results in these studies may be due to the confounding of a number of ecological variables. Some researchers (e.g., Hauser, 1972; Streitmatter, 1988) failed to control for SES. Abraham (1986) and Markstrom Adams and Adams (1995) used parental education as a measure of SES. Despite their effort to control for SES, they found that African American and Hispanic adolescents






9


scored significantly higher on foreclosure than the Caucasian adolescents in their sample, suggesting that either parental education is an inadequate measure of SES or that other factors account for the ethnic group differences in foreclosure.

Some research findings suggest other ecologicalvariables as factors that may account for the greater foreclosure among minority group members. Spencer and Markstrom-Adams (1990) pointed out that although both the Caucasian and African American males in Hauser's (1972) study were from low SES families, the African American males were more likely to live with single parents than were the Caucasian males. Hauser noted that the African American students whose fathers were absent were more foreclosed than the students whose fathers were present. Spencer and MarkstromAdams concluded that both family configuration and SES may confound the results of studies of the identity development of minority adolescents, but they also considered the alternative hypothesis that, because of acculturation or other cultural factors, foreclosure may be adaptive for minority youth. Supporting that conclusion, Markstrom-Adams and Adams (1995) proposed that African American and Hispanic families are likely to define maturity in terms of support of family values in contrast to Caucasian families who may define maturity in terms of independence. That is, autonomous choice may be the defining theme for the development of Caucasian identity, whereas acceptance of family values may be the defining theme for African American and Hispanic adolescents.

Purpose of the Study

In the comparison of minority and majority adolescent ego-identity development, conflicting results have been found. Some of the research suggests that minority adolescents are more likely to be foreclosed, but other studies have failed to support that





10


conclusion. Furthermore, the studies have for the most part failed to include important ecological variables that may account for the conflicting results. To account for the development of identity of ethnic minorities, the role of all of these variables must be examined. The purpose of this study, therefore, was to investigate whether the ecological variables of SES, acculturation, family characteristics, and gender are related to the egoidentity development of ethnic minority and Caucasian adolescents. Each of these ecological variables and their potential relationship to identity development will be discussed in the sections that follow.

Ecological Variables Related to Identity Status

Identity status has been found to be related to a number of ecological variables (Marcia, 1993). In response to findings that ethnic minority adolescents tend toward lss progressive identity statuses than their Caucasian peers, several influences on adolescent development have been offered as explanation: demographic variables, home culture variables, and the presence of an acculturation process in ethnic minority members. Demographic Variables

Differences in identity development may be explained, in part, by the

demographic variables gender, SES, and family configuration. Erikson (1963) argued that males and females experience identity development differently. Limited evidence suggests that identity development in low SES environments is not as progressive as in higher SES environments (Markstrom-Adams & Adams, 1995). As for family configuration, relations have been found between the construct and identity development, although the direction of the relationship is far from clear (Marcia, 1993).








Gender. In a discussion of male and female psychosocial differences, Erikson

(1968) attributed their differential development to socialization. That is, Erikson believed that because males possess a dominant role over women in society they are more likely to assert themselves than women. Women, on the other hand, due to their more submissive cultural role, are more likely to engage in activities that are passive and harmonious. According to Erikson, the male develops his sense of identity by competition, whereas the female resolves identity questions through communal, interdependent means.

The first studies to investigate the differences that Erikson proposed supported his claim (e.g., Marcia, 1966, 1976; Schenkel & Marcia, 1972). For example, Hodgson and Fischer (1979), in a study of 100 late adolescents' identity development, found that males tend to resolve identity issues via competence and knowledge. Males achieve competence by pursuing an education and obtaining a career, and knowledge by coming to understand the world around them. Females, however, were found to resolve identity issues through different mechanisms. Hodgson and Fischer stated that for females the response to "Who am I?" revolves around what they are in relation to others. Because of this foundational research, it was long accepted that gender differences exist in identity development. Matteson (1993) pointed out, however, that social changes in the past few decades may necessitate a reexamination of the differences between the identity development of men and women.

Archer (1989a), in a report of three studies, examined the identity processes of experimentation and commitment, the domains of identity, and the timing of identity changes for male and female students in early to late adolescence. Overall, Archer concluded that males and females undergo identity development in similar ways. As for






12


the identity processes, in two of the studies, males and females differed only in the foreclosure status; males were more likely to be foreclosed than females. In the third study no differences were found between males and females in the identity statuses.

Archer (1989a) found, however, that males were more likely than females to be diffused with regard to the family roles domain, whereas females were more likely to be classified in the foreclosure, moratorium, and achievement statuses. No differences were detected in the second study. In the third study, Archer found that males were more likely to be classified in the foreclosure status, and females in the diffusion status with regard to political ideology.

As for gender differences in the timing of identity development, Archer (1989a) found in two of the studies that no differences existed. In the third study, however, a Grade x Gender interaction was found. Females were more likely than males to be classified in the moratorium status in grades 6, 10, and 12. For identity achievement, males were more likely than females to be so at grade 12.

Overall, Archer (1989a) concluded that the differences between the genders were slight. On the basis of her studies, she argued that the assumption that males and females experience identity development differently should be discarded. Matteson (1993) proposed much the same argument as Archer. That is, Matteson pointed out that the relationships found between identity status and other psychological constructs for males were also found for females. For example, men and women classified in the moratorium status reported the highest levels of anxiety, and males and females in foreclosure were more likely than those in other statuses to defer to authority (Marcia & Friedman, 1970).





13


Other studies, however, have found difference between genders in identity

development. For example, Streitmatter (1987), in a study of 265 early adolescents, found that females scored significantly higher than males in total moratorium and ideological moratorium scales. Streitmatter concluded that such a difference indicated that females were "more involved in the questioning and searching process than males" (p. 184).

In a similar study, Streitmatter (1988) found that females were significantly more often found in ideological and interpersonal achievement and ideological moratorium than males. Given that identity achievement and moratorium are considered more advanced identity statuses than foreclosure and diffusion, Streitmatter concluded that females are considered more mature in their identity development than are males at the same age. Streitmatter (1988) also found a statistically significant Gender x Ethnicity interaction. Minority males were more likely than their non-minority peers to be in the moratorium status, a finding that Streitmatter cited as evidence that minority males are more actively seeking answers to their identity problems. Minority males were, however, more likely to be in the diffusion status than their peers. This finding is noteworthy, as it runs counter to the research that has found that minority males are more often in foreclosure than their Caucasian counterparts.

Markstrom-Adams and Adams (1995) found that females were not as likely as males to be classified in the diffusion status for ideological and interpersonal identity. A Gender x Grade interaction demonstrated that older females were less likely than their younger peers to exhibit lower diffusion scores. Markstrom-Adams and Adams, unlike Streitmatter (1988), did not find any Gender x Ethnicity interactions.





14


In summary, early theory and research of gender differences in identity

development pointed to clear differences (Erikson, 1968; Marcia, 1966, 1976). Males were perceived as focused on ideological identity development. Women were seen as focused on interpersonal identity development. Later studies (Archer, 1989a; MarkstromAdams & Adams, 1995; Streitmatter, 1988) suggest that differences between males and females in identity development, when present, are subtle.

Socioeconomic status. Minority adolescents are more likely to come from lowincome families than their Caucasian peers (Spencer & Dornbusch, 1990). Although few studies have considered the impact of low SES on identity development, limited evidence suggests a relation between the constructs. For example, Markstrom-Adams and Adams (1995) found that higher paternal and maternal education was related to lower ideological foreclosure (the correlation with maternal education was -.24 and the correlation with paternal education was -.22). No relationship was found between parents' education and interpersonal identity.

Berry (1980) hypothesized that minority group members with low SES are more likely to adopt an assimilation or separation style of acculturation rather than an accommodation style. Because of limited resources, these individuals tend to see the world as difficult to master. They may also avoid identification with the majority culture or reject their own culture in favor of the majority culture. Negy and Woods (1992), in a study of 339 Mexican American college students, found that acculturation was moderately related to SES (r = .44, 1? = .0001). That is, the greater the identification witn the majority culture, the higher the SES of the student tended to be. Negy and Woods, however, did not determine if those who identified with the majority culture also





15


identified with the minority culture. As a result, it was impossible to differentiate between assimilated and the accommodated respondents.

As a great number of minority children come from low SES backgrounds, many do not have access to the same economic resources as their higher SES peers and, as a result, may adopt a marginalized or rejected acculturation orientation. This puts these adolescents at greater risk of concluding that the expectations of society are incompatible with their own, which may result in the formation of a weaker identity than higher SES peers. Research is needed to investigate Berry's (1980) hypothesis and to determine whether acculturation status mediates the relationship between SES and adolescents' identity status.

Family configuration. African American and Hispanic minority adolescents are more likely to grow up in a single parent home than are their Caucasian peers (U. S. Bureau of the Census, 1994). Family configuration differences may be related to patterns of identity development among cultural groups, but research on the relationship between family configuration and identity has yielded inconsistent results. For example, Jordan (1970) found that male college students from divorced families were more likely to be identity diffused than were students from intact families. Other studies, however, do not support the claim that living in an intact family is related to progressive identity development. Grossman, Shea, and Adams (1980), in a study of 294 college students, found that males from divorced families possessed higher identity achievement scores than males in intact families and females in general. In a study of 467 students from ages 12 to 18, Jones and Streitmatter (1987) found that male and female high school students from intact families were more foreclosed on the ideological and interpersonal scales of






16


the EOM-EIS than peers living with a single parent. In their study of 80 Caucasian female high school students, using Marcia's (1966) Identity Status Interview, St. Clair and Day (1979) found that two-thirds of the 80 senior high school female adolescents in their study who were in the achieved status came from homes disrupted by death or divorce. In a study of 265 seventh graders, Streitmatter (1987) found that early adolescents from intact families were more likely to be classified in the foreclosure status than their peers from non-intact families. She also found that males seemed more affected by family disruption than females. Males were more likely to be classified as diffused than females. Streitmatter suggested that disruption might facilitate identity development in females, perhaps shortening their moratorium.

Jones and Streitmatter's (1987), St. Clair and Day's (1979), and Streitmatter's (1987) results may have been affected by their failure to discriminate between single parent homes due to divorce or death. Recognizing that research has shown that parentchild relations differ for children with a deceased parent compared to children whose parents are divorced (e.g., Hetherington, 1972), Streitmatter analyzed her data excluding children with a deceased parent. Finding only one change from the analysis with children with either a deceased parent or divorced parents, Streitmatter reported the data for the combined groups. She found a Diffusion x Gender x Family Status interaction but attributed the result to a loss in sample size.

St. Clair and Day (1979) suggested that the impact of family disruption may be different for male and female adolescents, and Streitmatter's (1987) study provides empirical support for the need to study the differential impact of family disruption on males and females. Streitmatter found that adolescents from intact families had higher






17


total scores (combined ideological and interpersonal domain) on foreclosure than adolescents from nonintact families. From post hoc analyses she found that the family status effect held for ideological foreclosure but not for interpersonal foreclosure, and she found a significant Family Status x Gender interaction indicating that female adolescents from intact homes had lower interpersonal foreclosure scores than female students from disrupted homes, whereas male students from intact homes had higher foreclosure scores than males in non-intact homes. She also found a significant Family Status x Gender interaction for diffusion. Male students from intact homes had lower ideological diffusion scores than male students from non-intact homes, and females from intact homes had higher ideological diffusion scores than females with only one parent in the home. She also found that the overall relationship between family status and identity development was greater for male students than for female students. She speculated that given the likelihood that most of these families were headed by women, the lower foreclosure scores may reflect the male adolescents' rejection of failed family values, whereas their higher diffusion scores may indicate that they were unable or unwilling to choose an identity. Streitmatter further speculated that family disruption may have facilitated identity development by thrusting the male adolescents into diffusion earlier than their peers in intact homes.

In sum, research has demonstrated that family configuration is related to identity development. However, the kind of household (i.e., intact, non-intact), as well as the gender of the child, plays a role in relationship between family configuration and identity development. Therefore, research that divides family configuration into separate categories for (a) single parent as the result of death, (b) single parent due to divorce, (c)





18


intact families, (d) single parent never married, and (e) reconstituted families and that analyzes the data separately for males and females and controls for SES is needed to clarify the relationship between family configuration and identity development. Family Characteristics

Differences in the family environment of minority and majority adolescents may explain, in part, the different identity development of the ethnic groups. Marcia (1993) argued that individuals classified in the high identity statuses often come from families where ideological and interpersonal differentiation between the child and parents is supported. Those classified in the lower identity statuses, in contrast, tend to come from families that either encourage conformity to family values or from families in which the parents distance themselves from the ideological and interpersonal development of the child. Significant differences have been found between minority cultures and the majority culture in parenting style (Holtzman, 1982) and familism (Mindel, 1980).

Parenting style. Baumrind (1971, 1991) identified three forms of parenting: (a) authoritarian whereby parents use a controlling and disciplinarian style to raise their children, (b) authoritative whereby parents encourage independence but a respect for rules in their children and, (c) laissez-faire whereby parents do not place limitations or hold expectations for their children's behavior. Maccoby and Martin (1983) further divided the laissez-faire style of parenting into two distinct styles: (a) permissiveindifferent, whereby parents are not involved in their children's lives and (b) permissiveindulgent, whereby parents give in to their children's demands and do not place restrictions on their behavior.





19


Parenting style is significantly related to SES (Hoff-Ginsberg & Tardif, 1995). Parents in the lower SES tend to use an authoritarian style of parenting, whereas middle SES parents tend to use an authoritative style of parenting (Hoff-Ginsberg & Tardif, 1995). As minority adolescents are more likely to have low-income families than their Caucasian peers (Spencer & Dornbusch, 1990), minority adolescents tend to be raised with an authoritarian style of parenting. Caucasian adolescents, in contrast, tend to be raised with an authoritative style. This difference may be related to patterns of identity development among cultural groups. For example, Waterman (1993) hypothesized that, because of the controlling power authoritarian-style parents exert, their children are likely to begin identity development in foreclosure. In contrast, children of permissive parents tend toward identity diffusion because of their parents' lack of expectations. Waterman was less certain about the likely outcomes of authoritative parenting, but he seemed to suggest that the support and caring these parents give their children might enable them to delay commitment, thus enabling them to experience moratorium.

Research supports Waterman's belief that an authoritative parenting style is related to moratorium. For example, Quintana and Lapsley (1990), in a study of 101 undergraduates, found an indirect relationship between the perceived parenting styles and identity statuses of adolescents mediated by individuation from parents. According to their results, a parenting style that encouraged independence tended to produce positive individuation from the family for adolescents. This type of attachment gave the adolescents the necessary freedom to differentiate from their families and succeed in forming high status identities. Controlling parenting styles, however, resulted in a





20


negative individuation from families, and these adolescents were prone to develop lower level identities than their peers from less restrictive homes.

Direct relationships have also been found between parenting style and identity development. La Voie (1976) studied 120 students in grades 10, 11, and 12 and found high identity males came from homes with less regulation and control and more praise by parents than their lower identity status peers. Females with a high identity status came from homes with less maternal control and greater encouragement to discuss personal problems with their parents than did their low identity peers. Quintana and Lapsley (1987) found similar results. In a study of 101 undergraduates, they found that the children of families that encouraged communication and were not controlling were better prepared to deal with their ensuing crises. Quintana and Lapsley presumed that their family background prepared them to freely explore identity alternatives. Adolescents from controlling homes were more likely to be classified in the lower identity statuses than were their peers from less controlling homes. However, in two studies of seventhand eleventh-grade students, Enright, Lapsley, Drivas, and Fehr (1980) found that only fathers' parenting style was related to identity development. In the first study of 262 respondents, Enright et al. found that males with fathers who practiced a democratic style of parenting-analogous to Baumrind's authoritative parenting-tended to be classified in higher identity statuses than their peers with autocratic or permissive fathers. Females whose fathers practiced an autocratic parenting style, however, were more often classified in the higher identity statuses than were peers with democratic or permissive fathers. No relationship was found between mothers' parenting style and identity development. In the second study of 168 respondents, Enright et al. found that both males





21


and females with democratic fathers were more often classified in the higher identity statuses than were their peers with autocratic or permissive fathers. As in the first study, no relationship was found between mothers' parenting style and identity development.

In all, research supports the idea that parenting style is related to identity development. Precisely how the two are related, however, is far from clear. An authoritative parenting style has been consistently demonstrated to be related to high identity status in males. Yet, both authoritative and authoritarian parenting styles have been found to be related to high identity status in females. The statistical relationship between identity status and permissive parenting, however, has not been considered. Fathers' parenting style has been shown to be related to identity development in adolescents. Mothers' parenting style has also been shown to be related to identity development in adolescents, yet not as consistently as a father's parenting style. Therefore, the gender of the adolescent and parent must be taken into account when assessing the relation between parenting style and identity development.

Familism. Markstrom-Adams and Adams (1995) pointed out that identity

differences among the ethnic groups they studied did not occur in interpersonal identity and suggested that the emphasis on family and connectedness among ethnic minority groups fosters the development of interpersonal identity. For example, Mexican Americans are characterized as living in close proximity to many relatives with whom they engage in frequent interaction and mutual support, and these behaviors are believed to be supported by strong emotional ties to family and acceptance of family values (Keefe, 1984). Caucasians, in contrast, are considered to be less likely to have extended family living nearby or to strongly value such relationships. Research, however, has





22


raised questions about the extent of familism among Mexican Americans. Although Farris and Glenn (1976) and Chandler (1979) found greater familism among Mexican American than Anglo American groups, the differences were not large. Also, contrary to the belief that Mexican Americans are high in familism, most of the Mexican Americans disagreed with most of the familism items.

Valenzuela and Dornbusch (1994) defined familism as the structural, attitudinal, and behavioral dimensions of family interactions. The structural aspect of familism is characterized by the presence of a nuclear and extended family, whereas the attitudinal aspect refers to concern for the welfare of the family. The behavioral dimension refers to the amount of contact with family members. Valenzuela and Dornbusch argued that familism can serve as a form of social capital (Coleman, 1988). That is, family connectedness provides children with supportive relationships that enable them to develop positive mental health and to perform well in school. Valenzuela and Dornbusch based their ideas on the work of Velez-Ibanez and Greenburg (1992), who argued that Mexican American children are raised in dense extended families that socialize the children to form strong emotional attachments with a wide variety of relatives who tend to live in nearby neighborhoods, and on research by Velez (as cited in Velez-Ibanez & Greenberg), who found that Mexican American mothers and their relatives had greater contact with infants than Caucasian mothers and relatives and the Mexican American children were in frequent contact with relatives.

Valenzuela and Dornbusch (1994) analyzed surveys of 2,666 Anglo and 492

Mexican students in 1987-88 and found that the Mexican American students were higher in familism than Caucasian students. They also found that for Mexican-origin students





23


neither parent occupation nor familism, when considered separately, was related to students' grades but when parent education was high, familism was significantly related to Mexican origin students' grades. Valenzuela and Dornbusch pnjecture that the patriarchal ideology of the traditional Mexican family may limit female students' development, thereby suggesting the need to examine gender differences in the effect of familism and parent education on development.

Other researchers have found significant differences in familism among African Americans, Hispanics, and Caucasians. Mindel (1980) reported that Hispanics demonstrated the highest degree of familism followed by African Americans and Caucasians, in that order. Farris and Glenn (1976) and Chandler (1979) also reported that Hispanics possess higher degrees of familism than Caucasians. The familism scale used by Chandler, however, consisted of the following four items: (a) Nothing in life is worth the sacrifice of moving away from your parents, (b) When the day comes for a young man to take a job, he should stay near his parents. even if it means losing a good job opportunity, (c) When young people get married, their main loyalty still belongs to their parents, and (d) When you need help of any kind, you can depend only on members of the family to help you out (p. 157).

Using an expanded familism measure consisting of nine items with a sample of 381 Mexican Americans and 163 Anglo Americans, Keefe (1984) obtained results that contradicted those of Chandler (1979) and Farris and Glenn (1976). She found that the Anglo Americans valued family as highly as the Mexican Americans, although the Anglo Americans were more likely to have a very small extended family or none at all, whereas






24


the Mexican Americans had comparatively large extended families who visited frequently.

Further, Keefe (1984) argued that differences between Hispanics and Caucasians in familism is a matter of quality instead of quantity. That is, she found that Hispanic and Caucasians had the same degree of familism, but Hispanic familism was associated with geographic stability, whereas Caucasian familism was associated with geographic mobility. In other words, Anglo and Mexican Americans are similar in their feelings of closeness to their families, but Anglos do not share the Mexican Americans' need for the physical presence of family members.

In a 1988 study, Valenzuela and Dornbusch (as cited in Valenzuela & Dornbusch, 1994) compared Mexican and Anglo adolescents on familistic structure, attitudes, and contact. They found the Mexican-origin adolescents tended to be higher than the Anglo students on all three dimensions of familism. Reguero (1991) investigated the relationship among demographic variables, familism, and ego-identity status with 180 Puerto Rican residents and 107 Puerto Ricans who had immigrated to the U.S. mainland. Using multiple regression analysis, Reguero found that the variables sex, age, parents' education and occupation, and immigrant status, and familism accounted for some of the variance in achievement for all groups. She also found that the variables partially explained the variance in moratorium for males living in Puerto Rico and male and female immigrants, as well as partially explaining variance in foreclosure for female immigrants. The variance in the diffusion status for female respondents, regardless of group, was explained, in part, by the variables.





25


As this review shows, research on the relationship between familism and identity development is far from clear. Though differences in familism have been found among African American, Hispanic, and Caucasian children, research is needed to determine the degree to which these differences affect identity development. Further, it appears that males and females experience familism differently. Gender, therefore, must be taken into account when investigating the relationship between familism and identity development. Acculturation

Phinney (1990) argued that to complete their identity development, members of minority groups must resolve two issues not faced by members of the majority culture:

(a) how to negotiate the different expectations of the two cultures and (b) how to deal with the negative perceptions the majority culture has toward the minority culture. According to Gonzalez and Cauce (1995), the way an adolescent member of a minority culture resolves these conflicts depends on the individual's style of acculturation into the majority lifestyle. On the basis of studies of members of minority groups, Berry (1980) identified four types of acculturation: (a) assimilation, whereby individuals give up their identification with their minority culture and identify with the majority culture, (b) biculturalism, whereby individuals develop positive identifications with their minority culture and the cultural majority, (c) rejection, whereby individuals identify with their minority culture and do not identify with the majority culture, and (d) marginalization, whereby individuals fail to identify with their minority culture or the majority culture.

According to Gonzalez and Cauce (1995) and Peretti and Wilson (1995),

individuals who adopt a bicultural model of acculturation should demonstrate more progressive identity development than those who adopt other models. That is, because





26


bicultural adolescents integrate the expectations of their minority culture and the majority culture and are less vulnerable to negative perceptions of their minority culture because of their identification with it, they should have less difficulty developing the integrated sense of self characteristic of the achieved status.

Although Streitmatter (1988) argued that minority adolescents "feel less comfort in the [majority culture] and as a result are more likely to conform to prescribed values and expectations .. of the adults in their lives" (p. 344), researchers have not examined the relationship between Marcia's (1976) identity statuses and the degree to which minority adolescents feel out of place in the majority culture. The findings of Mehan, Hubbard, and Villanueva (1994), however, support the idea that adolescents with a bicultural acculturation should demonstrate more progressive identity development than those with other acculturation orientations. Mehan et al. investigated the effects of a program to prepare underachieving African American and Hispanic adolescent students for college admission by placing them in classrooms with high achieving peers. Most of the students adopted what Gibson (1988) called accommodation without assimilation (similar to Berry's, 1980, bicultural orientation) toward academic tasks. That is, "they affirm[ed] their cultural identities while at the same time recognize[d] the need to develop certain cultural practices, notably achieving academically, that are acceptable to the mainstream" (Mehan et al., 1994, p. 105). By doing so these students resolved the two conditions set forth by Phinney (1990) for the development of an identity: (a) negotiate the different expectations of the cultures and (b) deal with the negative perceptions the majority culture has toward the minority culture. Also, as Gonzalez and Cauce (1995) and Peretti and Wilson (1995) argued, resolving the expectations of both the minority and





27


majority cultures allows for the development of an integrated sense of self, the hallmark of the achievement status.

In summary, a major task for minority adolescents is the resolution of conflicting expectations of the minority and majority cultures. As minority adolescents confront the identity crisis, they come to see themselves in one of four ways: estranged from the minority culture and solely a part of the majority (assimilated), identified with both the minority and majority culture (bicultured), estranged from the majority culture and solely a part of the minority (rejected), or estranged from both cultures (marginalized). As the development of a higher identity is built on an integrated sense of self (Phinney, 1990), it is plausible that minority adolescents who adopt the bicultural orientation of acculturation are more likely to reach the achieved identity status.

Research Questions

Some studies have reported a relationship between ethnicity and identity, and others have not. None of the studies has adequately controlled for the demographic variables of SES or family structure, which have been found to be related to identity status. Further, as this review shows, cultural and family process variables (i.e., acculturation, familism, and parenting style) may be related to identity status. The purpose of this study was to examine the relationships of these demographic, cultural, and family variables to identity status. The research questions addressed in this study were:

1. Is ethnicity related to identity status?
2. Is ethnicity related to identity status after gender, SES, and family status are
controlled?
3. Is identity status related to familism?
4. Is identity status related to parenting style?
5. Is identity status related to acculturation for African American and Hispanic
adolescents?





28


Significance of the Study

Formation of an identity is a defining aspect of personality development (Marcia, 1993). Marcia's (1966, 1976) extension of Erikson's (1963, 1968) theory of identity and its subsequent revisions (Archer, 1981; Grotevant, Thorbecke, & Meyer, 1982) have highlighted the importance of identity to adolescents' personality development (Marcia, 1993).

Theoretical Significance

Most attempts to apply the Marcia (1993) model of identity development to minority adolescents have found that adolescent ethnic minorities tend to have less progressive identities than do their Caucasian peers (Abraham, 1986; Hauser, 1972; Markstrom-Adams & Adams, 1995; Streitmatter, 1988). A number of factors have been proposed to account for this finding, including failure to control for differences in SES and family constellation as well as differences in acculturation and family values. This study is designed to determine whether these factors need to be incorporated into Marcia's model to provide an adequate conception of minority adolescent identity development.

Practical Significance

Identity research has revealed that the more progressive identity statuses

(achievement and moratorium) are related to more adaptive characteristics of adulthood than the less progressive statuses (foreclosure and diffusion) (Archer, 1989b). For example, individuals in the more progressive statuses have demonstrated higher levels of moral reasoning, autonomy, and internal locus of control (Josselson, 1994; Marcia, 1993). As a result, researchers have recommended the development of intervention





29


strategies to foster successful resolution of the identity crisis (Josselson, 1994; RotheramBorus & Wyche, 1994; Waterman, 1994).

It is not clear whether such intervention strategies are warranted for adolescents from minority groups. As Erikson (1963) noted, culture affects the development of identity. That is, cultural expectations and practices encourage and limit adolescents' opportunity for exploration and commitment in the various identity domains. Given that researchers have found that African American and Hispanic adolescents are more foreclosed than their Caucasian peers, Spencer and Markstrom-Adams (1990) stated that the foreclosure status may be adaptive in certain cultures: "Exploration of identity may not be desirable for such youth when social and ideological roles are clearly defined by the community" (p. 298). This study was designed to determine whether the relationship between ethnicity and identity development is attributable to SES, family configuration, and family values.

It may be inappropriate, however, to conclude that identity intervention is unnecessary because of differences in cultural expectations. Markstrom-Adams and Spencer (1994) argued that despite minority cultural support many minority adolescents are restricted in their ability to explore identity alternatives because of the limitations in their social environment. That is, a significant portion of minority adolescents grow up in such impoverished, discriminatory environments that they cannot take advantage of the role models available that are necessary for adequate identity exploration. Consequently, it is important to determine whether the foreclosure of minority adolescents is attributable to economic variables or to cultural values.














CHAPTER 2
METHOD

Research Participants

Three hundred students from a community college and a university in Florida were recruited to participate in this study. Respondents included 82 African American students (30 men and 52 women), 120 Hispanic students (50 men and 70 women), and 98 Caucasian (40 men and 58 women) late adolescents with a mean age of 20.8 years.

Instruments

Identity Status

Identity status was assessed using the Extended Objective Measure of Ego

Identity Status III (EOM-EIS III, Forbes & Ashton, 1996). The EOM-EIS III is a 64-item questionnaire. Respondents are instructed to indicate the degree to which an item is reflective of themselves using a 6-point Likert scale, ranging from strongly disagree to strongly agree. The scale assesses ideological identity in the areas of occupation, politics, religion, and philosophical lifestyle and interpersonal identity in the areas of gender roles, dating, friendship, and recreation. Each of the eight subdomains is measured by eight items with two items reflecting each of the four identity statuses: achievement, moratorium, foreclosure, and diffusion. Forbes and Ashton obtained the following Cronbach's alpha coefficients for scores on the four identity statuses: ideological achievement = .57; moratorium = .68; foreclosure = .64; diffusion = .53; interpersonal achievement = .62; moratorium = .63; foreclosure = .77; diffusion = .51.


30





31


Acculturation

Acculturation orientation was measured with an adaptation of a questionnaire developed by BerryhTrimble, and e~(9) to assess the four types of acculturation-assimilation, integration, separation, and marginality. The questionnaire consists of 24 items, with an item representing each acculturation type for each of the following domains: occupation, dating, friendship, life-style, media, and language. Responses were made on a 5-point Likert scale, ranging from strongly disagree to strongly agree.

Parenting Style

Parenting style was assessed using the two forms of Buri's (1991) Parental

Authority Questionnaire (PAQ). The PAQ measures parental authority of mothers and fathers. Each scale is a 30-item questionnaire designed to measure the degree to which an individual believes his or her parent(s) exhibited permissive, authoritarian, and authoritative parenting styles. Responses are made to the items on a 5-point Likert scale. With a sample of 61 psychology students who completed the questionnaire twice over a 2-week period, Buri obtained the following test-retest reliability coefficients: mother permissiveness = .75; authoritarianism = .85; authoritativeness = 82; father permissiveness = .74; authoritarianism = .87; authoritativeness = .85.

Buri (1991) obtained evidence for the construct validity of the scale in a study of 127 undergraduate psychology students. His results were consistent with predictions from Baumrind's (1971) theory: mother's authoritarianism items were negatively correlated with mother's authoritativeness and permissiveness items (r = -.38, P < .0005; r = -.48, P < .0005). Mother's authoritativeness and mother's permissiveness were not significantly





32


correlated (r = .07, p > .10). Father's authoritarianism was negatively correlated with father's authoritativeness and father's permissiveness (r = -.50, R < .05; r = -.52, p2 < .05). Father's authoritativeness and father's permissiveness items were not significantly correlated (r = .12, p > .10). In the same study, Buri reported the following data on criterion-related validity: The authoritative subscales were positively related to parental nurturance (mother r = .56, p < .0005; father r = .68, p < .0005), whereas the authoritarian subscale was negatively related to parental nurturance (mother r = -.36, p < .0005; father r = .-53, p < .0005). The permissiveness subscales were not significantly related to parental nurturance (mother r = .04, p < .10; father r = .13, p < .10), respectively.

In a study of 69 undergraduate psychology students, Buri (1991) assessed the

relationship between social desirability and respondents' answers to the PAQ subscales. None of the subscales was significantly related to the Marlowe-Crowne Social Desirability Scale.

Family Configuration

Family configuration was assessed by the respondents' self-report. Participants were asked to indicate the category that identified their family type on the questionniare. Categories included dual parent household without stepparents; dual parent household with stepparents; single-parent household-parents divorced; single-parent householdparent never married, single-parent household-parent deceased, and an Other category. Familism

Familism was assessed in each of the three domains of the construct. Familism behavior was measured using a single item taken from the work of Valenzuela and Dornbusch (1994) that asks respondents to indicate on a 6-point Likert scale "how often





33


[respondents] saw or talked on the telephone with adult relatives who do not live with them" (p. 24). Familism structure was also measured with a single item on a 6-point scale used by Valenzuela and Dornbusch. It asked the number of relatives that live within an hour's drive of the respondent. Familism attitudes were measured using a scale developed by Keefe (1984). The measure is a 9-item, 5-point Likert scale. Keefe (1984) obtained an internal consistency coefficient of.68 for the items in a study of 381 Mexican American and 163 Caucasians.

Socioeconomic Status (SES)

The SES of respondents' parents was assessed using lollinhshead's (1975) Four Factor Index of Social Status (FFISS). The factors of the scale include gender, occupation, education, and marital status. Gender, however, is not included in the calculation of social status, though it can be used with males or females. Occupation level is assessed on a 9-point scale of increasing complexity as determined by the 1970 United States Census. The educational factor is measured using a 7-point scale that ranges from less than seventh grade to graduate professional training. Marital status is assessed by first assigning respondents to one of the following classifications: unmarried individual (male or female), head of a household (male or female), or family in which both husband and wife are gainfully employed. Then social status is calculated by multiplying the scores of the occupation and education factors by factor weights. Scores range from 8 to 66 with higher scores indicating higher social status. When social status is measured for families with two employed persons, the occupation and education scores of both persons are considered in the calculation. The Four Factor Index is based on the belief that social status is a "multidimensional aggregate" (Gottfried, 1985, p. 92). That is, the four





34


factors-gender, occupation, education, and marital status-are combined into a single index rather than considered as separate factors. The FFISS has been described as a highly reliable and valid measure of social status (see Hollingshead, 1975). Gottfried found that the FFISS was more highly correlated with children's developmental outcomes in a longitudinal (30 months) study of 130 young children and their parents than were two other SES indices, the Siegel Prestige Scale and the Revised Duncan Socioeconomic Index. In addition, Gottfried pointed out that the Hollingshead Four Factor Index can be used to estimate the SES of unmarried heads of households as well as families.

Procedures

Each participant signed the consent form (see Appendix B) and anonymously

completed the instruments that consisted of the demographic questionnaire, EOM-EIS III, PAQ, acculturation, and familism questionnaires. Data were collected from summer, 1997 to spring, 1999. Respondents were told not to answer any items that they did not wish to and to ask questions at any time. Administration time took between 45 min to 1 hr.

Research Questions

Research Question #1

The first set of research hypotheses in this study was designed to determine if ethnic identification was related to identity status classification. Respondents' scores on the EOM-EIS were used as the dependent variable in the model. The independent variable was ethnic identification.





35


Research Question #2

The second group of research hypotheses was designed to determine if ethnic

identification was related to identity status when the demographic variables-SES, gender, family status-were controlled. The dependent variable was the EOM-EIS scores. Ethnic identification was the independent variable and SES, gender, and family status responses were covariates.

Research Question #3

The next set of hypotheses considered the relation of familism to identity status for each of the three ethnic groups. Separate analyses were conducted on the variables to determine their relationship with identity status. In both sets of analyses respondents' EOM-EIS scores were the dependent variable; whereas scores on the familism measures served as predictor variables and ethnic identity was a covariate. Research Question #4

The next set of hypotheses considered the relation of parenting style to identity status for each of the three ethnic groups. Separate analyses were conducted on the variables to determine their relationship with identity status. In both sets of analyses respondents' EOM-EIS scores were the dependent variable; whereas scores on the PAQ served as predictor variables and ethnic identity was a covariate. Research Question #5

This group of hypotheses considered the relation between the different forms of acculturation orientation available to African American and Hispanic adolescents and identity status.














CHAPTER 3
RESULTS

This study was designed to investigate the relation between etmc identification and identity status before and after controlling for the effect of theoretically significant demographic variables-gender, SES, and family status. It was also of interest to examii:e the relationships among parenting style, familism, and identity status classification. The final question of interest was whether acculturation orientation is related to identity status classification. This chapter presents a summary of the statistical findings of this study including (a) the summary statistics of demographic variables (age, gender, SES, family status), parenting style, familism, acculturation orientation, and identity status classification by ethnic group and (b) results of the analyses of the relationships among the variables of interest.

Summary Statistics

Demographic Variables

Of the 300 respondents who participated in this study, 82 identified themselves as African American, 120 as Hispanic, and 98 as Caucasian. All participants in this study were between the ages of 18 and 22. The mean age of African American respondents was 20.8 years. The mean ages of Hispanic and Caucasian respondents were 20.8 and 20.6 years, respectively. Forty percent of the respondents were men (n = 120); 60% of the respondents were women (n = 180). Of African American respondents, 62 (76%)



36





37


Table 1. Family Status Percentages by Ethnicity African American Hispanic Caucasian

Status % % %


Dual parent- 44 66 63
No step parent

Dual parent- 19 6 17
Stepparents

Single parent- 12 5 0
Never married

Single parent- 17 19 16
Parent divorced

Single parent- 4 3 2
Parent deceased

Other 4 1 2





identified themselves as women, and 20 (24%) identified themselves as men. Sixty-five Hispanic (54%) and 71(72%) Caucasian respondents identified themselves as womem; 55 (46%) Hispanic and 27 (28%) Caucasian respondents identified themselves as men.

Table 1 displays the percentage of respondents in each of the different family statuses by ethnic group. A chi-square analysis of the relationship between statuses by ethnic group resulted in a non-significant relationship.

Table 2 lists the means and standard deviations of respondents' scores on the Four Factor Index of Social Status (Hollingshead, 1975). An ANOVA indicated a significant





38


Table 2. Means and Standard Deviations of Socioeconomic Status by Ethnicity


African American Hispanic Caucasian

M S M S M


42.35. 12.17 43.08. 12.74 49.88b 8.96 Note. Means with different subscripts are significantly different.



difference among the means, F(2, 291) = 12.54, p < .0001. Post-hoc comparisons using the Tukey method revealed that Caucasian SES scores were significantly different from both African American and Hispanic SES scores (p < .05). The SES scores of African American and Hispanic respondents did not differ. Parenting Style

The means and standard deviations of the different parenting styles are listed by ethnicity in Table 3. One-way analyses of variance indicated significant differences in mean scores for mothers' authoritarian parenting style [F(2, 277) = 16.02, p < .0001] and fathers' authoritarian parenting style [(F(2, 244) = 5.73, R < .0037)].

The Tukey post-hoc comparison of scores indicated that African American and Hispanic adolescents' mean scores were significantly different from the Caucasian mean score on mother's and father's authoritarian parenting style (p < .05). The parenting-style scores of the African American adolescents and Hispanic adolescents did not differ.





39


Table 3. Parenting Style Mean Scores by Ethnicity African American Hispanic Caucasian Style M S M S M S Mother

Authoritative 41.17 7.32 42.55 6.65 43.47 7.57 Authoritarian 36.99. 6.94 35.54a 6.34 31.38b 7.18 Permissive 29.82 5.74 30.73 5.18 31.44 5.56 Father

Authoritative 37.06 6.93 38.40 6.65 39.23 9.58 Authoritarian 36.99a 6.94 35.53a 8.93 34.95b 11.29 Permissive 26.01 8.12 27.25 8.12 28.21 8.29 Note. Means with different subscripts are significantly different within the authoritarian rows.



Familism

The means and standard deviations of the forms of familism are listed by ethnic group in Table 4. Significant differences were found among structure means, F(2, 282)

9.79, p < .0001, for familism. Post-hoc analysis indicated that African American and Hispanic adolescents scores were significantly different from Caucasian scores on familism-structure (p < .05). No significant differences were found between African American and Hispanic adolescents.





40


Table 4. Means and Standard Deviations of Familism Type by Ethnicity African American Hispanic Caucasian Type M S M S M S Structure 20.01. 27.02 11.31, 19.68 6.92b 9.70 Behavior 4.78 1.70 4.83 1.41 4.85 1.52 Attitude 32.94 4.13 33.48 4.24 32.14 3.61 Note. Means with different subscripts are significantly different within the structure row.



Acculturation Orientation

Table 5 presents the means and standard deviations of the acculturation

orientations for African American and Hispanic adolescents. In the comparison of the means of acculturation orientation styles between African American and Hispanic adolescents, significant differences were found for the integration (t(1, 177) = 5.95, 1 < .0001) or marginalized ((1, 180) = 2.23, p < .05) styles. Identity Status Classification

Identity status was determined for each respondent using the method outlined by Adams, Shea, and Fitch (1979). The number of respondents classified in the statuses of each of the domains of identity is listed in Table 6 and Table 7. In the ideological and interpersonal domains the majority of respondents in each ethnic group were classified in a form of identity moratorium. Differences among the groups on identity variables will be presented in the next section.





41


Table 5. Means and Standard Deviations of Acculturation Orientations for African American and Hispanic Adolescents


African American Hispanic Orientation M S M S


Integration 26.99a 4.27 30.64b 3.89 Accommodation 10.56 3.69 9.90 3.75 Separated 14.01 4.01 12.77 5.06 Marginalized 13.59. 3.78 14.82b 3.62 Note. Row means with different subscripts are significantly different.





Table 6. Ideological Identity Status Percentages by Ethnicity African American Hispanic Caucasian Status (n=82) (n = 120) (n = 98)


Achieved 11 8 13 Moratorium 4 13 9 Foreclosure 16 15 8 Diffusion 2 5 14 Moratorium- 64 50 49
undifferentiated

Transition 4 7 6 Note. Column totals may not equal 100% due to rounding.





42


Table 7. Interpersonal Identity Status Percentages by Ethnicity


African American Hispanic Caucasian Status (n=82) (n = 120) (n = 98)


Achieved 7 12 10 Moratorium 13 11 12


Foreclosure 11 11 6 Diffusion 11 7 10 Moratorium- 49 51 52 undifferentiated

Transition 7 9 7 Note. Column totals may not equal 100% due to rounding.



Results of Statistical Analyses

For each question, eight analyses were conducted. A .05 familywise error rate was used to judge significance. However, in reporting analyses, results that were significant at a conventional .05 Type I error rate are noted. The research questions for this study were

1. Is ethnicity related to identity status?
2. Is ethnicity related to identity status when gender, SES, and family status are
controlled?
3. Is identity status related to parenting style when ethnicity is controlled?
4. Is identity status related to familism when ethnicity is controlled?
5. Is identity status related to acculturation for African American and Hispanic
adolescents?

Question #1

The means and standard deviations for the status analyses are reported by ethnic group in Table 8. To determine if ethnicity was related to identity status a one-way





43


Table 8. Mean Scores and Standard Deviations of Identity Statuses African American Hispanic Caucasian M S M S M S Ideological

Achieved 23.28 3.76 23.46 3.52 23.01 3.25 Moratorium 25.97 3.91 26.74 3.24 26.51 3.54 Foreclosed 20.44 4.74 22.53 4.72 21.63 3.95 Diffused 41.64 27.58 32.81 19.75 27.42 9.73 Interpersonal

Achieved 25.28 4.29 25.23 4.12 25.04 3.99 Moratorium 25.23 4.26 25.37 3.61 25.52 4.21 Foreclosure 25.60 3.89 26.21 3.88 24.70 3.63 Diffused 25.39 4.24 25.74 3.62 24.97 3.52





ANOVA was conducted on the means of each of the identity domain scores for African American, Hispanic, and Caucasian adolescents. When the Bonferroni-Holm procedure was used to control for the fact that ethnic group differences were tested on eight variables, none of the F statistics was significant. When a conventional .05 alpha level was used, ethnic differences among identity status scores were found for one statusdiffused, ideological, F(2, 287)= 4.13, 1 = .017.





44


Question #2

The general linear model was used to test ethnic group differences with the demographic variables controlled (gender, SES, family status). No significant ethnic group differences emerged when Bonferroni corrections were made for multiple comparisons, With a conventional .05 alpha level, ideological diffusion was most closely related to ethnicity, F(2, 256) = 3.63, p = .028. Of the ethnic groups, Caucasian adolescents reported the highest means on ideological diffusion.

In the research literature, gener.SES and family status have been identified as demographic variables related to identity development. However, when these variables were entered as covariates in the analysis of the relationship between identity status and ethnicity, and when corrections were made to the p value for multiple analyses, none of the covariates was found to be significantly related to the identity status domains (see Tables 9 and 10). The covariates with p values less than .10 were gender in ideological moratorium (p = .042), ideological foreclosure (p = 0.09), interpersonal achieved (p = 0.008), and interpersonal foreclosure (p = .026), and SES in ideological moratorium (P .042) and interpersonal foreclosure (p = .072). Question #3

The purpose of the third research question was to determine whether identity status variables were related to familism with ethnicity controlled. The covaate, ethnicity, was not found to be related to identity status. With Bonferroni corrections for multiple comparisons, no significant differences were found between ideological and interpersonal identity status and forms of familism, with ethnicity controlled. Results are





45


Table 9. ANCOVA Results for Ideological Identity Statuses


Source ss df Mean Square F p


Achieved

Race 47.51 2 23.75 0.72 0.49 Gender 9.66 1 9.66 0.29 0.59 SES 33.24 1 33.24 1.01 0.32 Family 125.40 5 25.08 0.76 0.58 Moratorium

Race 102.46 2 51.23 1.47 0.23 Gender 145.94 1 145.94 4.18 0.04 SES 29.63 1 29.63 0.85 0.36 Family 175.87 5 35.17 1.01 0.41 Foreclosure

Race 153.62 2 76.81 2.63 0.07 Gender 82.88 1 82.88 2.84 0.09 SES 0.99 1 0.99 0.03 0.85 Family 99.14 5 19.83 0.68 0.64 Diffusion

Race 258.75 2 129.38 3.63 0.28 Gender 88.73 1 88.73 2.49 0.12 SES 0.39 1 0.39 0.01 0.92 Family 180.64 5 36.18 1.01 0.41





46


Table 10. ANCOVA Results for Interpersonal Identity Statuses


Source ss df Mean Square F p


Achieved

Race 178.59 2 89.30 2.87 0.06 Gender 222.65 1 222.65 7.15 0.01 SES 12.35 1 12.35 0.40 0.53 Family 250.54 5 50.11 1.61 0.16 Moratorium

Race 2.27 2 1.18 0.04 0.96 Gender 0.08 1 0.08 0.00 0.96 SES 25.50 1 25.50 0.85 0.36 Family 32.47 5 6.49 0.22 0.96 Foreclosure

Race 44.83 2 22.42 0.73 0.49 Gender 154.18 1 154.18 4.99 0.03 SES 100.95 1 100.95 3.27 0.07 Family 113.68 5 22.73 0.74 0.60 Diffusion

Race 34.87 2 17.44 0.66 0.52 Gender 70.25 1 70.25 2.67 0.10 SES 0.99 1 0.99 0.04 0.85 Family 98.28 5 19.66 0.75 0.59





47


listed in Tables 11 (ideological identity statuses) and 12 (interpersonal identity status). No significant interactions between familism and ethnicity were found.



Table 11. Summary of GLM Analyses for Relationship between Ideological Identity Status and Familism, with Ethnicity Controlled


Source ss df Mean Square F 1 Achieved

Family structure 2.87 1 2.87 0.09 0.77 Family behavior 42.73 1 42.73 1.29 0.26 Family attitude 66.50 1 66.50 2.07 0.15 Moratorium

Family structure 41.90 1 41.90 1.14 0.29 Family behavior 105.16 1 105.16 1.03 0.32 Family attitude 49.46 1 49.46 1.39 0.24 Foreclosure

Family structure 103.91 1 103.91 3.51 0.06 Family behavior 45.70 1 45.70 1.53 0.22 Family attitude 169.17 1 169.17 5.75 0.02 Diffusion

Family structure 59.95 1 59.95 1.74 0.19 Family behavior 83.85 1 83.85 2.41 0.12 Family attitude 27.06 1 27.06 0.79 0.38





48


Table 12. Summary of GLM Analyses for Relationship between Interpersonal Identity Status and Familism, with Ethnicity Controlled


Source ss df Mean Square F p


Achieved

Family structure 0.41 1 0.41 0.01 0.91 Family behavior 15.95 1 15.95 0.48 0.49 Family attitude 47.39 1 47.39 2.48 0.22 Moratorium

Family structure 26.93 1 26.93 0.90 0.34 Family behavior 134.58 1 134.58 4.97 0.03 Family attitude 25.34 1 25.34 0.87 0.35 Foreclosure

Family structure 28.06 1 28.06 0.88 0.35 Family behavior 93.86 1 93.86 2.83 0.09 Family attitude 62.94 1 62.94 1.94 0.17 Diffusion

Family structure 3.08 1 3.08 0.12 0.73 Family behavior 3.94 1 3.94 0.14 0.71 Family attitude 99.08 1 99.08 3.85 0.05








Question #4

The aim of the fourth research question was to determine whether identity status variables were related to parenting style scores after controlling for ethnicity. The covariate, ethnicity, was not found to be related to identity status.

Mothers' parenting style. The F and probability values for the relationship between identity status and parenting style, with ethnicity controlled, are reported in Tables 13 and 14. Inspection of Table 13 reveals the following significant positive relationships between mother's parenting style and ideological identity status: authoritative style with ideological achievement, F (2, 265) = 8.12, p = .005, permissive style with ideological moratorium, F (2, 265) = 18.92, p = .0001, authoritarian style with ideological foreclosure, F (2, 265) = 13.48, p = .0003, and permissive style with ideological diffusion, F (2, 265) = 8.77, p = .003.

Significant relationships were also found between mothers' parenting style and interpersonal identity status with ethnicity controlled (see Table 14): authoritative style with achievement, F (2, 263) = 12.31, p = .0005, permissive style with moratorium, F (, 263) = 12.51, p = .0005, and diffusion, E (2, 263) = 12.20, p = .0006. Each of these relationships was also positive.

Analyses were performed to determine if significant interactions existed between ethnicity and mother's parenting style. The interaction between mother's authoritative style and ethnicity was significant, F (2, 265) = 5.43, p = .004, for the ideological achieved status. As a group, African Americans demonstrated the strongest positive relationship between ideological achievement and mother's authoritative parenting style, followed by Caucasians then Hispanics.





50


Table 13. Summary of GLM Analyses for the Relationship between Ideological Identity Status and Mother's Parenting Style, with Ethnicity Controlled Source ss df Mean Square F p Achieved

Authoritative 257.29 1 257.29 8.12 0.005 Authoritarian 95.78 1 95.78 3.01 0.08 Permissive 90.61 1 90.61 2.87 0.09 Moratorium

Authoritative 0.49 1 0.49 0.01 0.91 Authoritarian 101.88 1 101.88 2.84 0.09 Permissive 622.93 1 622.93 18.92 0.0001 Foreclosure

Authoritative 66.07 1 66.07 2.19 0.14 Authoritarian 389.76 1 389.76 13.48 0.0003 Permissive 15.21 1 15.21 0.95 0.41 Diffusion

Authoritative 296.27 1 296.27 3.34 0.02 Authoritarian 1.07 1 1.07 0.03 0.82 Permissive 290.78 1 290.78 8.77 0.0033





51


Table 14. Summary of GLM Analyses for the Relationship Between Interpersonal Identity Status and Mother's Parenting Style, with Ethnicity Controlled Source ss df Mean Square F R Achieved

Authoritative 368.50 1 368.50 12.31 0.0005 Authoritarian 103.82 1 103.82 3.27 0.07 Permissive 58.93 1 58.93 1.86 0.17 Moratorium

Authoritative 9.30 1 9.30 0.30 0.58 Authoritarian 2.00 1 2.00 0.07 0.80 Permissive 355.88 1 355.88 12.51 0.0005 Foreclosure

Authoritative 195.67 1 195.67 5.99 0.02 Authoritarian 80.85 1 80.85 2.50 0.12 Permissive 99.34 1 99.34 3.07 0.08 Diffusion

Authoritative 127.45 1 127.45 4.95 0.03 Authoritarian 70.63 1 70.63 2.78 0.10 Permissive 304.69 1 304.69 12.20 0.0006





52


Fathers' parenting style. Inspection of Table 15 reveals the following significant relationships between father's parenting style and ideological identity status with ethnicity controlled: authoritative style with foreclosure, F (2, 265) = 14.87, p = .0001, permissive style with diffusion, F (2, 234) = 11.31, P = .0009. Each relationship was positive. The same general findings were obtained for the relationship between interpersonal identity status and father's parenting style (see Table 16). That is, analyses revealed the following significant relationships between father's parenting and interpersonal identity status: authoritative style with foreclosure, F (2, 169) = 15.07, p = .0001, permissive style with diffusion, F (2, 232) = 30.02, p = .0001. Each relationship was positive. As with the ideological identity statuses, no relationship was found between father's parenting style and interpersonal achievement or moratorium. The interaction of father's parenting style and ethnicity was not significant. Question #5

The purpose of this question was to measure the relationship between identity status and forms of acculturation orientation. The results of the CG]Manalyses are presented in Table 17 (ideological identity) and Table 18 (interpersonal identity).

A significant, positive relationship was found between acculturation orientation and ideological identity status with ethnicity controlled-marginalized acculturation with diffusion, F (1, 170) = 12.27, p = .0006. The following positive significant relationships between acculturation orientation and interpersonal Identity were found: marginalized acculturation with diffusion, [(1, 173)=18.32, p = .0001, and separated acculturation foreclosure, F(1, 168) = 24.92, p = .0001. Analyses were conducted to determine whether





53


there was an interaction between acculturation orientation and ethnicity for identity statuses. No significant interactions were found.



Table 15. Summary of GLM Analyses for Relationship between Ideological Identity Status and Father's Parenting Style, with Ethnicity Controlled


Source ss df Mean Square F D Achieved

Authoritative 114.57 1 114.57 3.59 0.06 Authoritarian 179.03 1 179.03 5.65 0.02 Permissive 105.95 1 105.95 3.31 0.07 Moratorium

Authoritative 9.76 1 9.76 0.26 0.61 Authoritarian 9.15 1 9.15 0.24 0.63 Permissive 233.77 1 233.77 6.38 0.01 Foreclosure

Authoritative 440.73 1 440.73 14.87 0.0001 Authoritarian 51.21 1 51.21 1.63 0.20 Permissive 8.21 1 8.21 0.26 0.61 Diffusion

Authoritative 37.85 1 37.85 1.09 0.30 Authoritarian 29.49 1 29.49 0.79 0.38 Permissive 381.99 1 381.99 11.31 0.0009





54


Table 16. Summary of GLM Analyses for Relationship between Interpersonal Identity Status and Father's Parenting Style Controlling for Ethnicity Source ss df Mean Square F 9 Achieved

Authoritative 108.05 1 108.05 3.45 0.06 Authoritarian 196.54 1 196.54 6.37 0.01 Permissive 149.61 1 149.61 4.87 0.03 Moratorium

Authoritative 6.43 1 6.43 0.21 0.65 Authoritarian 0.21 1 0.21 0.01 0.93 Permissive 100.28 1 100.28 3.33 0.07 Foreclosure

Authoritative 482.96 1 482.96 15.07 0.0001 Authoritarian 9.47 1 9.47 0.28 0.60 Permissive 22.00 1 22.00 0.68 0.41 Diffusion

Authoritative 5.25 1 5.25 0.19 0.66 Authoritarian 45.52 1 45.52 1.74 0.19 Permissive 691.72 1 691.72 30.02 0.0001





55


Table 17. Summary of GLM Analyses for Relationship between Ideological Identity Status and Acculturation Orientation, with Ethnicity Controlled


Source ss df Mean Square F 1


Achieved

Accommodation 16.67 1 16.67 0.62 0.43 Separation 17.20 1 17.20 0.64 0.42 Marginalized 13.72 1 13.72 0.51 0.48 Integration 193.13 1 193.13 7.47 0.007 Moratorium

Accommodation 36.02 1 36.02 1.13 0.29 Separation 25.18 1 25.18 0.76 0.38 Marginalized 4.51 1 4.51 0.14 0.71 Integration 6.09 1 6.09 0.19 0.67 Foreclosure

Accommodation 37.63 1 37.63 1.29 0.26 Separation 154.99 1 154.99 5.46 0.02 Marginalized 52.65 1 52.65 1.81 0.18 Integration 29.06 1 29.06 1.00 0.32 Diffusion
Accommodation 50.35 1 50.35 1.68 0.20 Separation 1.68 1 1.68 0.06 0.81 Marginalized 321.88 1 321.88 12.27 0.0006 Integration 14.51 1 14.51 0.49 0.49





56



Table 18. Summary of GLM Analyses for Relationship between Interpersonal Identity Status and Acculturation Orientation with Ethnicity Controlled


Source ss df Mean Square F p


Achieved
Accommodation 0.04 1 0.04 0.00 0.97 Separation 123.06 1 123.06 3.73 0.06 Marginalized 114.39 1 114.39 3.53 0.06 Integration 21.34 1 21.34 0.65 0.42 Moratorium

Accommodation 123.17 1 123.17 4.19 0.04 Separation 73.57 1 73.57 2.35 0.13 Marginalized 86.26 1 86.26 2.86 0.09 Integration 54.35 1 54.35 1.79 0.18 Foreclosure

Accommodation 44.94 1 44.94 1.47 0.23 Separation 689.77 1 689.77 24.92 0.0001 Marginalized 12.92 1 12.92 0.41 0.52 Integration 96.19 1 96.19 3.08 0.08 Diffusion

Accommodation 73.64 1 73.64 2.65 0.11 Separation 6.24 1 6.24 0.22 0.64 Marginalized 472.82 1 472.82 18.32 0.0001 Integration 14.77 1 14.77 0.52 0.47














CHAPTER 4
DISCUSSION

Drawing from Erikson's theory of psychosocial development, Marcia (1966, 1993) operationalized ego-identity development to include four statuses of identityachieved, moratorium, foreclosure, and diffusion. Some researchers (Abraham, 1986; Hauser, 1972; Markstrom-Adams & Adams, 1995; Streitmatter, 1988), using Marcia's model, have found disparate rates of identity development among ethnic groups. Caucasian adolescents were found to have more progressive identities than their peers who were members of ethnic minorities do. Other studies, however, have failed to replicate these findings (Abraham, 1986; Rotheram-Borus, 1989). Explanations of these findings have centered on inadequate measures of the relation of demographic and cultural variables with identity development. It was the purpose of this study, therefore, to address the limitations of previous research to more accurately assess the identity development of minority adolescents. This final chapter, in part, is a discussion of the results obtained from the statistical analyses outlined in chapter 3. Implications and limitations of this study follow. The chapter concludes with recommendations for future research.

Discussion of Results of Statistical Analyses

Researchers who have found differences among ethnic groups in identity

development have largely concluded that the differences between non-Caucasian and Caucasian adolescents in identity development are due to the lack of opportunities for




57





58


social exploration available to minority groups. The proposed barriers to their exploration include SES, gender, family status, familism, and parenting style. The research questions examined in this study were designed to explore the relationship of these variables to identity development.

Ethnicity and Identity Development

The first research question was whether there is a difference among African

American, Caucasian, and Hispanic adolescents in identity development. Specifically, more Caucasian adolescents were expected to be in the progressive identity statuses in comparison to their African American and Hispanic peers. Tests of this hypothesis indicated no significant differences among the respondents' identity status scores.

Though the respondents of this study were late adolescents, they were also

students at institutions of higher education. College students may possess advantages that non-collegiate adolescents do not have in experiencing a crisis and exploring alternative roles and values.

Demographic Variables as Covariates of Ethnicity

The second question of this study was whether African American, Caucasian, and Hispanic adolescents differ in identity development, when demographic variables (SES, gender, family status) are controlled. A number of studies have suggested that the social and structural differences associated with ethnic identification are explained, in part, by SES, gender, and family status. That is, researchers have proposed that because ethnic minority adolescents tend to live in more impoverished and less stable family environments than their Caucasian peers and these differences inhibit progressive identity development for ethnic minority adolescents (Phinney, 1990). Research also supports the





59


contention that males and females experience identity development differently (Marcia, 1993).

In tests of the statistical hypothesis, no significant differences were found among the respondents' identity status scores, when SES, gender, and family status were controlled. Also, although there were significant differences among groups in their report of SES and family status, none of the demographic variables were found to be significant covariates of identity status.

Familism

The third question of this study was whether African American, Caucasian, and Hispanic adolescent identity development is related to familism. Familism is defined as the connectedness one has with immediate and extended family members. Current operationalization of the concept includes three dimensions: structural, attitudinal, and behavioral. Structural familism refers to the number of relatives living in close proximity to a person; attitudinal familism is a measure of a respondent's beliefs as to the benefit of contact with family members. Behavioral familism refers to the amount of time people stay in contact with family members. Valenzuela and Dornbusch (1994) proposed that a high degree of familism serves as a sort of social capital that enables adolescents to develop positive mental health characteristics. Because differences in familism have been found among African American, Caucasian, and Hispanic adolescents, it was of interest to measure the relationship between identity status and familism with ethnicity controlled. Testing the relationship between familism with ethnicity controlled resulted in no significant relationships.





60


Parenting Style

Only a few studies have considered the relationship between identity status and parenting style. This was the fourth question of the study. In previous studies an authoritative parenting style has been associated with the progressive identity statuses (Waterman, 1993), suggesting that experiencing a parenting style that encourages independence with guidance may enable adolescents to endure a crisis with the intention of working through it. In support of this premise, a restrictive parenting style has been found to be related to the regressive identity statuses (Quintana & Lapsley, 1987).

As for the relationship between a permissive parenting style and identity

development, no studies could be found that investigated this relationship. Inasmuch as a permissive parenting style encourages exploration, which may precipitate a crisis, this style also provides the least amount of parental direction, which may contribute to an inability to commit to identity options.

The statistical tests indicated a relationship between mother's parenting style and identity development. A significant relationship was found between mother's authoritative parenting style and ideological and interpersonal achievement and mother's authoritarian parenting style and ideological foreclosure. Mother's permissive parenting was significantly related to ideological and interpersonal moratorium, as well as diffusion in both domains.

Significant relationships were found between father's parenting style and identity development. Father's authoritarian parenting style was positively related to ideological and interpersonal achievement. Father's permissive parenting style was positively related to ideological and interpersonal diffusion.





61


In all, the results of this study support the idea that identity status is related to parenting style; however, the nature of the relation may differ for mothers and fathers. Analysis indicated that a mother's parenting style that is characterized as open and guiding (authoritative) is related to ideological and interpersonal achievement, whereas the same parenting style in fathers was related to ideological and interpersonal foreclosure. These conflicting findings may reflect the different social roles of mothers and fathers.

Identity Development and Acculturation Orientation

At the center of the discussion of identity development differences among ethnic groups is what effect, if any, does the experience of being a member of an ethnic minority have on the psychosocial development of adolescents. During identity development, all ethnic groups are exposed to the varying effects of SES, gender socialization, family status, parenting style, and familism. Acculturation theory proponents, however, argue that only members of a cultural minority must negotiate the often-competing demands of developing a bi-cultural existence. That is, because Caucasian adolescents are members of the cultural majority, they are unlikely to encounter expectations of themselves that differ from the macro-culture.

It has been proposed that there are four routes to reconciling this acculturation: (a) accommodation, whereby individuals give up their identification with their minority culture and identify with the majority culture, (b) integration, whereby individuals develop positive identifications with their minority culture and the cultural majority, (c) separation, whereby individuals identify with their minority culture and do not identify





62


with the majority culture, and (d) marginalization, whereby individuals fail to identify with their minority culture or the majority culture.

It was of interest, therefore, to determine the relationship between acculturation Drigtation-and identity development for African American and Hispanic adolescents. Statistical tests revealed significant relationships between acculturation and identity development with ethnicity controlled. A marginalized acculturation was positively related to both ideological and interpersonal diffusion, and a separation acculturation was positively related to interpersonal foreclosure. These findings support the contention that the forms of acculturation are differentially related to identity development.

These findings suggest that the results of earlier studies that found ethnic minorities more likely to be classified in the regressive identity statuses than their Caucasian peers might reflect the mediating effect of acculturation orientation. If nothing else, the findings lend cautious support to the argument that acculturation is related to identity development.

Limitations of the Study

The conclusions of this study are to be considered in light of the following limitations:

1. Previous studies of ethnicity and identity development were conducted with samples of high school students. The sample in this study was limited to volunteers at the University of Florida and Sante Fe Community College. The findings may not generalize to other students at the institutions, to students at other colleges and universities, or to high school students. In addition, participants were volunteers, which further limits the generalizability of the study.





63


2. This was a correlational study. Causal claims cannot be made. But even more, ego identity ever evolves. A snapshot inspection of the construct cannot document changes over time. A longitudinal design is better suited to address this limitation.

Implications of the Study

In this study differences in identity development among ethnic groups were not demonstrated, and, when the influence of SES, gender, and family status were controlled, differences were still not found among ethnic groups. Those researchers who reported no differences in identity development between students who were members of minority groups and Caucasian students tended to attribute their findings to the high degree of c utural iitegration in the schools of their samples of minority students. Earlier studies have focused on high school aged adolescents. This study investigated the identity development of late adolescent college students but a similar interpretation of the findings could be made. Students who are members of minority groups who attend universities may experience advantages that enable them to develop progressive identities. Also universities may offer minority students a degree of cultural integration not available to other minority adolescents. The findings of this study support the contention that ethnic differences in identity are not present in environments that support exploration and commitment.

A related finding was that an authoritarian parenting style indicative of control and harsh discipline was related to identity foreclosure; whereas an authoritative parenting style that promotes independence with limits, was related to identity achievement. The permissive parenting style was related to both identity moratorium






64


and diffusion. In each case the degree to which adolescents were given a secure foundation from which to explore was indicative of progressive identity development.

Acculturation orientation was also found to be related to identity development for African American and Hispanic adolescents. Respondents who endorsed a marginalized acculturation were more likely to experience identity diffusion than those who developed other acculturation orientations were. A separation acculturation was also related to interpersonal foreclosure. Again, it can be suggested that the degree to which an adolescent is not limited in avenues of exploration is related to the degree to which progressive identity development occurs.

Recommendations for Future Research

The findings of this study are to be considered in light of its limitations. First, the conclusions were based on a study of late adolescents enrolled in institutions of higher education. To expand the generalizability of these results, studies must be conducted with samples that are more representative of the late adolescent population. Second, as this was a correlational study of a single assessment of identity, no causal claims can be made. If these conditions were addressed, it would be of interest to consider the following:

1. Further research is needed to determine if exposure to an environment that supports identity exploration is related to progressive identity development in spite of ethnic differences.

2. In this study, a relationship between acculturation orientation and identity

development was demonstrated. This finding suggests that for minority group members negotiating a bicultural existence can be an impediment to identity development not





65


shared by those in the majority. Investigation is needed to assess the degree to which the need to resolve the conflicting expectations of the minority and majority cultures limits or expands the identity options available to adolescents.

3. Although significant relationships were found between parenting style and

identity development, no consistent pattern of relationships between parenting style and identity development has been established. Further research is needed to identify relationships of variables that may explain the inconsistent findings.














APPENDIX A
OVERVIEW OF IDENTITY STUDIES


Author/Date Respondents Results Interpretation Controls Measure

Hauser/1972 22 low SES AA more Social & None Q-sort
AA & C foreclosed economic male high conditions school may limit
students exploration of AAs
Abraham/ 273 MA & C C more Parenting None OM-EIS 1983 high school diffused/ style may
students ideological account for identity difference
Abraham/ 841 H & C MA more Parenting Parents' EOM-EIS 1986 high school foreclosed / style may education
students ideological account for identity difference Streitmatter/ 367 71 Minority Social & None EOM-EIS
1989 graders C & group more economic
combination foreclosed/ condition may
of ethnic both limit
minorities domains exploration of AAs
Markstrom- 123 AA, AI, Minority Socialization Parents' EOM-EIS Adams & MA & C 10th groups may account education Adams/ 12 graders more for difference 1995 foreclosed/ both
domains
Rotherham- 320 AA, No notable High degree Comparable EOM-EIS Borus / 1989 MA, NA, F, differences of integration economic
C high in school backgrounds
school
Students
Note. AA= African American, MA = Mexican American, C = Caucasian, F = Filipino, NA = Native American



66















APPENDIX B
INFORMED CONSENT

My name is Sean Forbes. I am a graduate student in the Foundations of Education Department in the College of Education. My faculty supervisor is Patricia Ashton, Ph.D. I am conducting a study to measure the relationship among identity development and personal characteristics. I am requesting that you complete these questionnaires. If you agree, you will complete the questionnaires anonymously.

The questionnaires will take you no more than an hour to complete and you do not have to answer any question that you do not wish to answer. If, at any time, you feel that you do not want to continue with the study you can stop.

As all data will be anonymous, you will not be able to obtain the results of the questionnaires. You will not be reimbursed for your time. However, your cooperation is extremely appreciated.

Please feel free to ask me questions at any time during the study. If you have questions about the study after completing the questionnaires feel free to contact me at 348 Norman Hall, 392-0726, ext 280. Questions or concerns about research participants' rights can be directed to the UFIRB Office, Box 112250, University ofFlorida, Gainesville, FL 326112250


Participant's signature Date

















67















REFERENCES

Abraham, K. G. (1983). The relation between identity status and locus of control among rural high school students. Journal of Early Adolescence, 3. 257-264.

Abraham, K. G. (1986). Ego-identity differences among Anglo-American and Mexican-American adolescents. Journal of Adolescence. 9. 151-166.

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

Adams, G. R., Shea, J., & Fitch, S. A. (1979). Toward the development of an objective assessment of ego identity status. Journal of Youth and Adolescence, 8. 223228.

Andrews, J. (1973). The relationship of values to identity achievement status. Journal of Youth and Adolescence, 2. 133-138.

Archer, S. L. (1981). Identity interview schedule for working women. Unpublished manuscript. Trenton State College, Trenton, NJ.

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

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

Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology. 4, (1, Pt. 2), 1-103.

Baumrind, D. (1991). Effective parenting during the early adolescent transition. In P. A. Cowan & E. M. Hetherington (Eds.), Advances in family research (Vol. 2, pp. 111163). Hillsdale, NJ: Erlbaum.

Berry, J. W. (1980). Acculturation as varieties of adaptation. In A. Padilla (Ed.), Acculturation: Theory, models, and some new findings (pp. 1-65). Boulder: Westview.





68





69


Berry, J. W., Trimble, J. E., & Olmedo, E. L. (1986). Assessment of acculturation. In W. J. Lonner & J. W. Berry (Eds.), Cross-cultural research and methodology series: Vol. 8. Field methods in cross-cultural research (pp. 291-101). Beverly Hills, CA: Sage.

Buri, J. R. (1991). Parental authority questionnaire. Journal of Personality Assessment, 57, 110-119.

Chandler, C. R. (1979). Traditionalism in a modern setting: A comparison of Anglo- and Mexican-American value orientations. Human Organization, 38, 153-159.

Coleman, J. S. (1988). Social capital in the creation of human capital. The American Journal of Sociology, 94, S95-S120.

Cross, R, & Allen, J. (1970). Ego identity status, adjustment, and academic achievement. Journal of Consulting and Clinical Psychology. 34. 288.

Dellas, M., & Jernigan, L. P. (1987). Occupational identity status development, gender comparisons, and internal-external control in first-year Air Force cadets. Journal of Youth and Adolescence, 16, 587-600.

Elkind, D. (1978). A sympathetic understanding of the child (2nd ed.). Boston: Allyn & Bacon.

Enright, R. D., Lapsley, D. K., Drivas, A. E., & Fehr, L. A. (1980). Parental influences on the development of adolescent autonomy and identity. Journal of Youth and Adolescence. 9. 529-545.

Erikson, E. H. (1963). Childhood and society (2nd ed.). New York: Norton.

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

Erikson, E. H. (1987). The human life cycle. In S. Schlein (Ed.), A way of looking at things (pp. 595-610). New York: Norton.

Farris, B. E., & Glenn, N. D. (1976). Fatalism and familism among Anglos and Mexican Americans in San Antonio. Sociology and Social Research, 60, 393-402.

Forbes, S., & Ashton, P. (1996). Improving the construct validity of scores on the Extended Objective Measure of Ego Identity Scale (EOM-EIS II). Manuscript submitted for publication.

Gibson, M. (1988). Accommodation without assimilation: Sikh immigrants in an American high school. New York: Cornell University.





70


Gonzalez, N. A., & Cauce, A. M. (1995). Ethnic identity and multicultural
competence: Dilemmas and challenges for minority youth. In W. D. Hawley & A. W. Jackson (Eds.), Toward a common destiny: Improving race and ethnic relations (pp. 131162). San Francisco: Jossey-Bass.

Gottfried, A. W. (1985). Measures of socioeconomic status in child development research: Data and recommendations. Merrill-Palmer Quarterly, 31, 85-92.

Grossman, S. M., Shea, J. A., & Adams, G. R. (1980). Effects of parental divorce during early childhood on ego development and identity formation of college students. Journal of Divorce, 3. 263-272.

Grotevant, H. D., & Adams, G. R. (1984). Development of an objective measure to assess ego identity in adolescence: Validation and replication. Journal of Youth and Adolescence. 13. 419-438.

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

Hauser, S. T. (1972). Black and white identity development: Aspects and perspectives. Journal of Youth and Adolescence, 1, 113-130.

Hetherington, E. M. (1972). Effects of father absence on personality development in adolescent daughters. Developmental Psychology, 7. 313-326.

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

Hoff-Ginsberg, E., & Tardiff, T. (1995). Socioeconomic status and parenting. In M. H. Bornstein (Ed.), Children and parenting (Vol. 2) (pp. 161-188). Hillsdale, NJ: Erlbaum.

Hollingshead, A. B. (1975). Four factor index of social status. Unpublished manuscript, Yale University, Department of Sociology, New Haven, CT.

Holtzman, W. H. (1982). Cross cultural comparisons of personality development in Mexico and the United States. In D. A. Wagner & H. W. Stevenson (Eds.), Cultural perspectives on child development (pp. 118-164). New York: Freeman.

Jones, R. M., & Streitmatter, J. L. (1987). Validity and reliability of the EOM-EIS for early adolescents. Adolescence, 22, 647-659.

Jordan, D. (1970). Parental antecedents of ego identity formation. Unpublished master's thesis, State University of New York at Buffalo.





71


Josselson, R. (1994). The theory of identity development and the question of intervention. In S. Archer (Ed.), Interventions for adolescent identity development (pp. 12-25). Thousand Oaks, CA: Sage.

Keefe, S. E. (1984). Real and ideal extended familism among Mexican Americans and Anglo Americans: On the meaning of close family ties. Human Organization, 38, 6569.

La Voie, J. C. (1976). Ego identity formation in middle adolescence. Journal of Youth and Adolescence. 5. 371-385.

Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent-child interaction. In P. H. Mussen (Ed.), Handbook of Child Psychology. (4 th ed., pp. 131-162). New York: Wiley.

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

Marcia, J. E. (1976). Identity six years after: A follow-up study. Journal of Youth and Adolescence, 5, 145-160.

Marcia, J. E. (1993). The status of the statuses: Research review. 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. 22-41). New York: Springer-Verlag.

Marcia, J. E., & Friedman, M. (1970). Ego identity status in college women. Journal of Personality,. 38, 249-262.

Markstrom-Adams, C., & Adams, G. R. (1995). Gender, ethnic group, and grade differences in psychosocial functioning during middle adolescence. Journal of Youth and Adolescence, 24, 397-417.

Markstrom-Adams, C., & Spencer, M. B. (1994). A model for identity
intervention with minority adolescents. In S. Archer (Ed.), Interventions for adolescent identity development (pp. 84-102). Thousand Oaks, CA: Sage.

Matteson, S. L. (1993). Differences within and between genders: A challenge to the theory. 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. 69-110). New York: Springer-Verlag.

Mehan, H., Hubbard, L., & Villanueva, I. (1994). Forming academic identities: Accommodation without assimilation among involuntary minorities. Anthropology and Education Ouarterly. 25. 91-117.





72


Mindel, C. H. (1980). Extended familism among Mexican Americans, Anglos, and Blacks. Hispanic Journal of Behavioral Sciences. 2. 21-34.

Negy, C., & Woods, D. J. (1992). A note on the relationship between
acculturation and socioeconomic status. Hispanic Journal of Behavioral Sciences, 2. 248251.

Nowicki, S., & Strickland, B. R. (1973). A locus of control scale for children. Journal of Consulting and Clinical Psychology, 40, 148-154.

Peretti, P. O., & Wilson, T. T. (1995). Unfavorable outcomes of the identity crisis among African American adolescents influenced by enforced acculturation. Social Behavior and Personality. 23. 171-175.

Phinney, J. S. (1990). Ethnic identity in adolescence and adulthood: A review of research. Psychological Bulletin, 108, 499-514.

-Quintana, S. M., & Lapsley, D. K. (1987). Adolescent attachment and ego -4 identity: A structural equations approach to the continuity of adaptation. Journal of Adolescent Research, 2. 393-409.

Quintana, S. M., & Lapsley, D. K. (1990). Rapprochement in late adolescent separation-individuation: A structural equations approach. Journal of Adolescence, 13, 371-385.

Raphael, D. (1977). Identity status in university women: A methodological note. Journal of Youth and Adolescence, 6. 57-62.

Reguero, J. T. (1991). Relationship between familism and ego identity development of Puerto Rican and immigrant Puerto Rican adolescents (Doctoral dissertation, Virginia Polytechnic Institute and State University, 1991). Dissertation Abstracts International, 52, 1750-B.

Rotheram-Borus, M. J. (1989). Ethnic differences in adolescents' identity status and associated behavior problems. Journal of Adolescence, 10, 525-537.

Rotheram-Borus, M. J., & Wyche, K. F. (1994). Ethnic differences in identity
development in the United States. In S. Archer (Ed.), Interventions for adolescent identity development (pp. 62-83). Thousand Oaks, CA: Sage.

Schenkel, S., & Marcia, J. E. (1972). Attitudes toward premarital intercourse in determining ego identity status in college women. Journal of Personality, 3, 472-482.

Spence, J. T., Helmreich, R., & Stapp, J. (1974). The Personal Attributes
Questionnaire: A measure of sex role stereotypes and masculinity and femininity. Catalog of Selected Documents in Psychology. 4, 43-44.





73


Spencer, M. B., & Dornbusch, S. M. (1990). Challenges in studying minority youth. In S. S. Feldman & G. R. Elliot (Eds.), At the threshold: The developing adolescent (pp. 123-146). Cambridge, MA: Harvard University Press.

Spencer, M. B., & Markstrom-Adams, C. (1990). Identity processes among racial and ethnic minority children in America. Child Development, 61. 290-310.

St. Clair, S., & Day, H. P. (1979). Ego identity status and values among young females. Journal of Youth and Adolescence, 8. 317-326.

Sterling, C. M., & Van Horn, K. R. (1989). Identity and death anxiety. Adolescence, 17, 895-899.

Streitmatter, J. L. (1987). The effect of gender and family status on ego identity development among early adolescents. Journal of Early Adolescence, 7. 179-189.

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

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

U. S. Bureau of the Census. Department of Commerce. (1994). Statistical abstract of the United States, 1994. Washington, DC: U. S. Government Printing Office.

Valenzuela, A., & Dornbusch, S. M. (1994). Familism and social capital in the academic achievement of Mexican origin and Anglo adolescents. Social Science Quarterly. 75. 18-36.

Velez-Ibanez, C. G., & Greenberg, J. B. (1992). Formation and transformation of funds of knowledge among U.S.-Mexican households. Anthropology and Education Quarterly, 23, 313-335.

Waterman, A. S. (1993). Developmental perspectives on identity formation: From adolescence to adulthood. 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. 42-68). New York: Springer-Verlag.

Waterman, A. S. (1994). Ethical consideration in interventions for promoting identity development. In S. Archer (Ed.), Interventions for adolescent identity development (pp. 231-244). Thousand Oaks, CA: Sage.

Waterman, C. K., & Waterman, A. S. (1975). Fathers and sons: A study of ego identity across two generations. Journal of Youth and Adolescence. 4, 331-338.














BIOGRAPHICAL SKETCH

Sean Forbes completed his secondary education in Shelby County, Tennessee, in 1988. In the spring of 1989 he enrolled at the University of Florida, earning a bachelor of arts degree in elementary education in 1992 and a master of arts degree in educational psychology in 1995.

After completing his doctoral studies in 1999, he accepted a position as an assistant professor of educational psychology in the Department of Educational Foundations, Leadership, and Technology at Auburn University.




























74














I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.


Patricia Ashton, Chair
Professor of Educational Psychology

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.


Ja esAlgina
Pr fessor of Educationl Psychology

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy



Associate l6fessor of Educational Psychology

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.


Joh anzler
Assba te Professor of Educational Psychology








I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.


Lee Mullally
Associate Professor of Teac ing and Learning

This dissertation was submitted to the Graduate Faculty of the College of Education and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy.



Dean, College of Educatio



Dean, Graduate School




Full Text
development among ethnic groups. In the studies investigating the topic, inconsistent
results have been found. Some researchers have reported that African American and
Hispanic adolescents were more likely to be foreclosed than their Caucasian peers. In
light of the differences among ethnic groups in parenting style and familism reported in
some studies, this study was designed to determine whether the inconsistent findings
regarding the relation between identity status and ethnic group could be explained by
parenting style, familism, SES, gender, and/or family configuration. Also, as adolescents
from cultural minorities must negotiate a bicultural existence, it was also of interest to
consider the relation between identity development and acculturation for African
American and Hispanic adolescents. Three hundred college aged adolescents participated
in this study. Of the total, 82 identified themselves as African American and 98 identified
themselves as Caucasian. One hundred twenty participants identified themselves as
Hispanic. Analysis of variance indicated no significant differences among groups in
identity development and no relationship between identity development and familism.
Significant relationships were found between parenting style and acculturation and
identity development.
vi


17
total scores (combined ideological and interpersonal domain) on foreclosure than
adolescents from nonintact families. From post hoc analyses she found that the family
status effect held for ideological foreclosure but not for interpersonal foreclosure, and she
found a significant Family Status x Gender interaction indicating that female adolescents
from intact homes had lower interpersonal foreclosure scores than female students from
disrupted homes, whereas male students from intact homes had higher foreclosure scores
than males in non-intact homes. She also found a significant Family Status x Gender
interaction for diffusion. Male students from intact homes had lower ideological diffusion
scores than male students from non-intact homes, and females from intact homes had
higher ideological diffusion scores than females with only one parent, in the home. She
also found that the overall relationship between family status and identity development
was greater for male students than for female students. She speculated that given the
likelihood that most of these families were headed by women, the lower foreclosure
scores may reflect the male adolescents' rejection of failed family values, whereas their
higher diffusion scores may indicate that they were unable or unwilling to choose an
identity. Streitmatter further speculated that family disruption may have facilitated
identity development by thrusting the male adolescents into diffusion earlier than their
peers in intact homes.
In sum, research has demonstrated that family configuration is related to identity
development. However, the kind of household (i.e., intact, non-intact), as well as the
gender of the child, plays a role in relationship between family configuration and identity
development. Therefore, research that divides family configuration into separate
categories for (a) single parent as the result of death, (b) single parent due to divorce, (c)


69
Berry, J. W., Trimble, J. E., & Olmedo, E. L. (1986). Assessment of acculturation.
In W. J. Lonner & J. W. Berry (Eds.), Cross-cultural research and methodology series:
Yol. 8, Field methods in cross-cultural research (pp. 291-101). Beverly Hills, CA: Sage.
Buri, J. R. (1991). Parental authority questionnaire. Journal of Personality
Assessment. 57. 110-119.
Chandler, C. R. (1979). Traditionalism in a modern setting: A comparison of
Anglo- and Mexican-American value orientations. Human Organization. 38. 153-159.
Coleman, J. S. (1988). Social capital in the creation of human capital. The
American Journal of Sociology. 94. S95-S120.
Cross, H, & Allen, J. (1970). Ego identity status, adjustment, and academic
achievement. Journal of Consulting and Clinical Psychology. 34. 288.
Dellas, M, & Jemigan, L. P. (1987). Occupational identity status development,
gender comparisons, and internal-external control in first-year Air Force cadets. Journal
of Youth and Adolescence. 16. 587-600.
Elkind, D. (1978). A sympathetic understanding of the child (2nd ed.). Boston:
Allyn & Bacon.
Enright, R. D., Lapsley, D. K., Drivas, A. E., & Fehr, L. A. (1980). Parental
influences on the development of adolescent autonomy and identity. Journal of Youth
and Adolescence. 9. 529-545.
Erikson, E. H. (1963). Childhood and society (2nd ed.). New York: Norton.
Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.
Erikson, E. H. (1987). The human life cycle. In S. Schlein (Ed.), A wav of
looking at things (pp. 595-610). New York: Norton.
Farris, B. E., & Glenn, N. D. (1976). Fatalism and familism among Anglos and
Mexican Americans in San Antonio. Sociology and Social Research. 60. 393-402.
Forbes, S., & Ashton, P. (1996). Improving the construct validity of scores on the
Extended Objective Measure of Ego Identity Scale (EOM-EIS ID. Manuscript submitted
for publication.
Gibson, M. (1988). Accommodation without assimilation: Sikh immigrants in an
American high school. New York: Cornell University.


18
intact families, (d) single parent never married, and (e) reconstituted families and that
analyzes the data separately for males and females and controls for SES is needed to
clarify the relationship between family configuration and identity development.
Family Characteristics
Differences in the family environment of minority and majority adolescents may
explain, in part, the different identity development of the ethnic groups. Marcia (1993)
argued that individuals classified in the high identity statuses often come from families
where ideological and interpersonal differentiation between the child and parents is
supported. Those classified in the lower identity statuses, in contrast, tend to come from
families that either encourage conformity to family values or from families in which the
parents distance themselves from the ideological and interpersonal development of the
child. Significant differences have been found between minority cultures and the majority
culture in parenting style (Holtzman, 1982) and familism (Mindel, 1980).
Parenting style. Baumrind (1971, 1991) identified three forms of parenting: (a)
authoritarian whereby parents use a controlling and disciplinarian style to raise their
children, (b) authoritative whereby parents encourage independence but a respect for
rules in their children and, (c) laissez-faire whereby parents do not place limitations or
hold expectations for their children's behavior. Maccoby and Martin (1983) further
divided the laissez-faire style of parenting into two distinct styles: (a) permissive-
indifferent, whereby parents are not involved in their children's lives and (b) permissive-
indulgent, whereby parents give in to their children's demands and do not place
restrictions on their behavior.


29
strategies to foster successful resolution of the identity crisis (Josselson, 1994; Rotheram-
Borus & Wyche, 1994; Waterman, 1994).
It is not clear whether such intervention strategies are warranted for adolescents
from minority groups. As Erikson (1963) noted, culture affects the development of
identity. That is, cultural expectations and practices encourage and limit adolescents'
opportunity for exploration and commitment in the various identity domains. Given that
researchers have found that African American and Hispanic adolescents are more
foreclosed than their Caucasian peers, Spencer and Markstrom-Adams (1990) stated that
the foreclosure status may be adaptive in certain cultures: "Exploration of identity may
not be desirable for such youth when social and ideological roles are clearly defined by
the community" (p. 298). This study was designed to determine whether the relationship
between ethnicity and identity development is attributable to SES, family configuration,
and family values.
It may be inappropriate, however, to conclude that identity intervention is
unnecessary because of differences in cultural expectations. Markstrom-Adams and
Spencer (1994) argued that despite minority cultural support many minority adolescents
are restricted in their ability to explore identity alternatives because of the limitations in
their social environment. That is, a significant portion of minority adolescents grow up in
such impoverished, discriminatory environments that they cannot take advantage of the
role models available that are necessary for adequate identity exploration. Consequently,
it is important to determine whether the foreclosure of minority adolescents is attributable
to economic variables or to cultural values.


32
correlated (r = .07, g > .10). Father's authoritarianism was negatively correlated with
father's authoritativeness and father's permissiveness (r = -.50, g < .05; r = -.52, g < .05).
Father's authoritativeness and father's permissiveness items were not significantly
correlated (r = 12, g > 10). In the same study, Buri reported the following data on
criterion-related validity: The authoritative subscales were positively related to parental
nurturance (mother r = .56, g < .0005; father r = .68, g < .0005), whereas the authoritarian
subscale was negatively related to parental nurturance (mother r = -.36, g < .0005; father
r = .-53, g < .0005). The permissiveness subscales were not significantly related to
parental nurturance (mother r = .04, g < 10; father r = 13, g < 10), respectively.
In a study of 69 undergraduate psychology students, Buri (1991) assessed the
relationship between social desirability and respondents' answers to the PAQ subscales.
None of the subscales was significantly related to the Marlowe-Crowne Social
Desirability Scale.
Family Configuration
Family configuration was assessed by the respondents' self-report. Participants
were asked to indicate the category that identified their family type on the questionniare.
Categories included dual parent household without stepparents; dual parent household
with stepparents; single-parent household-parents divorced; single-parent household-
parent never married, single-parent household-parent deceased, and an Other category.
Familism
Familism was assessed in each of the three domains of the construct. Familism
behavior was measured using a single item taken from the work of Valenzuela and
Dombusch (1994) that asks respondents to indicate on a 6-point Likert scale "how often


identified with the minority culture. As a result, it was impossible to differentiate
between assimilated and the accommodated respondents.
15
As a great number of minority children come from low SES backgrounds, many
do not have access to the same economic resources as their higher SES peers and, as a
result, may adopt a marginalized or rejected acculturation orientation. This puts these
adolescents at greater risk of concluding that the expectations of society are incompatible
with their own, which may result in the formation of a weaker identity than higher SES
peers. Research is needed to investigate Berry's (1980) hypothesis and to determine
whether acculturation status mediates the relationship between SES and adolescents'
identity status.
Family configuration. African American and Hispanic minority adolescents are
more likely to grow up in a single parent home than are their Caucasian peers (U. S.
Bureau of the Census, 1994). Family configuration differences may be related to patterns
of identity development among cultural groups, but research on the relationship between
family configuration and identity has yielded inconsistent results. For example, Jordan
(1970) found that male college students from divorced families were more likely to be
identity diffused than were students from intact families. Other studies, however, do not
support the claim that living in an intact family is related to progressive identity
development. Grossman, Shea, and Adams (1980), in a study of 294 college students,
found that males from divorced families possessed higher identity achievement scores
than males in intact families and females in general. In a study of 467 students from ages
12 to 18, Jones and Streitmatter (1987) found that male and female high school students
from intact families were more foreclosed on the ideological and interpersonal scales of


ACKNOWLEDGEMENTS
Patricia Ashton has guided me since I began graduate school. Her wisdom and
patience have not been lost on me. I thank her for her unwavering support and am
indebted for the lessons she has taught me. This dissertation is dedicated to her. I would
also like to thank the other members of my committee for their assistance. James Algina
was a sobering reminder of the diligence needed for any worthwhile pursuit. John
Kranzler, who has known me since I was an undergraduate student, has counseled me
many a time and his advice has served me well. Barry Guinagh has been the kindest
man I have ever met. Greg Neimeyer and Lee Mullally, my outside committee members,
have saved my skin on more than one occasion. Collectively, this committee has been
outstanding.
I must also recognize Randall Scott Hewittthe most natural philosopher I have-
known. Graduate school would not have been complete without him. He and his family
have opened my eyes to other worlds and I will miss their companionship.


34
factors-gender, occupation, education, and marital status-are combined into a single
index rather than considered as separate factors. The FFISS has been described as a
highly reliable and valid measure of social status (see Hollingshead, 1975). Gottfried
found that the FFISS was more highly correlated with children's developmental outcomes
in a longitudinal (30 months) study of 130 young children and their parents than were
two other SES indices, the Siegel Prestige Scale and the Revised Duncan Socioeconomic
Index. In addition, Gottfried pointed out that the Hollingshead Four Factor Index can be
used to estimate the SES of unmarried heads of households as well as families.
Procedures
Each participant signed the consent form (see Appendix B) and anonymously
completed the instruments that consisted of the demographic questionnaire, EOM-EIS III,
PAQ, acculturation, and familism questionnaires. Data were collected from summer,
1997 to spring, 1999. Respondents were told not to answer any items that they did not
wish to and to ask questions at any time. Administration time took between 45 min to 1
hr.
Research Questions
Research Question #1
The first set of research hypotheses in this study was designed to determine if
ethnic identification was related to identity status classification. Respondents' scores on
the EOM-EIS were used as the dependent variable in the model. The independent
variable was ethnic identification.


22
raised questions about the extent of familism among Mexican Americans. Although
Farris and Glenn (1976) and Chandler (1979) found greater familism among Mexican
American than Anglo American groups, the differences were not large. Also, contrary to
the belief that Mexican Americans are high in familism, most of the Mexican Americans
disagreed with most of the familism items.
Valenzuela and Dombusch (1994) defined familism as the structural, attitudinal,
and behavioral dimensions of family interactions. The structural aspect of familism is
characterized by the presence of a nuclear and extended family, whereas the attitudinal
aspect refers to concern for the welfare of the family. The behavioral dimension refers to
the amount of contact with family members. Valenzuela and Dombusch argued that
familism can serve as a form of social capital (Coleman, 1988). That is, family
connectedness provides children with supportive relationships that enable them to
develop positive mental health and to perform well in school. Valenzuela and Dombusch
based their ideas on the work of Velez-Ibanez and Greenburg (1992), who argued that
Mexican American children are raised in dense extended families that socialize the
children to form strong emotional attachments with a wide variety of relatives who tend
to live in nearby neighborhoods, and on research by Velez (as cited in Velez-Ibanez &
Greenberg), who found that Mexican American mothers and their relatives had greater
contact with infants than Caucasian mothers and relatives and the Mexican American
children were in frequent contact with relatives.
Valenzuela and Dombusch (1994) analyzed surveys of 2,666 Anglo and 492
Mexican students in 1987-88 and found that the Mexican American students were higher
in familism than Caucasian students. They also found that for Mexican-origin students


20
negative individuation from families, and these adolescents were prone to develop lower
level identities than their peers from less restrictive homes.
Direct relationships have also been found between parenting style and identity
development. La Voie (1976) studied 120 students in grades 10, 11, and 12 and found
high identity males came from homes with less regulation and control and more praise by
parents than their lower identity status peers. Females with a high identity status came
from homes with less maternal control and greater encouragement to discuss personal
problems with their parents than did their low identity peers. Quintana and Lapsley
(1987) found similar results. In a study of 101 undergraduates, they found that the
children of families that encouraged communication and were not controlling were better
prepared to deal with their ensuing crises. Quintana and Lapsley presumed that their
family background prepared them to freely explore identity alternatives. Adolescents
from controlling homes were more likely to be classified in the lower identity statuses
than were their peers from less controlling homes. However, in two studies of seventh-
and eleventh-grade students, Enright, Lapsley, Drivas, and Fehr (1980) found that only
fathers' parenting style was related to identity development. In the first study of 262
respondents, Enright et al. found that males with fathers who practiced a democratic style
of parenting-analogous to Baumrind's authoritative parenting-tended to be classified in
higher identity statuses than their peers with autocratic or permissive fathers. Females
whose fathers practiced an autocratic parenting style, however, were more often
classified in the higher identity statuses than were peers with democratic or permissive
fathers. No relationship was found between mothers' parenting style and identity
development. In the second study of 168 respondents, Enright et al. found that both males


46
Table 10. ANCOVA Results for Interpersonal Identity Statuses
Source
ss
df
Mean Square
F
R
Achieved
Race
178.59
2
89.30
2.87
0.06
Gender
222.65
1
222.65
7.15
0.01
SES
12.35
1
12.35
0.40
0.53
Family
250.54
5
50.11
1.61
0.16
Moratorium
Race
2.27
2
1.18
0.04
0.96
Gender
0.08
1
0.08
0.00
0.96
SES
25.50
1
25.50
0.85
0.36
Family
32.47
5
6.49
0.22
0.96
Foreclosure
Race
44.83
2
22.42
0.73
0.49
Gender
154.18
1
154.18
4.99
0.03
SES
100.95
1
100.95
3.27
0.07
Family
113.68
5
22.73
0.74
0.60
Diffusion
Race
34.87
2
17.44
0.66
0.52
Gender
70.25
1
70.25
2.67
0.10
SES
0.99
1
0.99
0.04
0.85
Family
98.28
5
19.66
0.75
0.59


37
Table 1. Family Status Percentages bv Ethnicity
African American
Hispanic
Caucasian
Status
%
%
%
Dual parent-
No step parent
44
66
63
Dual parent-
Stepparents
19
6
17
Single parent-
Never married
12
5
0
Single parent-
Parent divorced
17
19
16
Single parent-
Parent deceased
4
3
2
Other
4
1
2
identified themselves as women, and 20 (24%) identified themselves as men. Sixty-five
Hispanic (54%) and 71(72%) Caucasian respondents identified themselves as womem; 55
(46%) Hispanic and 27 (28%) Caucasian respondents identified themselves as men.
Table 1 displays the percentage of respondents in each of the different family
statuses by ethnic group. A chi-square analysis of the relationship between statuses by
ethnic group resulted in a non-significant relationship.
Table 2 lists the means and standard deviations of respondents' scores on the Four
Factor Index of Social Status (Hollingshead, 1975). An ANOVA indicated a significant


58
social exploration available to minority groups. The proposed barriers to their exploration
include SES, gender, family status, familism, and parenting style. The research questions
examined in this study were designed to explore the relationship of these variables to
identity development.
Ethnicity and Identity Development
The first research question was whether there is a difference among African
American, Caucasian, and Hispanic adolescents in identity development. Specifically,
more Caucasian adolescents were expected to be in the progressive identity statuses in
comparison to their African American and Hispanic peers. Tests of this hypothesis
indicated no significant differences among the respondents' identity status scores.
Though the respondents of this study were late adolescents, they were also
students at institutions of higher education. College students may possess advantages that
non-collegiate adolescents do not have in experiencing a crisis and exploring alternative
roles and values.
Demographic Variables as Covariates of Ethnicity
The second question of this study was whether African American, Caucasian, and
Hispanic adolescents differ in identity development, when demographic variables (SES,
gender, family status) are controlled. A number of studies have suggested that the social
and structural differences associated with ethnic identification are explained, in part, by
SES, gender, and family status. That is, researchers have proposed that because ethnic
minority adolescents tend to live in more impoverished and less stable family
environments than their Caucasian peers and these differences inhibit progressive identity
development for ethnic minority adolescents (Phinney, 1990). Research also supports the


philosophical lifestyle. No differences between ethnic groups were found on
interpersonal identity.
5
Abraham (1986) offered two alternative explanations for her findings. First, she
suggested that Hauser's belief that the students' ethnic minority status may have limited
the range of their exposure to ideological options. Second, she noted that differences in
parenting style might account for the results, though this variable was not included in the
analysis. Abraham concluded that the limitations in options and parenting style may have
contributed to the greater number of the Mexican American students in the foreclosure
status. Selection of participants from a single school raises questions, however, about the
generalizability of Abraham's findings.
Administering the EOM-EIS to 367 students in grades 7 and 8 in a junior high
school in the Southwest, Streitmatter (1988) compared the identity development of
Caucasians and an ethnic minority group composed of Hispanic Americans, African
Americans, Native Americans, and Asian Americans. Like Hauser (1972) and Abraham
(1986), Streitmatter found that the Caucasian adolescents were less foreclosed than the
minority respondents. Unlike the previous studies, however, differences were found in
the interpersonal domain as well as the ideological. Streitmatter suggested that the greater
incidence of foreclosure in the minority groups was the result of minority adolescents'
feeling more uncomfortable in the majority culture than their Caucasian peers. She
proposed that this discomfort led them to accept parental values and reduced their
willingness to experiment with ideas that conflicted with their parents' expectations.
Further, she pointed out that the minority students were bussed into the school, which
might have added to the students' discomfort and inhibited their exploration. She referred


7
diffused than their Caucasian counterparts. Also, relationships between the contextual
variables and identity status emerged. Higher parent education was related to lower
foreclosure rates for ideological identity. In concluding, Markstrom-Adams and Adams
(1995) agreed with the earlier researchers who argued that the greater foreclosure in non-
White adolescents might be because the minority groups are more accepting of
foreclosure than Caucasians, who may require experimentation and autonomous choice in
achieving an identity.
Two other studies, however, did not report differences in foreclosure in minority
and Caucasian adolescents (Abraham, 1983; Rotheram-Borus, 1989). Abraham (1983)
investigated the identity development of 223 Caucasian and Mexican American students
from a rural area in the southwest with the Objective Measure of Ego Identity Status
(OM-EIS), a 24-item measure of ideological identity, including the areas of occupation,
politics, and religion (Adams, Shea, & Fitch, 1979). She found that more Caucasian
students than Mexican American students were classified in diffusion. Although more
Mexican American students than Caucasian students were classified as foreclosed (18.8%
compared to 9.6%) and more Caucasian students were classified as identity achieved
(10.1% compared to 16.4%), these differences were not significantly different.
In her discussion, Abraham (1983) pointed out that SES level has been identified
as a variable related to identity development. A post hoc analysis of the parental
educational level of the students in her study revealed that the parents of the Mexican
American students were lower in educational level than the parents of the Caucasian
students. Because a higher SES level would be expected to be associated with lower
levels of diffusion, she proposed that parental socialization style might be the influential


CHAPTER 1
INTRODUCTION
According to Erikson (1963), the formation of an identity is the dominant
psychosocial crisis of adolescence. Commonly referred to as the identity crisis, this
process is characterized by a delay of adult commitments that permits the adolescent to
experiment with various roles and ultimately to commit to available identity alternatives
in ideological and interpersonal domains (Erikson, 1968, p. 159). As the goal of an
identity crisis is the synthesis of the identifications of childhood and the opportunities and
expectations of adulthood, identity resolution is the foundation upon which future adult
decisions are based (Erikson, 1987; Spencer & Markstrom-Adams, 1990). Moreover, the
individual who successfully resolves the crisis becomes a secure, independent adult
prepared to establish intimacy with others, whereas those who are unable to establish an
identity enter adulthood without a firm sense of how they fit into society (Elkind, 1978;
Peretti & Wilson, 1995).
Marcia (1966, 1976), in an extension of Erikson's theory, described four possible
outcomes of the crisis: identity achievement, moratorium, foreclosure, and diffusion.
Identity achievement is defined by a commitment to identity alternatives after a period of
exploration, whereas moratorium is defined as a period of exploration of alternative roles
during which commitment to an identity is delayed. Foreclosure is defined by identity
commitment without prior exploration, whereas diffusion is defined by the unwillingness
to commit to an identity with or without exploration.
l


Parenting style is significantly related to SES (HofF-Ginsberg & Tardif, 1995).
Parents in the lower SES tend to use an authoritarian style of parenting, whereas middle
19
SES parents tend to use an authoritative style of parenting (HofF-Ginsberg & Tardif,
1995). As minority adolescents are more likely to have low-income families than their
Caucasian peers (Spencer & Dombusch, 1990), minority adolescents tend to be raised
with an authoritarian style of parenting. Caucasian adolescents, in contrast, tend to be
raised with an authoritative style. This difference may be related to patterns of identity
development among cultural groups. For example, Waterman (1993) hypothesized that,
because of the controlling power authoritarian-style parents exert, their children are likely
to begin identity development in foreclosure. In contrast, children of permissive parents
tend toward identity diffusion because of their parents' lack of expectations. Waterman
was less certain about the likely outcomes of authoritative parenting, but he seemed to
suggest that the support and caring these parents give their children might enable them to
delay commitment, thus enabling them to experience moratorium.
Research supports Waterman's belief that an authoritative parenting style is
related to moratorium. For example, Quintana and Lapsley (1990), in a study of 101
undergraduates, found an indirect relationship between the perceived parenting styles and
identity statuses of adolescents mediated by individuation from parents. According to
their results, a parenting style that encouraged independence tended to produce positive
individuation from the family for adolescents. This type of attachment gave the
adolescents the necessary freedom to differentiate from their families and succeed in
forming high status identities. Controlling parenting styles, however, resulted in a


43
Table 8. Mean Scores and Standard Deviations of Identity Statuses
African American Hispanic Caucasian
M
S
M
S
M
S
Ideological
Achieved
23.28
3.76
23.46
3.52
23.01
3.25
Moratorium
25.97
3.91
26.74
3.24
26.51
3.54
Foreclosed
20.44
4.74
22.53
4.72
21.63
3.95
Diffused
41.64
27.58
32.81
19.75
27.42
9.73
Interoersonal
Achieved
25.28
4.29
25.23
4.12
25.04
3.99
Moratorium
25.23
4.26
25.37
3.61
25.52
4.21
Foreclosure
25.60
3.89
26.21
3.88
24.70
3.63
Diffused
25.39
4.24
25.74
3.62
24.97
3.52
ANOVA was conducted on the means of each of the identity domain scores for African
American, Hispanic, and Caucasian adolescents. When the Bonferroni-Holm procedure
was used to control for the fact that ethnic group differences were tested on eight
variables, none of the F statistics was significant. When a conventional .05 alpha level
was used, ethnic differences among identity status scores were found for one status-
diffused, ideological, F(2, 287) = 4.13, p = .017.


12
the identity processes, in two of the studies, males and females differed only in the
foreclosure status; males were more likely to be foreclosed than females. In the third
study no differences were found between males and females in the identity statuses.
Archer (1989a) found, however, that males were more likely than females to be
diffused with regard to the family roles domain, whereas females were more likely to be
classified in the foreclosure, moratorium, and achievement statuses. No differences were
detected in the second study. In the third study, Archer found that males were more likely
to be classified in the foreclosure status, and females in the diffusion status with regard to
political ideology.
As for gender differences in the timing of identity development, Archer (1989a)
found in two of the studies that no differences existed. In the third study, however, a
Grade x Gender interaction was found. Females were more likely than males to be
classified in the moratorium status in grades 6, 10, and 12. For identity achievement,
males were more likely than females to be so at grade 12.
Overall, Archer (1989a) concluded that the differences between the genders were
slight. On the basis of her studies, she argued that the assumption that males and females
experience identity development differently should be discarded. Matteson (1993)
proposed much the same argument as Archer. That is, Matteson pointed out that the
relationships found between identity status and other psychological constructs for males
were also found for females. For example, men and women classified in the moratorium
status reported the highest levels of anxiety, and males and females in foreclosure were
more likely than those in other statuses to defer to authority (Marcia & Friedman, 1970).


8
variable in her study. She pointed out that Adams and Jones (1983) found that a maternal
style characterized by control, regulation, and strong encouragement of independence and
a paternal style characterized by praise, approval, and less fair discipline were related to
diffusion, parental styles that might have been more common among the parents of the
Caucasian students compared to the Mexican American students.
Using the OM-EIS (Adams et al., 1979), Rotheram-Borus (1989), in an
investigation of the identity status of 330 African American, Caucasian, Filipino,
Hispanic, and Native American high school students, found that there was little relation
between identity status and ethnicity. Her only significant finding was that Caucasian
students in the 11th and 12th grades were more likely than the other students to have high
moratorium scores on the OM-EIS. The students, however, were not classified in the
moratorium status at a rate different from others. Rotheram-Borus argued that because of
the school's high degree of integration, ethnic differences in the stages of identity did not
exist.
In summary, Hauser (1972), Abraham (1986), Streitmatter (1988), and
Markstrom-Adams and Adams (1995) reported that the minority students in their studies
were more foreclosed that their Caucasian peers. Abraham (1983) and Rotherham-Borus
(1989) reported no significant differences in foreclosure (see Appendix A for a
comparison of these studies). The inconsistency of results in these studies may be due to
the confounding of a number of ecological variables. Some researchers (e g., Hauser,
1972; Streitmatter, 1988) failed to control for SES. Abraham (1986) and Markstrom
Adams and Adams (1995) used parental education as a measure of SES. Despite their
effort to control for SES, they found that African American and Hispanic adolescents


28
Significance of the Study
Formation of an identity is a defining aspect of personality development (Marcia,
1993). Marcia's (1966, 1976) extension of Erikson's (1963, 1968) theory of identity and
its subsequent revisions (Archer, 1981; Grotevant, Thorbecke, & Meyer, 1982) have
highlighted the importance of identity to adolescents' personality development (Marcia,
1993).
Theoretical Significance
Most attempts to apply the Marcia (1993) model of identity development to
minority adolescents have found that adolescent ethnic minorities tend to have less
progressive identities than do their Caucasian peers (Abraham, 1986; Hauser, 1972;
Markstrom-Adams & Adams, 1995; Streitmatter, 1988). A number of factors have been
proposed to account for this finding, including failure to control for differences in SES
and family constellation as well as differences in acculturation and family values. This
study is designed to determine whether these factors need to be incorporated into
Marcia's model to provide an adequate conception of minority adolescent identity
development.
Practical Significance
Identity research has revealed that the more progressive identity statuses
(achievement and moratorium) are related to more adaptive characteristics of adulthood
than the less progressive statuses (foreclosure and diffusion) (Archer, 1989b). For
example, individuals in the more progressive statuses have demonstrated higher levels of
moral reasoning, autonomy, and internal locus of control (Josselson, 1994; Marcia,
1993). As a result, researchers have recommended the development of intervention


Table 5. Means and Standard Deviations of Acculturation Orientations for African
American and Hispanic Adolescents
41
Orientation
African American
M S
Hispanic
M S
Integration
26.99.
4.27
30.64b
3.89
Accommodation
10.56
3.69
9.90
3.75
Separated
14.01
4.01
12.77
5.06
Marginalized
13.59.
3.78
14.82b
3.62
Note. Row means with different subscripts are significantly different.
Table 6. Ideological Identity Status Percentages bv Ethnicity
African American
Hispanic
Caucasian
Status
(n=82)
(n= 120)
(0 = 98)
Achieved
11
8
13
Moratorium
4
13
9
Foreclosure
16
15
8
Diffusion
2
5
14
Moratorium
64
50
49
undifferentiated
Transition
4
7
6
Note. Column totals may not equal 100% due to rounding.


11
Gender. In a discussion of male and female psychosocial differences, Erikson
(1968) attributed their differential development to socialization. That is, Erikson believed
that because males possess a dominant role over women in society they are more likely to
assert themselves than women. Women, on the other hand, due to their more submissive
cultural role, are more likely to engage in activities that are passive and harmonious.
According to Erikson, the male develops his sense of identity by competition, whereas
the female resolves identity questions through communal, interdependent means.
The first studies to investigate the differences that Erikson proposed supported his
claim (e g., Marcia, 1966, 1976; Schenkel & Marcia, 1972). For example, Hodgson and
Fischer (1979), in a study of 100 late adolescents' identity development, found that males
tend to resolve identity issues via competence and knowledge. Males achieve competence
by pursuing an education and obtaining a career, and knowledge by coming to understand
the world around them. Females, however, were found to resolve identity issues through
different mechanisms. Hodgson and Fischer stated that for females the response to "Who
am I?" revolves around what they are in relation to others. Because of this foundational
research, it was long accepted that gender differences exist in identity development.
Matteson (1993) pointed out, however, that social changes in the past few decades may
necessitate a reexamination of the differences between the identity development of men
and women.
Archer (1989a), in a report of three studies, examined the identity processes of
experimentation and commitment, the domains of identity, and the timing of identity
changes for male and female students in early to late adolescence. Overall, Archer
concluded that males and females undergo identity development in similar ways. As for


25
As this review shows, research on the relationship between familism and identity
development is far from clear. Though differences in familism have been found among
African American, Hispanic, and Caucasian children, research is needed to determine the
degree to which these differences affect identity development. Further, it appears that
males and females experience familism differently. Gender, therefore, must be taken into
account when investigating the relationship between familism and identity development.
Acculturation
Phinney (1990) argued that to complete their identity development, members of
minority groups must resolve two issues not faced by members of the majority culture:
(a) how to negotiate the different expectations of the two cultures and (b) how to deal
with the negative perceptions the majority culture has toward the minority culture.
According to Gonzalez and Cauce (1995), the way an adolescent member of a minority
culture resolves these conflicts depends on the individual's style of acculturation into the
majority lifestyle. On the basis of studies of members of minority groups, Berry (1980)
identified four types of acculturation: (a) assimilation, whereby individuals give up their
identification with their minority culture and identify with the majority culture, (b)
biculturalism, whereby individuals develop positive identifications with their minority
culture and the cultural majority, (c) rejection, whereby individuals identify with their
minority culture and do not identify with the majority culture, and (d) marginalization,
whereby individuals fail to identify with their minority culture or the majority culture.
According to Gonzalez and Cauce (1995) and Peretti and Wilson (1995),
individuals who adopt a bicultural model of acculturation should demonstrate more
progressive identity development than those who adopt other models. That is, because


16
the EOM-EIS than peers living with a single parent. In their study of 80 Caucasian
female high school students, using Marcia's (1966) Identity Status Interview, St. Clair
and Day (1979) found that two-thirds of the 80 senior high school female adolescents in
their study who were in the achieved status came from homes disrupted by death or
divorce. In a study of 265 seventh graders, Streitmatter (1987) found that early
adolescents from intact families were more likely to be classified in the foreclosure status
than their peers from non-intact families. She also found that males seemed more affected
by family disruption than females. Males were more likely to be classified as diffused
than females. Streitmatter suggested that disruption might facilitate identity development
in females, perhaps shortening their moratorium.
Jones and Streitmatter's (1987), St. Clair and Day's (1979), and Streitmatter's
(1987) results may have been affected by their failure to discriminate between single
parent homes due to divorce or death. Recognizing that research has shown that parent-
child relations differ for children with a deceased parent compared to children whose
parents are divorced (e g., Hetherington, 1972), Streitmatter analyzed her data excluding
children with a deceased parent. Finding only one change from the analysis with children
with either a deceased parent or divorced parents, Streitmatter reported the data for the
combined groups. She found a Diffusion x Gender x Family Status interaction but
attributed the result to a loss in sample size.
St. Clair and Day (1979) suggested that the impact of family disruption may be
different for male and female adolescents, and Streitmatter's (1987) study provides
empirical support for the need to study the differential impact of family disruption on
males and females. Streitmatter found that adolescents from intact families had higher


13
Other studies, however, have found difference between genders in identity
development. For example, Streitmatter (1987), in a study of 265 early adolescents, found
that females scored significantly higher than males in total moratorium and ideological
moratorium scales. Streitmatter concluded that such a difference indicated that females
were "more involved in the questioning and searching process than males" (p. 184).
In a similar study, Streitmatter (1988) found that females were significantly more
often found in ideological and interpersonal achievement and ideological moratorium
than males. Given that identity achievement and moratorium are considered more
advanced identity statuses than foreclosure and diffusion, Streitmatter concluded that
females are considered more mature in their identity development than are males at the
same age. Streitmatter (1988) also found a statistically significant Gender x Ethnicity
interaction. Minority males were more likely than their non-minority peers to be in the
moratorium status, a finding that Streitmatter cited as evidence that minority males are
more actively seeking answers to their identity problems. Minority males were, however,
more likely to be in the diffusion status than their peers. This finding is noteworthy, as it
runs counter to the research that has found that minority males are more often in
foreclosure than their Caucasian counterparts.
Markstrom-Adams and Adams (1995) found that females were not as likely as
males to be classified in the diffusion status for ideological and interpersonal identity. A
Gender x Grade interaction demonstrated that older females were less likely than their
younger peers to exhibit lower diffusion scores. Markstrom-Adams and Adams, unlike
Streitmatter (1988), did not find any Gender x Ethnicity interactions.


and diffusion. In each case the degree to which adolescents were given a secure
foundation from which to explore was indicative of progressive identity development.
64
Acculturation orientation was also found to be related to identity development for
African American and Hispanic adolescents. Respondents who endorsed a marginalized
acculturation were more likely to experience identity diffusion than those who developed
other acculturation orientations were. A separation acculturation was also related to
interpersonal foreclosure. Again, it can be suggested that the degree to which an
adolescent is not limited in avenues of exploration is related to the degree to which
progressive identity development occurs.
Recommendations for Future Research
The findings of this study are to be considered in light of its limitations. First, the
conclusions were based on a study of late adolescents enrolled in institutions of higher
education. To expand the generalizability of these results, studies must be conducted with
samples that are more representative of the late adolescent population. Second, as this
was a correlational study of a single assessment of identity, no causal claims can be
made. If these conditions were addressed, it would be of interest to consider the
following:
1. Further research is needed to determine if exposure to an environment that
supports identity exploration is related to progressive identity development in spite of
ethnic differences.
2. In this study, a relationship between acculturation orientation and identity
development was demonstrated. This finding suggests that for minority group members
negotiating a bicultural existence can be an impediment to identity development not


the Mexican Americans had comparatively large extended families who visited
frequently.
24
Further, Keefe (1984) argued that differences between Hispanics and Caucasians
in familism is a matter of quality instead of quantity. That is, she found that Hispanic and
Caucasians had the same degree of familism, but Hispanic familism was associated with
geographic stability, whereas Caucasian familism was associated with geographic
mobility. In other words, Anglo and Mexican Americans are similar in their feelings of
closeness to their families, but Anglos do not share the Mexican Americans' need for the
physical presence of family members.
In a 1988 study, Valenzuela and Dombusch (as cited in Valenzuela & Dombusch,
1994) compared Mexican and Anglo adolescents on familistic structure, attitudes, and
contact. They found the Mexican-origin adolescents tended to be higher than the Anglo
students on all three dimensions of familism. Reguero (1991) investigated the
relationship among demographic variables, familism, and ego-identity status with 180
Puerto Rican residents and 107 Puerto Ricans who had immigrated to the U.S. mainland.
Using multiple regression analysis, Reguero found that the variables sex, age, parents
education and occupation, and immigrant status, and familism accounted for some of the
variance in achievement for all groups. She also found that the variables partially
explained the variance in moratorium for males living in Puerto Rico and male and
female immigrants, as well as partially explaining variance in foreclosure for female
immigrants. The variance in the diffusion status for female respondents, regardless of
group, was explained, in part, by the variables.


CHAPTER 4
DISCUSSION
Drawing from Erikson's theory of psychosocial development, Marcia (1966,
1993) operationalized ego-identity development to include four statuses of identity-
achieved, moratorium, foreclosure, and diffusion. Some researchers (Abraham, 1986;
Hauser, 1972; Markstrom-Adams & Adams, 1995; Streitmatter, 1988), using Marcia's
model, have found disparate rates of identity development among ethnic groups. Cauca
sian adolescents were found to have more progressive identities than their peers who
were members of ethnic minorities do. Other studies, however, have failed to replicate
these findings (Abraham, 1986; Rotheram-Borus, 1989). Explanations of these findings
have centered on inadequate measures of the relation of demographic and cultural
variables with identity development. It was the purpose of this study, therefore, to address
the limitations of previous research to more accurately assess the identity development of
minority adolescents. This final chapter, in part, is a discussion of the results obtained
from the statistical analyses outlined in chapter 3. Implications and limitations of this
study follow. The chapter concludes with recommendations for future research.
Discussion of Results of Statistical Analyses
Researchers who have found differences among ethnic groups in identity
development have largely concluded that the differences between non-Caucasian and
Caucasian adolescents in identity development are due to the lack of opportunities for
57


40
Table 4. Means and Standard Deviations of Familism Type bv Ethnicity
African American Hispanic Caucasian
Type
M
S
M
S
M
S
Structure
20.01a
27.02
11.31.
19.68
6.92b
9.70
Behavior
4.78
1.70
4.83
1.41
4.85
1.52
Attitude
32.94
4.13
33.48
4.24
32.14
3.61
Note. Means with different subscripts are significantly different within the structure row.
Acculturation Orientation
Table 5 presents the means and standard deviations of the acculturation
orientations for African American and Hispanic adolescents. In the comparison of the
means of acculturation orientation styles between African American and Hispanic
adolescents, significant differences were found for the integration (t(l, 177) = 5.95, p <
.0001) or marginalized (t(l, 180) = 2.23, p < .05) styles.
Identity Status Classification
Identity status was determined for each respondent using the method outlined by
Adams, Shea, and Fitch (1979). The number of respondents classified in the statuses of
each of the domains of identity is listed in Table 6 and Table 7. In the ideological and
interpersonal domains the majority of respondents in each ethnic group were classified in
a form of identity moratorium. Differences among the groups on identity variables will be
presented in the next section.


50
Table 13. Summary of GLM Analyses for the Relationship between Ideological Identity
Status and Mother's Parenting Style, with Ethnicity Controlled
Source
ss
df
Mean Sauare
F
E
Achieved
Authoritative
257.29
1
257.29
8.12
0.005
Authoritarian
95.78
1
95.78
3.01
0.08
Permissive
90.61
1
90.61
2.87
0.09
Moratorium
Authoritative
0.49
1
0.49
0.01
0.91
Authoritarian
101.88
1
101.88
2.84
0.09
Permissive
622.93
1
622.93
18.92
0.0001
Foreclosure
Authoritative
66.07
1
66.07
2.19
0.14
Authoritarian
389.76
1
389.76
13.48
0.0003
Permissive
15.21
1
15.21
0.95
0.41
Diffusion
Authoritative
296.27
1
296.27
3.34
0.02
Authoritarian
1.07
1
1.07
0.03
0.82
Permissive
290.78
1
290.78
8.77
0.0033


Table 14. Summary of GLM Analyses for the Relationship Between Interpersonal
Identity Status and Mother's Parenting Style, with Ethnicity Controlled
51
Source
ss
df
Mean Sauare
F
E
Achieved
Authoritative
368.50
1
368.50
12.31
0.0005
Authoritarian
103.82
1
103.82
3.27
0.07
Permissive
58.93
1
58.93
1.86
0.17
Moratorium
Authoritative
9.30
1
9.30
0.30
0.58
Authoritarian
2.00
1
2.00
0.07
0.80
Permissive
355.88
1
355.88
12.51
0.0005
Foreclosure
Authoritative
195.67
1
195.67
5.99
0.02
Authoritarian
80.85
1
80.85
2.50
0.12
Permissive
99.34
1
99.34
3.07
0.08
Diffusion
Authoritative
127.45
1
127.45
4.95
0.03
Authoritarian
70.63
1
70.63
2.78
0.10
Permissive
304.69
1
304.69
12.20
0.0006


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY
DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND HISPANIC
ADOLESCENTS
By
Sean Alan Forbes
December 1999
Chair: Patricia Ashton
Major Department: Educational Psychology
Adolescence is a critical period in the development of psychosocial identity. That
is, as the transition is made from child to adult, the adolescent must reconcile childhood
identifications with the coming expectations of adulthood. The goal is to establish a sense
of individuality in the larger society. This goal is attained through the development of an
identity. The identity crisis includes four statuses: (a) achievementwherein an
adolescent experiences an identity crisis and makes a commitment to alternatives, (b)
moratoriumwherein an adolescent experiences an identity crisis but makes no
commitments, (c) foreclosure wherein an adolescent bypasses an identity crisis by
adopting the commitments of parents or other significant figures, and (d) diffusion
wherein an adolescent avoids the experience of a crisis and postpones commitments. One
of the less considered areas in identity status research has been the comparison of identity
v


ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY
DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND HISPANIC
ADOLESCENTS
By
SEAN ALAN FORBES
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1999

ACKNOWLEDGEMENTS
Patricia Ashton has guided me since I began graduate school. Her wisdom and
patience have not been lost on me. I thank her for her unwavering support and am
indebted for the lessons she has taught me. This dissertation is dedicated to her. I would
also like to thank the other members of my committee for their assistance. James Algina
was a sobering reminder of the diligence needed for any worthwhile pursuit. John
Kranzler, who has known me since I was an undergraduate student, has counseled me
many a time and his advice has served me well. Barry Guinagh has been the kindest
man I have ever met. Greg Neimeyer and Lee Mullally, my outside committee members,
have saved my skin on more than one occasion. Collectively, this committee has been
outstanding.
I must also recognize Randall Scott Hewittthe most natural philosopher I have-
known. Graduate school would not have been complete without him. He and his family
have opened my eyes to other worlds and I will miss their companionship.

TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS ii
ABSTRACT v
CHAPTERS
1 INTRODUCTION 1
Statement of the Problem 3
Purpose of the Study 9
Ecological Variables Related to Identity Status 10
Research Questions 27
Significance of Study 28
2 METHOD 30
Research Participants 30
Instruments 30
Procedures 34
Research Questions 34
3 RESULTS 36
Summary Statistics 36
Results of Statistical Analyses 42
4 DISCUSSION 5/
Discussion of Results of Statistical Analyses 57
Limitations of the Study 62
Implications of the Study 63
Recommendations for Future Research 64
APPENDICES
A OVERVIEW OF IDENTITY STUDIES 66
ui

B INFORMED CONSENT
67
REFERENCES 68
BIOGRAPHICAL SKETCH 74
IV

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY
DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND HISPANIC
ADOLESCENTS
By
Sean Alan Forbes
December 1999
Chair: Patricia Ashton
Major Department: Educational Psychology
Adolescence is a critical period in the development of psychosocial identity. That
is, as the transition is made from child to adult, the adolescent must reconcile childhood
identifications with the coming expectations of adulthood. The goal is to establish a sense
of individuality in the larger society. This goal is attained through the development of an
identity. The identity crisis includes four statuses: (a) achievementwherein an
adolescent experiences an identity crisis and makes a commitment to alternatives, (b)
moratoriumwherein an adolescent experiences an identity crisis but makes no
commitments, (c) foreclosure wherein an adolescent bypasses an identity crisis by
adopting the commitments of parents or other significant figures, and (d) diffusion
wherein an adolescent avoids the experience of a crisis and postpones commitments. One
of the less considered areas in identity status research has been the comparison of identity
v

development among ethnic groups. In the studies investigating the topic, inconsistent
results have been found. Some researchers have reported that African American and
Hispanic adolescents were more likely to be foreclosed than their Caucasian peers. In
light of the differences among ethnic groups in parenting style and familism reported in
some studies, this study was designed to determine whether the inconsistent findings
regarding the relation between identity status and ethnic group could be explained by
parenting style, familism, SES, gender, and/or family configuration. Also, as adolescents
from cultural minorities must negotiate a bicultural existence, it was also of interest to
consider the relation between identity development and acculturation for African
American and Hispanic adolescents. Three hundred college aged adolescents participated
in this study. Of the total, 82 identified themselves as African American and 98 identified
themselves as Caucasian. One hundred twenty participants identified themselves as
Hispanic. Analysis of variance indicated no significant differences among groups in
identity development and no relationship between identity development and familism.
Significant relationships were found between parenting style and acculturation and
identity development.
vi

CHAPTER 1
INTRODUCTION
According to Erikson (1963), the formation of an identity is the dominant
psychosocial crisis of adolescence. Commonly referred to as the identity crisis, this
process is characterized by a delay of adult commitments that permits the adolescent to
experiment with various roles and ultimately to commit to available identity alternatives
in ideological and interpersonal domains (Erikson, 1968, p. 159). As the goal of an
identity crisis is the synthesis of the identifications of childhood and the opportunities and
expectations of adulthood, identity resolution is the foundation upon which future adult
decisions are based (Erikson, 1987; Spencer & Markstrom-Adams, 1990). Moreover, the
individual who successfully resolves the crisis becomes a secure, independent adult
prepared to establish intimacy with others, whereas those who are unable to establish an
identity enter adulthood without a firm sense of how they fit into society (Elkind, 1978;
Peretti & Wilson, 1995).
Marcia (1966, 1976), in an extension of Erikson's theory, described four possible
outcomes of the crisis: identity achievement, moratorium, foreclosure, and diffusion.
Identity achievement is defined by a commitment to identity alternatives after a period of
exploration, whereas moratorium is defined as a period of exploration of alternative roles
during which commitment to an identity is delayed. Foreclosure is defined by identity
commitment without prior exploration, whereas diffusion is defined by the unwillingness
to commit to an identity with or without exploration.
l

2
The achievement and moratorium statuses, by involving exploration, are more
progressive identity statuses than foreclosure and diffusion, which do not involve
exploration (Waterman, 1993). That is, adolescents must play an active role in the
resolution of childhood identifications with adult opportunities and expectations; they
must experiment with the identity alternatives available if a positive resolution of the
crisis is to be achieved.
Taking an active part in resolving their identity crisis often leaves the adolescent
open for problems. For example, adolescents classified in the moratorium status are more
prone to anxiety than those in foreclosure because of the lack of commitment in their
lives (Marcia, 1976; Sterling & Van Horn, 1989). Yet, the overall outcome of being
identity achieved or in moratorium is positive. That is, by taking an active role in their
identity crises, adolescents in the moratorium or achievement status are psychologically
healthier than those in diffusion and foreclosure. For example, those in moratorium and
achievement are less likely than those in foreclosure to be authoritarian or to use
stereotypical thinking (Marcia & Friedman, 1970; Streitmatter & Pate, 1989). Those in
achievement and moratorium are more likely to possess an internal locus of control
(Dellas & Jernigan, 1987), and to be more autonomous than those in foreclosure and
diffusion (Andrews, 1973). In the areas of cognitive performance and cognitive style,
those in the achieved and moratorium statuses often perform at a higher level than those
in foreclosure and diffusion. For example, adolescents in the achievement status have
been found to have better study habits (Waterman & Waterman, 1975) and higher grade
point averages than those in other statuses (Cross & Allen, 1970, Raphael, 1977).
Waterman and Waterman (1975) found that individuals in achievement and moratorium

3
were more reflective and less impulsive than those in foreclosure and diffusion, and
Adams, Abraham, and Markstrom (1987) found that individuals in the higher statuses
were less self-conscious than those in foreclosure and diffusion.
Statement of the Problem
As individuals move through adolescence, most tend toward the progressive
statuses of achievement and moratorium rather than the regressive statuses of diffusion
and foreclosure. Several studies, however, have suggested that progressive identity
development is more common for Caucasian adolescents than for their African American
and Hispanic peers.
In a longitudinal study of ideological identity in 22 low socioeconomic status
(SES) African American and Caucasian high school males, Hauser (1972) administered a
Q-sort annually for 3 years and obtained correlations within each annual Q-sort and
between each annual Q-sort to estimate the consistency within aspects of the students'
self-image and change over time in their self-image. From these correlations, Hauser
concluded that, compared to the Caucasian students, the African American students were
more foreclosed than the Caucasian students.
Several problems in the Hauser (1972) study raise questions about the validity of
his findings. First, the study consisted of a convenience sample of 22 matched African
American and Caucasian students. The small size of the sample and the inability to
determine the adequacy of the matching contribute to doubts about the validity of the
findings. Second, the author used a Q-sort of undetermined reliability rather than the
traditional interview or questionnaire measures of identity. Third, the reliability of the
differences between the correlations in the African American and Caucasian groups is

4
unclear. Given these weaknesses in Hauser's study, the validity of his conclusion that the
African American students were more foreclosed than their Caucasian counterparts must
be questioned. Furthermore, the study was conducted over 30 years ago (between 1962
and 1967). Hauser referred to students' comments during the interview that suggested that
racial restrictions on the African American students' daily lives (e g., restrictions in jobs,
social life, recreation, and housing), limitations in work opportunities, and the lack of role
models who had succeeded in the majority culture may have contributed to their sense of
foreclosure. The greater social and economic opportunities available to African
Americans today may have reduced that sense of foreclosure.
Despite the weaknesses in Hauser's study, a more recent study by Abraham
(1986) supported Hauser's findings regarding the relationship between ethnic-minority
status and identity foreclosure. Abraham investigated the ideological and interpersonal
identity of 841 Mexican American and Caucasian students in grades 9-12 at a high school
in the southwest United States. Respondents completed the Extended Objective Measure
of Ego Identity Status (EOM-EIS) (Grotevant & Adams, 1984), a 64-item measure of
identity in four areas of ideological identity (occupation, religion, politics, and lifestyle)
and interpersonal identity (gender roles, dating, friendship, and recreation). In a
multivariate analysis of covariance, Abraham used mothers' and fathers' education as
covariates to control for differences in SES in the two groups. The results indicated
Caucasians and Mexican Americans differed in ideological identity. A greater percentage
of Mexican American respondents were foreclosed than their Caucasian counterparts.
Abraham concluded, therefore, that the Mexican American adolescents seemed to be
more receptive to their parents' views and ideals on politics, religion, occupation, and

philosophical lifestyle. No differences between ethnic groups were found on
interpersonal identity.
5
Abraham (1986) offered two alternative explanations for her findings. First, she
suggested that Hauser's belief that the students' ethnic minority status may have limited
the range of their exposure to ideological options. Second, she noted that differences in
parenting style might account for the results, though this variable was not included in the
analysis. Abraham concluded that the limitations in options and parenting style may have
contributed to the greater number of the Mexican American students in the foreclosure
status. Selection of participants from a single school raises questions, however, about the
generalizability of Abraham's findings.
Administering the EOM-EIS to 367 students in grades 7 and 8 in a junior high
school in the Southwest, Streitmatter (1988) compared the identity development of
Caucasians and an ethnic minority group composed of Hispanic Americans, African
Americans, Native Americans, and Asian Americans. Like Hauser (1972) and Abraham
(1986), Streitmatter found that the Caucasian adolescents were less foreclosed than the
minority respondents. Unlike the previous studies, however, differences were found in
the interpersonal domain as well as the ideological. Streitmatter suggested that the greater
incidence of foreclosure in the minority groups was the result of minority adolescents'
feeling more uncomfortable in the majority culture than their Caucasian peers. She
proposed that this discomfort led them to accept parental values and reduced their
willingness to experiment with ideas that conflicted with their parents' expectations.
Further, she pointed out that the minority students were bussed into the school, which
might have added to the students' discomfort and inhibited their exploration. She referred

6
to research on parenting style that has suggested that a warm, supportive, and controlling
parenting style is related to the foreclosure status and proposed that the Mexican
American parents may be more likely to use such a style with their children. In light of
her findings, Streitmatter suggested that studies must be done to identify mediators of the
relationships between ethnicity and identity development.
Like Hauser (1972) and Abraham (1986), the Streitmatter study is seriously
flawed. She combined the four ethnic groups into one group in her analysis. Citing
problems such as poor sampling techniques and a lack of control for SES in previous
studies, Markstrom-Adams and Adams (1995) designed a study comparing the identity
development of African American, American Indian, Mexican American, and Caucasian
adolescents to address these design flaws. Their respondent pool roughly approximated
the racial composition of the area from which the respondents were sampled, and SES
was controlled by obtaining the respondents' parents' educational level. Markstrom-
Adams and Adams also investigated the role that age and gender roles play in the
development of identity.
For the Markstrom-Adams and Adams (1995) study, 123 adolescents in grades
10-12 in a high school in the rural Southwest completed the EOM-EIS (Grotevant &
Adams, 1984), the Personal Attributes Questionnaire (Spence, Helmreich, & Stapp,
1974), a gender role measure, the Nowicki-Strickland Locus of Control Scale for children
(Nowicki & Strickland, 1973), and questions about parents' level of education. Like the
previous researchers, Markstrom-Adams and Adams found that the minority adolescents
were more foreclosed than their Caucasian peers. In addition, they found a nonsignificant
trend suggesting that the minority adolescents were also more likely to be classified as

7
diffused than their Caucasian counterparts. Also, relationships between the contextual
variables and identity status emerged. Higher parent education was related to lower
foreclosure rates for ideological identity. In concluding, Markstrom-Adams and Adams
(1995) agreed with the earlier researchers who argued that the greater foreclosure in non-
White adolescents might be because the minority groups are more accepting of
foreclosure than Caucasians, who may require experimentation and autonomous choice in
achieving an identity.
Two other studies, however, did not report differences in foreclosure in minority
and Caucasian adolescents (Abraham, 1983; Rotheram-Borus, 1989). Abraham (1983)
investigated the identity development of 223 Caucasian and Mexican American students
from a rural area in the southwest with the Objective Measure of Ego Identity Status
(OM-EIS), a 24-item measure of ideological identity, including the areas of occupation,
politics, and religion (Adams, Shea, & Fitch, 1979). She found that more Caucasian
students than Mexican American students were classified in diffusion. Although more
Mexican American students than Caucasian students were classified as foreclosed (18.8%
compared to 9.6%) and more Caucasian students were classified as identity achieved
(10.1% compared to 16.4%), these differences were not significantly different.
In her discussion, Abraham (1983) pointed out that SES level has been identified
as a variable related to identity development. A post hoc analysis of the parental
educational level of the students in her study revealed that the parents of the Mexican
American students were lower in educational level than the parents of the Caucasian
students. Because a higher SES level would be expected to be associated with lower
levels of diffusion, she proposed that parental socialization style might be the influential

8
variable in her study. She pointed out that Adams and Jones (1983) found that a maternal
style characterized by control, regulation, and strong encouragement of independence and
a paternal style characterized by praise, approval, and less fair discipline were related to
diffusion, parental styles that might have been more common among the parents of the
Caucasian students compared to the Mexican American students.
Using the OM-EIS (Adams et al., 1979), Rotheram-Borus (1989), in an
investigation of the identity status of 330 African American, Caucasian, Filipino,
Hispanic, and Native American high school students, found that there was little relation
between identity status and ethnicity. Her only significant finding was that Caucasian
students in the 11th and 12th grades were more likely than the other students to have high
moratorium scores on the OM-EIS. The students, however, were not classified in the
moratorium status at a rate different from others. Rotheram-Borus argued that because of
the school's high degree of integration, ethnic differences in the stages of identity did not
exist.
In summary, Hauser (1972), Abraham (1986), Streitmatter (1988), and
Markstrom-Adams and Adams (1995) reported that the minority students in their studies
were more foreclosed that their Caucasian peers. Abraham (1983) and Rotherham-Borus
(1989) reported no significant differences in foreclosure (see Appendix A for a
comparison of these studies). The inconsistency of results in these studies may be due to
the confounding of a number of ecological variables. Some researchers (e g., Hauser,
1972; Streitmatter, 1988) failed to control for SES. Abraham (1986) and Markstrom
Adams and Adams (1995) used parental education as a measure of SES. Despite their
effort to control for SES, they found that African American and Hispanic adolescents

9
scored significantly higher on foreclosure than the Caucasian adolescents in their sample,
suggesting that either parental education is an inadequate measure of SES or that other
factors account for the ethnic group differences in foreclosure.
Some research findings suggest other ecological variables as factors that may
account for the greater foreclosure among minority group members. Spencer and
Markstrom- Adams (1990) pointed out that although both the Caucasian and African
American males in Hauser's (1972) study were from low SES families, the African
American males were more likely to live with single parents than were the Caucasian
males. Hauser noted that the African American students whose fathers were absent were
more foreclosed than the students whose fathers were present. Spencer and Markstrom-
Adams concluded that both family configuration and SES may confound the results of
studies of the identity development of minority adolescents, but they also considered the
alternative hypothesis that, because of acculturation or other cultural factors, foreclosure
may be adaptive for minority youth. Supporting that conclusion, Markstrom-Adams and
Adams (1995) proposed that African American and Hispanic families are likely to define
maturity in terms of support of family values in contrast to Caucasian families who may
define maturity in terms of independence. That is, autonomous choice may be the
defining theme for the development of Caucasian identity, whereas acceptance of family
values may be the defining theme for African American and Hispanic adolescents.
Purpose of the Study
In the comparison of minority and majority adolescent ego-identity development,
conflicting results have been found. Some of the research suggests that minority
adolescents are more likely to be foreclosed, but other studies have failed to support that

10
conclusion. Furthermore, the studies have for the most part failed to include important
ecological variables that may account for the conflicting results. To account for the
development of identity of ethnic minorities, the role of all of these variables must be
examined. The purpose of this study, therefore, was to investigate whether the ecological
variables of SES, acculturation, family characteristics, and gender are related to the ego-
identity development of ethnic minority and Caucasian adolescents. Each of these
ecological variables and their potential relationship to identity development will be
discussed in the sections that follow.
Ecological Variables Related to Identity Status
Identity status has been found to be related to a number of ecological variables
(Marcia, 1993). In response to findings that ethnic minority adolescents tend toward less
progressive identity statuses than their Caucasian peers, several influences on adolescent
development have been offered as explanation: demographic variables, home culture
variables, and the presence of an acculturation process in ethnic minority members.
Demographic Variables
Differences in identity development may be explained, in part, by the
demographic variables gender, SES, and family configuration. Erikson (1963) argued that
males and females experience identity development differently. Limited evidence
suggests that identity development in low SES environments is not as progressive as in
higher SES environments (Markstrom-Adams & Adams, 1995). As for family
configuration, relations have been found between the construct and identity development,
although the direction of the relationship is far from clear (Marcia, 1993).

11
Gender. In a discussion of male and female psychosocial differences, Erikson
(1968) attributed their differential development to socialization. That is, Erikson believed
that because males possess a dominant role over women in society they are more likely to
assert themselves than women. Women, on the other hand, due to their more submissive
cultural role, are more likely to engage in activities that are passive and harmonious.
According to Erikson, the male develops his sense of identity by competition, whereas
the female resolves identity questions through communal, interdependent means.
The first studies to investigate the differences that Erikson proposed supported his
claim (e g., Marcia, 1966, 1976; Schenkel & Marcia, 1972). For example, Hodgson and
Fischer (1979), in a study of 100 late adolescents' identity development, found that males
tend to resolve identity issues via competence and knowledge. Males achieve competence
by pursuing an education and obtaining a career, and knowledge by coming to understand
the world around them. Females, however, were found to resolve identity issues through
different mechanisms. Hodgson and Fischer stated that for females the response to "Who
am I?" revolves around what they are in relation to others. Because of this foundational
research, it was long accepted that gender differences exist in identity development.
Matteson (1993) pointed out, however, that social changes in the past few decades may
necessitate a reexamination of the differences between the identity development of men
and women.
Archer (1989a), in a report of three studies, examined the identity processes of
experimentation and commitment, the domains of identity, and the timing of identity
changes for male and female students in early to late adolescence. Overall, Archer
concluded that males and females undergo identity development in similar ways. As for

12
the identity processes, in two of the studies, males and females differed only in the
foreclosure status; males were more likely to be foreclosed than females. In the third
study no differences were found between males and females in the identity statuses.
Archer (1989a) found, however, that males were more likely than females to be
diffused with regard to the family roles domain, whereas females were more likely to be
classified in the foreclosure, moratorium, and achievement statuses. No differences were
detected in the second study. In the third study, Archer found that males were more likely
to be classified in the foreclosure status, and females in the diffusion status with regard to
political ideology.
As for gender differences in the timing of identity development, Archer (1989a)
found in two of the studies that no differences existed. In the third study, however, a
Grade x Gender interaction was found. Females were more likely than males to be
classified in the moratorium status in grades 6, 10, and 12. For identity achievement,
males were more likely than females to be so at grade 12.
Overall, Archer (1989a) concluded that the differences between the genders were
slight. On the basis of her studies, she argued that the assumption that males and females
experience identity development differently should be discarded. Matteson (1993)
proposed much the same argument as Archer. That is, Matteson pointed out that the
relationships found between identity status and other psychological constructs for males
were also found for females. For example, men and women classified in the moratorium
status reported the highest levels of anxiety, and males and females in foreclosure were
more likely than those in other statuses to defer to authority (Marcia & Friedman, 1970).

13
Other studies, however, have found difference between genders in identity
development. For example, Streitmatter (1987), in a study of 265 early adolescents, found
that females scored significantly higher than males in total moratorium and ideological
moratorium scales. Streitmatter concluded that such a difference indicated that females
were "more involved in the questioning and searching process than males" (p. 184).
In a similar study, Streitmatter (1988) found that females were significantly more
often found in ideological and interpersonal achievement and ideological moratorium
than males. Given that identity achievement and moratorium are considered more
advanced identity statuses than foreclosure and diffusion, Streitmatter concluded that
females are considered more mature in their identity development than are males at the
same age. Streitmatter (1988) also found a statistically significant Gender x Ethnicity
interaction. Minority males were more likely than their non-minority peers to be in the
moratorium status, a finding that Streitmatter cited as evidence that minority males are
more actively seeking answers to their identity problems. Minority males were, however,
more likely to be in the diffusion status than their peers. This finding is noteworthy, as it
runs counter to the research that has found that minority males are more often in
foreclosure than their Caucasian counterparts.
Markstrom-Adams and Adams (1995) found that females were not as likely as
males to be classified in the diffusion status for ideological and interpersonal identity. A
Gender x Grade interaction demonstrated that older females were less likely than their
younger peers to exhibit lower diffusion scores. Markstrom-Adams and Adams, unlike
Streitmatter (1988), did not find any Gender x Ethnicity interactions.

14
In summary, early theory and research of gender differences in identity
development pointed to clear differences (Erikson, 1968; Marcia, 1966, 1976). Males
were perceived as focused on ideological identity development. Women were seen as
focused on interpersonal identity development. Later studies (Archer, 1989a; Markstrom-
Adams & Adams, 1995; Streitmatter, 1988) suggest that differences between males and
females in identity development, when present, are subtle.
Socioeconomic status. Minority adolescents are more likely to come from low-
income families than their Caucasian peers (Spencer & Dombusch, 1990). Although few
studies have considered the impact of low SES on identity development, limited evidence
suggests a relation between the constructs. For example, Markstrom-Adams and Adams
(1995) found that higher paternal and maternal education was related to lower ideological
foreclosure (the correlation with maternal education was -.24 and the correlation with
paternal education was -.22). No relationship was found between parents' education and
interpersonal identity.
Berry (1980) hypothesized that minority group members with low SES are more
likely to adopt an assimilation or separation style of acculturation rather than an
accommodation style. Because of limited resources, these individuals tend to see the
world as difficult to master. They may also avoid identification with the majority culture
or reject their own culture in favor of the majority culture. Negy and Woods (1992), in a
study of 339 Mexican American college students, found that acculturation was
moderately related to SES (r = .44, p = .0001). That is, the greater the identification witn
the majority culture, the higher the SES of the student tended to be. Negy and Woods,
however, did not determine if those who identified with the majority culture also

identified with the minority culture. As a result, it was impossible to differentiate
between assimilated and the accommodated respondents.
15
As a great number of minority children come from low SES backgrounds, many
do not have access to the same economic resources as their higher SES peers and, as a
result, may adopt a marginalized or rejected acculturation orientation. This puts these
adolescents at greater risk of concluding that the expectations of society are incompatible
with their own, which may result in the formation of a weaker identity than higher SES
peers. Research is needed to investigate Berry's (1980) hypothesis and to determine
whether acculturation status mediates the relationship between SES and adolescents'
identity status.
Family configuration. African American and Hispanic minority adolescents are
more likely to grow up in a single parent home than are their Caucasian peers (U. S.
Bureau of the Census, 1994). Family configuration differences may be related to patterns
of identity development among cultural groups, but research on the relationship between
family configuration and identity has yielded inconsistent results. For example, Jordan
(1970) found that male college students from divorced families were more likely to be
identity diffused than were students from intact families. Other studies, however, do not
support the claim that living in an intact family is related to progressive identity
development. Grossman, Shea, and Adams (1980), in a study of 294 college students,
found that males from divorced families possessed higher identity achievement scores
than males in intact families and females in general. In a study of 467 students from ages
12 to 18, Jones and Streitmatter (1987) found that male and female high school students
from intact families were more foreclosed on the ideological and interpersonal scales of

16
the EOM-EIS than peers living with a single parent. In their study of 80 Caucasian
female high school students, using Marcia's (1966) Identity Status Interview, St. Clair
and Day (1979) found that two-thirds of the 80 senior high school female adolescents in
their study who were in the achieved status came from homes disrupted by death or
divorce. In a study of 265 seventh graders, Streitmatter (1987) found that early
adolescents from intact families were more likely to be classified in the foreclosure status
than their peers from non-intact families. She also found that males seemed more affected
by family disruption than females. Males were more likely to be classified as diffused
than females. Streitmatter suggested that disruption might facilitate identity development
in females, perhaps shortening their moratorium.
Jones and Streitmatter's (1987), St. Clair and Day's (1979), and Streitmatter's
(1987) results may have been affected by their failure to discriminate between single
parent homes due to divorce or death. Recognizing that research has shown that parent-
child relations differ for children with a deceased parent compared to children whose
parents are divorced (e g., Hetherington, 1972), Streitmatter analyzed her data excluding
children with a deceased parent. Finding only one change from the analysis with children
with either a deceased parent or divorced parents, Streitmatter reported the data for the
combined groups. She found a Diffusion x Gender x Family Status interaction but
attributed the result to a loss in sample size.
St. Clair and Day (1979) suggested that the impact of family disruption may be
different for male and female adolescents, and Streitmatter's (1987) study provides
empirical support for the need to study the differential impact of family disruption on
males and females. Streitmatter found that adolescents from intact families had higher

17
total scores (combined ideological and interpersonal domain) on foreclosure than
adolescents from nonintact families. From post hoc analyses she found that the family
status effect held for ideological foreclosure but not for interpersonal foreclosure, and she
found a significant Family Status x Gender interaction indicating that female adolescents
from intact homes had lower interpersonal foreclosure scores than female students from
disrupted homes, whereas male students from intact homes had higher foreclosure scores
than males in non-intact homes. She also found a significant Family Status x Gender
interaction for diffusion. Male students from intact homes had lower ideological diffusion
scores than male students from non-intact homes, and females from intact homes had
higher ideological diffusion scores than females with only one parent, in the home. She
also found that the overall relationship between family status and identity development
was greater for male students than for female students. She speculated that given the
likelihood that most of these families were headed by women, the lower foreclosure
scores may reflect the male adolescents' rejection of failed family values, whereas their
higher diffusion scores may indicate that they were unable or unwilling to choose an
identity. Streitmatter further speculated that family disruption may have facilitated
identity development by thrusting the male adolescents into diffusion earlier than their
peers in intact homes.
In sum, research has demonstrated that family configuration is related to identity
development. However, the kind of household (i.e., intact, non-intact), as well as the
gender of the child, plays a role in relationship between family configuration and identity
development. Therefore, research that divides family configuration into separate
categories for (a) single parent as the result of death, (b) single parent due to divorce, (c)

18
intact families, (d) single parent never married, and (e) reconstituted families and that
analyzes the data separately for males and females and controls for SES is needed to
clarify the relationship between family configuration and identity development.
Family Characteristics
Differences in the family environment of minority and majority adolescents may
explain, in part, the different identity development of the ethnic groups. Marcia (1993)
argued that individuals classified in the high identity statuses often come from families
where ideological and interpersonal differentiation between the child and parents is
supported. Those classified in the lower identity statuses, in contrast, tend to come from
families that either encourage conformity to family values or from families in which the
parents distance themselves from the ideological and interpersonal development of the
child. Significant differences have been found between minority cultures and the majority
culture in parenting style (Holtzman, 1982) and familism (Mindel, 1980).
Parenting style. Baumrind (1971, 1991) identified three forms of parenting: (a)
authoritarian whereby parents use a controlling and disciplinarian style to raise their
children, (b) authoritative whereby parents encourage independence but a respect for
rules in their children and, (c) laissez-faire whereby parents do not place limitations or
hold expectations for their children's behavior. Maccoby and Martin (1983) further
divided the laissez-faire style of parenting into two distinct styles: (a) permissive-
indifferent, whereby parents are not involved in their children's lives and (b) permissive-
indulgent, whereby parents give in to their children's demands and do not place
restrictions on their behavior.

Parenting style is significantly related to SES (HofF-Ginsberg & Tardif, 1995).
Parents in the lower SES tend to use an authoritarian style of parenting, whereas middle
19
SES parents tend to use an authoritative style of parenting (HofF-Ginsberg & Tardif,
1995). As minority adolescents are more likely to have low-income families than their
Caucasian peers (Spencer & Dombusch, 1990), minority adolescents tend to be raised
with an authoritarian style of parenting. Caucasian adolescents, in contrast, tend to be
raised with an authoritative style. This difference may be related to patterns of identity
development among cultural groups. For example, Waterman (1993) hypothesized that,
because of the controlling power authoritarian-style parents exert, their children are likely
to begin identity development in foreclosure. In contrast, children of permissive parents
tend toward identity diffusion because of their parents' lack of expectations. Waterman
was less certain about the likely outcomes of authoritative parenting, but he seemed to
suggest that the support and caring these parents give their children might enable them to
delay commitment, thus enabling them to experience moratorium.
Research supports Waterman's belief that an authoritative parenting style is
related to moratorium. For example, Quintana and Lapsley (1990), in a study of 101
undergraduates, found an indirect relationship between the perceived parenting styles and
identity statuses of adolescents mediated by individuation from parents. According to
their results, a parenting style that encouraged independence tended to produce positive
individuation from the family for adolescents. This type of attachment gave the
adolescents the necessary freedom to differentiate from their families and succeed in
forming high status identities. Controlling parenting styles, however, resulted in a

20
negative individuation from families, and these adolescents were prone to develop lower
level identities than their peers from less restrictive homes.
Direct relationships have also been found between parenting style and identity
development. La Voie (1976) studied 120 students in grades 10, 11, and 12 and found
high identity males came from homes with less regulation and control and more praise by
parents than their lower identity status peers. Females with a high identity status came
from homes with less maternal control and greater encouragement to discuss personal
problems with their parents than did their low identity peers. Quintana and Lapsley
(1987) found similar results. In a study of 101 undergraduates, they found that the
children of families that encouraged communication and were not controlling were better
prepared to deal with their ensuing crises. Quintana and Lapsley presumed that their
family background prepared them to freely explore identity alternatives. Adolescents
from controlling homes were more likely to be classified in the lower identity statuses
than were their peers from less controlling homes. However, in two studies of seventh-
and eleventh-grade students, Enright, Lapsley, Drivas, and Fehr (1980) found that only
fathers' parenting style was related to identity development. In the first study of 262
respondents, Enright et al. found that males with fathers who practiced a democratic style
of parenting-analogous to Baumrind's authoritative parenting-tended to be classified in
higher identity statuses than their peers with autocratic or permissive fathers. Females
whose fathers practiced an autocratic parenting style, however, were more often
classified in the higher identity statuses than were peers with democratic or permissive
fathers. No relationship was found between mothers' parenting style and identity
development. In the second study of 168 respondents, Enright et al. found that both males

21
and females with democratic fathers were more often classified in the higher identity
statuses than were their peers with autocratic or permissive fathers. As in the first study,
no relationship was found between mothers' parenting style and identity development.
In all, research supports the idea that parenting style is related to identity
development. Precisely how the two are related, however, is far from clear. An
authoritative parenting style has been consistently demonstrated to be related to high
identity status in males. Yet, both authoritative and authoritarian parenting styles have
been found to be related to high identity status in females. The statistical relationship
between identity status and permissive parenting, however, has not been considered.
Fathers' parenting style has been shown to be related to identity development in
adolescents. Mothers' parenting style has also been shown to be related to identity
development in adolescents, yet not as consistently as a father's parenting style.
Therefore, the gender of the adolescent and parent must be taken into account when
assessing the relation between parenting style and identity development.
Familism. Markstrom-Adams and Adams (1995) pointed out that identity
differences among the ethnic groups they studied did not occur in interpersonal identity
and suggested that the emphasis on family and connectedness among ethnic minority
groups fosters the development of interpersonal identity. For example, Mexican
Americans are characterized as living in close proximity to many relatives with whom
they engage in frequent interaction and mutual support, and these behaviors are believed
to be supported by strong emotional ties to family and acceptance of family values
(Keefe, 1984). Caucasians, in contrast, are considered to be less likely to have extended
family living nearby or to strongly value such relationships. Research, however, has

22
raised questions about the extent of familism among Mexican Americans. Although
Farris and Glenn (1976) and Chandler (1979) found greater familism among Mexican
American than Anglo American groups, the differences were not large. Also, contrary to
the belief that Mexican Americans are high in familism, most of the Mexican Americans
disagreed with most of the familism items.
Valenzuela and Dombusch (1994) defined familism as the structural, attitudinal,
and behavioral dimensions of family interactions. The structural aspect of familism is
characterized by the presence of a nuclear and extended family, whereas the attitudinal
aspect refers to concern for the welfare of the family. The behavioral dimension refers to
the amount of contact with family members. Valenzuela and Dombusch argued that
familism can serve as a form of social capital (Coleman, 1988). That is, family
connectedness provides children with supportive relationships that enable them to
develop positive mental health and to perform well in school. Valenzuela and Dombusch
based their ideas on the work of Velez-Ibanez and Greenburg (1992), who argued that
Mexican American children are raised in dense extended families that socialize the
children to form strong emotional attachments with a wide variety of relatives who tend
to live in nearby neighborhoods, and on research by Velez (as cited in Velez-Ibanez &
Greenberg), who found that Mexican American mothers and their relatives had greater
contact with infants than Caucasian mothers and relatives and the Mexican American
children were in frequent contact with relatives.
Valenzuela and Dombusch (1994) analyzed surveys of 2,666 Anglo and 492
Mexican students in 1987-88 and found that the Mexican American students were higher
in familism than Caucasian students. They also found that for Mexican-origin students

23
neither parent occupation nor familism, when considered separately, was related to
students' grades but when parent education was high, familism was significantly related
to Mexican origin students' grades. Valenzuela and Dombusch Qonjectured that the
patriarchal ideology of the traditional Mexican family may limit female students'
development, thereby suggesting the need to examine gender differences in the effect of
familism and parent education on development.
Other researchers have found significant differences in familism among African
Americans, Hispanics, and Caucasians. Mindel (1980) reported that Hispanics
demonstrated the highest degree of familism followed by African Americans and
Caucasians, in that order. Farris and Glenn (1976) and Chandler (1979) also reported that
Hispanics possess higher degrees of familism than Caucasians. The familism scale used
by Chandler, however, consisted of the following four items: (a) Nothing in life is worth
the sacrifice of moving away from vour parents, (b) When the day comes for a voung
man to take a job, he should stay near his parents, even if it means losing a good job
opportunity, (c) When young people get married, their main loyalty still belongs to their
parents, and (d) When you need help of any kind, you can depend only on members of
the family to help you out (p. 157).
Using an expanded familism measure consisting of nine items with a sample of
381 Mexican Americans and 163 Anglo Americans, Keefe (1984) obtained results that
contradicted those of Chandler (1979) and Farris and Glenn (1976). She found that the
Anglo Americans valued family as highly as the Mexican Americans, although the Anglo
Americans were more likely to have a very small extended family or none at all, whereas

the Mexican Americans had comparatively large extended families who visited
frequently.
24
Further, Keefe (1984) argued that differences between Hispanics and Caucasians
in familism is a matter of quality instead of quantity. That is, she found that Hispanic and
Caucasians had the same degree of familism, but Hispanic familism was associated with
geographic stability, whereas Caucasian familism was associated with geographic
mobility. In other words, Anglo and Mexican Americans are similar in their feelings of
closeness to their families, but Anglos do not share the Mexican Americans' need for the
physical presence of family members.
In a 1988 study, Valenzuela and Dombusch (as cited in Valenzuela & Dombusch,
1994) compared Mexican and Anglo adolescents on familistic structure, attitudes, and
contact. They found the Mexican-origin adolescents tended to be higher than the Anglo
students on all three dimensions of familism. Reguero (1991) investigated the
relationship among demographic variables, familism, and ego-identity status with 180
Puerto Rican residents and 107 Puerto Ricans who had immigrated to the U.S. mainland.
Using multiple regression analysis, Reguero found that the variables sex, age, parents
education and occupation, and immigrant status, and familism accounted for some of the
variance in achievement for all groups. She also found that the variables partially
explained the variance in moratorium for males living in Puerto Rico and male and
female immigrants, as well as partially explaining variance in foreclosure for female
immigrants. The variance in the diffusion status for female respondents, regardless of
group, was explained, in part, by the variables.

25
As this review shows, research on the relationship between familism and identity
development is far from clear. Though differences in familism have been found among
African American, Hispanic, and Caucasian children, research is needed to determine the
degree to which these differences affect identity development. Further, it appears that
males and females experience familism differently. Gender, therefore, must be taken into
account when investigating the relationship between familism and identity development.
Acculturation
Phinney (1990) argued that to complete their identity development, members of
minority groups must resolve two issues not faced by members of the majority culture:
(a) how to negotiate the different expectations of the two cultures and (b) how to deal
with the negative perceptions the majority culture has toward the minority culture.
According to Gonzalez and Cauce (1995), the way an adolescent member of a minority
culture resolves these conflicts depends on the individual's style of acculturation into the
majority lifestyle. On the basis of studies of members of minority groups, Berry (1980)
identified four types of acculturation: (a) assimilation, whereby individuals give up their
identification with their minority culture and identify with the majority culture, (b)
biculturalism, whereby individuals develop positive identifications with their minority
culture and the cultural majority, (c) rejection, whereby individuals identify with their
minority culture and do not identify with the majority culture, and (d) marginalization,
whereby individuals fail to identify with their minority culture or the majority culture.
According to Gonzalez and Cauce (1995) and Peretti and Wilson (1995),
individuals who adopt a bicultural model of acculturation should demonstrate more
progressive identity development than those who adopt other models. That is, because

2(¡
bicultural adolescents integrate the expectations of their minority culture and the majority
culture and are less vulnerable to negative perceptions of their minority culture because
of their identification with it, they should have less difficulty developing the integrated
sense of self characteristic of the achieved status.
Although Streitmatter (1988) argued that minority adolescents "feel less comfort
in the [majority culture] and as a result are more likely to conform to prescribed values
and expectations ... of the adults in their lives" (p. 344), researchers have not examined
the relationship between Marcia's (1976) identity statuses and the degree to which
minority adolescents feel out of place in the majority culture. The findings of Mehan,
Hubbard, and Villanueva (1994), however, support the idea that adolescents with a
bicultural acculturation should demonstrate more progressive identity development than
those with other acculturation orientations. Mehan et al. investigated the effects of a
program to prepare underachieving African American and Hispanic adolescent students
for college admission by placing them in classrooms with high achieving peers. Most of
the students adopted what Gibson (1988) called accommodation without assimilation
(similar to Berry's, 1980, bicultural orientation) toward academic tasks. That is, "they
affirm[ed] their cultural identities while at the same time recognize[d] the need to
develop certain cultural practices, notably achieving academically, that are acceptable to
the mainstream" (Mehan et al., 1994, p. 105). By doing so these students resolved the two
conditions set forth by Phinney (1990) for the development of an identity: (a) negotiate
the different expectations of the cultures and (b) deal with the negative perceptions the
majority culture has toward the minority culture. Also, as Gonzalez and Cauce (1995)
and Peretti and Wilson (1995) argued, resolving the expectations of both the minority and

27
majority cultures allows for the development of an integrated sense of self, the hallmark
of the achievement status.
In summary, a major task for minority adolescents is the resolution of conflicting
expectations of the minority and majority cultures. As minority adolescents confront the
identity crisis, they come to see themselves in one of four ways: estranged from the
minority culture and solely a part of the majority (assimilated), identified with both the
minority and majority culture (bicultured), estranged from the majority culture and solely
a part of the minority (rejected), or estranged from both cultures (marginalized). As the
development of a higher identity is built on an integrated sense of self (Phinney, 1990), it
is plausible that minority adolescents who adopt the bicultural orientation of acculturation
are more likely to reach the achieved identity status.
Research Questions
Some studies have reported a relationship between ethnicity and identity, and
others have not. None of the studies has adequately controlled for the demographic
variables of SES or family structure, which have been found to be related to identity
status. Further, as this review shows, cultural and family process variables (i.e., accul
turation, familism, and parenting style) may be related to identity status. The purpose of
this study was to examine the relationships of these demographic, cultural, and family
variables to identity status. The research questions addressed in this study were:
1. Is ethnicity related to identity status?
2. Is ethnicity related to identity status after gender, SES, and family status are
controlled?
3. Is identity status related to familism?
4. Is identity status related to parenting style?
5. Is identity status related to acculturation for African American and Hispanic
adolescents?

28
Significance of the Study
Formation of an identity is a defining aspect of personality development (Marcia,
1993). Marcia's (1966, 1976) extension of Erikson's (1963, 1968) theory of identity and
its subsequent revisions (Archer, 1981; Grotevant, Thorbecke, & Meyer, 1982) have
highlighted the importance of identity to adolescents' personality development (Marcia,
1993).
Theoretical Significance
Most attempts to apply the Marcia (1993) model of identity development to
minority adolescents have found that adolescent ethnic minorities tend to have less
progressive identities than do their Caucasian peers (Abraham, 1986; Hauser, 1972;
Markstrom-Adams & Adams, 1995; Streitmatter, 1988). A number of factors have been
proposed to account for this finding, including failure to control for differences in SES
and family constellation as well as differences in acculturation and family values. This
study is designed to determine whether these factors need to be incorporated into
Marcia's model to provide an adequate conception of minority adolescent identity
development.
Practical Significance
Identity research has revealed that the more progressive identity statuses
(achievement and moratorium) are related to more adaptive characteristics of adulthood
than the less progressive statuses (foreclosure and diffusion) (Archer, 1989b). For
example, individuals in the more progressive statuses have demonstrated higher levels of
moral reasoning, autonomy, and internal locus of control (Josselson, 1994; Marcia,
1993). As a result, researchers have recommended the development of intervention

29
strategies to foster successful resolution of the identity crisis (Josselson, 1994; Rotheram-
Borus & Wyche, 1994; Waterman, 1994).
It is not clear whether such intervention strategies are warranted for adolescents
from minority groups. As Erikson (1963) noted, culture affects the development of
identity. That is, cultural expectations and practices encourage and limit adolescents'
opportunity for exploration and commitment in the various identity domains. Given that
researchers have found that African American and Hispanic adolescents are more
foreclosed than their Caucasian peers, Spencer and Markstrom-Adams (1990) stated that
the foreclosure status may be adaptive in certain cultures: "Exploration of identity may
not be desirable for such youth when social and ideological roles are clearly defined by
the community" (p. 298). This study was designed to determine whether the relationship
between ethnicity and identity development is attributable to SES, family configuration,
and family values.
It may be inappropriate, however, to conclude that identity intervention is
unnecessary because of differences in cultural expectations. Markstrom-Adams and
Spencer (1994) argued that despite minority cultural support many minority adolescents
are restricted in their ability to explore identity alternatives because of the limitations in
their social environment. That is, a significant portion of minority adolescents grow up in
such impoverished, discriminatory environments that they cannot take advantage of the
role models available that are necessary for adequate identity exploration. Consequently,
it is important to determine whether the foreclosure of minority adolescents is attributable
to economic variables or to cultural values.

CHAPTER 2
METHOD
Research Participants
Three hundred students from a community college and a university in Florida
were recruited to participate in this study. Respondents included 82 African American
students (30 men and 52 women), 120 Hispanic students (50 men and 70 women), and 98
Caucasian (40 men and 58 women) late adolescents with a mean age of 20.8 years.
Instruments
Identity Status
Identity status was assessed using the Extended Objective Measure of Ego
Identity Status III (EOM-EIS HI, Forbes & Ashton, 1996). The EOM-EIS III is a 64-item
questionnaire. Respondents are instructed to indicate the degree to which an item is
reflective of themselves using a 6-point Likert scale, ranging from strongly disagree to
strongly agree. The scale assesses ideological identity in the areas of occupation, politics,
religion, and philosophical lifestyle and interpersonal identity in the areas of gender roles,
dating, friendship, and recreation. Each of the eight subdomains is measured by eight
items with two items reflecting each of the four identity statuses: achievement,
moratorium, foreclosure, and diffusion. Forbes and Ashton obtained the following
Cronbachs alpha coefficients for scores on the four identity statuses: ideological
achievement = .57; moratorium = .68; foreclosure = .64; diffusion = .53; interpersonal
achievement = .62; moratorium = .63; foreclosure = .77; diffusion = .51.
30

31
Acculturation
Acculturation orientation was measured with an adaptation of a questionnaire
developed by Berry, Trimble, and Olmedo (1986) to assess the four types of
acculturation-assimilation, integration, separation, and marginality. The questionnaire
consists of 24 items, with an item representing each acculturation type for each of the
following domains: occupation, dating, friendship, life-style, media, and language.
Responses were made on a 5-point Likert scale, ranging from strongly disagree to
strongly agree.
Parenting Style
Parenting style was assessed using the two forms of Buri's (1991) Parental
Authority Questionnaire (PAQ). The PAQ measures parental authority of mothers and
fathers. Each scale is a 30-item questionnaire designed to measure the degree to which an
individual believes his or her parent(s) exhibited permissive, authoritarian, and
authoritative parenting styles. Responses are made to the items on a 5-point Likert scale.
With a sample of 61 psychology students who completed the questionnaire twice over a
2-week period, Buri obtained the following test-retest reliability coefficients: mother
permissiveness = .75; authoritarianism = .85; authoritativeness = 82; father
permissiveness = .74; authoritarianism = .87; authoritativeness = .85.
Buri (1991) obtained evidence for the construct validity of the scale in a study of
127 undergraduate psychology students. His results were consistent with predictions from
Baumrind's (1971) theory: mother's authoritarianism items were negatively correlated
with mother's authoritativeness and permissiveness items (r = -.38, g < .0005; r = -.48, g
< .0005). Mother's authoritativeness and mother's permissiveness were not significantly

32
correlated (r = .07, g > .10). Father's authoritarianism was negatively correlated with
father's authoritativeness and father's permissiveness (r = -.50, g < .05; r = -.52, g < .05).
Father's authoritativeness and father's permissiveness items were not significantly
correlated (r = 12, g > 10). In the same study, Buri reported the following data on
criterion-related validity: The authoritative subscales were positively related to parental
nurturance (mother r = .56, g < .0005; father r = .68, g < .0005), whereas the authoritarian
subscale was negatively related to parental nurturance (mother r = -.36, g < .0005; father
r = .-53, g < .0005). The permissiveness subscales were not significantly related to
parental nurturance (mother r = .04, g < 10; father r = 13, g < 10), respectively.
In a study of 69 undergraduate psychology students, Buri (1991) assessed the
relationship between social desirability and respondents' answers to the PAQ subscales.
None of the subscales was significantly related to the Marlowe-Crowne Social
Desirability Scale.
Family Configuration
Family configuration was assessed by the respondents' self-report. Participants
were asked to indicate the category that identified their family type on the questionniare.
Categories included dual parent household without stepparents; dual parent household
with stepparents; single-parent household-parents divorced; single-parent household-
parent never married, single-parent household-parent deceased, and an Other category.
Familism
Familism was assessed in each of the three domains of the construct. Familism
behavior was measured using a single item taken from the work of Valenzuela and
Dombusch (1994) that asks respondents to indicate on a 6-point Likert scale "how often

33
[respondents] saw or talked on the telephone with adult relatives who do not live with
them" (p. 24). Familism structure was also measured with a single item on a 6-point scale
used by Valenzuela and Dombusch. It asked the number of relatives that live within an
hour's drive of the respondent. Familism attitudes were measured using a scale developed
by Keefe (1984). The measure is a 9-item, 5-point Likert scale. Keefe (1984) obtained an
internal consistency coefficient of .68 for the items in a study of 381 Mexican American
and 163 Caucasians.
Socioeconomic Status (SESI
The SES of respondents' parents was assessed using Hollingshead's (1975) Four
Factor Index of Social Status (FFISS). The factors of the scale include gender,
occupation, education, and marital status. Gender, however, is not included in the
calculation of social status, though it can be used with males or females. Occupation level
is assessed on a 9-point scale of increasing complexity as determined by the 1970 United
States Census. The educational factor is measured using a 7-point scale that ranges from
less than seventh grade to graduate professional training. Marital status is assessed by
first assigning respondents to one of the following classifications: unmarried individual
(male or female), head of a household (male or female), or family in which both husband
and wife are gainfully employed. Then social status is calculated by multiplying the
scores of the occupation and education factors by factor weights. Scores range from 8 to
66 with higher scores indicating higher social status. When social status is measured for
families with two employed persons, the occupation and education scores of both persons
are considered in the calculation. The Four Factor Index is based on the belief that social
status is a "multidimensional aggregate" (Gottfried, 1985, p. 92). That is, the four

34
factors-gender, occupation, education, and marital status-are combined into a single
index rather than considered as separate factors. The FFISS has been described as a
highly reliable and valid measure of social status (see Hollingshead, 1975). Gottfried
found that the FFISS was more highly correlated with children's developmental outcomes
in a longitudinal (30 months) study of 130 young children and their parents than were
two other SES indices, the Siegel Prestige Scale and the Revised Duncan Socioeconomic
Index. In addition, Gottfried pointed out that the Hollingshead Four Factor Index can be
used to estimate the SES of unmarried heads of households as well as families.
Procedures
Each participant signed the consent form (see Appendix B) and anonymously
completed the instruments that consisted of the demographic questionnaire, EOM-EIS III,
PAQ, acculturation, and familism questionnaires. Data were collected from summer,
1997 to spring, 1999. Respondents were told not to answer any items that they did not
wish to and to ask questions at any time. Administration time took between 45 min to 1
hr.
Research Questions
Research Question #1
The first set of research hypotheses in this study was designed to determine if
ethnic identification was related to identity status classification. Respondents' scores on
the EOM-EIS were used as the dependent variable in the model. The independent
variable was ethnic identification.

35
Research Question #2
The second group of research hypotheses was designed to determine if ethnic
identification was related to identity status when the demographic variables-SES, gender,
family status-were controlled. The dependent variable was the EOM-EIS scores. Ethnic
identification was the independent variable and SES, gender, and family status responses
were covariates.
Research Question #3
The next set of hypotheses considered the relation of familism to identity status
for each of the three ethnic groups. Separate analyses were conducted on the variables to
determine their relationship with identity status. In both sets of analyses respondents'
EOM-EIS scores were the dependent variable; whereas scores on the familism measures
served as predictor variables and ethnic identity was a covariate.
Research Question #4
The next set of hypotheses considered the relation of parenting style to identity
status for each of the three ethnic groups. Separate analyses were conducted on the
variables to determine their relationship with identity status. In both sets of analyses
respondents' EOM-EIS scores were the dependent variable; whereas scores on the PAQ
served as predictor variables and ethnic identity was a covariate.
Research Question #5
This group of hypotheses considered the relation between the different forms of
acculturation orientation available to African American and Hispanic adolescents and
identity status.

CHAPTER 3
RESULTS
This study was designed to investigate the relation between ethnic identification
and identity status before and after controlling for the effect of theoretically significant
demographic variables-gender, SES, and family status. It was also of interest to examine
the relationships among parenting style, familism, and identity status classification. The
final question of interest was whether acculturation orientation is related to identity status
classification. This chapter presents a summary of the statistical findings of this study
including (a) the summary statistics of demographic variables (age, gender, SES, family
status), parenting style, familism, acculturation orientation, and identity status
classification by ethnic group and (b) results of the analyses of the relationships among
the variables of interest.
Summary Statistics
Demographic Variables
Of the 300 respondents who participated in this study, 82 identified themselves as
African American, ¡20 as Hispanic, and 98 as Caucasian. All participants in this study
were between the ages of 18 and 22. The mean age of African American respondents was
20.8 years. The mean ages of Hispanic and Caucasian respondents were 20.8 and 20.6
years, respectively. Forty percent of the respondents were men (n = 120); 60% of the
respondents were women (n = 180). Of African American respondents, 62 (76%)
36

37
Table 1. Family Status Percentages bv Ethnicity
African American
Hispanic
Caucasian
Status
%
%
%
Dual parent-
No step parent
44
66
63
Dual parent-
Stepparents
19
6
17
Single parent-
Never married
12
5
0
Single parent-
Parent divorced
17
19
16
Single parent-
Parent deceased
4
3
2
Other
4
1
2
identified themselves as women, and 20 (24%) identified themselves as men. Sixty-five
Hispanic (54%) and 71(72%) Caucasian respondents identified themselves as womem; 55
(46%) Hispanic and 27 (28%) Caucasian respondents identified themselves as men.
Table 1 displays the percentage of respondents in each of the different family
statuses by ethnic group. A chi-square analysis of the relationship between statuses by
ethnic group resulted in a non-significant relationship.
Table 2 lists the means and standard deviations of respondents' scores on the Four
Factor Index of Social Status (Hollingshead, 1975). An ANOVA indicated a significant

38
Table 2. Means and Standard Deviations of Socioeconomic Status bv Ethnicity
African American
HisDanic
Caucasian
M
S
M S
M
s
42.35a
12.17
43.08, 12.74
49.88b
8.96
Note. Means with different subscripts are significantly different.
difference among the means, F(2, 291) = 12.54, p < .0001. Post-hoc comparisons using
the Tukey method revealed that Caucasian SES scores were significantly different from
both African American and Hispanic SES scores (p < .05). The SES scores of African
American and Hispanic respondents did not differ.
Parenting Style
The means and standard deviations of the different parenting styles are listed by
ethnicity in Table 3. One-way analyses of variance indicated significant differences in
mean scores for mothers' authoritarian parenting style [F(2, 277) = 16.02, p < .0001] and
fathers' authoritarian parenting style [(F(2, 244) = 5.73, p < .0037)].
The Tukey post-hoc comparison of scores indicated that African American and
Hispanic adolescents' mean scores were significantly different from the Caucasian mean
score on mother's and father's authoritarian parenting style (p < .05). The parenting-style
scores of the African American adolescents and Hispanic adolescents did not differ.

39
Table 3. Parenting Style Mean Scores bv Ethnicity
African American Hispanic Caucasian
Style
M
S
M
S
M
S
Mother
Authoritative
41.17
7.32
42.55
6.65
43.47
7.57
Authoritarian
36.99,
6.94
35.54,
6.34
31.38b
7.18
Permissive
29.82
5.74
30.73
5.18
31.44
5.56
Father
Authoritative
37.06
6.93
38.40
6.65
39.23
9.58
Authoritarian
36.99,
6.94
35.53,
8.93
34.95b
11.29
Permissive
26.01
8.12
27.25
8.12
28.21
8.29
Note. Means with different subscripts are significantly different within the authoritarian
rows.
Familism
The means and standard deviations of the forms of familism are listed by ethnic
group in Table 4. Significant differences were found among structure means, F(2, 282) =
9.79, p < .0001, for familism. Post-hoc analysis indicated that African American and
Hispanic adolescents scores were significantly different from Caucasian scores on
familism-structure (p < .05). No significant differences were found between African
American and Hispanic adolescents.

40
Table 4. Means and Standard Deviations of Familism Type bv Ethnicity
African American Hispanic Caucasian
Type
M
S
M
S
M
S
Structure
20.01a
27.02
11.31.
19.68
6.92b
9.70
Behavior
4.78
1.70
4.83
1.41
4.85
1.52
Attitude
32.94
4.13
33.48
4.24
32.14
3.61
Note. Means with different subscripts are significantly different within the structure row.
Acculturation Orientation
Table 5 presents the means and standard deviations of the acculturation
orientations for African American and Hispanic adolescents. In the comparison of the
means of acculturation orientation styles between African American and Hispanic
adolescents, significant differences were found for the integration (t(l, 177) = 5.95, p <
.0001) or marginalized (t(l, 180) = 2.23, p < .05) styles.
Identity Status Classification
Identity status was determined for each respondent using the method outlined by
Adams, Shea, and Fitch (1979). The number of respondents classified in the statuses of
each of the domains of identity is listed in Table 6 and Table 7. In the ideological and
interpersonal domains the majority of respondents in each ethnic group were classified in
a form of identity moratorium. Differences among the groups on identity variables will be
presented in the next section.

Table 5. Means and Standard Deviations of Acculturation Orientations for African
American and Hispanic Adolescents
41
Orientation
African American
M S
Hispanic
M S
Integration
26.99.
4.27
30.64b
3.89
Accommodation
10.56
3.69
9.90
3.75
Separated
14.01
4.01
12.77
5.06
Marginalized
13.59.
3.78
14.82b
3.62
Note. Row means with different subscripts are significantly different.
Table 6. Ideological Identity Status Percentages bv Ethnicity
African American
Hispanic
Caucasian
Status
(n=82)
(n= 120)
(0 = 98)
Achieved
11
8
13
Moratorium
4
13
9
Foreclosure
16
15
8
Diffusion
2
5
14
Moratorium
64
50
49
undifferentiated
Transition
4
7
6
Note. Column totals may not equal 100% due to rounding.

42
Table 7, Interpersonal Identity Status Percentages bv Ethnicity
African American
Hispanic
Caucasian
Status
(n=82)
(n= 120)
(D = 98)
Achieved
7
12
10
Moratorium
13
11
12
Foreclosure
11
11
6
Diffusion
11
7
10
Moratorium
49
51
52
undifferentiated
Transition
7
9
7
Note. Column totals may not equal 100% due to rounding.
Results of Statistical Analyses
For each question, eight analyses were conducted. A .05 familywise error rate was
used to judge significance. However, in reporting analyses, results that were significant at
a conventional .05 Type I error rate are noted. The research questions for this study were
1. Is ethnicity related to identity status?
2. Is ethnicity related to identity status when gender, SES, and family status are
controlled?
3. Is identity status related to parenting style when ethnicity is controlled?
4. Is identity status related to familism when ethnicity is controlled?
5. Is identity status related to acculturation for African American and Hispanic
adolescents?
Question #1
The means and standard deviations for the status analyses are reported by ethnic
group in Table 8. To determine if ethnicity was related to identity status a one-way

43
Table 8. Mean Scores and Standard Deviations of Identity Statuses
African American Hispanic Caucasian
M
S
M
S
M
S
Ideological
Achieved
23.28
3.76
23.46
3.52
23.01
3.25
Moratorium
25.97
3.91
26.74
3.24
26.51
3.54
Foreclosed
20.44
4.74
22.53
4.72
21.63
3.95
Diffused
41.64
27.58
32.81
19.75
27.42
9.73
Interoersonal
Achieved
25.28
4.29
25.23
4.12
25.04
3.99
Moratorium
25.23
4.26
25.37
3.61
25.52
4.21
Foreclosure
25.60
3.89
26.21
3.88
24.70
3.63
Diffused
25.39
4.24
25.74
3.62
24.97
3.52
ANOVA was conducted on the means of each of the identity domain scores for African
American, Hispanic, and Caucasian adolescents. When the Bonferroni-Holm procedure
was used to control for the fact that ethnic group differences were tested on eight
variables, none of the F statistics was significant. When a conventional .05 alpha level
was used, ethnic differences among identity status scores were found for one status-
diffused, ideological, F(2, 287) = 4.13, p = .017.

44
Question #2
The general linear model was used to test ethnic group differences with the
demographic variables controlled (gender, SES, family status). No significant ethnic
group differences emerged when Bonferroni corrections were made for multiple
comparisons, With a conventional .05 alpha level, ideological diffusion was most closely
related to ethnicity, F(2, 256) = 3.63, p = .028. Of the ethnic groups, Caucasian
adolescents reported the highest means on ideological diffusion.
In the research literature, gender, SES, and family status have been identified as
demographic variables related to identity development. However, when these variables
were entered as covariates in the analysis of the relationship between identity status and
ethnicity, and when corrections were made to the p value for multiple analyses, none of
the covariates was found to be significantly related to the identity status domains (see
Tables 9 and 10). The covariates with p values less than .10 were gender in ideological
moratorium (p = .042), ideological foreclosure (p = 0.09), interpersonal achieved (p =
0.008), and interpersonal foreclosure (p = .026), and SES in ideological moratorium (p =
.042) and interpersonal foreclosure (p = .072).
Question #3
The purpose of the third research question was to determine whether identity
status variables were related to familism with ethnicity controlled. The, covariate,
ethnicity, was not found to be related to identity status. With Bonferroni corrections for
multiple comparisons, no significant differences were found between ideological and
interpersonal identity status and forms of familism, with ethnicity controlled. Results are

45
Table 9. ANCOVA Results for Ideological Identity Statuses
Source
ss
df
Mean Sauare
F
E
Achieved
Race
47.51
2
23.75
0.72
0.49
Gender
9.66
1
9.66
0.29
0.59
SES
33.24
1
33.24
1.01
0.32
Family
125.40
5
25.08
0.76
0.58
Moratorium
Race
102.46
2
51.23
1.47
0.23
Gender
145.94
1
145.94
4.18
0.04
SES
29.63
1
29.63
0.85
0.36
Family
175.87
5
35.17
1.01
0.41
Foreclosure
Race
153.62
2
76.81
2.63
0.07
Gender
82.88
1
82.88
2.84
0.09
SES
0.99
1
0.99
0.03
0.85
Family
99.14
5
19.83
0.68
0.64
Diffusion
Race
258.75
2
129.38
3.63
0.28
Gender
88.73
1
88.73
2.49
0.12
SES
0.39
1
0.39
0.01
0.92
Family
180.64
5
36.18
1.01
0.41

46
Table 10. ANCOVA Results for Interpersonal Identity Statuses
Source
ss
df
Mean Square
F
R
Achieved
Race
178.59
2
89.30
2.87
0.06
Gender
222.65
1
222.65
7.15
0.01
SES
12.35
1
12.35
0.40
0.53
Family
250.54
5
50.11
1.61
0.16
Moratorium
Race
2.27
2
1.18
0.04
0.96
Gender
0.08
1
0.08
0.00
0.96
SES
25.50
1
25.50
0.85
0.36
Family
32.47
5
6.49
0.22
0.96
Foreclosure
Race
44.83
2
22.42
0.73
0.49
Gender
154.18
1
154.18
4.99
0.03
SES
100.95
1
100.95
3.27
0.07
Family
113.68
5
22.73
0.74
0.60
Diffusion
Race
34.87
2
17.44
0.66
0.52
Gender
70.25
1
70.25
2.67
0.10
SES
0.99
1
0.99
0.04
0.85
Family
98.28
5
19.66
0.75
0.59

47
listed in Tables 11 (ideological identity statuses) and 12 (interpersonal identity status).
No significant interactions between familism and ethnicity were found.
Table 11. Summary of GLM Analyses for Relationship between Ideological Identity
Status and Familism. with Ethnicity Controlled
Source
ss
df
Mean Square
F
E
Achieved
Family structure
2.87
1
2.87
0.09
0.77
Family behavior
42.73
1
42.73
1.29
0.26
Family attitude
66.50
1
66.50
2.07
0.15
Moratorium
Family structure
41.90
1
41.90
1.14
0.29
Family behavior
105.16
1
105.16
1.03
0.32
Family attitude
49.46
1
49.46
1.39
0.24
Foreclosure
Family structure
103.91
1
103.91
3.51
0.06
Family behavior
45.70
1
45.70
1.53
0.22
Family attitude
169.17
1
169.17
5.75
0.02
Diffusion
Family structure
59.95
1
59.95
1.74
0.19
Family behavior
83.85
1
83.85
2.41
0.12
Family attitude
27.06
1
27.06
0.79
0.38

48
Status and Familism. with Ethnicitv Controlled
Source
ss
df
Mean Square
F
2
Achieved
Family structure
0.41
1
0.41
0.01
0.91
Family behavior
15.95
1
15.95
0.48
0.49
Family attitude
47.39
1
47.39
2.48
0.22
Moratorium
Family structure
26.93
1
26.93
0.90
0.34
Family behavior
134.58
1
134.58
4.97
0.03
Family attitude
25.34
1
25.34
0.87
0.35
Foreclosure
Family structure
28.06
1
28.06
0.88
0.35
Family behavior
93.86
1
93.86
2.83
0.09
Family attitude
62.94
1
62.94
1.94
0.17
Diffusion
Family structure
3.08
1
3.08
0.12
0.73
Family behavior
3.94
1
3.94
0.14
0.71
Family attitude
99.08
1
99.08
3.85
0.05

Question #4
The aim of the fourth research question was to determine whether identity status
variables were related to parenting style scores after controlling for ethnicity. The
covariate, ethnicity, was not found to be related to identity status.
Mothers' parenting style. The F and probability values for the relationship
between identity status and parenting style, with ethnicity controlled, are reported in
Tables 13 and 14. Inspection of Table 13 reveals the following significant positive
relationships between mothers parenting style and ideological identity status:
authoritative style with ideological achievement, F (2, 265) = 8.12, p = .005, permissive
style with ideological moratorium, F (2, 265) = 18.92, p = .0001, authoritarian style with
ideological foreclosure, F (2, 265) = 13 .48, p = .0003, and permissive style with
ideological diffusion, F (2, 265) = 8.77, p = .003.
Significant relationships were also found between mothers parenting style and
interpersonal identity status with ethnicity controlled (see Table 14): authoritative style
with achievement, F (2, 263) = 12.31, p = .0005, permissive style with moratorium, F (2,
263) = 12.51, p = .0005, and diffusion, F (2, 263) = 12.20, p = .0006. Each of these
relationships was also positive.
Analyses were performed to determine if significant interactions existed between
ethnicity and mother's parenting style. The interaction between mother's authoritative
style and ethnicity was significant, F (2, 265) = 5.43, p = .004, for the ideological
achieved status. As a group, African Americans demonstrated the strongest positive
relationship between ideological achievement and mothers authoritative parenting style,
followed by Caucasians then Hispanics.

50
Table 13. Summary of GLM Analyses for the Relationship between Ideological Identity
Status and Mother's Parenting Style, with Ethnicity Controlled
Source
ss
df
Mean Sauare
F
E
Achieved
Authoritative
257.29
1
257.29
8.12
0.005
Authoritarian
95.78
1
95.78
3.01
0.08
Permissive
90.61
1
90.61
2.87
0.09
Moratorium
Authoritative
0.49
1
0.49
0.01
0.91
Authoritarian
101.88
1
101.88
2.84
0.09
Permissive
622.93
1
622.93
18.92
0.0001
Foreclosure
Authoritative
66.07
1
66.07
2.19
0.14
Authoritarian
389.76
1
389.76
13.48
0.0003
Permissive
15.21
1
15.21
0.95
0.41
Diffusion
Authoritative
296.27
1
296.27
3.34
0.02
Authoritarian
1.07
1
1.07
0.03
0.82
Permissive
290.78
1
290.78
8.77
0.0033

Table 14. Summary of GLM Analyses for the Relationship Between Interpersonal
Identity Status and Mother's Parenting Style, with Ethnicity Controlled
51
Source
ss
df
Mean Sauare
F
E
Achieved
Authoritative
368.50
1
368.50
12.31
0.0005
Authoritarian
103.82
1
103.82
3.27
0.07
Permissive
58.93
1
58.93
1.86
0.17
Moratorium
Authoritative
9.30
1
9.30
0.30
0.58
Authoritarian
2.00
1
2.00
0.07
0.80
Permissive
355.88
1
355.88
12.51
0.0005
Foreclosure
Authoritative
195.67
1
195.67
5.99
0.02
Authoritarian
80.85
1
80.85
2.50
0.12
Permissive
99.34
1
99.34
3.07
0.08
Diffusion
Authoritative
127.45
1
127.45
4.95
0.03
Authoritarian
70.63
1
70.63
2.78
0.10
Permissive
304.69
1
304.69
12.20
0.0006

52
Fathers' parenting style. Inspection of Table 15 reveals the following significant
relationships between fathers parenting style and ideological identity status with
ethnicity controlled: authoritative style with foreclosure, F (2, 265) = 14.87, p = .0001,
permissive style with diffusion, F (2, 234) = 11.31, p = .0009. Each relationship was
positive. The same general findings were obtained for the relationship between
interpersonal identity status and father's parenting style (see Table 16). That is, analyses
revealed the following significant relationships between father's parenting and
interpersonal identity status: authoritative style with foreclosure, F (2, 169) = 15.07, p =
.0001, permissive style with diffusion, F (2, 232) = 30.02, p = .0001 Each relationship
was positive. As with the ideological identity statuses, no relationship was found between
father's parenting style and interpersonal achievement or moratorium. The interaction of
father's parenting style and ethnicity was not significant.
Question #5
The purpose of this question was to measure the relationship between identity
status and forms of acculturation orientation. The results of the GLM analyses are
presented in Table 17 (ideological identity) and Table 18 (interpersonal identity).
A significant, positive relationship was found between acculturation orientation
and ideological identity status with ethnicity controlled-marginalized acculturation with
diffusion, F (1, 170) = 12.27, p = .0006. The following positive significant relationships
between acculturation orientation and interpersonal Identity were found: marginalized
acculturation with diffusion, F(l, 173)=18.32, p = .0001, and separated acculturation
foreclosure, F(l, 168) = 24.92, p = .0001. Analyses were conducted to determine whether

53
there was an interaction between acculturation orientation and ethnicity for identity
statuses. No significant interactions were found.
Table 15. Summary of GLM Analyses for Relationship between Ideological Identity
Status and Father's Parenting Style, with Ethnicity Controlled
Source
ss
df
Mean Sauare
F
E
Achieved
Authoritative
114.57
1
114.57
3.59
0.06
Authoritarian
179.03
1
179.03
5.65
0.02
Permissive
105.95
1
105.95
3.31
0.07
Moratorium
Authoritative
9.76
1
9.76
0.26
0.61
Authoritarian
9.15
1
9.15
0.24
0.63
Permissive
233.77
1
233.77
6.38
0.01
Foreclosure
Authoritative
440.73
1
440.73
14.87
0.0001
Authoritarian
51.21
1
51.21
1.63
0.20
Permissive
8.21
1
8.21
0.26
0.61
Diffusion
Authoritative
37.85
1
37.85
1.09
0.30
Authoritarian
29.49
1
29.49
0.79
0.38
Permissive
381.99
1
381.99
11.31
0.0009

54
Table 16. Summary of GLM Analyses for Relationship between Interpersonal Identity
Status and Father's Parenting Style Controlling for Ethnicity
Source
ss
df
Mean Square
F
E
Achieved
Authoritative
108.05
1
108.05
3.45
0.06
Authoritarian
196.54
1
196.54
6.37
0.01
Permissive
149.61
1
149.61
4.87
0.03
Moratorium
Authoritative
6.43
1
6.43
0.21
0.65
Authoritarian
0.21
1
0.21
0.01
0.93
Permissive
100.28
1
100.28
3.33
0.07
Foreclosure
Authoritative
482.96
1
482.96
15.07
0.0001
Authoritarian
9.47
1
9.47
0.28
0.60
Permissive
22.00
1
22.00
0.68
0.41
Diffusion
Authoritative
5.25
1
5.25
0.19
0.66
Authoritarian
45.52
1
45.52
1.74
0.19
Permissive
691.72
1
691.72
30.02
0.0001

Table 17, Summary of GLM Analyses for Relationship between Ideological Identity
Status and Acculturation Orientation, with Ethnicity Controlled
55
Source
ss
df
Mean Sauare
F
e
Achieved
Accommodation
16.67
1
16.67
0.62
0.43
Separation
17.20
1
17.20
0.64
0.42
Marginalized
13.72
1
13.72
0.51
0.48
Integration
193.13
1
193.13
7.47
0.007
Moratorium
Accommodation
36.02
1
36.02
1.13
0.29
Separation
25.18
1
25.18
0.76
0.38
Marginalized
4.51
1
4.51
0.14
0.71
Integration
6.09
1
6.09
0.19
0.67
Foreclosure
Accommodation
37.63
1
37.63
1.29
0.26
Separation
154.99
1
154.99
5.46
0.02
Marginalized
52.65
1
52.65
1.81
0.18
Integration
29.06
1
29.06
1.00
0.32
Diffusion
Accommodation
50.35
1
50.35
1.68
0.20
Separation
1.68
1
1.68
0.06
0.81
Marginalized
321.88
1
321.88
12.27
0.0006
Integration
14.51
1
14.51
0.49
0.49

56
Table 18. Summary of GLM Analyses for Relationship between Interpersonal Identity
Status and Acculturation Orientation with Ethnicity Controlled
Source
ss
df
Mean Square
F
E
Achieved
Accommodation
0.04
1
0.04
0.00
0.97
Separation
123.06
1
123.06
3.73
0.06
Marginalized
114.39
1
114.39
3.53
0.06
Integration
21.34
1
21.34
0.65
0.42
Moratorium
Accommodation
123.17
1
123.17
4.19
0.04
Separation
73.57
1
73.57
2.35
0.13
Marginalized
86.26
1
86.26
2.86
0.09
Integration
54.35
1
54.35
1.79
0.18
Foreclosure
Accommodation
44.94
1
44.94
1.47
0.23
Separation
689.77
1
689.77
24.92
0.0001
Marginalized
12.92
1
12.92
0.41
0.52
Integration
96.19
1
96.19
3.08
0.08
Diffusion
Accommodation
73.64
1
73.64
2.65
0.11
Separation
6.24
1
6.24
0.22
0.64
Marginalized
472.82
1
472.82
18.32
0.0001
Integration
14.77
1
14.77
0.52
0.47

CHAPTER 4
DISCUSSION
Drawing from Erikson's theory of psychosocial development, Marcia (1966,
1993) operationalized ego-identity development to include four statuses of identity-
achieved, moratorium, foreclosure, and diffusion. Some researchers (Abraham, 1986;
Hauser, 1972; Markstrom-Adams & Adams, 1995; Streitmatter, 1988), using Marcia's
model, have found disparate rates of identity development among ethnic groups. Cauca
sian adolescents were found to have more progressive identities than their peers who
were members of ethnic minorities do. Other studies, however, have failed to replicate
these findings (Abraham, 1986; Rotheram-Borus, 1989). Explanations of these findings
have centered on inadequate measures of the relation of demographic and cultural
variables with identity development. It was the purpose of this study, therefore, to address
the limitations of previous research to more accurately assess the identity development of
minority adolescents. This final chapter, in part, is a discussion of the results obtained
from the statistical analyses outlined in chapter 3. Implications and limitations of this
study follow. The chapter concludes with recommendations for future research.
Discussion of Results of Statistical Analyses
Researchers who have found differences among ethnic groups in identity
development have largely concluded that the differences between non-Caucasian and
Caucasian adolescents in identity development are due to the lack of opportunities for
57

58
social exploration available to minority groups. The proposed barriers to their exploration
include SES, gender, family status, familism, and parenting style. The research questions
examined in this study were designed to explore the relationship of these variables to
identity development.
Ethnicity and Identity Development
The first research question was whether there is a difference among African
American, Caucasian, and Hispanic adolescents in identity development. Specifically,
more Caucasian adolescents were expected to be in the progressive identity statuses in
comparison to their African American and Hispanic peers. Tests of this hypothesis
indicated no significant differences among the respondents' identity status scores.
Though the respondents of this study were late adolescents, they were also
students at institutions of higher education. College students may possess advantages that
non-collegiate adolescents do not have in experiencing a crisis and exploring alternative
roles and values.
Demographic Variables as Covariates of Ethnicity
The second question of this study was whether African American, Caucasian, and
Hispanic adolescents differ in identity development, when demographic variables (SES,
gender, family status) are controlled. A number of studies have suggested that the social
and structural differences associated with ethnic identification are explained, in part, by
SES, gender, and family status. That is, researchers have proposed that because ethnic
minority adolescents tend to live in more impoverished and less stable family
environments than their Caucasian peers and these differences inhibit progressive identity
development for ethnic minority adolescents (Phinney, 1990). Research also supports the

contention that males and females experience identity development differently (Marcia,
1993).
59
In tests of the statistical hypothesis, no significant differences were found among
the respondents' identity status scores, when SES, gender, and family status were
controlled. Also, although there were significant differences among groups in their report
of SES and family status, none of the demographic variables were found to be significant
covariates of identity status.
Familism
The third question of this study was whether African American, Caucasian, and
Hispanic adolescent identity development is related to familism. Familism is defined as
the connectedness one has with immediate and extended family members. Current
operationalization of the concept includes three dimensions: structural, attitudinal, and
behavioral. Structural familism refers to the number of relatives living in close proximity
to a person; attitudinal familism is a measure of a respondent's beliefs as to the benefit of
contact with family members. Behavioral familism refers to the amount of time people
stay in contact with family members. Valenzuela and Dombusch (1994) proposed that a
high degree of familism serves as a sort of social capital that enables adolescents to
develop positive mental health characteristics. Because differences in familism have been
found among African American, Caucasian, and Hispanic adolescents, it was of interest
to measure the relationship between identity status and familism with ethnicity
controlled. Testing the relationship between familism with ethnicity controlled resulted in
no significant relationships.

60
Parenting Style
Only a few studies have considered the relationship between identity status and
parenting style. This was the fourth question of the study. In previous studies an
authoritative parenting style has been associated with the progressive identity statuses
(Waterman, 1993), suggesting that experiencing a parenting style that encourages
independence with guidance may enable adolescents to endure a crisis with the intention
of working through it. In support of this premise, a restrictive parenting style has been
found to be related to the regressive identity statuses (Quintana & Lapsley, 1987).
As for the relationship between a permissive parenting style and identity
development, no studies could be found that investigated this relationship. Inasmuch as a
permissive parenting style encourages exploration, which may precipitate a crisis, this
style also provides the least amount of parental direction, which may contribute to an
inability to commit to identity options.
The statistical tests indicated a relationship between mothers parenting style and
identity development. A significant relationship was found between mother's
authoritative parenting style and ideological and interpersonal achievement and mother's
authoritarian parenting style and ideological foreclosure. Mother's permissive parenting
was significantly related to ideological and interpersonal moratorium, as well as diffusion
in both domains.
Significant relationships were found between fathers parenting style and identity
development. Fathers authoritarian parenting style was positively related to ideological
and interpersonal achievement. Fathers permissive parenting style was positively related
to ideological and interpersonal diffusion.

In all, the results of this study support the idea that identity status is related to
parenting style; however, the nature of the relation may differ for mothers and fathers.
61
Analysis indicated that a mothers parenting style that is characterized as open and
guiding (authoritative) is related to ideological and interpersonal achievement, whereas
the same parenting style in fathers was related to ideological and interpersonal
foreclosure. These conflicting findings may reflect the different social roles of mothers
and fathers.
Identity Development and Acculturation Orientation
At the center of the discussion of identity development differences among ethnic
groups is what effect, if any, does the experience of being a member of an ethnic minority
have on the psychosocial development of adolescents. During identity development, all
ethnic groups are exposed to the varying effects of SES, gender socialization, family
status, parenting style, and familism. Acculturation theory proponents, however, argue
that only members of a cultural minority must negotiate the often-competing demands of
developing a bi-cultural existence. That is, because Caucasian adolescents are members
of the cultural majority, they are unlikely to encounter expectations of themselves that
differ from the macro-culture.
It has been proposed that there are four routes to reconciling this acculturation: (a)
accommodation, whereby individuals give up their identification with their minority
culture and identify with the majority culture, (b) integration, whereby individuals
develop positive identifications with their minority culture and the cultural majority, (c)
separation, whereby individuals identify with their minority culture and do not identify

with the majority culture, and (d) marginalization, whereby individuals fail to identify
with their minority culture or the majority culture.
It was of interest, therefore, to determine the relationship between acculturation
orientation and identity development for African American and Hispanic adolescents.
Statistical tests revealed significant relationships between acculturation and identity
development with ethnicity controlled. A marginalized acculturation was positively
related to both ideological and interpersonal diffusion, and a separation acculturation was
positively related to interpersonal foreclosure. These findings support the contention that
the forms of acculturation are differentially related to identity development.
These findings suggest that the results of earlier studies that found ethnic
minorities more likely to be classified in the regressive identity statuses than their
Caucasian peers might reflect the mediating effect of acculturation orientation. If nothing
else, the findings lend cautious support to the argument that acculturation is related to
identity development.
Limitations of the Study
The conclusions of this study are to be considered in light of the following
limitations:
1. Previous studies of ethnicity and identity development were conducted with samples of
high school students. The sample in this study was limited to volunteers at the University
of Florida and Sante Fe Community College. The findings may not generalize to other
students at the institutions, to students at other colleges and universities, or to high school
students. In addition, participants were volunteers, which further limits the
generalizability of the study.

63
2. This was a correlational study. Causal claims cannot be made. But even more, ego
identity ever evolves. A snapshot inspection of the construct cannot document changes
over time. A longitudinal design is better suited to address this limitation.
Implications of the Study
In this study differences in identity development among ethnic groups were not
demonstrated, and, when the influence of SES, gender, and family status were controlled,
differences were still not found among ethnic groups. Those researchers who reported no
differences in identity development between students who were members of minority
groups and Caucasian students tended to attribute their findings to the high degree of
cultural integration in the schools of their samples of minority students. Earlier studies
have focused on high school aged adolescents. This study investigated the identity
development of late adolescent college students but a similar interpretation of the
findings could be made. Students who are members of minority groups who attend
universities may experience advantages that enable them to develop progressive
identities. Also universities may offer minority students a degree of cultural integration
not available to other minority adolescents. The findings of this study support the
contention that ethnic differences in identity are not present in environments that support
exploration and commitment.
A related finding was that an authoritarian parenting style indicative of control
and harsh discipline was related to identity foreclosure; whereas an authoritative
parenting style that promotes independence with limits, was related to identity
achievement. The permissive parenting style was related to both identity moratorium

and diffusion. In each case the degree to which adolescents were given a secure
foundation from which to explore was indicative of progressive identity development.
64
Acculturation orientation was also found to be related to identity development for
African American and Hispanic adolescents. Respondents who endorsed a marginalized
acculturation were more likely to experience identity diffusion than those who developed
other acculturation orientations were. A separation acculturation was also related to
interpersonal foreclosure. Again, it can be suggested that the degree to which an
adolescent is not limited in avenues of exploration is related to the degree to which
progressive identity development occurs.
Recommendations for Future Research
The findings of this study are to be considered in light of its limitations. First, the
conclusions were based on a study of late adolescents enrolled in institutions of higher
education. To expand the generalizability of these results, studies must be conducted with
samples that are more representative of the late adolescent population. Second, as this
was a correlational study of a single assessment of identity, no causal claims can be
made. If these conditions were addressed, it would be of interest to consider the
following:
1. Further research is needed to determine if exposure to an environment that
supports identity exploration is related to progressive identity development in spite of
ethnic differences.
2. In this study, a relationship between acculturation orientation and identity
development was demonstrated. This finding suggests that for minority group members
negotiating a bicultural existence can be an impediment to identity development not

63
shared by those in the majority. Investigation is needed to assess the degree to which the
need to resolve the conflicting expectations of the minority and majority cultures limits or
expands the identity options available to adolescents.
3. Although significant relationships were found between parenting style and
identity development, no consistent pattern of relationships between parenting style and
identity development has been established. Further research is needed to identify
relationships of variables that may explain the inconsistent findings.

APPENDIX A
OVERVIEW OF IDENTITY STUDIES
Author/Date
Respondents
Results
Interpretation
Controls
Measure
Hauser/1972
22 low SES
AA&C
male high
school
students
AA more
foreclosed
Social &
economic
conditions
may limit
exploration of
AAs
None
Q-sort
Abraham/
1983
273 MA&C
high school
students
C more
diffused/
ideological
identity
Parenting
style may
account for
difference
None
OM-EIS
Abraham/
1986
841 H&C
high school
students
MA more
foreclosed /
ideological
identity
Parenting
style may
account for
difference
Parents
education
EOM-EIS
Streitmatter/
1989
367 7th
graders C &
combination
of ethnic
minorities
Minority
group more
foreclosed/
both
domains
Social &
economic
condition may
limit
exploration of
AAs
None
EOM-EIS
Markstrom-
Adams &
Adams/
1995
123 A A, AI,
MA&C 10th
- 12th graders
Minority
groups
more
foreclosed/
both
domains
Socialization
may account
for difference
Parents
education
EOM-EIS
Rotherham-
Borus / 1989
320 AA,
MA, NA, F,
Chigh
school
Students
No notable
differences
High degree
of integration
in school
Comparable
economic
backgrounds
EOM-EIS
Note. AA= African American, MA = Mexican American, C = Caucasian, F = Filipino,
NA = Native American
66

APPENDIX B
INFORMED CONSENT
My name is Sean Forbes. I am a graduate student in the Foundations of Education
Department in the College of Education. My faculty supervisor is Patricia Ashton, PhD.
I am conducting a study to measure the relationship among identity development and
personal characteristics. I am requesting that you complete these questionnaires. If you
agree, you will complete the questionnaires anonymously.
The questionnaires will take you no more than an hour to complete and you do not have
to answer any question that you do not wish to answer. If, at any time, you feel that you
do not want to continue with the study you can stop.
As all data will be anonymous, you will not be able to obtain the results of the
questionnaires. You will not be reimbursed for your time. However, your cooperation is
extremely appreciated.
Please feel free to ask me questions at any time during the study. If you have questions
about the study after completing the questionnaires feel free to contact me at 348 Norman
Hall, 392-0726, ext 280. Questions or concerns about research participants' rights can be
directed to the UFIRB Office, Box 112250, University of Florida, Gainesville, FL 32611-
2250
Participants signature
Date
67

REFERENCES
Abraham, K. G. (1983). The relation between identity status and locus of control
among rural high school students. Journal of Early Adolescence. 3. 257-264.
Abraham, K. G. (1986). Ego-identity differences among Anglo-American and
Mexican-American adolescents. Journal of Adolescence. 9. 151-166.
Adams, G. R., & Jones, R. M. (1983). Female adolescents' identity development:
Age comparisons and perceived child-rearing experience. Developmental Psychology,
19,249-256.
Adams, G. R., Shea, J., & Fitch, S. A. (1979). Toward the development of an
objective assessment of ego identity status. Journal of Youth and Adolescence. 8. 223-
228.
Andrews, J. (1973). The relationship of values to identity achievement status.
Journal of Youth and Adolescence. 2. 133-138.
Archer, S. L. (1981). Identity interview schedule for working women.
Unpublished manuscript. Trenton State College, Trenton, NJ.
Archer, S. L. (1989a). Gender differences in identity development: Issues of
process, domain, and timing. Journal of Adolescence. 12. 117-138.
Archer, S. L. (1989b). The status of identity: Reflections on the need for
intervention. Journal of Adolescence. 12. 345-359.
Baumrind, D. (1971). Current patterns of parental authority. Developmental
Psychology. 4. (1, Pt. 2), 1-103.
Baumrind, D. (1991). Effective parenting during the early adolescent transition. In
P. A. Cowan & E. M. Hetherington (Eds ), Advances in family research (Vol. 2, pp. 111-
163). Hillsdale, NJ: Erlbaum.
Berry, J. W. (1980). Acculturation as varieties of adaptation. In A. Padilla (Ed.),
Acculturation: Theory, models, and some new findings (pp. 1-65). Boulder: Westview.
68

69
Berry, J. W., Trimble, J. E., & Olmedo, E. L. (1986). Assessment of acculturation.
In W. J. Lonner & J. W. Berry (Eds.), Cross-cultural research and methodology series:
Yol. 8, Field methods in cross-cultural research (pp. 291-101). Beverly Hills, CA: Sage.
Buri, J. R. (1991). Parental authority questionnaire. Journal of Personality
Assessment. 57. 110-119.
Chandler, C. R. (1979). Traditionalism in a modern setting: A comparison of
Anglo- and Mexican-American value orientations. Human Organization. 38. 153-159.
Coleman, J. S. (1988). Social capital in the creation of human capital. The
American Journal of Sociology. 94. S95-S120.
Cross, H, & Allen, J. (1970). Ego identity status, adjustment, and academic
achievement. Journal of Consulting and Clinical Psychology. 34. 288.
Dellas, M, & Jemigan, L. P. (1987). Occupational identity status development,
gender comparisons, and internal-external control in first-year Air Force cadets. Journal
of Youth and Adolescence. 16. 587-600.
Elkind, D. (1978). A sympathetic understanding of the child (2nd ed.). Boston:
Allyn & Bacon.
Enright, R. D., Lapsley, D. K., Drivas, A. E., & Fehr, L. A. (1980). Parental
influences on the development of adolescent autonomy and identity. Journal of Youth
and Adolescence. 9. 529-545.
Erikson, E. H. (1963). Childhood and society (2nd ed.). New York: Norton.
Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.
Erikson, E. H. (1987). The human life cycle. In S. Schlein (Ed.), A wav of
looking at things (pp. 595-610). New York: Norton.
Farris, B. E., & Glenn, N. D. (1976). Fatalism and familism among Anglos and
Mexican Americans in San Antonio. Sociology and Social Research. 60. 393-402.
Forbes, S., & Ashton, P. (1996). Improving the construct validity of scores on the
Extended Objective Measure of Ego Identity Scale (EOM-EIS ID. Manuscript submitted
for publication.
Gibson, M. (1988). Accommodation without assimilation: Sikh immigrants in an
American high school. New York: Cornell University.

70
Gonzalez, N. A., & Cauce, A. M. (1995). Ethnic identity and multicultural
competence: Dilemmas and challenges for minority youth. In W. D. Hawley & A. W.
Jackson (Eds.), Toward a common destiny: Improving race and ethnic relations (pp. Ol
i!). San Francisco: Jossey-Bass.
Gottfried, A. W. (1985). Measures of socioeconomic status in child development
research: Data and recommendations. Merrill-Palmer Quarterly. 31. 85-92.
Grossman, S. M., Shea, J. A., & Adams, G. R. (1980). Effects of parental divorce
during early childhood on ego development and identity formation of college students.
Journal of Divorce. 3. 263-272.
Grotevant, H. D., & Adams, G. R. (1984). Development of an objective measure
to assess ego identity in adolescence: Validation and replication. Journal of Youth and
Adolescence. 13, 419-438.
Grotevant, H. D., Thorbecke, W. L., & Meyer, M. L. (1982). An extension of
Marcia's identity status interview into the interpersonal domain. Journal of Youth and
Adolescence. 11. 33-47.
Hauser, S. T. (1972). Black and white identity development: Aspects and
perspectives. Journal of Youth and Adolescence. 1. 113-130.
Hetherington, E. M. (1972). Effects of father absence on personality development
in adolescent daughters. Developmental Psychology, 7. 313-326.
Hodgson, J. W., & Fischer, J. L. ((1979). Sex differences in identity and intimacy
development in college youth. Journal of Youth and Adolescence. 8, 37-50.
Hoff-Ginsberg, E., & Tardiff, T. (1995). Socioeconomic status and parenting. In
M. H. Bomstein (Ed ), Children and parenting (Vol. 2) (pp. 161-188). Hillsdale, NJ:
Erlbaum.
Hollingshead, A. B. (1975). Four factor index of social status. Unpublished
manuscript, Yale University, Department of Sociology, New Haven, CT.
Holtzman, W. H. (1982). Cross cultural comparisons of personality development
in Mexico and the United States. In D. A. Wagner & H. W. Stevenson (Eds ), Cultural
perspectives on child development (pp. 118-164). New York: Freeman.
Jones, R. M., & Streitmatter, J. L. (1987). Validity and reliability of the EOM-EIS
for early adolescents. Adolescence. 22. 647-659.
Jordan, D. (1970). Parental antecedents of ego identity formation. Unpublished
master's thesis, State University of New York at Buffalo.

71
Josselson, R. (1994). The theory of identity development and the question of
intervention. In S. Archer (Ed.), Interventions for adolescent identity development (pp.
12-25). Thousand Oaks, CA: Sage.
Keefe, S. E. (1984). Real and ideal extended familism among Mexican Americans
and Anglo Americans: On the meaning of close family ties. Human Organization. 38, 65-
69.
La Voie, J. C. (1976). Ego identity formation in middle adolescence. Journal of
Youth and Adolescence. 5. 371-385.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family:
Parent-child interaction. In P. H. Mussen (Ed.), Handbook of Child Psychology. (4 th ed.,
pp. 131-162). New York: Wiley.
Marcia, J. E. (1966). Development and validation of ego identity status. Journal of
Personality and Social Psychology. 3. 551-558.
Marcia, J. E. (1976). Identity six years after: A follow-up study. Journal of Youth
and Adolescence. 5. 145-160.
Marcia, J. E. (1993). The status of the statuses: Research review. 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. 22-41). New York: Springer-Verlag.
Marcia, J. E., & Friedman, M. (1970). Ego identity status in college women.
Journal of Personality, 38. 249-262.
Markstrom-Adams, C., & Adams, G. R. (1995). Gender, ethnic group, and grade
differences in psychosocial functioning during middle adolescence. Journal of Youth and
Adolescence. 24, 397-417.
Markstrom-Adams, C., & Spencer, M. B. (1994). A model for identity
intervention with minority adolescents. In S. Archer (Ed.), Interventions for adolescent
identity development (pp. 84-102). Thousand Oaks, CA: Sage.
Matteson, S. L. (1993). Differences within and between genders: A challenge to
the theory. 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. 69-110). New
York: Springer-Verlag.
Mehan, H., Hubbard, L., & Villanueva, I. (1994). Forming academic identities: ^
Accommodation without assimilation among involuntary minorities. Anthropology and
Education Quarterly, 25. 91-117.

72
Mindel, C. H. (1980). Extended familism among Mexican Americans, Anglos,
and Blacks. Hispanic Journal of Behavioral Sciences. 2. 21-34.
Negy, C., & Woods, D. J. (1992). A note on the relationship between
acculturation and socioeconomic status. Hispanic Journal of Behavioral Sciences, 2, 248-
251.
Nowicki, S., & Strickland, B. R. (1973). A locus of control scale for children.
Journal of Consulting and Clinical Psychology. 40. 148-154.
^ Peretti, P. O., & Wilson, T. T. (1995). Unfavorable outcomes of the identity crisis
among African American adolescents influenced by enforced acculturation. Social
Behavior and Personality. 23. 171-175.
Phinney, J. S. (1990). Ethnic identity in adolescence and adulthood: A review of
research. Psychological Bulletin. 108. 499-514.
Quintana, S. M., & Lapsley, D. K. (1987). Adolescent attachment and ego
identity: A structural equations approach to the continuity of adaptation. Journal of
Adolescent Research. 2. 393-409.
Quintana, S. M., & Lapsley, D. K. (1990). Rapprochement in late adolescent
separation-individuation: A structural equations approach. Journal of Adolescence. 13.
371-385.
Raphael, D. (1977). Identity status in university women: A methodological note.
Journal of Youth and Adolescence. 6. 57-62.
Reguero, J. T. (1991). Relationship between familism and ego identity
development of Puerto Rican and immigrant Puerto Rican adolescents (Doctoral
dissertation, Virginia Polytechnic Institute and State University, 1991). Dissertation
Abstracts International. 52. 1750-B.
Rotheram-Borus, M. J. (1989). Ethnic differences in adolescents' identity status
and associated behavior problems. Journal of Adolescence. 10. 525-537.
Rotheram-Borus, M. J., & Wyche, K. F. (1994). Ethnic differences in identity
development in the United States. In S. Archer (Ed.), Interventions for adolescent identity
development (pp. 62-83). Thousand Oaks, CA: Sage.
Schenkel, S., & Marcia, J. E. (1972). Attitudes toward premarital intercourse in
determining ego identity status in college women. Journal of Personality. 3. 472-482.
Spence, J. T., Helmreich, R., & Stapp, J. (1974). The Personal Attributes
Questionnaire: A measure of sex role stereotypes and masculinity and femininity. Catalog
of Selected Documents in Psychology, 4. 43-44.

73
Spencer, M. B., & Dombusch, S. M. (1990). Challenges in studying minority
youth. In S. S. Feldman & G. R. Elliot (Eds.), At the threshold: The developing
adolescent (pp. 123-146). Cambridge, MA: Harvard University Press.
Spencer, M. B., & Markstrom-Adams, C. (1990). Identity processes among racial
and ethnic minority children in America. Child Development. 61. 290-310.
St. Clair, S., & Day, H. P. (1979). Ego identity status and values among young
females. Journal of Youth and Adolescence. 8. 317-326.
Sterling, C. M., & Van Horn, K. R. (1989). Identity and death anxiety.
Adolescence. 17, 895-899.
Streitmatter, J. L. (1987). The effect of gender and family status on ego identity
development among early adolescents. Journal of Early Adolescence. 7. 179-189.
Streitmatter, J. L. (1988). Ethnicity as a mediating variable of early adolescent
identity development. Journal of Adolescence. 11. 335-346.
Streitmatter, J. L., & Pate, G. (1989). Identity status development and cognitive
prejudice in early adolescents. Journal of Early Adolescence. 9. 142-152.
U. S. Bureau of the Census. Department of Commerce. (1994). Statistical abstract
of the United States. 1994, Washington, DC: U. S. Government Printing Office.
Valenzuela, A., & Dombusch, S. M. (1994). Familism and social capital in the
academic achievement of Mexican origin and Anglo adolescents. Social Science
Quarterly. 75. 18-36.
Velez-Ibanez, C. G., & Greenberg, J. B. (1992). Formation and transformation of
funds of knowledge among U.S.-Mexican households. Anthropology and Education
Quarterly. 23.313-335.
Waterman, A. S. (1993). Developmental perspectives on identity formation: From
adolescence to adulthood. 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. 42-68).
New York: Springer-Verlag.
Waterman, A. S. (1994). Ethical consideration in interventions for promoting
identity development. In S. Archer (Ed.), Interventions for adolescent identity
development (pp. 231-244). Thousand Oaks, CA: Sage.
Waterman, C. K., & Waterman, A. S. (1975). Fathers and sons: A study of ego
identity across two generations. Journal of Youth and Adolescence. 4, 331-338.

BIOGRAPHICAL SKETCH
Sean Forbes completed his secondary education in Shelby County, Tennessee, in
1988. In the spring of 1989 he enrolled at the University of Florida, earning a bachelor of
arts degree in elementary education in 1992 and a master of arts degree in educational
psychology in 1995.
After completing his doctoral studies in 1999, he accepted a position as an
assistant professor of educational psychology in the Department of Educational
Foundations, Leadership, and Technology at Auburn University.
74

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Patricia Ashton, Chair
Professor of Educational Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Jarhes Algina
Professor of Educatic
il Psychology
I certify that f have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy^
iarry Gui
Associate Prfessor of Educational
Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Joh ttranzler
Associate Professor of Educational
Psychology

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Lee Mullally
Associate Professor of Teaci
Learning
ig and
This dissertation was submitted to the Graduate Faculty of the College of
Education and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
Dean, Graduate School



Table 17, Summary of GLM Analyses for Relationship between Ideological Identity
Status and Acculturation Orientation, with Ethnicity Controlled
55
Source
ss
df
Mean Sauare
F
e
Achieved
Accommodation
16.67
1
16.67
0.62
0.43
Separation
17.20
1
17.20
0.64
0.42
Marginalized
13.72
1
13.72
0.51
0.48
Integration
193.13
1
193.13
7.47
0.007
Moratorium
Accommodation
36.02
1
36.02
1.13
0.29
Separation
25.18
1
25.18
0.76
0.38
Marginalized
4.51
1
4.51
0.14
0.71
Integration
6.09
1
6.09
0.19
0.67
Foreclosure
Accommodation
37.63
1
37.63
1.29
0.26
Separation
154.99
1
154.99
5.46
0.02
Marginalized
52.65
1
52.65
1.81
0.18
Integration
29.06
1
29.06
1.00
0.32
Diffusion
Accommodation
50.35
1
50.35
1.68
0.20
Separation
1.68
1
1.68
0.06
0.81
Marginalized
321.88
1
321.88
12.27
0.0006
Integration
14.51
1
14.51
0.49
0.49


CHAPTER 2
METHOD
Research Participants
Three hundred students from a community college and a university in Florida
were recruited to participate in this study. Respondents included 82 African American
students (30 men and 52 women), 120 Hispanic students (50 men and 70 women), and 98
Caucasian (40 men and 58 women) late adolescents with a mean age of 20.8 years.
Instruments
Identity Status
Identity status was assessed using the Extended Objective Measure of Ego
Identity Status III (EOM-EIS HI, Forbes & Ashton, 1996). The EOM-EIS III is a 64-item
questionnaire. Respondents are instructed to indicate the degree to which an item is
reflective of themselves using a 6-point Likert scale, ranging from strongly disagree to
strongly agree. The scale assesses ideological identity in the areas of occupation, politics,
religion, and philosophical lifestyle and interpersonal identity in the areas of gender roles,
dating, friendship, and recreation. Each of the eight subdomains is measured by eight
items with two items reflecting each of the four identity statuses: achievement,
moratorium, foreclosure, and diffusion. Forbes and Ashton obtained the following
Cronbachs alpha coefficients for scores on the four identity statuses: ideological
achievement = .57; moratorium = .68; foreclosure = .64; diffusion = .53; interpersonal
achievement = .62; moratorium = .63; foreclosure = .77; diffusion = .51.
30


Question #4
The aim of the fourth research question was to determine whether identity status
variables were related to parenting style scores after controlling for ethnicity. The
covariate, ethnicity, was not found to be related to identity status.
Mothers' parenting style. The F and probability values for the relationship
between identity status and parenting style, with ethnicity controlled, are reported in
Tables 13 and 14. Inspection of Table 13 reveals the following significant positive
relationships between mothers parenting style and ideological identity status:
authoritative style with ideological achievement, F (2, 265) = 8.12, p = .005, permissive
style with ideological moratorium, F (2, 265) = 18.92, p = .0001, authoritarian style with
ideological foreclosure, F (2, 265) = 13 .48, p = .0003, and permissive style with
ideological diffusion, F (2, 265) = 8.77, p = .003.
Significant relationships were also found between mothers parenting style and
interpersonal identity status with ethnicity controlled (see Table 14): authoritative style
with achievement, F (2, 263) = 12.31, p = .0005, permissive style with moratorium, F (2,
263) = 12.51, p = .0005, and diffusion, F (2, 263) = 12.20, p = .0006. Each of these
relationships was also positive.
Analyses were performed to determine if significant interactions existed between
ethnicity and mother's parenting style. The interaction between mother's authoritative
style and ethnicity was significant, F (2, 265) = 5.43, p = .004, for the ideological
achieved status. As a group, African Americans demonstrated the strongest positive
relationship between ideological achievement and mothers authoritative parenting style,
followed by Caucasians then Hispanics.


B INFORMED CONSENT
67
REFERENCES 68
BIOGRAPHICAL SKETCH 74
IV


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Lee Mullally
Associate Professor of Teaci
Learning
ig and
This dissertation was submitted to the Graduate Faculty of the College of
Education and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
Dean, Graduate School


63
2. This was a correlational study. Causal claims cannot be made. But even more, ego
identity ever evolves. A snapshot inspection of the construct cannot document changes
over time. A longitudinal design is better suited to address this limitation.
Implications of the Study
In this study differences in identity development among ethnic groups were not
demonstrated, and, when the influence of SES, gender, and family status were controlled,
differences were still not found among ethnic groups. Those researchers who reported no
differences in identity development between students who were members of minority
groups and Caucasian students tended to attribute their findings to the high degree of
cultural integration in the schools of their samples of minority students. Earlier studies
have focused on high school aged adolescents. This study investigated the identity
development of late adolescent college students but a similar interpretation of the
findings could be made. Students who are members of minority groups who attend
universities may experience advantages that enable them to develop progressive
identities. Also universities may offer minority students a degree of cultural integration
not available to other minority adolescents. The findings of this study support the
contention that ethnic differences in identity are not present in environments that support
exploration and commitment.
A related finding was that an authoritarian parenting style indicative of control
and harsh discipline was related to identity foreclosure; whereas an authoritative
parenting style that promotes independence with limits, was related to identity
achievement. The permissive parenting style was related to both identity moratorium


72
Mindel, C. H. (1980). Extended familism among Mexican Americans, Anglos,
and Blacks. Hispanic Journal of Behavioral Sciences. 2. 21-34.
Negy, C., & Woods, D. J. (1992). A note on the relationship between
acculturation and socioeconomic status. Hispanic Journal of Behavioral Sciences, 2, 248-
251.
Nowicki, S., & Strickland, B. R. (1973). A locus of control scale for children.
Journal of Consulting and Clinical Psychology. 40. 148-154.
^ Peretti, P. O., & Wilson, T. T. (1995). Unfavorable outcomes of the identity crisis
among African American adolescents influenced by enforced acculturation. Social
Behavior and Personality. 23. 171-175.
Phinney, J. S. (1990). Ethnic identity in adolescence and adulthood: A review of
research. Psychological Bulletin. 108. 499-514.
Quintana, S. M., & Lapsley, D. K. (1987). Adolescent attachment and ego
identity: A structural equations approach to the continuity of adaptation. Journal of
Adolescent Research. 2. 393-409.
Quintana, S. M., & Lapsley, D. K. (1990). Rapprochement in late adolescent
separation-individuation: A structural equations approach. Journal of Adolescence. 13.
371-385.
Raphael, D. (1977). Identity status in university women: A methodological note.
Journal of Youth and Adolescence. 6. 57-62.
Reguero, J. T. (1991). Relationship between familism and ego identity
development of Puerto Rican and immigrant Puerto Rican adolescents (Doctoral
dissertation, Virginia Polytechnic Institute and State University, 1991). Dissertation
Abstracts International. 52. 1750-B.
Rotheram-Borus, M. J. (1989). Ethnic differences in adolescents' identity status
and associated behavior problems. Journal of Adolescence. 10. 525-537.
Rotheram-Borus, M. J., & Wyche, K. F. (1994). Ethnic differences in identity
development in the United States. In S. Archer (Ed.), Interventions for adolescent identity
development (pp. 62-83). Thousand Oaks, CA: Sage.
Schenkel, S., & Marcia, J. E. (1972). Attitudes toward premarital intercourse in
determining ego identity status in college women. Journal of Personality. 3. 472-482.
Spence, J. T., Helmreich, R., & Stapp, J. (1974). The Personal Attributes
Questionnaire: A measure of sex role stereotypes and masculinity and femininity. Catalog
of Selected Documents in Psychology, 4. 43-44.



PAGE 1

ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND HISPANIC ADOLESCENTS By SEAN ALAN FORBES A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1999

PAGE 2

ACKNOWLEDGEMENTS Patricia Ashton has guided me since I began graduate school. Her wisdom and patience have not been lost on me. I thank her for her unwavering support and am indebted for the lessons she has taught me. This dissertation is dedicated to her. I would also Uke to thank the other members of my committee for their assistance. James Algina was a sobering reminder of the diligence needed for any worthwhile pursuit. John Kranzler, who has known me since I was an undergraduate student, has counseled me many a time and his advice has served me well. Barry Guinagh has been the kindest man I have ever met. Greg Neimeyer and Lee Mullally, my outside committee members, have saved my skin on more than one occasion. Collectively, this conmiittee has been outstanding. I must also recognize Randall Scott Hewitt — the most natural philosopher I havo known. Graduate school would not have been complete without him. He and his family have opened my eyes to other worlds and I will miss their companionship. ii

PAGE 3

TABLE OF CONTENTS page ACKNOWLEDGEMENTS i ABSTRACT > CHAPTERS 1 INTRODUCTION Statement of the Problem 3 Purpose of the Study 9 Ecological Variables Related to Identity Status 10 Research Questions 2" Significance of Study 28 2 METHOD 30 Research Participants 30 Instruments 30 Procedures 34 Research Questions 34 3 RESULTS 36 Summary Statistics 36 Results of Statistical Analyses 42 4 DISCUSSION 5/ Discussion of Results of Statistical Analyses 57 Limitations of the Study 62 Implications of the Study 63 Recommendations for Future Research 64 APPENDICES A OVERVIEW OF IDENTITY STUDIES 66 111

PAGE 4

B INFORMED CONSENT 67 REFERENCES 68 BIOGRAPfflCAL SKETCH 74 iv

PAGE 5

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND fflSPANIC ADOLESCENTS By Sean Alan Forbes December 1999 Chair: Patricia Ashton Major Department: Educational Psychology Adolescence is a critical period in the development of psychosocial identity. That is, as the transition is made from child to adult, the adolescent must reconcile childhood identifications with the coming expectations of adulthood. The goal is to establish a sense of individuality in the larger society. This goal is attained through the development of an identity. The identity crisis includes four statuses: (a) achievement — wherein an adolescent experiences an identity crisis and makes a commitment to alternatives, (b) moratorium — wherein an adolescent experiences an identity crisis but makes no commitments, (c) foreclosure — wherein an adolescent bypasses an identity crisis by adopting the commitments of parents or other significant figures, and (d) diffusion — wherein an adolescent avoids the experience of a crisis and postpones commitments. One of the less considered areas in identity status research has been the comparison of identity V

PAGE 6

development among ethnic groups. In the studies investigating the topic, inconsistent results have been found. Some researchers have reported that African American and Hispanic adolescents were more likely to be foreclosed than their Caucasian peers. In light of the differences among ethnic groups in parenting style andiamilism reported in some studies, this study was designed to determine whether the inconsistent findings regarding the relation between identity status and ethnic group could be explained by parenting style, familism, SES, gender, and/or family configuration. Also, as adolescents from cultural minorities must negotiate a bicultural existence, it was also of interest to consider the relation between identity development and acculturation for African American and Hispanic adolescents. Three hundred college aged adolescents participated in this study. Of the total, 82 identified themselves as African American and 98 identified themselves as Caucasian. One hundred twenty participants identified themselves as Hispanic. Analysis of variance indicated no significant differences among groups in identity development and no relationship between identity development and familism. Significant relationships were found between parenting ^yle and acculturation and identity development. vi

PAGE 7

CHAPTER 1 INTRODUCTION According to Erikson (1963), the formation of an identity is the dominant psychosocial crisis of adolescence. Commonly referred to as the identity crisis this process is characterized by a delay of adult commitments that permits the adolescent to experiment with various roles and ultimately to commit to available identity alternatives in ideological and interpersonal domains (Erikson, 1968, p. 159). As the goal of an identity crisis is the synthesis of the identifications of childhood and the opportunities and expectations of adulthood, identity resolution is the foundation upon which future aduh decisions are based (Erikson, 1987; Spencer & MarkstromAdams, 1990). Moreover, the individual who successfully resolves the crisis becomes a secure, independent adult prepared to establish intimacy with others, whereas those who are unable to establish an identity enter adulthood without a firm sense of how they fit into society (Elkind, 1978, Peretti & Wilson, 1995). Marcia (1966, 1976), in an extension of Erikson's theory, described four possible outcomes of the crisis: identity achievement, moratorium, foreclosure, and diffusion. Identity achievement is defined by a commitment to identity alternatives after a period of exploration, whereas moratorium is defined as a period of exploration of alternative roles during which commitment to an identity is delayed. Foreclosure is defined by identity commitment without prior exploration, whereas diffusion is defined by the unwillingness to commit to an identity with or without exploration. 1

PAGE 8

2 The achievement and moratorium statuses, by involving exploration, are more progressive identity statuses than foreclosure and diffusion, which do not involve exploration (Waterman, 1993). That is, adolescents must play an active role in the resolution of childhood identifications with adult opportunities and expectations; they must experiment with the identity alternatives available if a positive resolution of the crisis is to be achieved. Taking an active part in resolving their identity crisis often leaves the adolescent open for problems. For example, adolescents classified in the moratorium status are more prone to anxiety than those in foreclosure because of the lack of commitment in their lives (Marcia, 1976; Steriing & Van Horn, 1989). Yet, the overall outcome of being identity achieved or in moratorium is positive. That is, by taking an active role in their identity crises, adolescents in the moratorium or achievement status are psychologically heahhier than those in diffusion and foreclosure. For example, those in moratorium and achievement are less likely than those in foreclosure to be authoritarian or to use stereotypical thinking (Marcia & Friedman, 1970; Streitmatter & Pate, 1989). Those in achievement and moratorium are more likely to possess an internal locus of control (Delias & Jemigan, 1987), and to be more autonomous than those in foreclosure and diffusion (Andrews, 1973). In the areas of cognitive performance and cognitive style, those in the achieved and moratorium statuses often perform at a higher level than those in foreclosure and diffusion. For example, adolescents in the achievement status have been found to have better study habits (Waterman & Waterman, 1975) and higher grade point averages than those in other statuses (Cross & Allen, 1970, Raphael, 1977). Waterman and Waterman (1975) found that individuals in achievement and moratorium

PAGE 9

3 were more reflective and less impulsive than those in foreclosure and diffusion, and Adams, Abraham, and Markstrom (1987) found that individuals in the higher statuses were less self-conscious than those in foreclosure and diffusion. Statement of the Problem As individuals move through adolescence, most tend toward the progressive statuses of achievement and moratorium rather than the regressive statuses of diffusion and foreclosure. Several studies, however, have suggested that progressive identity development is more common for Caucasian adolescents than for their African American and Hispanic peers. In a longitudinal study of ideological identity in 22 low socioeconomic status (SES) African American and Caucasian high school males, Hauser (1972) administered a Q-sort annually for 3 years and obtained correlations within each annual Q-sort and between each annual Q-sort to estimate the consistency within aspects of the students' self-image and change over time in their self-image. From these correlations, Hauser concluded that, compared to the Caucasian students, the African American students were more foreclosed than the Caucasian students. Several problems in the Hauser (1972) study raise questions about the validity of his findings. First, the study consisted of a convenience sample of 22 matched African American and Caucasian students. The small size of the sample and the inability to determine the adequacy of the matching contribute to doubts about the validity of the findings. Second, the author used a Q-sort of undetermined reliability rather than the traditional interview or questionnaire measures of identity. Third, the reUability of the differences between the correlations in the African American and Caucasian groups is

PAGE 10

4 unclear. Given these weaknesses in Hauser's study, the validity of his conclusion that the African American students were more foreclosed than their Caucasian counterparts must be questioned. Furthermore, the study was conducted over 30 years ago (between 1962 and 1967). Hauser referred to students' comments during the interview that suggested that racial restrictions on the African American students' daily lives (e.g., restrictions in jobs, social life, recreation, and housing), limitations in work opportunities, and the lack of role models who had succeeded in the majority cuhure may have contributed to their sense of foreclosure. The greater social and economic opportunities available to African Americans today may have reduced that sense of foreclosure. Despite the weaknesses in Hauser's study, a more recent study by Abraham (1986) supported Hauser's findings regarding the relationship between ethnic-minority status and identity foreclosure. Abraham investigated the ideological and interpersonal identity of 841 Mexican American and Caucasian students in grades 9-12 at a high school in the southwest United States. Respondents completed the Extended Objective Measure of Ego Identity Status (EOM-EIS) (Grotevant & Adams, 1984), a 64-item measure of identity in four areas of ideological identity (occupation, religion, politics, and lifestyle) and interpersonal identity (gender roles, dating, friendship, and recreation). In a multivariate analysis of covariance, Abraham used mothers' and fathers' education as covariates to control for differences in SES in the two groups. The results indicated Caucasians and Mexican Americans differed in ideological identity. A greater percentage of Mexican American respondents were foreclosed than their Caucasian counterparts. Abraham concluded, therefore, that the Mexican American adolescents seemed to be more receptive to their parents' views and ideals on politics, religion, occupation, and

PAGE 11

5 philosophical lifestyle. No differences between ethnic groups were found on interpersonal identity. Abraham (1986) offered two alternative explanations for her findings. First, she suggested that Hauser's belief that the students' ethnic minority status may have limited the range of their exposure to ideological options. Second, she noted that differences in parenting style might account for the results, though this variable was not included in the analysis. Abraham concluded that the limitations in options and parenting style may have contributed to the greater number of the Mexican American students in the foreclosure status. Selection of participants from a single school raises questions, however, about the generalizability of Abraham's findings. Administering the EOM-EIS to 367 students in grades 7 and 8 in a junior high school in the Southwest, Streitmatter (1988) compared the identity development of Caucasians and an ethnic minority group composed of Hispanic Americans, African Americans, Native Americans, and Asian Americans. Like Hauser (1972) and Abraham (1986), Streitmatter found that the Caucasian adolescents were less foreclosed than the minority respondents. Unlike the previous studies, however, differences were found in the interpersonal domain as well as the ideological. Streitmatter suggested that the greater incidence of foreclosure in the minority groups was the resuh of minority adolescents' feeling more uncomfortable in the majority culture than their Caucasian peers. She proposed that this discomfort led them to accept parental values and reduced their willingness to experiment with ideas that conflicted with their parents' expectations. Further, she pointed out that the minority students were bussed into the school, which might have added to the students' discomfort and inhibited their exploration. She referred

PAGE 12

6 to research on parenting style that has suggested that a warm, supportive, and controlHng parenting style is related to the foreclosure status and proposed that the Mexican American parents may be more likely to use such a style with their children. In light of her findings, Streitmatter suggested that studies must be done to identify mediators of the relationships between ethnicity and identity development. Like Hauser (1972) and Abraham (1986), the Streitmatter study is seriously flawed. She combined the four ethnic groups into one group in her analysis. Citing problems such as poor sampling techniques and a lack of control for SES in previous studies, MarkstromAdams and Adams (1995) designed a study comparing the identity development of African American, American Indian, Mexican American, and Caucasian adolescents to address these design flaws. Their respondent pool roughly approximated the racial composition of the area from which the respondents were sampled, and SES was controlled by obtaining the respondents' parents' educational level. MarkstromAdams and Adams also investigated the role that age and gender roles play in the development of identity. For the MarkstromAdams and Adams (1995) study, 123 adolescents in grades 10-12 in a high school in the rural Southwest completed the EOM-EIS (Grotevant & Adams, 1984), the Personal Attributes Questionnaire (Spence, Helmreich, & Stapp, 1974), a gender role measure, the Nowicki-Strickland Locus of Control Scale for children (Nowicki & Strickland, 1973), and questions about parents' level of education. Like the previous researchers, MarkstromAdams and Adams found that the minority adolescents were more foreclosed than their Caucasian peers. In addition, they found a nonsignificant trend suggesting that the minority adolescents were also more likely to be classified as I

PAGE 13

7 diffused than their Caucasian counterparts. Also, relationships between the contextual variables and identity status emerged. Higher parent education was related to lower foreclosure rates for ideological identity. In concluding, MarkstromAdams and Adams (1995) agreed with the earlier researchers who argued that the greater foreclosure in nonWhite adolescents might be because the minority groups are more accepting of foreclosure than Caucasians, who may require experimentation and autonomous choice in achieving an identity. Two other studies, however, did not report differences in foreclosure in minority and Caucasian adolescents (Abraham, 1983; Rotheram-Borus, 1989). Abraham (1983) investigated the identity development of 223 Caucasian and Mexican American students from a rural area in the southwest with the Objective Measure of Ego Identity Status (OM-EIS), a 24-item measure of ideological identity, including the areas of occupation, politics, and religion (Adams, Shea, & Fitch, 1979). She found that more Caucasian students than Mexican American students were classified in diffusion. Although more Mexican American students than Caucasian students were classified as foreclosed (18.8% compared to 9.6%) and more Caucasian students were classified as identity achieved (10.1% compared to 16.4%), these differences were not significantly different. In her discussion, Abraham (1983) pointed out that SES level has been identified as a variable related to identity development. A post hoc analysis of the parental educational level of the students in her study revealed that the parents of the Mexican American students were lower in educational level than the parents of the Caucasian students. Because a higher SES level would be expected to be associated with lower levels of diffusion, she proposed that parental socialization style might be the influential

PAGE 14

8 variable in her study. She pointed out that Adams and Jones (1983) found that a maternal style characterized by control, regulation, and strong encouragement of independence and a paternal style characterized by praise, approval, and less fair discipline were related to diffusion, parental styles that might have been more common among the parents of the Caucasian students compared to the Mexican American students. Using the OM-EIS (Adams et al., 1979), Rotheram-Borus (1989), in an investigation of the identity status of 330 African American, Caucasian, Filipino, Hispanic, and Native American high school students, found that there was little relation between identity status and ethnicity. Her only significant finding was that Caucasian students in the 1 1th and 12th grades were more likely than the other students to have high moratorium scores on the OM-EIS. The students, however, were not classified in the moratorium status at a rate different fi-om others. Rotheram-Borus argued that because of the school's high degree of integration, ethnic differences in the stages of identity did not exist. In summary, Hauser (1972), Abraham (1986), Streitmatter (1988), and MarkstromAdams and Adams (1995) reported that the minority students in their studies were more foreclosed that their Caucasian peers. Abraham (1983) and Rotherham-Borus (1989) reported no significant differences in foreclosure (see Appendix A for a comparison of these studies). The inconsistency of results in these studies may be due to the confounding of a number of ecological variables. Some researchers (e.g., Hauser, 1972; Streitmatter, 1988) failed to control for SES. Abraham (1986) and Markstrom Adams and Adams (1995) used parental education as a measure of SES. Despite their effort to control for SES, they found that African American and Hispanic adolescents

PAGE 15

scored significantly higher on foreclosure than the Caucasian adolescents in their sample, suggesting that either parental education is an inadequate measure of SES or that other factors account for the ethnic group differences in foreclosure. Some research findings suggest other ecological variables as factors that may account for the greater foreclosure among minority group members. Spencer and MarkstromAdams (1990) pointed out that although both the Caucasian and Afi-ican American males in Mauser's (1972) study were from low SES families, the Afincan American males were more likely to live with single parents than were the Caucasian males. Hauser noted that the African American students whose fathers were absent were more foreclosed than the students whose fathers were present. Spencer and MarkstromAdams concluded that both family configuration and SES may confound the results of studies of the identity development of minority adolescents, but they also considered the ahemative hypothesis that, because of acculturation or other cultural factors, foreclosure may be adaptive for minority youth. Supporting that conclusion, Markstrom-Adams and Adams (1995) proposed that Afi^ican American and Hispanic families are likely to define maturity in terms of support of family values in contrast to Caucasian families who may define maturity in terms of independence. That is, autonomous choice may be the defining theme for the development of Caucasian identity, whereas acceptance of family values may be the defining theme for Afiican American and Hispanic adolescents. Purpose of the Study In the comparison of minority and majority adolescent ego-identity development, conflicting resuhs have been found. Some of the research suggests that minority adolescents are more likely to be foreclosed, but other studies have failed to support that

PAGE 16

10 conclusion. Furthermore, the studies have for the most part failed to include important ecological variables that may account for the conflicting resuhs. To account for the development of identity of ethnic minorities, the role of all of these variables must be examined. The purpose of this study, therefore, was to investigate whether the ecological Variables of SES, acculturation, family characteristics, and gender are related to the egoidentity development of ethnic minority and Caucasian adolescents. Each of these ecological variables and their potential relationship to identity development will be discussed in the sections that follow. Ecological Variables Related to Identity Status Identity status has been found to be related to a number of ecological variables (Marcia, 1993). In response to findings that ethnic minority adolescents tend toward less progressive identity statuses than their Caucasian peers, several influences on adolescent development have been offered as explanation: demographic variables, home culture variables, and the presence of an acculturation process in ethnic minority members. Demographic Variables Differences in identity development may be explained, in part, by the demographic variables gender, SES, and family configuration. Erikson (1963) argued that males and females experience identity development differently. Limited evidence suggests that identity development in low SES environments is not as progressive as in higher SES environments (MarkstromAdams & Adams, 1995). As for family configuration, relations have been found between the construct and identity development, although the direction of the relationship is far fi-om clear (Marcia, 1993).

PAGE 17

11 Gender. In a discussion of male and female psychosocial differences, Erikson (1968) attributed their differential development to socialization. That is, Erikson believed that because males possess a dominant role over women in society they are more likely to assert themselves than women. Women, on the other hand, due to their more submissive cultural role, are more likely to engage in activities that are passive and harmonious. According to Erikson, the male develops his sense of identity by competition, whereas the female resolves identity questions through communal, interdependent means. The first studies to investigate the differences that Erikson proposed supported his claim (e.g., Marcia, 1966, 1976; Schenkel & Marcia, 1972). For example, Hodgson and Fischer (1979), in a study of 100 late adolescents' identity development, found that males tend to resolve identity issues via competence and knowledge. Males achieve competence by pursuing an education and obtaining a career, and knowledge by coming to understand the world around them. Females, however, were found to resolve identity issues through different mechanisms. Hodgson and Fischer stated that for females the response to "Who am I?" revolves around what they are in relation to others. Because of this foundational research, it was long accepted that gender differences exist in identity development. Matteson (1993) pointed out, however, that social changes in the past few decades may necessitate a reexamination of the differences between the identity development of men and women. Archer (1989a), in a report of three studies, examined the identity processes of experimentation and commitment, the domains of identity, and the timing of identity changes for male and female students in early to late adolescence. Overall, Archer concluded that males and females undergo identity development in similar ways. As for

PAGE 18

12 the identity processes, in two of the studies, males and females differed only in the foreclosure status; males were more likely to be foreclosed than females. In the third study no differences were found between males and females in the identity statuses. Archer (1989a) found, however, that males were more likely than females to be diffused with regard to the family roles domain, whereas females were more likely to be classified in the foreclosure, moratorium, and achievement statuses. No differences were detected in the second study. In the third study. Archer found that males were more likely to be classified in the foreclosure status, and females in the diffusion status with regard to political ideology. As for gender differences in the timing of identity development. Archer (1989a) found in two of the studies that no differences existed. In the third study, however, a Grade x Gender interaction was found. Females were more likely than males to be classified in the moratorium status in grades 6, 10, and 12. For identity achievement, males were more likely than females to be so at grade 12. Overall, Archer (1989a) concluded that the differences between the genders were slight. On the basis of her studies, she argued that the assumption that males and females experience identity development differently should be discarded. Matteson (1993) proposed much the same argument as Archer. That is, Matteson pointed out that the relationships found between identity status and other psychological constructs for males were also found for females. For example, men and women classified in the moratorium status reported the highest levels of anxiety, and males and females in foreclosure were more likely than those in other statuses to defer to authority (Marcia & Friedman, 1970).

PAGE 19

13 Other studies, however, have found difference between genders in identity development. For example, Streitmatter (1987), in a study of 265 early adolescents, found that females scored significantly higher than males in total moratorium and ideological moratorium scales. Streitmatter concluded that such a difference indicated that females were "more involved in the questioning and searching process than males" (p. 184). In a similar study, Streitmatter (1988) found that females were significantly more often found in ideological and interpersonal achievement and ideological moratorium than males. Given that identity achievement and moratorium are considered more advanced identity statuses than foreclosure and diffusion, Streitmatter concluded that females are considered more mature in their identity development than are males at the same age. Streitmatter (1988) also found a statistically significant Gender x Ethnicity interaction. Minority males were more likely than their non-minority peers to be in the moratorium status, a finding that Streitmatter cited as evidence that minority males are more actively seeking answers to their identity problems. Minority males were, however, more likely to be in the diffusion status than their peers. This finding is noteworthy, as it runs counter to the research that has found that minority males are more often in foreclosure than their Caucasian counterparts. MarkstromAdams and Adams (1995) found that females were not as likely as males to be classified in the diffusion status for ideological and interpersonal identity. A Gender x Grade interaction demonstrated that older females were less likely than their younger peers to exhibit lower diffusion scores. Markstrom-Adams and Adams, unlike Streitmatter (1988), did not find any Gender x Ethnicity interactions.

PAGE 20

In summary, early theory and research of gender differences in identity development pointed to clear differences (Erikson, 1968; Marcia, 1966, 1976). Males were perceived as focused on ideological identity development. Women were seen as focused on interpersonal identity development. Later studies (Archer, 1989a; MarkstromAdams & Adams, 1995; Streitmatter, 1988) suggest that diflFerences between males and females in identity development, when present, are subtle. Socioeconomic status. Minority adolescents are more likely to come from lowincome families than their Caucasian peers (Spencer & Dombusch, 1990). Although few studies have considered the impact of low SES on identity development, limited evidence suggests a relation between the constructs. For example, MarkstromAdams and Adams (1995) found that higher paternal and maternal education was related to lower ideological foreclosure (the correlation with maternal education was -.24 and the correlation with paternal education was -.22). No relationship was found between parents' education and interpersonal identity. Berry (1980) hypothesized that minority group members with low SES are more likely to adopt an assimilation or separation style of acculturation rather than an accommodation style. Because of limited resources, these individuals tend to see the world as difificuU to master. They may also avoid identification with the majority culture or reject their own culture in favor of the majority culture. Negy and Woods (1992), in a study of 339 Mexican American college students, found that acculturation was moderately related to SES (r = .44, p = .0001). That is, the greater the identification witn the majority culture, the higher the SES of the student tended to be. Negy and Woods, however, did not determine if those who identified with the majority culture also

PAGE 21

15 identified with the minority culture. As a result, it was impossible to differentiate between assimilated and the accommodated respondents. As a great number of minority children come from low SES backgrounds, many do not have access to the same economic resources as their higher SES peers and, as a result, may adopt a marginalized or rejected acculturation orientation. This puts these adolescents at greater risk of concluding that the expectations of society are incompatible with their own, which may resuh in the formation of a weaker identity than higher SES peers. Research is needed to investigate Berry's (1980) hypothesis and to determine whether acculturation status mediates the relationship between SES and adolescents' identity status. Family configuration. African American and Hispanic minority adolescents are more likely to grow up in a single parent home than are their Caucasian peers (U. S. Bureau of the Census, 1994). Family configuration differences may be related to patterns of identity development among cultural groups, but research on the relationship between family configuration and identity has yielded inconsistent resuUs. For example, Jordan (1970) found that male college students from divorced families were more likely to be identity diffused than were students fi^om intact families. Other studies, however, do not support the claim that living in an intact family is related to progressive identity development. Grossman, Shea, and Adams (1980), in a study of 294 college students, found that males fi'om divorced families possessed higher identity achievement scores than males in intact families and females in general. In a study of 467 students from ages 12 to 18, Jones and Streitmatter (1987) found that male and female high school students fi-om intact families were more foreclosed on the ideological and interpersonal scales of

PAGE 22

16 the EOM-EIS than peers living with a single parent. In their study of 80 Caucasian female high school students, using Marcia's (1966) Identity Status Interview, St. Clair and Day (1979) found that two-thirds of the 80 senior high school female adolescents in their study who were in the achieved status came from homes disrupted by death or divorce. In a study of 265 seventh graders, Streitmatter (1987) found that early adolescents from intact families were more likely to be classified in the foreclosure status than their peers from non-intact families. She also found that males seemed more affected by family disruption than females. Males were more likely to be classified as diffused than females. Streitmatter suggested that disruption might facilitate identity development in females, perhaps shortening their moratorium. Jones and Streitmatter's (1987), St. Clair and Day's (1979), and Streitmatter's (1987) results may have been affected by their failure to discriminate between single parent homes due to divorce or death. Recognizing that research has shown that parentchild relations differ for children with a deceased parent compared to children whose parents are divorced (e.g., Hetherington, 1972), Streitmatter analyzed her data excluding children with a deceased parent. Finding only one change from the analysis with children with either a deceased parent or divorced parents, Streitmatter reported the data for the combined groups. She found a Diffusion x Gender x Family Status interaction but attributed the resuh to a loss in sample size. St. Clair and Day (1979) suggested that the impact of family disruption may be different for male and female adolescents, and Streitmatter's (1987) study provides empirical support for the need to study the differential impact of family disruption on males and females. Streitmatter found that adolescents from intact families had higher

PAGE 23

17 total scores (combined ideological and interpersonal domain) on foreclosure than adolescents from nonintact families. From post hoc analyses she found that the family status effect held for ideological foreclosure but not for interpersonal foreclosure, and she found a significant Family Status x Gender interaction indicating that female adolescents from intact homes had lower interpersonal foreclosure scores than female students from disrupted homes, whereas male students from intact homes had higher foreclosure scores than males in non-intact homes. She also found a significant Family Status x Gender interaction for diffusion. Male students from intact homes had lower ideological diffusion scores than male students from non-intact homes, and females from intact homes had higher ideological diffusion scores than females with only one parent in the home. She also found that the overall relationship between family status and identity development was greater for male students than for female students. She speculated that given the likelihood that most of these families were headed by women, the lower foreclosure scores may reflect the male adolescents' rejection of failed family values, whereas their higher diffusion scores may indicate that they were unable or unwilling to choose an identity. Streitmatter further speculated that family disruption may have facilitated identity development by thrusting the male adolescents into diffrision earlier than their peers in intact homes. In sum, research has demonstrated that family configuration is related to identity development. However, the kind of household (i.e., intact, non-intact), as well as the gender of the child, plays a role in relationship between family configuration and identity development. Therefore, research that divides family configuration into separate categories for (a) single parent as the result of death, (b) single parent due to divorce, (c)

PAGE 24

18 intact families, (d) single parent never married, and (e) reconstituted families and that analyzes the data separately for males and females and controls for SES is needed to clarify the relationship between family configuration and identity development. Family Characteristics Differences in the family environment of minority and majority adolescents may explain, in part, the different identity development of the ethnic groups. Marcia (1993) argued that individuals classified in the high identity statuses often come from families where ideological and interpersonal differentiation between the child and parents is supported. Those classified in the lower identity statuses, in contrast, tend to come from families that either encourage conformity to family values or from families in which the parents distance themselves from the ideological and interpersonal development of the child. Significant differences have been found between minority cultures and the majority culture in parenting style (Holtzman, 1982) and familism (Mindel, 1980). Parenting stvle. Baumrind (1971, 1991) identified three forms of parenting: (a) authoritarian whereby parents use a controlling and disciplinarian style to raise their children, (b) authoritative whereby parents encourage independence but a respect for rules in their children and, (c) laissez-faire whereby parents do not place limitations or hold expectations for their children's behavior. Maccoby and Martin (1983) further divided the laissez-faire style of parenting into two distinct styles: (a) permissiveindifferent, whereby parents are not involved in their children's hves and (b) permissiveindulgent, whereby parents give in to their children's demands and do not place restrictions on their behavior.

PAGE 25

19 Parenting style is significantly related to SES (HoflF-Ginsberg & Tardif, 1995). Parents in the lower SES tend to use an authoritarian style of parenting, whereas middle SES parents tend to use an authoritative style of parenting (HofF-Ginsberg & Tardif 1995). As minority adolescents are more likely to have low-income families than their Caucasian peers (Spencer & Dombusch, 1990), minority adolescents tend to be raised with an authoritarian style of parenting. Caucasian adolescents, in contrast, tend to be raised with an authoritative style. This difference may be related to patterns of identity development among cultural groups. For example. Waterman (1993) hypothesized that, because of the controlling power authoritarian-style parents exert, their children are likely to begin identity development in foreclosure. In contrast, children of permissive parents tend toward identity diffusion because of their parents' lack of expectations. Waterman was less certain about the likely outcomes of authoritative parenting, but he seemed to suggest that the support and caring these parents give their children might enable them to delay commitment, thus enabling them to experience moratorium. Research supports Waterman's belief that an authoritative parenting style is related to moratorium. For example, Quintana and Lapsley (1990), in a study of 101 undergraduates, found an indirect relationship between the perceived parenting styles and identity statuses of adolescents mediated by individuation from parents. According to their resuhs, a parenting style that encouraged independence tended to produce positive individuation from the family for adolescents. This type of attachment gave the adolescents the necessary freedom to differentiate from their families and succeed in forming high status identities. Controlling parenting styles, however, resulted in a

PAGE 26

20 negative individuation from families, and these adolescents were prone to develop lower level identities than their peers from less restrictive homes. Direct relationships have also been found between parenting style and identity development. La Voie (1976) studied 120 students in grades 10, 1 1, and 12 and found high identity males came from homes with less regulation and control and more praise by parents than their lower identity status peers. Females with a high identity status came from homes with less maternal control and greater encouragement to discuss personal problems with their parents than did their low identity peers. Quintana and Lapsley (1987) found similar results. In a study of 101 undergraduates, they found that the children of families that encouraged communication and were not controlling were better prepared to deal with their ensuing crises. Quintana and Lapsley presumed that their family background prepared them to freely explore identity alternatives. Adolescents from controlling homes were more likely to be classified in the lower identity statuses than were their peers from less controlling homes. However, in two studies of seventhand eleventh-grade students, Enright, Lapsley, Drivas, and Fehr (1980) found that only fathers' parenting style was related to identity development. In the first study of 262 respondents, Enright et al. found that males with fathers who practiced a democratic style of parenting-analogous to Baumrind's authoritative parenting-tended to be classified in higher identity statuses than their peers with autocratic or permissive fathers. Females whose fathers practiced an autocratic parenting style, however, were more often classified in the higher identity statuses than were peers with democratic or permissive fathers. No relationship was found between mothers' parenting style and identity development. In the second study of 168 respondents, Enright et al. found that both males

PAGE 27

and females with democratic fathers were more often classified in the higher identity statuses than were their peers with autocratic or permissive fathers. As in the first study, no relationship was found between mothers' parenting style and identity development. In all, research supports the idea that parenting style is related to identity development. Precisely how the two are related, however, is far fi-om clear. An authoritative parenting style has been consistently demonstrated to be related to high identity status in males. Yet, both authoritative and authoritarian parenting styles have been found to be related to high identity status in females. The statistical relationship between identity status and permissive parenting, however, has not been considered. Fathers' parenting style has been shown to be related to identity development in adolescents. Mothers' parenting style has also been shown to be related to identity development in adolescents, yet not as consistently as a father's parenting style. Therefore, the gender of the adolescent and parent must be taken into account when assessing the relation between parenting style and identity development. Familism. MarkstromAdams and Adams (1995) pointed out that identity differences among the ethnic groups they studied did not occur in interpersonal identity and suggested that the emphasis on family and connectedness among ethnic minority groups fosters the development of interpersonal identity. For example, Mexican Americans are characterized as living in close proximity to many relatives with whom they engage in frequent interaction and mutual support, and these behaviors are believed to be supported by strong emotional ties to family and acceptance of family values (Keefe, 1984). Caucasians, in contrast, are considered to be less likely to have extended family living nearby or to strongly value such relationships. Research, however, has

PAGE 28

raised questions about the extent of familism among Mexican Americans. Although Farris and Glenn (1976) and Chandler (1979) found greater familism among Mexican American than Anglo American groups, the differences were not large. Also, contrary to the belief that Mexican Americans are high in familism, most of the Mexican Americans disagreed with most of the familism items. Valenzuela and Dombusch (1994) defined familism as the structural, attitudinal, and behavioral dimensions of family interactions. The structural aspect of familism is characterized by the presence of a nuclear and extended family, whereas the attitudinal aspect refers to concern for the welfare of the family. The behavioral dimension refers to the amount of contact with family members. Valenzuela and Dombusch argued that familism can serve as a form of social capital (Coleman, 1988). That is, family connectedness provides children with supportive relationships that enable them to develop positive mental health and to perform well in school. Valenzuela and Dombusch based their ideas on the work of Velez-Ibanez and Greenburg (1992), who argued that Mexican American children are raised in dense extended families that socialize the children to form strong emotional attachments with a wide variety of relatives who tend to live in nearby neighborhoods, and on research by Velez (as cited in Velez-Ibanez & Greenberg), who found that Mexican American mothers and their relatives had greater contact with infants than Caucasian mothers and relatives and the Mexican American children were in frequent contact with relatives. Valenzuela and Dombusch (1994) analyzed surveys of 2,666 Anglo and 492 Mexican students in 1987-88 and found that the Mexican American students were higher in familism than Caucasian students. They also found that for Mexican-origin students

PAGE 29

23 neither parent occupation nor familism, when considered separately, was related to students' grades but when parent education was high, familism was significantly related to Mexican origin students' grades. Valenzuela and Dombusch Qonjectured that the patriarchal ideology of the traditional Mexican family may limit female students' development, thereby suggesting the need to examine gender differences in the effect of familism and parent education on development. Other researchers have found significant differences in familism among African Americans, Hispanics, and Caucasians. Mindel (1980) reported that Hispanics demonstrated the highest degree of familism followed by AfHcan Americans and Caucasians, in that order. Farris and Glenn (1976) and Chandler (1979) also reported that Hispanics possess higher degrees of familism than Caucasians. The familism scale used by Chandler, however, consisted of the following four items: (a) Nothing in life is worth the sacrifice of moving away from your parents (b) When the day comes for a young man to take a job, he should stay near his parents, even if it means losing a good job o pportunity (c) When young people get married, their main loyalty still belongs to their parents and (d) When you need help of any kind, you can depend only on members of the family to help you out (p. 157). Using an expanded familism measure consisting of nine items with a sample of 381 Mexican Americans and 163 Anglo Americans, Keefe (1984) obtained resuhs that contradicted those of Chandler (1979) and Farris and Glenn (1976). She found that the Anglo Americans valued family as highly as the Mexican Americans, although the Anglo Americans were more likely to have a very small extended family or none at all, whereas

PAGE 30

24 the Mexican Americans had comparatively large ejctended families who visited frequently. Further, Keefe (1984) argued that differences between Hispanics and Caucasians in familism is a matter of quality instead of quantity. That is, she found that Hispanic and Caucasians had the same degree of familism, but Hispanic familism was associated v^th geographic stability, whereas Caucasian familism was associated with geographic mobility. In other words, Anglo and Mexican Americans are similar in their feelings of closeness to their families, but Anglos do not share the Mexican Americans' need for the physical presence of family members. In a 1988 study, Valenzuela and Dombusch (as cited in Valenzuela & Dombusch, 1994) compared Mexican and Anglo adolescents on familistic structure, attitudes, and contact. They found the Mexican-origin adolescents tended to be higher than the Anglo students on all three dimensions of familism. Reguero (1991) investigated the relationship among demographic variables, familism, and ego-identity status with 180 Puerto Rican residents and 107 Puerto Ricans who had immigrated to the U.S. mainland. Using multiple regression analysis, Reguero found that the variables sex, age, parents' education and occupation, and immigrant status, and familism accounted for some of the variance in achievement for all groups. She also found that the variables partially explained the variance in moratorium for males living in Puerto Rico and male and female immigrants, as well as partially explaining variance in foreclosure for female immigrants. The variance in the diffusion status for female respondents, regardless of group, was explained, in part, by the variables.

PAGE 31

25 As this review shows, research on the relationship between familism and identity development is far from clear. Though differences in familism have been found among African American, Hispanic, and Caucasian children, research is needed to determine the degree to which these differences affect identity development. Further, it appears that males and females experience familism differently. Gender, therefore, must be taken into account when investigating the relationship between familism and identity development. Acculturation Phinney (1990) argued that to complete their identity development, members of minority groups must resolve two issues not faced by members of the majority culture: (a) how to negotiate the different expectations of the two cultures and (b) how to deal with the negative perceptions the majority culture has toward the minority culture. According to Gonzalez and Cauce (1995), the way an adolescent member of a minority culture resolves these conflicts depends on the individual's style of acculturation into the majority lifestyle. On the basis of studies of members of minority groups. Berry (1980) identified four types of acculturation: (a) assimilation, whereby individuals give up their identification with their minority culture and identify with the majority culture, (b) biculturalism, whereby individuals develop positive identifications with their minority culture and the cultural majority, (c) rejection, whereby individuals identify with their minority culture and do not identify with the majority culture, and (d) marginalization, whereby individuals fail to identify with their minority culture or the majority culture. According to Gonzalez and Cauce (1995) and Peretti and Wilson (1995), individuals who adopt a bicultural model of acculturation should demonstrate more progressive identity development than those who adopt other models. That is, because

PAGE 32

26 bicultural adolescents integrate the expectations of their minority culture and the majority culture and are less vulnerable to negative perceptions of their minority culture because of their identification with it, they should have less difficulty developing the integrated sense of self characteristic of the achieved status. Although Streitmatter (1988) argued that minority adolescents "feel less comfort in the [majority culture] and as a result are more likely to conform to prescribed values and expectations of the adults in their lives" (p. 344), researchers have not examined the relationship between Marcia's (1976) identity statuses and the degree to which minority adolescents feel out of place in the majority culture. The findings of Mehan, Hubbard, and Villanueva (1994), however, support the idea that adolescents with a bicultural acculturation should demonstrate more progressive identity development than those with other acculturation orientations. Mehan et al. investigated the effects of a program to prepare underachieving African American and Hispanic adolescent students for college admission by placing them in classrooms with high achieving peers. Most of the students adopted what Gibson (1988) called accommodation without assimilation (similar to Berry's, 1980, bicultural orientation) toward academic tasks. That is, "they affirm[ed] their cultural identities while at the same time recognize[d] the need to develop certain cultural practices, notably achieving academically, that are acceptable to the mainstream" (Mehan et al., 1994, p. 105). By doing so these students resolved the two conditions set forth by Phinney (1990) for the development of an identity: (a) negotiate the different expectations of the cultures and (b) deal with the negative perceptions the majority culture has toward the minority culture. Also, as Gonzalez and Cauce (1995) and Peretti and Wilson (1995) argued, resolving the expectations of both the minority and

PAGE 33

27 majority cultures allows for the development of an integrated sense of self, the hallmark of the achievement status. In summary, a major task for minority adolescents is the resolution of conflicting expectations of the minority and majority cultures. As minority adolescents confront the identity crisis, they come to see themselves in one of four ways: estranged from the minority culture and solely a part of the majority (assimilated), identified with both the minority and majority culture (bicultured), estranged from the majority culture and solely a part of the minority (rejected), or estranged from both cultures (marginalized). As the development of a higher identity is built on an integrated sense of self (Phinney, 1990), it is plausible that minority adolescents who adopt the bicultural orientation of acculturation are more likely to reach the achieved identity status. Research Questions Some studies have reported a relationship between ethnicity and identity, and others have not. None of the studies has adequately controlled for the demographic variables of SES or family structure, which have been found to be related to identity status. Further, as this review shows, cultural and family process variables (i.e., acculturation, familism, and parenting style) may be related to identity status. The purpose of this study was to examine the relationships of these demographic, cultural, and family variables to identity status. The research questions addressed in this study were: 1 Is ethnicity related to identity status? 2. Is ethnicity related to identity status after gender, SES, and family status are controlled? 3. Is identity status related to familism? 4. Is identity status related to parenting style? 5. Is identity status related to acculturation for African American and Hispanic adolescents?

PAGE 34

28 Significance of the Study Formation of an identity is a defining aspect of personality development (Marcia, 1993). Marcia's (1966, 1976) extension of Erikson's (1963, 1968) theory of identity and its subsequent revisions (Archer, 1981; Grotevant, Thorbecke, & Meyer, 1982) have highlighted the importance of identity to adolescents' personality development (Marcia, 1993). Theoretical Significance Most attempts to apply the Marcia (1993) model of identity development to minority adolescents have found that adolescent ethnic minorities tend to have less progressive identities than do their Caucasian peers (Abraham, 1986; Hauser, 1972; MarkstromAdams & Adams, 1995; Streitmatter, 1988). A number of factors have been proposed to account for this finding, including failure to control for differences in SES and family constellation as well as differences in acculturation and family values. This study is designed to determine whether these factors need to be incorporated into Marcia's model to provide an adequate conception of minority adolescent identity development. Practical Significance Identity research has revealed that the more progressive identity statuses (achievement and moratorium) are related to more adaptive characteristics of adulthood than the less progressive statuses (foreclosure and diffusion) (Archer, 1989b). For example, individuals in the more progressive statuses have demonstrated higher levels of moral reasoning, autonomy, and internal locus of control (Josselson, 1994; Marcia, 1993). As a result, researchers have recommended the development of intervention

PAGE 35

29 Strategies to foster successful resolution of the identity crisis (Josselson, 1994; RotheramBorus & Wyche, 1994; Waterman, 1994). It is not clear whether such intervention strategies are warranted for adolescents from minority groups. As Erikson (1963) noted, culture affects the development of identity. That is, cultural expectations and practices encourage and limit adolescents' opportunity for exploration and commitment in the various identity domains. Given that researchers have found that African American and Hispanic adolescents are more foreclosed than their Caucasian peers, Spencer and MarkstromAdams (1990) stated that the foreclosure status may be adaptive in certain cultures: "Exploration of identity may not be desirable for such youth when social and ideological roles are clearly defined by the community" (p. 298). This study was designed to determine whether the relationship between ethnicity and identity development is attributable to SES, family configuration, and family values. It may be inappropriate, however, to conclude that identity intervention is imnecessary because of differences in cultural expectations. Markstrom-Adams and Spencer (1994) argued that despite minority cultural support many minority adolescents are restricted in their ability to explore identity alternatives because of the limitations in their social environment. That is, a significant portion of minority adolescents grow up in such impoverished, discriminatory environments that they cannot take advantage of the role models available that are necessary for adequate identity exploration. Consequently, it is important to determine whether the foreclosure of minority adolescents is attributable to economic variables or to cultural values. i

PAGE 36

CHAPTER 2 METHOD Research Participants Three hundred students from a community college and a university in Florida were recruited to participate in this study. Respondents included 82 African American students (30 men and 52 women), 120 Hispanic students (50 men and 70 women), and 98 Caucasian (40 men and 58 women) late adolescents with a mean age of 20.8 years. Instruments Identity Status Identity status was assessed using the Extended Objective Measure of Ego Identity Status HI (EOM-EIS EI, Forbes & Ashton, 1996). The EOM-EIS IH is a 64-item questionnaire. Respondents are instructed to indicate the degree to which an item is reflective of themselves using a 6-point Likert scale, ranging from strongly disagree to strongly agree. The scale assesses ideological identity in the areas of occupation, politics, religion, and philosophical lifestyle and interpersonal identity in the areas of gender roles, dating, friendship, and recreation. Each of the eight subdomains is measured by eight items with two items reflecting each of the four identity statuses: achievement, moratorium, foreclosure, and diffusion. Forbes and Ashton obtained the following Cronbach's alpha coefficients for scores on the four identity statuses: ideological achievement = .57; moratorium = .68; foreclosure = .64; diffusion = .53; interpersonal achievement = .62; moratorium = .63; foreclosure = .77; diffusion = .51. 30

PAGE 37

31 Acculturation Acculturation orientation was measured with an adaptation of a questionnaire developed by Berry/ Trimble and Ol medo (19 8^1 to assess the four types of acculturation-assimilation, integration, separation, and marginality. The questionnaire consists of 24 items, with an item representing each acculturation type for each of the following domains: occupation, dating, friendship, life-style, media, and language. Responses were made on a 5-point Likert scale, ranging from strongly disagree to strongly agree. Parenting Style Parenting style was assessed using the two forms of Buri's (1991) Parental Authority Questionnaire (PAQ). The PAQ measures parental authority of mothers and fathers. Each scale is a 30-item questionnaire designed to measure the degree to which an individual believes his or her parent(s) exhibited permissive, authoritarian, and authoritative parenting styles. Responses are made to the items on a 5-point Likert scale. With a sample of 61 psychology students who completed the questionnaire twice over a 2-week period, Buri obtained the following test-retest reliability coefficients: mother permissiveness = .75; authoritarianism = .85; authoritativeness = 82; father permissiveness = .74; authoritarianism = .87; authoritativeness = .85. Buri (1991) obtained evidence for the construct validity of the scale in a study of 127 undergraduate psychology students. His results were consistent with predictions from Baumrind's (1971) theory: mother's authoritarianism items were negatively correlated with mother's authoritativeness and permissiveness items (r = -.38, p < .0005; r = -.48, p < .0005). Mother's authoritativeness and mother's permissiveness were not significantly

PAGE 38

32 correlated (r = .07, p > .10). Father's authoritarianism was negatively correlated with father's authoritativeness and father's permissiveness (r = -.50, p < .05; r = -.52, p < .05). Father's authoritativeness and father's permissiveness items were not significantly correlated (r = 12, p > 10). In the same study, Buri reported the following data on criterion-related validity: The authoritative subscales were positively related to parental nurturance (mother r = .56, p < .0005; father r = .68, p < .0005), whereas the authoritarian subscale was negatively related to parental nurturance (mother r = -.36, p < .0005; father r = .-53, p < .0005). The permissiveness subscales were not significantly related to parental nurturance (mother r = .04, p < 10; father r = 13, p < 10), respectively. In a study of 69 undergraduate psychology students, Buri (1991) assessed the relationship between social desirability and respondents' answers to the PAQ subscales. None of the subscales was significantly related to the Marlowe-Crowne Social Desirability Scale. Family Configuration Family configuration was assessed by the respondents' self-report. Participants were asked to indicate the category that identified their family type on the questionniare. Categories included dual parent household without stepparents; dual parent household with stepparents; single-parent household-parents divorced; single-parent householdparent never married, single-parent household-parent deceased, and an Other category. Familism Familism was assessed in each of the three domains of the construct. Familism behavior was measured using a single item taken from the work of Valenzuela and Dombusch (1994) that asks respondents to indicate on a 6-point Likert scale "how often

PAGE 39

33 [respondents] saw or talked on the telephone with adult relatives who do not live with them" (p. 24). Familism structure was also measured with a single item on a 6-point scale used by Valenzuela and Dombusch. It asked the number of relatives that live within an hour's drive of the respondent. Familism attitudes were measured using a scale developed by Keefe (1984). The measure is a 9-item, 5-point Likert scale. Keefe (1984) obtained an internal consistency coefficient of .68 for the items in a study of 381 Mexican American and 163 Caucasians. Socioeconomic Status (SES) The SES of respondents' parents was assessed using ffoUincshead's (1975) Four Factor Index of Social Status (FFISS). The factors of the scale include gender, occupation, education, and marital status. Gender, however, is not included in the calculation of social status, though it can be used with males or females. Occupation level is assessed on a 9-point scale of increasing complexity as determined by the 1970 United States Census. The educational factor is measured using a 7-point scale that ranges from less than seventh grade to graduate professional training Marital status is assessed by first assigning respondents to one of the following classifications: unmarried individual (male or female), head of a household (male or female), or family in which both husband and wife are gainfully employed Then social status is calculated by mukiplying the scores of the occupation and education factors by factor weights. Scores range from 8 to 66 with higher scores indicating higher social status. When social status is measured for families with two employed persons, the occupation and education scores of both persons are considered in the calculation. The Four Factor Index is based on the belief that social status is a "multidimensional aggregate" (Gottfried, 1985, p. 92). That is, the four

PAGE 40

34 factors-gender, occupation, education, and marital status-are combined into a single index rather than considered as separate factors. The FFISS has been described as a highly reliable and valid measure of social status (see Hollingshead, 1975). Gottfried found that the FFISS was more highly correlated with children's developmental outcomes in a longitudinal (30 months) study of 130 young children and their parents than were two other SES indices, the Siegel Prestige Scale and the Revised Duncan Socioeconomic Index. In addition, Gottfried pointed out that the Hollingshead Four Factor Index can be used to estimate the SES of unmarried heads of households as well as families. Procedures Each participant signed the consent form (see Appendix B) and anonymously completed the instruments that consisted of the demographic questionnaire, EOM-EIS ID, PAQ, acculturation, and familism questionnaires. Data were collected from summer, 1997 to spring, 1999. Respondents were told not to answer any items that they did not wish to and to ask questions at any time. Administration time took between 45 min to 1 hr. Research Questions Research Question #1 The first set of research hypotheses in this study was designed to determine if ethnic identification was related to identity status classification. Respondents' scores on the EOM-EIS were used as the dependent variable in the model. The independent variable was ethnic identification.

PAGE 41

35 Research Question #2 The second group of research hypotheses was designed to determine if ethnic identification was related to identity status when the demographic variables-SES, gender, family status-were controlled. The dependent variable was the EOM-EIS scores. Ethnic identification was the independent variable and SES, gender, and family status responses were covariates. Research Question #3 The next set of hypotheses considered the relation of familism to identity status for each of the three ethnic groups. Separate analyses were conducted on the variables to determine their relationship with identity status. In both sets of analyses respondents' EQM-EIS scores were the dependent variable; whereas scores on the familism measures served as predictor variables and ethnic identity was a covariate. Research Question #4 The next set of hypotheses considered the relation of parenting style to identity status for each of the three ethnic groups. Separate analyses were conducted on the variables to determine their relationship with identity status. In both sets of analyses respondents' EQM-EIS scores were the dependent variable; whereas scores on the PAQ served as predictor variables and ethnic identity was a covariate. Research Question #5 This group of hypotheses considered the relation between the different forms of acculturation orientation available to African American and Hispanic adolescents and identity status.

PAGE 42

CHAPTER 3 RESULTS This study was designed to investigate the relation between ethnic identification and identity status before and after controlling for the effect of theoretically significant demographic variables-gender, SES, and family status. It was also of interest to examine the relationships among parenting style, familism, and identity status classification. The final question of interest was whether acculturation orientation is related to identity status classification. This chapter presents a summary of the statistical findings of this study including (a) the summary statistics of demographic variables (age, gender, SES, family status), parenting style, familism, acculturation orientation, and identity status classification by ethnic group and (b) resuhs of the analyses of the relationships among the variables of interest. Summary Statistics Demographic Variables Of the 300 respondents who participated in this study, 82 identified themselves as African American, 120 as Hispanic, and 98 as Caucasian. All participants in this study were between the ages of 18 and 22. The mean age of African American respondents was 20.8 years. The mean ages of Hispanic and Caucasian respondents were 20.8 and 20.6 years, respectively. Forty percent of the respondents were men (n = 120); 60% of the respondents were women (n = 180). Of African American respondents, 62 (76%) 36

PAGE 43

37 Table 1. Family Status Percentages bv Ethnicity African American Hispanic Caucasian Status % % % Dual parent44 66 63 No step parent Dual parent19 6 17 Stepparents Single parent12 5 0 Never married Single parent17 19 16 Parent divorced Single parent4 3 2 Parent deceased Other 4 12 identified themselves as women, and 20 (24%) identified themselves as men. Sixty-five Hispanic (54%) and 71(72%) Caucasian respondents identified themselves as womem; 55 (46%) Hispanic and 27 (28%) Caucasian respondents identified themselves as men. Table 1 displays the percentage of respondents in each of the different family statuses by ethnic group. A chi-square analysis of the relationship between statuses by ethnic group resuhed in a non-significant relationship. Table 2 lists the means and standard deviations of respondents' scores on the Four Factor Index of Social Status (Hollingshead, 1975). An ANOVA indicated a significant

PAGE 44

38 Table 2. Means and Standard Deviations of Socioeconomic Status bv Ethnicity African American Hispanic Caucasian M S M M 42.35. 12.17 43.08. 12.74 49.88b 8.96 Note. Means with different subscripts are significantly different. difference among the means, F(2, 291) = 12.54, p < .0001. Post-hoc comparisons using the Tukey method revealed that Caucasian SES scores were significantly different from both African American and Hispanic SES scores (p < .05). The SES scores of African American and Hispanic respondents did not differ. Parenting Style The means and standard deviations of the different parenting styles are listed by ethnicity in Table 3. One-way analyses of variance indicated significant differences in mean scores for mothers' authoritarian parenting style [F(2, 277) = 16.02, g < .0001] and fathers' authoritarian parenting style [(F(2, 244) = 5.73, p < .0037)]. The Tukey post-hoc comparison of scores indicated that African American and Hispanic adolescents' mean scores were significantly different from the Caucasian mean score on mother's and father's authoritarian parenting style (p < .05). The parenting-style scores of the African American adolescents and Hispanic adolescents did not differ.

PAGE 45

39 Table 3. Parenting Style Mean Scores bv Ethnicity African American Hispanic Caucasian Style M c M Q IVi c /\UlllUi ILoU Vv 41.17 7.32 42.55 6.65 7.57 /vuinoniandn 36.99. 6.94 35.54a 6.34 31.38b 7.18 rCl IIllSSIVC 29.82 J. /4 jV. Ij ^ 18 J.lO 11 44 5 56 Father Authoritatiye 37.06 6.93 38.40 6.65 39.23 9.58 Authoritarian 36.99, 6.94 35.53. 8.93 34.95b 11.29 Permissive 26.01 8.12 27.25 8.12 28.21 8.29 Note. Means with different subscripts are significantly different within the authoritarian rows. Familism The means and standard deviations of the forms of familism are listed by ethnic group in Table 4. Significant differences were found among structure means, F(2, 282) = 9.79, p < .0001, for familism. Post-hoc analysis indicated that Afiican American and Hispanic adolescents scores were significantly different fi"om Caucasian scores on familism-structure (p < .05). No significant differences were found between African American and Hispanic adolescents.

PAGE 46

40 Table 4. Means and Standard Deviations of Familism Type bv Ethnicity African American Hispanic Caucasian Type M S M S M S Stmcture 20.01, 27.02 11.31, 19.68 6.92b 9.70 Behayior 4.78 1.70 4.83 1.41 4.85 1.52 Attitude 32.94 4.13 33.48 4.24 32.14 3.61 Note. Means with different subscripts are significantly different within the structure row. Acculturation Orientation Table 5 presents the means and standard deviations of the acculturation orientations for African American and Hispanic adolescents. In the comparison of the means of acculturation orientation styles between African American and Hispanic adolescents, significant differences were found for the integration (t(l, 177) = 5.95, p < .0001) or marginalized (t(l, 180) = 2.23, p < .05) styles. Identity Status Classification Identity status was determined for each respondent using the method outlined by Adams, Shea, and Fitch (1979). The number of respondents classified in the statuses of each of the domains of identity is listed in Table 6 and Table 7. In the ideological and interpersonal domains the majority of respondents in each ethnic group were classified in a form of identity moratorium. Differences among the groups on identity variables will be presented in the next section.

PAGE 47

41 Table 5. Means and Standard Deviations of Acculturation Orientations for African American and Hispanic Adolescents African American Hispanic Orientation c M s Integration 26.99a 4.27 30.64b 3.89 Accommodation 10.56 3.69 9.90 3.75 Separated 14.01 4.01 12.77 S 06 J .\J\J Marginalized 13.59a 3.78 14.82b 3.62 Note. Row means with different subscripts are significantly different. Table 6 Tdeoloeical Identity Status Percentages bv Ethnicitv Status African American (n=82) Hispanic (n= 120) Caucasian (n = 98) Achieved 11 8 13 Moratorium 4 13 9 Foreclosure 16 15 8 Diffusion 2 5 14 Moratorium — undifferentiated 64 50 49 Transition 4 7 6 Note Column totals may not equal 100% due to rounding.

PAGE 48

42 Table 7. Interpersonal Identity Status Percentages by Ethnicity Status African American (n=82) Hispanic (n= 120) Caucasian (n = 98) Achieved 7 12 10 Moratorium 13 11 12 Foreclosure 11 11 6 Diffusion 11 7 10 Moratorium — undifferentiated 49 51 52 Transition 7 9 7 Note. Column totals may not equal 100% due to rounding. Results of Statistical Analyses For each question, eight analyses were conducted. A .05 familywise error rate was used to judge significance. However, in reporting analyses, resuhs that were significant at a conventional .05 Type I error rate are noted. The research questions for this study were 1. Is ethnicity related to identity status? 2. Is ethnicity related to identity status when gender, SES, and family status are controlled? 3. Is identity status related to parenting style when ethnicity is controlled? 4. Is identity status related to familism when ethnicity is controlled? 5. Is identity status related to acculturation for African American and Hispanic adolescents? Question #1 The means and standard deviations for the status analyses are reported by ethnic group in Table 8. To determine if ethnicity was related to identity status a one-way

PAGE 49

43 Table 8. Mean Scores and Standard Deviations of Identity Statuses African American Hispanic Caucasian M S M S M S Ideological Achieved 23.28 3.76 23.46 3.52 23.01 3.25 Moratorium 25.97 3.91 26.74 3.24 26.51 3.54 Foreclosed 20.44 4.74 22.53 4.72 21.63 3.95 Diffused 41.64 27.58 32.81 19.75 27.42 9.73 Interpersonal Achieved 25.28 4.29 25.23 4.12 25.04 3.99 Moratorium 25.23 4.26 25.37 3.61 25.52 4.21 Foreclosure 25.60 3.89 26.21 3.88 24.70 3.63 Diffused 25.39 4.24 25.74 3.62 24.97 3.52 ANOVA was conducted on the means of each of the identity domain scores for African American, Hispanic, and Caucasian adolescents. When the Bonferroni-Holm procedure was used to confrol for the fact that ethnic group differences were tested on eight variables, none of the F statistics was significant. When a conventional .05 alpha level was used, ethnic differences among identity status scores were found for one statusdiffused, ideological, F(2, 287) = 4.13, p = .017.

PAGE 50

44 Question #2 The general linear model was used to test ethnic group differences with the demographic variables controlled (gender, SES, family status). No significant ethnic group differences emerged when Bonferroni corrections were made for multiple comparisons. With a conventional .05 alpha level, ideological diffusion was most closely related to ethnicity, F(2, 256) = 3.63, p = .028. Of the ethnic groups, Caucasian adolescents reported the highest means on ideological diffusion. In the research literature, gender-SES_and family status have been identified as demographic variables related to identity development. However, when these variables were entered as covariates in the analysis of the relationship between identity status and ethnicity, and when corrections were made to the p value for multiple analyses, none of the covariates was found to be significantly related to the identity status domains (see Tables 9 and 10). The covariates with p values less than 10 were gender in ideological moratorium (p = .042), ideological foreclosure (p = 0.09), interpersonal achieved (p = 0.008), and interpersonal foreclosure (p = .026), and SES in ideological moratorium (p = .042) and interpersonal foreclosure (p = .072). Question #3 The purpose of the third research question was to determine whether identity status variables were related to familism with ethnicity controlled. The covariate. ethnicity, was not found to be related to identity status. With Bonferroni corrections for multiple comparisons, no significant differences were found between ideological and interpersonal identity status and forms of familism, with ethnicity controlled. Results are

PAGE 51

45 Table 9. ANCOVA Results for Ideological Identity Statuses Source ss df Mean Square F /\.cnievcu Race 47.51 2 23.75 0.72 0.49 VJCllUCi 9.66 1 9.66 0.29 0.59 33 24 1 33.24 1.01 0.32 raiiiiiy 125 40 5 25.08 0.76 0.58 JViUralUI lUlll Race 102.46 2 51.23 1.47 0.23 145.94 1 145.94 4.18 0.04 29 63 1 29.63 0.85 0.36 Fnmilv ITttlillljf 175.87 5 35.17 1.01 0.41 P" r^r pf* 1 n Qi 1 rp Race 153.62 2 76.81 2.63 0.07 82.88 1 82.88 2.84 0.09 0 99 1 0.99 0.03 0.85 F amilv 99 14 5 19.83 0.68 0.64 Race 258.75 2 129.38 3.63 0.28 Gender 88.73 1 88.73 2.49 0.12 SES 0.39 1 0.39 0.01 0.92 Family 180.64 5 36.18 1.01 0.41

PAGE 52

46 Table 10. ANCOVA Results for Interpersonal Identity Statuses Source ss df Mean Square F B Achieved Race 178.59 2 Gender 222.65 1 SES 12.35 1 Family 250., 54 5 Moratorium Race 2.27 2 Gender 0.08 1 SES 25.50 1 Family 32.47 5 Foreclosure Race 44.83 2 Gender 154.18 1 SES 100.95 1 Family 113.68 5 Diffusion Race 34.87 2 Gender 70.25 1 SES 0.99 1 Family 98.28 5 89.30 2.87 0.06 222.65 7.15 0.01 12.35 0.40 0.53 50.11 1.61 0.16 1.18 0.04 0.96 0.08 0.00 0.96 25.50 0.85 0.36 6.49 0.22 0.96 22.42 0.73 0.49 154.18 4.99 0.03 100.95 3.27 0.07 22.73 0.74 0.60 17.44 0.66 0.52 70.25 2.67 0.10 0.99 0.04 0.85 19.66 0.75 0.59

PAGE 53

47 listed in Tables 1 1 (ideological identity statuses) and 12 (interpersonal identity status). No significant interactions between familism and ethnicity were found. Table 1 1 Summary of GLM Analyses for Relationship betw een Ideological Identity Status and Familism. with Ethnicity Controlled Source ss df Mean Square F Achieyed Family structure 2.87 1 2.87 0.09 0.77 Family behayior 42.73 1 42.73 1.29 0.26 Family attitude 66.50 1 66.50 2.07 0.15 Moratorium Family structure 41.90 1 41.90 1.14 0.29 Family behayior 105.16 1 105.16 1.03 0.32 Family attitude 49.46 1 49.46 1.39 0.24 Foreclosure Family structure 103.91 1 103.91 3.51 0.06 Family behayior 45.70 1 45.70 1.53 0.22 Family attitude 169.17 1 169.17 5.75 0.02 Diffusion Family structure 59.95 1 59.95 1.74 0.19 Family behayior 83.85 1 83.85 2.41 0.12 Family attitude 27.06 1 27.06 0.79 0.38

PAGE 54

48 Table 12. Summary of GLM Analyses for Relationship between Interpers onal Identity Status and Familism. with Ethnicity Controlled Source Mean Square F Achieved Family structure 0.41 0.41 0.01 0.91 Family behavior 15.95 15.95 0.48 0.49 Family attitude 47.39 47.39 2.48 0.22 Moratorium Family structure 26.93 26.93 0.90 0.34 Family behavior 134.58 134.58 4.97 0.03 Family attitude 25.34 25.34 0.87 0.35 Foreclosure Family structure 28.06 28.06 0.88 0.35 Family behavior 93.86 93.86 2.83 0.09 Family attitude 62.94 62.94 1.94 0.17 Diffusion Family structure 3.08 3.08 0.12 0.73 Family behavior 3.94 3.94 0.14 0.71 Family attitude 99.08 99.08 3.85 0.05

PAGE 55

Question #4 The aim of the fourth research question was to determine whether identity status variables were related to parenting style scores after controlling for ethnicity. The covariate, ethnicity, was not found to be related to identity status. Mothers' parenting style. The F and probability values for the relationship between identity status and parenting style, with ethnicity controlled, are reported in Tables 13 and 14. Inspection of Table 13 reveals the following significant positive relationships between mother's parenting style and ideological identity status: authoritative style with ideological achievement, F (2, 265) = 8.12, p = .005, permissive style with ideological moratorium, F (2, 265) = 18.92, p = .0001, authoritarian style with ideological foreclosure, F (2, 265) = 13.48, p = .0003, and permissive style with ideological diffusion, F (2, 265) = 8.77, p = .003. Significant relationships were also found between mothers' parenting style and interpersonal identity status with ethnicity controlled (see Table 14): authoritative style with achievement, F (2, 263) = 12.3 1, p = .0005, permissive style with moratorium, F (':;, 263) = 12.51, p = .0005, and diffusion, F (2, 263) = 12.20, p = .0006. Each of these relationships was also positive. Analyses were performed to determine if significant interactions existed between ethnicity and mother's parenting style. The interaction between mother's authoritative style and ethnicity was significant, F (2, 265) = 5.43, p = .004, for the ideological achieved status. As a group, African Americans demonstrated the strongest positive relationship between ideological achievement and mother's authoritative parenting style, followed by Caucasians then Hispanics.

PAGE 56

50 Table 13. Summary of GLM Analyses for the Relationship between Ideological Identity Status and Mother's Parenting Style, with Ethnicity Controlled Source jif Mean Square Achieved Authoritative 257.29 Authoritarian 95.78 Permissive Moratorium Authoritative Permissive Foreclosure Permissive Diffusion Authoritarian Permissive 90.61 0.49 Authoritarian 101.88 622.93 Authoritative 66.07 Authoritarian 389.76 15.21 Authoritative 296.27 1.07 290.78 257.29 95.78 90.61 0.49 101.88 622.93 66.07 389.76 15.21 296.27 1.07 290.78 8.12 3.01 2.87 0.01 2.84 18.92 2.19 13.48 0.95 3.34 0.03 8.77 0.005 0.08 0.09 0.91 0.09 0.0001 0.14 0.0003 0.41 0.02 0.82 0.0033

PAGE 57

51 Table 14. Summary of GLM Analyses for the Relationship Between Interpersonal Identity Status and Mother's Parenting Style, v^ith Ethnicity Controlled Source df Mean Square Achieyed Authoritatiye 368.50 Authoritarian 103.82 Permissiye 58.93 Moratorium Authoritative 9.30 Authoritarian 2.00 Permissive 355.88 Foreclosure Authoritative 195.67 Authoritarian 80.85 Permissive Diffusion 99.34 Authoritatiye 127.45 Authoritarian 70.63 Permissive 304.69 368.50 103.82 58.93 9.30 2.00 355.88 195.67 80.85 99.34 127.45 70.63 304.69 12.31 3.27 1.86 0.30 0.07 12.51 5.99 2.50 3.07 4.95 2.78 12.20 0.0005 0.07 0.17 0.58 0.80 0.0005 0.02 0.12 0.08 0.03 0.10 0.0006

PAGE 58

52 Fathers' parenting style. Inspection of Table 15 reveals the following significant relationships between father's parenting style and ideological identity status with ethnicity controlled: authoritative style with foreclosure, F (2, 265) = 14.87, p = .0001, permissive style with diffusion, F (2, 234) = 11.31, p = .0009. Each relationship was positive. The same general findings were obtained for the relationship between interpersonal identity status and father's parenting style (see Table 16). That is, analyses revealed the following significant relationships between father's parenting and interpersonal identity status: authoritative style with foreclosure, F (2, 169) = 15.07, p = .0001, permissive style with diffusion, F (2, 232) = 30.02, p = .0001. Each relationship was positive. As with the ideological identity statuses, no relationship was found between father's parenting style and interpersonal achievement or moratorium. The interaction of father's parenting style and ethnicity was not significant. Question #5 The purpose of this question was to measure the relationship between identity status and forms of acculturation orientation. The resuUs of the Q] M analysp.*; are presented in Table 17 (ideological identity) and Table 18 (interpersonal identity). A significant, positive relationship was found between acculturation orientation and ideological identity status with ethnicity controlled-marginalized acculturation with diffusion, F (1, 170) = 12.27, p = .0006. The following positive significant relationships between acculturation orientation and interpersonal Identity were found: marginalized acculturation with diffusion, F(l, 173)=18.32, p = .0001, and separated acculturation foreclosure, F(l, 168) = 24.92, p = .0001. Analyses were conducted to determine whether

PAGE 59

53 there was an interaction between acculturation orientation and ethnicity for identity statuses. No significant interactions were found. Table 15. Summary of GLM Analyses for Relationship between Ideological Identity Status and Father's Parenting Style, with Ethnicity Controlled Source SSt Mean Square F Achieyed Authoritatiye 114.57 114.57 3.59 0.06 Authoritarian 179.03 179.03 5.65 0.02 Permissiye 105.95 105.95 3.31 0.07 Moratorium Authoritative 9.76 9.76 0.26 0.61 Authoritarian 9.15 9.15 0.24 0.63 Permissiye 233.77 233.77 6.38 0.01 Foreclosure Authoritatiye 440.73 440.73 14.87 0.0001 Authoritarian 51.21 51.21 1.63 0.20 Permissiye 8.21 8.21 0.26 0.61 Diffusion Authoritatiye 37.85 37.85 1.09 0.30 Authoritarian 29.49 29.49 0.79 0.38 Permissiye 381.99 381.99 11.31 0.0009

PAGE 60

54 Table 16. Summary of GLM Analyses for Relationship between Interpersonal Identity Status and Father's Parenting Style Controlling for Ethnicity Source df Mean Square Achieyed Authoritative 108.05 Authoritarian 196.54 Permissive 149.61 Moratorium Authoritative 6.43 Authoritarian 0.21 Permissive 100.28 Foreclosure Authoritative 482.96 Authoritarian Permissive Diffusion Authoritative 9.47 22.00 5.25 Authoritarian 45.52 Permissive 691.72 108.05 196.54 149.61 6.43 0.21 100.28 482.96 9.47 22.00 5.25 45.52 691.72 3.45 6.37 4.87 0.21 0.01 3.33 15.07 0.28 0.68 0.19 1.74 30.02 0.06 0.01 0.03 0.65 0.93 0.07 0.0001 0.60 0.41 0.66 0.19 0.0001

PAGE 61

55 Table 17. Summary of GLM Analyses for Relationship between Ideological Identity Status and Acculturation Orientation, with Ethnicity Controlled Source ss df Mean Square F c Achieved Accommodation 16.67 16.67 0.62 0.43 Separation 17.20 17.20 0.64 0.42 Marginalized 13.72 13.72 0.51 0.48 Integration 193.13 193.13 7.47 0.007 Moratorium Accommodation 36.02 36.02 1.13 0.29 Separation 25.18 25.18 0.76 0.38 Marginalized 4.51 4.51 0.14 0.71 Integration 6.09 6.09 0.19 0.67 Foreclosure Accommodation 37.63 1 37.63 1.29 0.26 Separation 154.99 1 154.99 5.46 0.02 Marginalized 52.65 1 52.65 1.81 0.18 Integration 29.06 1 29.06 1.00 0.32 Diffusion Accommodation 50.35 50.35 1.68 0.20 Separation 1.68 1.68 0.06 0.81 Marginalized 321.88 321.88 12.27 0.0006 Integration 14.51 14.51 0.49 0.49

PAGE 62

56 Table 18. Summary of GLM Analyses for Relationship between Interpers onal Identity Status and Acculturation Orientation with Ethnicity Controlled Source ss Mean Square E C Achieved Acconmiodation 0.04 0.04 0.00 0.97 Separation 123.06 1 123.06 3.73 0.06 Marginalized 114.39 1 114.39 3.53 0.06 Integration 21.34 21.34 0.65 0.42 Moratorium Accommodation 123.17 123.17 4.19 0.04 Separation 73.57 1 73.57 2.35 0.13 Marginalized 86.26 86.26 2.86 0.09 Integration 54.35 54.35 1.79 0.18 Foreclosure Accommodation 44.94 1 44.94 1.47 0.23 Separation 689.77 1 689.77 24.92 0.0001 Marginalized 12.92 12.92 0.41 0.52 Integration 96.19 96.19 3.08 0.08 Diffusion Accommodation 73.64 73.64 2.65 0.11 Separation 6.24 6.24 0.22 0.64 Marginalized 472.82 472.82 18.32 0.0001 Integration 14.77 14.77 0.52 0.47

PAGE 63

CHAPTER 4 DISCUSSION Drawing from Erikson's theory of psychosocial development, Marcia (1966, 1993) operationalized ego-identity development to include four statuses of identityachieved, moratorium, foreclosure, and diffusion. Some researchers (Abraham, 1986; Hauser, 1972; MarkstromAdams & Adams, 1995; Streitmatter, 1988), using Marcia's model, have found disparate rates of identity development among ethnic groups. Caucasian adolescents were found to have more progressive identities than their peers who were members of ethnic minorities do. Other studies, however, have failed to replicate these findings (Abraham, 1986; Rotheram-Borus, 1989). Explanations of these findings have centered on inadequate measures of the relation of demographic and cultural variables with identity development. It was the purpose of this study, therefore, to address the limitations of previous research to more accurately assess the identity development of minority adolescents. This final chapter, in part, is a discussion of the results obtained from the statistical analyses outlined in chapter 3. Implications and limitations of this study follow. The chapter concludes with recommendations for future research. Discussion of Resuhs of Statistical Analyses Researchers who have found differences among ethnic groups in identity development have largely concluded that the differences between non-Caucasian and Caucasian adolescents in identity development are due to the lack of opportunities for 57

PAGE 64

58 social exploration available to minority groups. The proposed barriers to their exploration include SES, gender, family status, familism, and parenting style. The research questions examined in this study were designed to explore the relationship of these variables to identity development. Ethnicity and Identity Development The first research question was whether there is a difference among Afi-ican American, Caucasian, and Hispanic adolescents in identity development. Specifically, more Caucasian adolescents were expected to be in the progressive identity statuses in comparison to their African American and Hispanic peers. Tests of this hypothesis indicated no significant differences among the respondents' identity status scores. Though the respondents of this study were late adolescents, they were also students at institutions of higher education. College students may possess advantages that non-collegiate adolescents do not have in experiencing a crisis and exploring alternative roles and values. Demographic Variables as Covariates of Ethnicity The second question of this study was whether Afiican American, Caucasian, and Hispanic adolescents differ in identity development, when demographic variables (SES, gender, family status) are controlled. A number of studies have suggested that the social and structural differences associated with ethnic identification are explained, in part, by SES, gender, and family status. That is, researchers have proposed that because ethnic minority adolescents tend to live in more impoverished and less stable family environments than their Caucasian peers and these differences inhibit progressive identity development for ethnic minority adolescents (Phinney, 1990). Research also supports the

PAGE 65

59 contention that males and females experience identity development differently (Marcia, 1993). In tests of the statistical hypothesis, no significant differences were found among the respondents' identity status scores, when SES, gender, and family status were controlled. Also, although there were significant differences among groups in their report of SES and family status, none of the demographic variables were found to be significant covariates of identity status. Familism The third question of this study was whether African American, Caucasian, and Hispanic adolescent identity development is related to familism. Familism is defined as the connectedness one has with immediate and extended family members. Current operationalization of the concept includes three dimensions: structural, attitudinal, and behavioral. Structural familism refers to the number of relatives living in close proximity to a person; attitudinal familism is a measure of a respondent's beliefs as to the benefit of contact with family members. Behavioral familism refers to the amount of time people stay in contact with family members. Valenzuela and Dombusch (1994) proposed that a high degree of familism serves as a sort of social capital that enables adolescents to develop positive mental heaUh characteristics. Because differences in familism have been found among Afi-ican American, Caucasian, and Hispanic adolescents, it was of interest to measure the relationship between identity status and familism with ethnicity controlled. Testing the relationship between familism with ethnicity controlled resuhed in no significant relationships.

PAGE 66

60 Parenting Style Only a few studies have considered the relationship between identity status and parenting style. This was the fourth question of the study. In previous studies an authoritative parenting style has been associated with the progressive identity statuses (Waterman, 1993), suggesting that experiencing a parenting style that encourages independence with guidance may enable adolescents to endure a crisis with the intention of working through it. In support of this premise, a restrictive parenting style has been found to be related to the regressive identity statuses (Quintana & Lapsley, 1987). As for the relationship between a permissive parenting style and identity development, no studies could be found that investigated this relationship. Inasmuch as a permissive parenting style encourages exploration, which may precipitate a crisis, this style also provides the least amount of parental direction, which may contribute to an inability to commit to identity options. The statistical tests indicated a relationship between mother's parenting style and identity development. A significant relationship was found between mother's authoritative parenting style and ideological and interpersonal achievement and mother's authoritarian parenting style and ideological foreclosure. Mother's permissive parenting was significantly related to ideological and interpersonal moratorium, as well as diffusion in both domains. Significant relationships were found between father's parenting style and identity development. Father's authoritarian parenting style was positively related to ideological and interpersonal achievement. Father's permissive parenting style was positively related to ideological and interpersonal diffusion.

PAGE 67

61 In all, the results of this study support the idea that identity status is related to parenting style; however, the nature of the relation may differ for mothers and fathers. Analysis indicated that a mother's parenting style that is characterized as open and guiding (authoritative) is related to ideological and interpersonal achievement, whereas the same parenting style in fathers was related to ideological and interpersonal foreclosure. These conflicting findings may reflect the different social roles of mothers and fathers. Identity Development and Accultu ration Orientation At the center of the discussion of identity development differences among ethnic groups is what effect, if any, does the experience of being a member of an ethnic minority have on the psychosocial development of adolescents. During identity development, all ethnic groups are exposed to the varying effects of SES, gender socialization, family status, parenting style, and familism. Acculturation theory proponents, however, argue that only members of a cultural minority must negotiate the often-competing demands of developing a bi-cultural existence. That is, because Caucasian adolescents are members of the cultural majority, they are unlikely to encounter expectations of themselves that differ fi"om the macro-culture. It has been proposed that there are four routes to reconciling this acculturation: (a) accommodation, whereby individuals give up their identification with their minority culture and identify with the majority culture, (b) integration, whereby individuals develop positive identifications with their minority culture and the cultural majority, (c) separation, whereby individuals identify with their minority culture and do not identify

PAGE 68

SI with the majority culture, and (d) marginalization, whereby individuals fail to identify with their minority culture or the majority culture. It was of interest, therefore, to determine the relationship between acculturation fttientation and identity development for African American and Hispanic adolescents. Statistical tests revealed significant relationships between acculturation and identity development with ethnicity controlled. A marginalized acculturation was positively related to both ideological and interpersonal diffusion, and a separation acculturation was positively related to interpersonal foreclosure. These findings support the contention that the forms of acculturation are differentially related to identity development. These findings suggest that the results of earlier studies that found ethnic minorities more likely to be classified in the regressive identity statuses than their Caucasian peers might reflect the mediating effect of acculturation orientation. If nothing else, the findings lend cautious support to the argument that acculturation is related to identity development. Limitations of the Study The conclusions of this study are to be considered in light of the following limitations: 1. Previous studies of ethnicity and identity development were conducted with samples of high school students. The sample in this study was limited to volunteers at the University of Florida and Sante Fe Community College. The findings may not generalize to other students at the institutions, to students at other colleges and universities, or to high school students. In addition, participants were volunteers, which fiirther limits the generalizability of the study.

PAGE 69

63 2. This was a correlational study. Causal claims cannot be made. But even more, ego identity ever evolves. A snapshot inspection of the construct cannot document changes over time. A longitudinal design is better suited to address this limitation. Implications of the Study In this study differences in identity development among ethnic groups were not demonstrated, and, when the influence of SES, gender, and family status were controlled, differences were still not found among ethnic groups. Those researchers who reported no differences in identity development between students who were members of minority groups and Caucasian students tended to attribute their findings to the high degree of cultural integration in the schools of their samples of minority students. Earlier studies have focused on high school aged adolescents. This study investigated the identity development of late adolescent college students but a similar interpretation of the findings could be made. Students who are members of minority groups who attend universities may experience advantages that enable them to develop progressive identities. Also universities may offer minority students a degree of cultural integration not available to other minority adolescents. The findings of this study support the contention that ethnic differences in identity are not present in environments that support exploration and commitment. A related finding was that an authoritarian parenting style indicative of control and harsh discipline was related to identity foreclosure; whereas an authoritative parenting style that promotes independence with limits, was related to identity achievement. The permissive parenting style was related to both identity moratorium

PAGE 70

64 and difRision. In each case the degree to which adolescents were given a secure foundation from which to explore was indicative of progressive identity development. Acculturation orientation was also found to be related to identity development for African American and Hispanic adolescents. Respondents who endorsed a marginalized acculturation were more likely to experience identity diffusion than those who developed other acculturation orientations were. A separation acculturation was also related to interpersonal foreclosure. Again, it can be suggested that the degree to which an adolescent is not limited in avenues of exploration is related to the degree to which progressive identity development occurs. Recommendations for Future Research The findings of this study are to be considered in light of its limitations. First, the conclusions were based on a study of late adolescents enrolled in institutions of higher education. To expand the generalizability of these resuUs, studies must be conducted with samples that are more representative of the late adolescent population. Second, as this was a correlational study of a single assessment of identity, no causal claims can be made. If these conditions were addressed, it would be of interest to consider the following: 1. Further research is needed to determine if exposure to an environment that supports identity exploration is related to progressive identity development in spite of ethnic differences. _2. In this study, a relationship between acculturation orientation and identity development was demonstrated. This finding suggests that for minority group members negotiating a bicultural existence can be an impediment to identity development not

PAGE 71

65 shared by those in the majority. Investigation is needed to assess the degree to which the need to resolve the conflicting expectations of the minority and majority cultures limits or expands the identity options available to adolescents. 3. Although significant relationships were found between parenting style and identity development, no consistent pattern of relationships between parenting style and identity development has been established. Further research is needed to identify relationships of variables that may explain the inconsistent findings.

PAGE 72

APPENDIX A OVERVffiW OF IDENTITY STUDIES Author/Date Respondents Af*1 lite ivcSUilS lllLCl Ul datlUit Controls Measure nauser/iy /z ZZ low dJ^O Social & None Q-sort AA & U male high cunuiiiuiis school *vtoir limit may iiiiiii students CXpiUl allLfll KJl A Ac AAS Abraham/ 273 MA & C C more Parenting None 1983 high school dinused/ style may students ideological account for identity difference Abraham/ 841 H&C MA more Parenting ^ i Parents JiUJVi-nfiij 1986 high school foreclosed / style may education students ideological account for identity difference Streitmatter/ 367 7"" Minority Social & r^^ None 1989 graders C & group more economic combination foreclosed/ condition may of ethnic both limit minorities domains exploration of AAs Markstrom123 AA, AI, Minority Socialization Parents' EOM-EIS Adams & MA&C 10*^ groups may account education Adams/ 12*** graders more for difference 1995 foreclosed/ both domains Rotherham320 AA, No notable High degree Comparable EOM-EIS Borus/1989 MA, NA, F, differences of integration economic Chigh in school backgrounds school Students Note. AA= African American, MA = Mexican American, C = Caucasian, F = Filipino, NA = Native American 66

PAGE 73

APPENDIX B INFORMED CONSENT My name is Sean Forbes. I am a graduate student in the Foundations of Education Department in the College of Education. My faculty supervisor is Patricia Ashton, Ph.D. I am conducting a study to measure the relationship among identity development and personal characteristics. I am requesting that you complete these questionnaires. If you agree, you will complete the questionnaires anonymously. The questionnaires will take you no more than an hour to complete and you do not have to answer any question that you do not wish to answer. If, at any time, you feel that you do not want to continue with the study you can stop. As all data will be anonymous, you will not be able to obtain the results of the questionnaires. You will not be reimbursed for your time. However, your cooperation is extremely appreciated. Please feel free to ask me questions at any time during the study. If you have questions about the study after completing the questionnaires feel free to contact me at 348 Norman Hall, 392-0726, ext 280. Questions or concerns about research participants' rights can be directed to the UFIRB Office, Box 1 12250, University of Florida, Gainesville, FL 3261 12250 Participant's signature Date 67

PAGE 74

REFERENCES Abraham, K. G. (1983). The relation between identity status and locus of control among rural high school students. Journal of Ea rl v Adolescence. 3. 257-264. Abraham, K. G. (1986). Ego-identity differences among Anglo-American and MexicanAmerican adolescents. Journal of A dolescence. 9. 151-166. Adams, G. R., & Jones, R. M. (1983). Female adolescents' identity development: Age comparisons and perceived child-rearing experience. Developmental Psychology, 19, 249-256. Adams, G. R., Shea, J., & Fitch, S. A. (1979). Toward the development of an objective assessment of ego identity status. Journal of Youth and Adoles cence, 8, 223228. Andrews, J. (1973). The relationship of values to identity achievement status. Journal of Youth and Adolescence. 2. 133-138. Archer, S. L. (1981). Tdentitv interview schedu le for working women. Unpublished manuscript. Trenton State College, Trenton, NJ. Archer, S. L. (1989a). Gender differences in identity development: Issues of process, domain, and timing. Journal of A dolescence. 12. 1 17-138. Archer, S. L. (1989b). The status of identity: Reflections on the need for intervention. Journal of Adolescence. 12. 345-359. Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology. 4. (1, Pt. 2), 1-103. Baumrind, D. (1991). Effective parenting during the early adolescent transition. In P. A. Cowan & E. M. Hetherington (Eds ), Advances in family research (Vol. 2, pp. 111163). Hillsdale, NJ: Erlbaum. Berry, J. W. (1980). Acculturation as varieties of adaptation. In A. Padilla OEd ), Acculturation: Theory, models, and some new findings (pp. 1-65). Boulder: Westview. 68

PAGE 75

69 Berry, J. W., Trimble, J. E., & Olmedo, E. L. (1986). Assessment of acculturation. In W J Lonner & J W. Berry (Eds.), Cross-cultural research and m ethodology series: Vol 8. Field methods in cross-cultural research (pp. 291-101). Beverly Hills, CA: Sage. Buri, J. R. (1991). Parental authority questionnaire. Journal of Personality Assessment. 57. 110-119. Chandler, C. K (1979). Traditionalism in a modem setting: A comparison of Angloand Mexican-American value orientations. Human Organization, 38, 153-159. Coleman, J. S. (1988). Social capital in the creation of human capital. The American Journal of Sociology. 94. S95-S120. Cross, H., & Allen, J. (1970). Ego identity status, adjustment, and academic arhiftvftment Journal of Consulting an d Clinical Psychology, 34, 288. Delias, M., & Jemigan, L. P. (1987). Occupational identity status development, gender comp^isons, and internal-external control in first-year Air Force cadets. Journal of Youth and Adolescence. 16. 587-600. Elkind, D. (1978). A sympathetic understand ing of the child (2nd ed.). Boston: Allyn & Bacon. Enright, R. D., Lapsley, D. K., Drivas, A. E., & Fehr, L. A. (1980). Parental — influences on the development of adolescent autonomy and identity. Journal of Youth and Adolescence. 9. 529-545. Erikson, E. H. (1963). Childhood and society (2nd ed.). New York: Norton. Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton. Erikson, E. H. (1987). The human life cycle. In S. Schlein (Ed ), A way of looking at things (pp. 595-610). New York: Norton. Farris, B. E., & Glenn, N. D. (1976). Fatalism and familism among Anglos and Mexican Americans in San Antonio. Sociology and Social Research. 60, 393-402. Forbes, S., & Ashton, P. (1996). Improving the construct v alidity of scores on the Extended Objective Measure of Eeo Identity Scale (EOM-EIS ID. Manuscript submitted for publication. Gibson, M. (1988). Accommodation without assimilation: Sikh imm igrants in an American high school. New York: Cornell University.

PAGE 76

70 Gonzalez, N. A., & Cauce, A. M. (1995). Ethnic identity and multicultural competence: Dilemmas and challenges for minority youth. In W. D. Hawley & A. W Jackson (Eds.), Toward a common destin y Tmprovine race and ethnic relations (pp. 131162). San Francisco: Jossey-Bass. Gottfried, A. W. (1985). Measures of socioeconomic status in child development research: Data and recommendations. Merrill-Palmer Quarterly, 3 1 85-92. Grossman, S. M., Shea, J. A., & Adams, G. R. (1980). Effects of parental divorce during early childhood on ego development and identity formation of college students. Journal of Divorce. 3. 263-272. Grotevant, H. D., & Adams, G. R. (1984). Development of an objective measure to assess ego identity in adolescence: Validation and replication. Journal of Youth and Adolescence. 13. 419-438. Grotevant, H. D., Thorbecke, W. L., & Meyer, M. L. (1982). An extension of Marcia's identity status interview into the interpersonal domain. Journal of Youth and Adolescence. 11. 33-47. Hauser, S. T. (1972). Black and white identity development: Aspects and perspectives. Journal of Youth and Adolescence. 1. 113-130. Hetherington, E. M. (1972). Effects of father absence on personality development in adolescent daughters. Developmen tal P.svcholoov. 7. 313-326. Hodgson, J. W., & Fischer, J. L. ((1979). Sex differences in identity and intimacy development in college youth. Journal of Youth and A dolescence. 8. 37-50. Hoflf-Ginsberg, E., & Tardiflf, T. (1995). Socioeconomic status and parenting. In M. H. Bomstein (Ed ), Children and parenting (Vol. 2) (pp. 161-188). Hillsdale, NJ: Erlbaum. HoUingshead, A. B. (1975). Four factor index of social status. Unpublished manuscript, Yale University, Department of Sociology, New Haven, CT. Holtzman, W. H. (1982). Cross cultural comparisons of personality development in Mexico and the United States. In D. A. Wagner & H. W. Stevenson (Eds.), Cultural perspectives on child development (pp. 1 18-164). New York: Freeman. Jones, R. M., & Streitmatter, J. L. (1987). Validity and reliability of the EOM-EIS for early adolescents. Adolescence. 22. 647-659. Jordan, D. n970Y Parental antecedents of eeo identitv formation. Unpublished master's thesis. State University of New York at Buffalo.

PAGE 77

71 Josselson, R. (1994). The theory of identity development and the question of intervention. In S. Archer (Ed.), Interventions for adolescent i dentity development (pp. 12-25). Thousand Oalcs, CA: Sage. Keefe S. E. (1984). Real and ideal extended familism among Mexican Americans and Anglo Americans: On the meaning of close family ties. Human Organization, 38, 6569. La Voie, J. C. (1976). Ego identity formation in middle adolescence. Journal of — Youth and Adolescence. 5. 371-385. Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent-child interaction. In P. H. Mussen (Ed ), Handhook of Child Psychology, (4 th ed., pp. 131-162). New York: Wiley. Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personalitv and Social Ps vcholoev. 3. 551-558. Marcia, J. E. (1976). Identity six years after: A follow-up study. Journal of Youth and Adolescence. 5. 145-160. Marcia, J E (1993). The status of the statuses: Research review. In J. E. Marcia, A. S Waterman, D. R. Matteson, S. L. Archer, & J. L. Orlofsky (Eds ), Ego identity: A handhook for psychosocial research (pp. 22-41). New York: Springer-Verlag. Marcia, J. E., & Friedman, M. (1970). Ego identity status in college women. Journal of Personalitv. 38. 249-262. Markstrom-Adams, C, & Adams, G. R. (1995). Gender, ethnic group, and grade differences in psychosocial functioning during middle adolescence. Journal of Youth and Adolescence. 24. 397-417. Markstrom-Adams, C, & Spencer, M. B. (1994). A model for identity intervention with minority adolescents. In S. Archer (Ed.), Interventions for adolescent identitv development (pp. 84-102). Thousand Oaks, CA: Sage. Matteson, S. L. (1993). Differences within and between genders: A challenge to the theory. In J. E. Marcia, A. S. Waterman, D. R. Matteson, S. L. Archer, & J. L. Orlofsky (Eds.), E go identitv A handbook for psvcho social research (pp. 69-1 10). New York: Springer-Verlag. Mehan, H., Hubbard, L., & Villanueva, I. (1994). Forming academic identities: Accommodation without assimilation among involuntary minorities. Anthropology and Education Ouarterlv. 25. 91-117.

PAGE 78

72 Mindel, C. H. (1980). Extended familism among Mexican Americans, Anglos, and Blacks. His panic Journal of Behavioral Scien ces, 2, 21-34. Negy C & Woods, D. J. (1992). A note on the relationship between acculturation and socioeconomic status. His panic Journal of Behavioral Sciences, 2, 248251. Nowicki, S., & Strickland, B. R. (1973). A locus of control scale for children. Toumal of Con sultin g and Cl inir.al Psychology. 40. 148-154. Peretti P O & Wilson, T. T. (1995). Unfavorable outcomes of the identity crisis among African American adolescents influenced by enforced acculturation. Social Behavior and Personality. 23. 171-175. Phinney, J. S. (1990). Ethnic identity in adolescence and adulthood: A review of research. Ps ychological Bulletin. 108. 499-5 14. -Quintana, S. M., & Lapsley, D. K. (1987). Adolescent attachment and ego ^ identity: A structural equations approach to the continuity of adaptation. Journal of AHnlRscent Research. 2. 393-409. Quintana, S. M., & Lapsley, D. K. (1990). Rapprochement in late adolescent separation-individuation: A structural equations approach, .loumal of Adolescence, 13. 371-385. Raphael, D. (1977). Identity status in university women: A methodological note. Journal of Youth and Adolescence. 6. 57-62. Reguero, J. T. (1991). Relationship between familism and ego identity development of Puerto Rican and immigrant Puerto Rican adolescents (Doctoral dissertation, Virginia Polytechnic Institute and State University, 1991). Dissertation Abstracts International. 52. 1750-B. Rotheram-Borus, M. J. (1989). Ethnic differences in adolescents' identity status and associated behavior problems. Journal of A dolescence. 10. 525-537. Rotheram-Borus, M. J., & Wyche, K. F. (1994). Ethnic differences in identity development in the United States. In S. Archer (Ed ), Interventions for adolescent identity development (pp. 62-83). Thousand Oaks, CA: Sage. Schenkel, S., & Marcia, J. E. (1972). Attitudes toward premarital intercourse in determining ego identity status in college women. Journal of Personality, 3, 472-482. Spence, J. T., Helmreich, R., & Stapp, J. (1974). The Personal Attributes Questionnaire: A measure of sex role stereotypes and masculinity and femininity. Catalog of Selected Documents in Psvcholoev. 4. 43-44.

PAGE 79

73 Spencer M B.. & Dombusch, S. M. (1990). Challenges in studying minonty youth In S S. Feldman & G. R. Elliot (Eds ), At the threshold: The developing adolescent (pp. 123-146). Cambridge, MA: Harvard University Press. Spencer, M. B., & Markstrom-Adams, C. (1990). Identity processes among racial and ethnic minority children in America. Child Development, 61, 290-310. St. Clair, S., & Day, H. P. (1979). Ego identity status and values among young females. Jnumal of Youth and Ad olescence. 8. 317-326. Sterling, C. M., & Van Horn, K. R. (1989). Identity and death anxiety. Adolescence. 17. 895-899. Streitmatter, J. L. (1987). The effect of gender and family status on ego identity development among early adolescents. .Journal of Early Adolescence, 7, 179-189. Streitmatter, J. L. (1988). Ethnicity as a mediating variable of early adolescent identity development. Tnnmal of A dolescence. 11. 335-346. Streitmatter, J. L., & Pate, G. (1989). Identity status development and cognitive prejudice in early adolescents. Jnumal of Early Adolescence. 9, 142-152. U. S. Bureau of the Census. Department of Commerce. (1994). Statistical abstract ofthe United States. 1994. Washington, DC: U. S. Government Printing Office. Valenzuela, A., & Dombusch, S. M. (1994). Familism and social capital in the academic achievement of Mexican origin and Anglo adolescents. Social Science Ouarterlv. 75. 18-36. Velez-Ibanez, C. G., & Greenberg, J. B. (1992). Formation and transformation of funds of knowledge among U.S.-Mexican households. Anthropology and Education Ouarterlv. 23. 313-335. Waterman, A. S. (1993). Developmental perspectives on identity formation: From adolescence to adulthood. In J. E. Marcia, A. S. Waterman, D. R. Matteson, S. L. Archer, & J. L. Orlofsky (Eds.), E go identitv: A handbook for psvchosocial research (pp. 42-68). New York: SpringerVerlag. Waterman, A. S. (1994). Ethical consideration in interventions for promoting identity development. In S. Archer (Ed.), Interventions for adolescent identitv development (pp. 231-244). Thousand Oaks, CA: Sage. Waterman, C. K., & Waterman, A. S. (1975). Fathers and sons: A study of ego identity across two generations. Joumal of Y outh and Adolescence. 4. 331-338.

PAGE 80

BIOGRAPHICAL SKETCH Sean Forbes completed his secondary education in Shelby County, Tennessee, i 1988. In the spring of 1989 he enrolled at the University of Florida, earning a bachelor arts degree in elementary education in 1992 and a master of arts degree in educational psychology in 1995. After completing his doctoral studies in 1999, he accepted a position as an assistant professor of educational psychology in the Department of Educational Foundations, Leadership, and Technology at Auburn University. 74

PAGE 81

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Patncia Ashton, Chair Professor of Educational Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. les Algina Ifessor of EducatioiiAl Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy^ any Gu Associate I^fessor of Educational Psychology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. John iffanzler AssQs^ate Professor of Educational Psychology

PAGE 82

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Lee MuUally Associate Professor of Teac'hing and Learning This dissertation was submitted to the Graduate Faculty of the College of Education and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. Dean, College of EducatioA

14
In summary, early theory and research of gender differences in identity
development pointed to clear differences (Erikson, 1968; Marcia, 1966, 1976). Males
were perceived as focused on ideological identity development. Women were seen as
focused on interpersonal identity development. Later studies (Archer, 1989a; Markstrom-
Adams & Adams, 1995; Streitmatter, 1988) suggest that differences between males and
females in identity development, when present, are subtle.
Socioeconomic status. Minority adolescents are more likely to come from low-
income families than their Caucasian peers (Spencer & Dombusch, 1990). Although few
studies have considered the impact of low SES on identity development, limited evidence
suggests a relation between the constructs. For example, Markstrom-Adams and Adams
(1995) found that higher paternal and maternal education was related to lower ideological
foreclosure (the correlation with maternal education was -.24 and the correlation with
paternal education was -.22). No relationship was found between parents' education and
interpersonal identity.
Berry (1980) hypothesized that minority group members with low SES are more
likely to adopt an assimilation or separation style of acculturation rather than an
accommodation style. Because of limited resources, these individuals tend to see the
world as difficult to master. They may also avoid identification with the majority culture
or reject their own culture in favor of the majority culture. Negy and Woods (1992), in a
study of 339 Mexican American college students, found that acculturation was
moderately related to SES (r = .44, p = .0001). That is, the greater the identification witn
the majority culture, the higher the SES of the student tended to be. Negy and Woods,
however, did not determine if those who identified with the majority culture also


TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS ii
ABSTRACT v
CHAPTERS
1 INTRODUCTION 1
Statement of the Problem 3
Purpose of the Study 9
Ecological Variables Related to Identity Status 10
Research Questions 27
Significance of Study 28
2 METHOD 30
Research Participants 30
Instruments 30
Procedures 34
Research Questions 34
3 RESULTS 36
Summary Statistics 36
Results of Statistical Analyses 42
4 DISCUSSION 5/
Discussion of Results of Statistical Analyses 57
Limitations of the Study 62
Implications of the Study 63
Recommendations for Future Research 64
APPENDICES
A OVERVIEW OF IDENTITY STUDIES 66
ui


53
there was an interaction between acculturation orientation and ethnicity for identity
statuses. No significant interactions were found.
Table 15. Summary of GLM Analyses for Relationship between Ideological Identity
Status and Father's Parenting Style, with Ethnicity Controlled
Source
ss
df
Mean Sauare
F
E
Achieved
Authoritative
114.57
1
114.57
3.59
0.06
Authoritarian
179.03
1
179.03
5.65
0.02
Permissive
105.95
1
105.95
3.31
0.07
Moratorium
Authoritative
9.76
1
9.76
0.26
0.61
Authoritarian
9.15
1
9.15
0.24
0.63
Permissive
233.77
1
233.77
6.38
0.01
Foreclosure
Authoritative
440.73
1
440.73
14.87
0.0001
Authoritarian
51.21
1
51.21
1.63
0.20
Permissive
8.21
1
8.21
0.26
0.61
Diffusion
Authoritative
37.85
1
37.85
1.09
0.30
Authoritarian
29.49
1
29.49
0.79
0.38
Permissive
381.99
1
381.99
11.31
0.0009


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Patricia Ashton, Chair
Professor of Educational Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Jarhes Algina
Professor of Educatic
il Psychology
I certify that f have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy^
iarry Gui
Associate Prfessor of Educational
Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Joh ttranzler
Associate Professor of Educational
Psychology


45
Table 9. ANCOVA Results for Ideological Identity Statuses
Source
ss
df
Mean Sauare
F
E
Achieved
Race
47.51
2
23.75
0.72
0.49
Gender
9.66
1
9.66
0.29
0.59
SES
33.24
1
33.24
1.01
0.32
Family
125.40
5
25.08
0.76
0.58
Moratorium
Race
102.46
2
51.23
1.47
0.23
Gender
145.94
1
145.94
4.18
0.04
SES
29.63
1
29.63
0.85
0.36
Family
175.87
5
35.17
1.01
0.41
Foreclosure
Race
153.62
2
76.81
2.63
0.07
Gender
82.88
1
82.88
2.84
0.09
SES
0.99
1
0.99
0.03
0.85
Family
99.14
5
19.83
0.68
0.64
Diffusion
Race
258.75
2
129.38
3.63
0.28
Gender
88.73
1
88.73
2.49
0.12
SES
0.39
1
0.39
0.01
0.92
Family
180.64
5
36.18
1.01
0.41


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EI858ARMF_UU3FYQ INGEST_TIME 2014-11-17T21:21:01Z PACKAGE AA00026383_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


2
The achievement and moratorium statuses, by involving exploration, are more
progressive identity statuses than foreclosure and diffusion, which do not involve
exploration (Waterman, 1993). That is, adolescents must play an active role in the
resolution of childhood identifications with adult opportunities and expectations; they
must experiment with the identity alternatives available if a positive resolution of the
crisis is to be achieved.
Taking an active part in resolving their identity crisis often leaves the adolescent
open for problems. For example, adolescents classified in the moratorium status are more
prone to anxiety than those in foreclosure because of the lack of commitment in their
lives (Marcia, 1976; Sterling & Van Horn, 1989). Yet, the overall outcome of being
identity achieved or in moratorium is positive. That is, by taking an active role in their
identity crises, adolescents in the moratorium or achievement status are psychologically
healthier than those in diffusion and foreclosure. For example, those in moratorium and
achievement are less likely than those in foreclosure to be authoritarian or to use
stereotypical thinking (Marcia & Friedman, 1970; Streitmatter & Pate, 1989). Those in
achievement and moratorium are more likely to possess an internal locus of control
(Dellas & Jernigan, 1987), and to be more autonomous than those in foreclosure and
diffusion (Andrews, 1973). In the areas of cognitive performance and cognitive style,
those in the achieved and moratorium statuses often perform at a higher level than those
in foreclosure and diffusion. For example, adolescents in the achievement status have
been found to have better study habits (Waterman & Waterman, 1975) and higher grade
point averages than those in other statuses (Cross & Allen, 1970, Raphael, 1977).
Waterman and Waterman (1975) found that individuals in achievement and moratorium


71
Josselson, R. (1994). The theory of identity development and the question of
intervention. In S. Archer (Ed.), Interventions for adolescent identity development (pp.
12-25). Thousand Oaks, CA: Sage.
Keefe, S. E. (1984). Real and ideal extended familism among Mexican Americans
and Anglo Americans: On the meaning of close family ties. Human Organization. 38, 65-
69.
La Voie, J. C. (1976). Ego identity formation in middle adolescence. Journal of
Youth and Adolescence. 5. 371-385.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family:
Parent-child interaction. In P. H. Mussen (Ed.), Handbook of Child Psychology. (4 th ed.,
pp. 131-162). New York: Wiley.
Marcia, J. E. (1966). Development and validation of ego identity status. Journal of
Personality and Social Psychology. 3. 551-558.
Marcia, J. E. (1976). Identity six years after: A follow-up study. Journal of Youth
and Adolescence. 5. 145-160.
Marcia, J. E. (1993). The status of the statuses: Research review. 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. 22-41). New York: Springer-Verlag.
Marcia, J. E., & Friedman, M. (1970). Ego identity status in college women.
Journal of Personality, 38. 249-262.
Markstrom-Adams, C., & Adams, G. R. (1995). Gender, ethnic group, and grade
differences in psychosocial functioning during middle adolescence. Journal of Youth and
Adolescence. 24, 397-417.
Markstrom-Adams, C., & Spencer, M. B. (1994). A model for identity
intervention with minority adolescents. In S. Archer (Ed.), Interventions for adolescent
identity development (pp. 84-102). Thousand Oaks, CA: Sage.
Matteson, S. L. (1993). Differences within and between genders: A challenge to
the theory. 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. 69-110). New
York: Springer-Verlag.
Mehan, H., Hubbard, L., & Villanueva, I. (1994). Forming academic identities: ^
Accommodation without assimilation among involuntary minorities. Anthropology and
Education Quarterly, 25. 91-117.


BIOGRAPHICAL SKETCH
Sean Forbes completed his secondary education in Shelby County, Tennessee, in
1988. In the spring of 1989 he enrolled at the University of Florida, earning a bachelor of
arts degree in elementary education in 1992 and a master of arts degree in educational
psychology in 1995.
After completing his doctoral studies in 1999, he accepted a position as an
assistant professor of educational psychology in the Department of Educational
Foundations, Leadership, and Technology at Auburn University.
74


44
Question #2
The general linear model was used to test ethnic group differences with the
demographic variables controlled (gender, SES, family status). No significant ethnic
group differences emerged when Bonferroni corrections were made for multiple
comparisons, With a conventional .05 alpha level, ideological diffusion was most closely
related to ethnicity, F(2, 256) = 3.63, p = .028. Of the ethnic groups, Caucasian
adolescents reported the highest means on ideological diffusion.
In the research literature, gender, SES, and family status have been identified as
demographic variables related to identity development. However, when these variables
were entered as covariates in the analysis of the relationship between identity status and
ethnicity, and when corrections were made to the p value for multiple analyses, none of
the covariates was found to be significantly related to the identity status domains (see
Tables 9 and 10). The covariates with p values less than .10 were gender in ideological
moratorium (p = .042), ideological foreclosure (p = 0.09), interpersonal achieved (p =
0.008), and interpersonal foreclosure (p = .026), and SES in ideological moratorium (p =
.042) and interpersonal foreclosure (p = .072).
Question #3
The purpose of the third research question was to determine whether identity
status variables were related to familism with ethnicity controlled. The, covariate,
ethnicity, was not found to be related to identity status. With Bonferroni corrections for
multiple comparisons, no significant differences were found between ideological and
interpersonal identity status and forms of familism, with ethnicity controlled. Results are


60
Parenting Style
Only a few studies have considered the relationship between identity status and
parenting style. This was the fourth question of the study. In previous studies an
authoritative parenting style has been associated with the progressive identity statuses
(Waterman, 1993), suggesting that experiencing a parenting style that encourages
independence with guidance may enable adolescents to endure a crisis with the intention
of working through it. In support of this premise, a restrictive parenting style has been
found to be related to the regressive identity statuses (Quintana & Lapsley, 1987).
As for the relationship between a permissive parenting style and identity
development, no studies could be found that investigated this relationship. Inasmuch as a
permissive parenting style encourages exploration, which may precipitate a crisis, this
style also provides the least amount of parental direction, which may contribute to an
inability to commit to identity options.
The statistical tests indicated a relationship between mothers parenting style and
identity development. A significant relationship was found between mother's
authoritative parenting style and ideological and interpersonal achievement and mother's
authoritarian parenting style and ideological foreclosure. Mother's permissive parenting
was significantly related to ideological and interpersonal moratorium, as well as diffusion
in both domains.
Significant relationships were found between fathers parenting style and identity
development. Fathers authoritarian parenting style was positively related to ideological
and interpersonal achievement. Fathers permissive parenting style was positively related
to ideological and interpersonal diffusion.


31
Acculturation
Acculturation orientation was measured with an adaptation of a questionnaire
developed by Berry, Trimble, and Olmedo (1986) to assess the four types of
acculturation-assimilation, integration, separation, and marginality. The questionnaire
consists of 24 items, with an item representing each acculturation type for each of the
following domains: occupation, dating, friendship, life-style, media, and language.
Responses were made on a 5-point Likert scale, ranging from strongly disagree to
strongly agree.
Parenting Style
Parenting style was assessed using the two forms of Buri's (1991) Parental
Authority Questionnaire (PAQ). The PAQ measures parental authority of mothers and
fathers. Each scale is a 30-item questionnaire designed to measure the degree to which an
individual believes his or her parent(s) exhibited permissive, authoritarian, and
authoritative parenting styles. Responses are made to the items on a 5-point Likert scale.
With a sample of 61 psychology students who completed the questionnaire twice over a
2-week period, Buri obtained the following test-retest reliability coefficients: mother
permissiveness = .75; authoritarianism = .85; authoritativeness = 82; father
permissiveness = .74; authoritarianism = .87; authoritativeness = .85.
Buri (1991) obtained evidence for the construct validity of the scale in a study of
127 undergraduate psychology students. His results were consistent with predictions from
Baumrind's (1971) theory: mother's authoritarianism items were negatively correlated
with mother's authoritativeness and permissiveness items (r = -.38, g < .0005; r = -.48, g
< .0005). Mother's authoritativeness and mother's permissiveness were not significantly


47
listed in Tables 11 (ideological identity statuses) and 12 (interpersonal identity status).
No significant interactions between familism and ethnicity were found.
Table 11. Summary of GLM Analyses for Relationship between Ideological Identity
Status and Familism. with Ethnicity Controlled
Source
ss
df
Mean Square
F
E
Achieved
Family structure
2.87
1
2.87
0.09
0.77
Family behavior
42.73
1
42.73
1.29
0.26
Family attitude
66.50
1
66.50
2.07
0.15
Moratorium
Family structure
41.90
1
41.90
1.14
0.29
Family behavior
105.16
1
105.16
1.03
0.32
Family attitude
49.46
1
49.46
1.39
0.24
Foreclosure
Family structure
103.91
1
103.91
3.51
0.06
Family behavior
45.70
1
45.70
1.53
0.22
Family attitude
169.17
1
169.17
5.75
0.02
Diffusion
Family structure
59.95
1
59.95
1.74
0.19
Family behavior
83.85
1
83.85
2.41
0.12
Family attitude
27.06
1
27.06
0.79
0.38


73
Spencer, M. B., & Dombusch, S. M. (1990). Challenges in studying minority
youth. In S. S. Feldman & G. R. Elliot (Eds.), At the threshold: The developing
adolescent (pp. 123-146). Cambridge, MA: Harvard University Press.
Spencer, M. B., & Markstrom-Adams, C. (1990). Identity processes among racial
and ethnic minority children in America. Child Development. 61. 290-310.
St. Clair, S., & Day, H. P. (1979). Ego identity status and values among young
females. Journal of Youth and Adolescence. 8. 317-326.
Sterling, C. M., & Van Horn, K. R. (1989). Identity and death anxiety.
Adolescence. 17, 895-899.
Streitmatter, J. L. (1987). The effect of gender and family status on ego identity
development among early adolescents. Journal of Early Adolescence. 7. 179-189.
Streitmatter, J. L. (1988). Ethnicity as a mediating variable of early adolescent
identity development. Journal of Adolescence. 11. 335-346.
Streitmatter, J. L., & Pate, G. (1989). Identity status development and cognitive
prejudice in early adolescents. Journal of Early Adolescence. 9. 142-152.
U. S. Bureau of the Census. Department of Commerce. (1994). Statistical abstract
of the United States. 1994, Washington, DC: U. S. Government Printing Office.
Valenzuela, A., & Dombusch, S. M. (1994). Familism and social capital in the
academic achievement of Mexican origin and Anglo adolescents. Social Science
Quarterly. 75. 18-36.
Velez-Ibanez, C. G., & Greenberg, J. B. (1992). Formation and transformation of
funds of knowledge among U.S.-Mexican households. Anthropology and Education
Quarterly. 23.313-335.
Waterman, A. S. (1993). Developmental perspectives on identity formation: From
adolescence to adulthood. 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. 42-68).
New York: Springer-Verlag.
Waterman, A. S. (1994). Ethical consideration in interventions for promoting
identity development. In S. Archer (Ed.), Interventions for adolescent identity
development (pp. 231-244). Thousand Oaks, CA: Sage.
Waterman, C. K., & Waterman, A. S. (1975). Fathers and sons: A study of ego
identity across two generations. Journal of Youth and Adolescence. 4, 331-338.


56
Table 18. Summary of GLM Analyses for Relationship between Interpersonal Identity
Status and Acculturation Orientation with Ethnicity Controlled
Source
ss
df
Mean Square
F
E
Achieved
Accommodation
0.04
1
0.04
0.00
0.97
Separation
123.06
1
123.06
3.73
0.06
Marginalized
114.39
1
114.39
3.53
0.06
Integration
21.34
1
21.34
0.65
0.42
Moratorium
Accommodation
123.17
1
123.17
4.19
0.04
Separation
73.57
1
73.57
2.35
0.13
Marginalized
86.26
1
86.26
2.86
0.09
Integration
54.35
1
54.35
1.79
0.18
Foreclosure
Accommodation
44.94
1
44.94
1.47
0.23
Separation
689.77
1
689.77
24.92
0.0001
Marginalized
12.92
1
12.92
0.41
0.52
Integration
96.19
1
96.19
3.08
0.08
Diffusion
Accommodation
73.64
1
73.64
2.65
0.11
Separation
6.24
1
6.24
0.22
0.64
Marginalized
472.82
1
472.82
18.32
0.0001
Integration
14.77
1
14.77
0.52
0.47


27
majority cultures allows for the development of an integrated sense of self, the hallmark
of the achievement status.
In summary, a major task for minority adolescents is the resolution of conflicting
expectations of the minority and majority cultures. As minority adolescents confront the
identity crisis, they come to see themselves in one of four ways: estranged from the
minority culture and solely a part of the majority (assimilated), identified with both the
minority and majority culture (bicultured), estranged from the majority culture and solely
a part of the minority (rejected), or estranged from both cultures (marginalized). As the
development of a higher identity is built on an integrated sense of self (Phinney, 1990), it
is plausible that minority adolescents who adopt the bicultural orientation of acculturation
are more likely to reach the achieved identity status.
Research Questions
Some studies have reported a relationship between ethnicity and identity, and
others have not. None of the studies has adequately controlled for the demographic
variables of SES or family structure, which have been found to be related to identity
status. Further, as this review shows, cultural and family process variables (i.e., accul
turation, familism, and parenting style) may be related to identity status. The purpose of
this study was to examine the relationships of these demographic, cultural, and family
variables to identity status. The research questions addressed in this study were:
1. Is ethnicity related to identity status?
2. Is ethnicity related to identity status after gender, SES, and family status are
controlled?
3. Is identity status related to familism?
4. Is identity status related to parenting style?
5. Is identity status related to acculturation for African American and Hispanic
adolescents?


35
Research Question #2
The second group of research hypotheses was designed to determine if ethnic
identification was related to identity status when the demographic variables-SES, gender,
family status-were controlled. The dependent variable was the EOM-EIS scores. Ethnic
identification was the independent variable and SES, gender, and family status responses
were covariates.
Research Question #3
The next set of hypotheses considered the relation of familism to identity status
for each of the three ethnic groups. Separate analyses were conducted on the variables to
determine their relationship with identity status. In both sets of analyses respondents'
EOM-EIS scores were the dependent variable; whereas scores on the familism measures
served as predictor variables and ethnic identity was a covariate.
Research Question #4
The next set of hypotheses considered the relation of parenting style to identity
status for each of the three ethnic groups. Separate analyses were conducted on the
variables to determine their relationship with identity status. In both sets of analyses
respondents' EOM-EIS scores were the dependent variable; whereas scores on the PAQ
served as predictor variables and ethnic identity was a covariate.
Research Question #5
This group of hypotheses considered the relation between the different forms of
acculturation orientation available to African American and Hispanic adolescents and
identity status.


33
[respondents] saw or talked on the telephone with adult relatives who do not live with
them" (p. 24). Familism structure was also measured with a single item on a 6-point scale
used by Valenzuela and Dombusch. It asked the number of relatives that live within an
hour's drive of the respondent. Familism attitudes were measured using a scale developed
by Keefe (1984). The measure is a 9-item, 5-point Likert scale. Keefe (1984) obtained an
internal consistency coefficient of .68 for the items in a study of 381 Mexican American
and 163 Caucasians.
Socioeconomic Status (SESI
The SES of respondents' parents was assessed using Hollingshead's (1975) Four
Factor Index of Social Status (FFISS). The factors of the scale include gender,
occupation, education, and marital status. Gender, however, is not included in the
calculation of social status, though it can be used with males or females. Occupation level
is assessed on a 9-point scale of increasing complexity as determined by the 1970 United
States Census. The educational factor is measured using a 7-point scale that ranges from
less than seventh grade to graduate professional training. Marital status is assessed by
first assigning respondents to one of the following classifications: unmarried individual
(male or female), head of a household (male or female), or family in which both husband
and wife are gainfully employed. Then social status is calculated by multiplying the
scores of the occupation and education factors by factor weights. Scores range from 8 to
66 with higher scores indicating higher social status. When social status is measured for
families with two employed persons, the occupation and education scores of both persons
are considered in the calculation. The Four Factor Index is based on the belief that social
status is a "multidimensional aggregate" (Gottfried, 1985, p. 92). That is, the four


APPENDIX A
OVERVIEW OF IDENTITY STUDIES
Author/Date
Respondents
Results
Interpretation
Controls
Measure
Hauser/1972
22 low SES
AA&C
male high
school
students
AA more
foreclosed
Social &
economic
conditions
may limit
exploration of
AAs
None
Q-sort
Abraham/
1983
273 MA&C
high school
students
C more
diffused/
ideological
identity
Parenting
style may
account for
difference
None
OM-EIS
Abraham/
1986
841 H&C
high school
students
MA more
foreclosed /
ideological
identity
Parenting
style may
account for
difference
Parents
education
EOM-EIS
Streitmatter/
1989
367 7th
graders C &
combination
of ethnic
minorities
Minority
group more
foreclosed/
both
domains
Social &
economic
condition may
limit
exploration of
AAs
None
EOM-EIS
Markstrom-
Adams &
Adams/
1995
123 A A, AI,
MA&C 10th
- 12th graders
Minority
groups
more
foreclosed/
both
domains
Socialization
may account
for difference
Parents
education
EOM-EIS
Rotherham-
Borus / 1989
320 AA,
MA, NA, F,
Chigh
school
Students
No notable
differences
High degree
of integration
in school
Comparable
economic
backgrounds
EOM-EIS
Note. AA= African American, MA = Mexican American, C = Caucasian, F = Filipino,
NA = Native American
66


63
shared by those in the majority. Investigation is needed to assess the degree to which the
need to resolve the conflicting expectations of the minority and majority cultures limits or
expands the identity options available to adolescents.
3. Although significant relationships were found between parenting style and
identity development, no consistent pattern of relationships between parenting style and
identity development has been established. Further research is needed to identify
relationships of variables that may explain the inconsistent findings.


23
neither parent occupation nor familism, when considered separately, was related to
students' grades but when parent education was high, familism was significantly related
to Mexican origin students' grades. Valenzuela and Dombusch Qonjectured that the
patriarchal ideology of the traditional Mexican family may limit female students'
development, thereby suggesting the need to examine gender differences in the effect of
familism and parent education on development.
Other researchers have found significant differences in familism among African
Americans, Hispanics, and Caucasians. Mindel (1980) reported that Hispanics
demonstrated the highest degree of familism followed by African Americans and
Caucasians, in that order. Farris and Glenn (1976) and Chandler (1979) also reported that
Hispanics possess higher degrees of familism than Caucasians. The familism scale used
by Chandler, however, consisted of the following four items: (a) Nothing in life is worth
the sacrifice of moving away from vour parents, (b) When the day comes for a voung
man to take a job, he should stay near his parents, even if it means losing a good job
opportunity, (c) When young people get married, their main loyalty still belongs to their
parents, and (d) When you need help of any kind, you can depend only on members of
the family to help you out (p. 157).
Using an expanded familism measure consisting of nine items with a sample of
381 Mexican Americans and 163 Anglo Americans, Keefe (1984) obtained results that
contradicted those of Chandler (1979) and Farris and Glenn (1976). She found that the
Anglo Americans valued family as highly as the Mexican Americans, although the Anglo
Americans were more likely to have a very small extended family or none at all, whereas


6
to research on parenting style that has suggested that a warm, supportive, and controlling
parenting style is related to the foreclosure status and proposed that the Mexican
American parents may be more likely to use such a style with their children. In light of
her findings, Streitmatter suggested that studies must be done to identify mediators of the
relationships between ethnicity and identity development.
Like Hauser (1972) and Abraham (1986), the Streitmatter study is seriously
flawed. She combined the four ethnic groups into one group in her analysis. Citing
problems such as poor sampling techniques and a lack of control for SES in previous
studies, Markstrom-Adams and Adams (1995) designed a study comparing the identity
development of African American, American Indian, Mexican American, and Caucasian
adolescents to address these design flaws. Their respondent pool roughly approximated
the racial composition of the area from which the respondents were sampled, and SES
was controlled by obtaining the respondents' parents' educational level. Markstrom-
Adams and Adams also investigated the role that age and gender roles play in the
development of identity.
For the Markstrom-Adams and Adams (1995) study, 123 adolescents in grades
10-12 in a high school in the rural Southwest completed the EOM-EIS (Grotevant &
Adams, 1984), the Personal Attributes Questionnaire (Spence, Helmreich, & Stapp,
1974), a gender role measure, the Nowicki-Strickland Locus of Control Scale for children
(Nowicki & Strickland, 1973), and questions about parents' level of education. Like the
previous researchers, Markstrom-Adams and Adams found that the minority adolescents
were more foreclosed than their Caucasian peers. In addition, they found a nonsignificant
trend suggesting that the minority adolescents were also more likely to be classified as


48
Status and Familism. with Ethnicitv Controlled
Source
ss
df
Mean Square
F
2
Achieved
Family structure
0.41
1
0.41
0.01
0.91
Family behavior
15.95
1
15.95
0.48
0.49
Family attitude
47.39
1
47.39
2.48
0.22
Moratorium
Family structure
26.93
1
26.93
0.90
0.34
Family behavior
134.58
1
134.58
4.97
0.03
Family attitude
25.34
1
25.34
0.87
0.35
Foreclosure
Family structure
28.06
1
28.06
0.88
0.35
Family behavior
93.86
1
93.86
2.83
0.09
Family attitude
62.94
1
62.94
1.94
0.17
Diffusion
Family structure
3.08
1
3.08
0.12
0.73
Family behavior
3.94
1
3.94
0.14
0.71
Family attitude
99.08
1
99.08
3.85
0.05


contention that males and females experience identity development differently (Marcia,
1993).
59
In tests of the statistical hypothesis, no significant differences were found among
the respondents' identity status scores, when SES, gender, and family status were
controlled. Also, although there were significant differences among groups in their report
of SES and family status, none of the demographic variables were found to be significant
covariates of identity status.
Familism
The third question of this study was whether African American, Caucasian, and
Hispanic adolescent identity development is related to familism. Familism is defined as
the connectedness one has with immediate and extended family members. Current
operationalization of the concept includes three dimensions: structural, attitudinal, and
behavioral. Structural familism refers to the number of relatives living in close proximity
to a person; attitudinal familism is a measure of a respondent's beliefs as to the benefit of
contact with family members. Behavioral familism refers to the amount of time people
stay in contact with family members. Valenzuela and Dombusch (1994) proposed that a
high degree of familism serves as a sort of social capital that enables adolescents to
develop positive mental health characteristics. Because differences in familism have been
found among African American, Caucasian, and Hispanic adolescents, it was of interest
to measure the relationship between identity status and familism with ethnicity
controlled. Testing the relationship between familism with ethnicity controlled resulted in
no significant relationships.


10
conclusion. Furthermore, the studies have for the most part failed to include important
ecological variables that may account for the conflicting results. To account for the
development of identity of ethnic minorities, the role of all of these variables must be
examined. The purpose of this study, therefore, was to investigate whether the ecological
variables of SES, acculturation, family characteristics, and gender are related to the ego-
identity development of ethnic minority and Caucasian adolescents. Each of these
ecological variables and their potential relationship to identity development will be
discussed in the sections that follow.
Ecological Variables Related to Identity Status
Identity status has been found to be related to a number of ecological variables
(Marcia, 1993). In response to findings that ethnic minority adolescents tend toward less
progressive identity statuses than their Caucasian peers, several influences on adolescent
development have been offered as explanation: demographic variables, home culture
variables, and the presence of an acculturation process in ethnic minority members.
Demographic Variables
Differences in identity development may be explained, in part, by the
demographic variables gender, SES, and family configuration. Erikson (1963) argued that
males and females experience identity development differently. Limited evidence
suggests that identity development in low SES environments is not as progressive as in
higher SES environments (Markstrom-Adams & Adams, 1995). As for family
configuration, relations have been found between the construct and identity development,
although the direction of the relationship is far from clear (Marcia, 1993).


9
scored significantly higher on foreclosure than the Caucasian adolescents in their sample,
suggesting that either parental education is an inadequate measure of SES or that other
factors account for the ethnic group differences in foreclosure.
Some research findings suggest other ecological variables as factors that may
account for the greater foreclosure among minority group members. Spencer and
Markstrom- Adams (1990) pointed out that although both the Caucasian and African
American males in Hauser's (1972) study were from low SES families, the African
American males were more likely to live with single parents than were the Caucasian
males. Hauser noted that the African American students whose fathers were absent were
more foreclosed than the students whose fathers were present. Spencer and Markstrom-
Adams concluded that both family configuration and SES may confound the results of
studies of the identity development of minority adolescents, but they also considered the
alternative hypothesis that, because of acculturation or other cultural factors, foreclosure
may be adaptive for minority youth. Supporting that conclusion, Markstrom-Adams and
Adams (1995) proposed that African American and Hispanic families are likely to define
maturity in terms of support of family values in contrast to Caucasian families who may
define maturity in terms of independence. That is, autonomous choice may be the
defining theme for the development of Caucasian identity, whereas acceptance of family
values may be the defining theme for African American and Hispanic adolescents.
Purpose of the Study
In the comparison of minority and majority adolescent ego-identity development,
conflicting results have been found. Some of the research suggests that minority
adolescents are more likely to be foreclosed, but other studies have failed to support that


42
Table 7, Interpersonal Identity Status Percentages bv Ethnicity
African American
Hispanic
Caucasian
Status
(n=82)
(n= 120)
(D = 98)
Achieved
7
12
10
Moratorium
13
11
12
Foreclosure
11
11
6
Diffusion
11
7
10
Moratorium
49
51
52
undifferentiated
Transition
7
9
7
Note. Column totals may not equal 100% due to rounding.
Results of Statistical Analyses
For each question, eight analyses were conducted. A .05 familywise error rate was
used to judge significance. However, in reporting analyses, results that were significant at
a conventional .05 Type I error rate are noted. The research questions for this study were
1. Is ethnicity related to identity status?
2. Is ethnicity related to identity status when gender, SES, and family status are
controlled?
3. Is identity status related to parenting style when ethnicity is controlled?
4. Is identity status related to familism when ethnicity is controlled?
5. Is identity status related to acculturation for African American and Hispanic
adolescents?
Question #1
The means and standard deviations for the status analyses are reported by ethnic
group in Table 8. To determine if ethnicity was related to identity status a one-way


70
Gonzalez, N. A., & Cauce, A. M. (1995). Ethnic identity and multicultural
competence: Dilemmas and challenges for minority youth. In W. D. Hawley & A. W.
Jackson (Eds.), Toward a common destiny: Improving race and ethnic relations (pp. Ol
i!). San Francisco: Jossey-Bass.
Gottfried, A. W. (1985). Measures of socioeconomic status in child development
research: Data and recommendations. Merrill-Palmer Quarterly. 31. 85-92.
Grossman, S. M., Shea, J. A., & Adams, G. R. (1980). Effects of parental divorce
during early childhood on ego development and identity formation of college students.
Journal of Divorce. 3. 263-272.
Grotevant, H. D., & Adams, G. R. (1984). Development of an objective measure
to assess ego identity in adolescence: Validation and replication. Journal of Youth and
Adolescence. 13, 419-438.
Grotevant, H. D., Thorbecke, W. L., & Meyer, M. L. (1982). An extension of
Marcia's identity status interview into the interpersonal domain. Journal of Youth and
Adolescence. 11. 33-47.
Hauser, S. T. (1972). Black and white identity development: Aspects and
perspectives. Journal of Youth and Adolescence. 1. 113-130.
Hetherington, E. M. (1972). Effects of father absence on personality development
in adolescent daughters. Developmental Psychology, 7. 313-326.
Hodgson, J. W., & Fischer, J. L. ((1979). Sex differences in identity and intimacy
development in college youth. Journal of Youth and Adolescence. 8, 37-50.
Hoff-Ginsberg, E., & Tardiff, T. (1995). Socioeconomic status and parenting. In
M. H. Bomstein (Ed ), Children and parenting (Vol. 2) (pp. 161-188). Hillsdale, NJ:
Erlbaum.
Hollingshead, A. B. (1975). Four factor index of social status. Unpublished
manuscript, Yale University, Department of Sociology, New Haven, CT.
Holtzman, W. H. (1982). Cross cultural comparisons of personality development
in Mexico and the United States. In D. A. Wagner & H. W. Stevenson (Eds ), Cultural
perspectives on child development (pp. 118-164). New York: Freeman.
Jones, R. M., & Streitmatter, J. L. (1987). Validity and reliability of the EOM-EIS
for early adolescents. Adolescence. 22. 647-659.
Jordan, D. (1970). Parental antecedents of ego identity formation. Unpublished
master's thesis, State University of New York at Buffalo.


21
and females with democratic fathers were more often classified in the higher identity
statuses than were their peers with autocratic or permissive fathers. As in the first study,
no relationship was found between mothers' parenting style and identity development.
In all, research supports the idea that parenting style is related to identity
development. Precisely how the two are related, however, is far from clear. An
authoritative parenting style has been consistently demonstrated to be related to high
identity status in males. Yet, both authoritative and authoritarian parenting styles have
been found to be related to high identity status in females. The statistical relationship
between identity status and permissive parenting, however, has not been considered.
Fathers' parenting style has been shown to be related to identity development in
adolescents. Mothers' parenting style has also been shown to be related to identity
development in adolescents, yet not as consistently as a father's parenting style.
Therefore, the gender of the adolescent and parent must be taken into account when
assessing the relation between parenting style and identity development.
Familism. Markstrom-Adams and Adams (1995) pointed out that identity
differences among the ethnic groups they studied did not occur in interpersonal identity
and suggested that the emphasis on family and connectedness among ethnic minority
groups fosters the development of interpersonal identity. For example, Mexican
Americans are characterized as living in close proximity to many relatives with whom
they engage in frequent interaction and mutual support, and these behaviors are believed
to be supported by strong emotional ties to family and acceptance of family values
(Keefe, 1984). Caucasians, in contrast, are considered to be less likely to have extended
family living nearby or to strongly value such relationships. Research, however, has


39
Table 3. Parenting Style Mean Scores bv Ethnicity
African American Hispanic Caucasian
Style
M
S
M
S
M
S
Mother
Authoritative
41.17
7.32
42.55
6.65
43.47
7.57
Authoritarian
36.99,
6.94
35.54,
6.34
31.38b
7.18
Permissive
29.82
5.74
30.73
5.18
31.44
5.56
Father
Authoritative
37.06
6.93
38.40
6.65
39.23
9.58
Authoritarian
36.99,
6.94
35.53,
8.93
34.95b
11.29
Permissive
26.01
8.12
27.25
8.12
28.21
8.29
Note. Means with different subscripts are significantly different within the authoritarian
rows.
Familism
The means and standard deviations of the forms of familism are listed by ethnic
group in Table 4. Significant differences were found among structure means, F(2, 282) =
9.79, p < .0001, for familism. Post-hoc analysis indicated that African American and
Hispanic adolescents scores were significantly different from Caucasian scores on
familism-structure (p < .05). No significant differences were found between African
American and Hispanic adolescents.


APPENDIX B
INFORMED CONSENT
My name is Sean Forbes. I am a graduate student in the Foundations of Education
Department in the College of Education. My faculty supervisor is Patricia Ashton, PhD.
I am conducting a study to measure the relationship among identity development and
personal characteristics. I am requesting that you complete these questionnaires. If you
agree, you will complete the questionnaires anonymously.
The questionnaires will take you no more than an hour to complete and you do not have
to answer any question that you do not wish to answer. If, at any time, you feel that you
do not want to continue with the study you can stop.
As all data will be anonymous, you will not be able to obtain the results of the
questionnaires. You will not be reimbursed for your time. However, your cooperation is
extremely appreciated.
Please feel free to ask me questions at any time during the study. If you have questions
about the study after completing the questionnaires feel free to contact me at 348 Norman
Hall, 392-0726, ext 280. Questions or concerns about research participants' rights can be
directed to the UFIRB Office, Box 112250, University of Florida, Gainesville, FL 32611-
2250
Participants signature
Date
67


with the majority culture, and (d) marginalization, whereby individuals fail to identify
with their minority culture or the majority culture.
It was of interest, therefore, to determine the relationship between acculturation
orientation and identity development for African American and Hispanic adolescents.
Statistical tests revealed significant relationships between acculturation and identity
development with ethnicity controlled. A marginalized acculturation was positively
related to both ideological and interpersonal diffusion, and a separation acculturation was
positively related to interpersonal foreclosure. These findings support the contention that
the forms of acculturation are differentially related to identity development.
These findings suggest that the results of earlier studies that found ethnic
minorities more likely to be classified in the regressive identity statuses than their
Caucasian peers might reflect the mediating effect of acculturation orientation. If nothing
else, the findings lend cautious support to the argument that acculturation is related to
identity development.
Limitations of the Study
The conclusions of this study are to be considered in light of the following
limitations:
1. Previous studies of ethnicity and identity development were conducted with samples of
high school students. The sample in this study was limited to volunteers at the University
of Florida and Sante Fe Community College. The findings may not generalize to other
students at the institutions, to students at other colleges and universities, or to high school
students. In addition, participants were volunteers, which further limits the
generalizability of the study.


REFERENCES
Abraham, K. G. (1983). The relation between identity status and locus of control
among rural high school students. Journal of Early Adolescence. 3. 257-264.
Abraham, K. G. (1986). Ego-identity differences among Anglo-American and
Mexican-American adolescents. Journal of Adolescence. 9. 151-166.
Adams, G. R., & Jones, R. M. (1983). Female adolescents' identity development:
Age comparisons and perceived child-rearing experience. Developmental Psychology,
19,249-256.
Adams, G. R., Shea, J., & Fitch, S. A. (1979). Toward the development of an
objective assessment of ego identity status. Journal of Youth and Adolescence. 8. 223-
228.
Andrews, J. (1973). The relationship of values to identity achievement status.
Journal of Youth and Adolescence. 2. 133-138.
Archer, S. L. (1981). Identity interview schedule for working women.
Unpublished manuscript. Trenton State College, Trenton, NJ.
Archer, S. L. (1989a). Gender differences in identity development: Issues of
process, domain, and timing. Journal of Adolescence. 12. 117-138.
Archer, S. L. (1989b). The status of identity: Reflections on the need for
intervention. Journal of Adolescence. 12. 345-359.
Baumrind, D. (1971). Current patterns of parental authority. Developmental
Psychology. 4. (1, Pt. 2), 1-103.
Baumrind, D. (1991). Effective parenting during the early adolescent transition. In
P. A. Cowan & E. M. Hetherington (Eds ), Advances in family research (Vol. 2, pp. 111-
163). Hillsdale, NJ: Erlbaum.
Berry, J. W. (1980). Acculturation as varieties of adaptation. In A. Padilla (Ed.),
Acculturation: Theory, models, and some new findings (pp. 1-65). Boulder: Westview.
68


ACCULTURATION, FAMILY CHARACTERISTICS, GENDER, AND IDENTITY
DEVELOPMENT OF AFRICAN AMERICAN, CAUCASIAN, AND HISPANIC
ADOLESCENTS
By
SEAN ALAN FORBES
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1999


38
Table 2. Means and Standard Deviations of Socioeconomic Status bv Ethnicity
African American
HisDanic
Caucasian
M
S
M S
M
s
42.35a
12.17
43.08, 12.74
49.88b
8.96
Note. Means with different subscripts are significantly different.
difference among the means, F(2, 291) = 12.54, p < .0001. Post-hoc comparisons using
the Tukey method revealed that Caucasian SES scores were significantly different from
both African American and Hispanic SES scores (p < .05). The SES scores of African
American and Hispanic respondents did not differ.
Parenting Style
The means and standard deviations of the different parenting styles are listed by
ethnicity in Table 3. One-way analyses of variance indicated significant differences in
mean scores for mothers' authoritarian parenting style [F(2, 277) = 16.02, p < .0001] and
fathers' authoritarian parenting style [(F(2, 244) = 5.73, p < .0037)].
The Tukey post-hoc comparison of scores indicated that African American and
Hispanic adolescents' mean scores were significantly different from the Caucasian mean
score on mother's and father's authoritarian parenting style (p < .05). The parenting-style
scores of the African American adolescents and Hispanic adolescents did not differ.


In all, the results of this study support the idea that identity status is related to
parenting style; however, the nature of the relation may differ for mothers and fathers.
61
Analysis indicated that a mothers parenting style that is characterized as open and
guiding (authoritative) is related to ideological and interpersonal achievement, whereas
the same parenting style in fathers was related to ideological and interpersonal
foreclosure. These conflicting findings may reflect the different social roles of mothers
and fathers.
Identity Development and Acculturation Orientation
At the center of the discussion of identity development differences among ethnic
groups is what effect, if any, does the experience of being a member of an ethnic minority
have on the psychosocial development of adolescents. During identity development, all
ethnic groups are exposed to the varying effects of SES, gender socialization, family
status, parenting style, and familism. Acculturation theory proponents, however, argue
that only members of a cultural minority must negotiate the often-competing demands of
developing a bi-cultural existence. That is, because Caucasian adolescents are members
of the cultural majority, they are unlikely to encounter expectations of themselves that
differ from the macro-culture.
It has been proposed that there are four routes to reconciling this acculturation: (a)
accommodation, whereby individuals give up their identification with their minority
culture and identify with the majority culture, (b) integration, whereby individuals
develop positive identifications with their minority culture and the cultural majority, (c)
separation, whereby individuals identify with their minority culture and do not identify


2(¡
bicultural adolescents integrate the expectations of their minority culture and the majority
culture and are less vulnerable to negative perceptions of their minority culture because
of their identification with it, they should have less difficulty developing the integrated
sense of self characteristic of the achieved status.
Although Streitmatter (1988) argued that minority adolescents "feel less comfort
in the [majority culture] and as a result are more likely to conform to prescribed values
and expectations ... of the adults in their lives" (p. 344), researchers have not examined
the relationship between Marcia's (1976) identity statuses and the degree to which
minority adolescents feel out of place in the majority culture. The findings of Mehan,
Hubbard, and Villanueva (1994), however, support the idea that adolescents with a
bicultural acculturation should demonstrate more progressive identity development than
those with other acculturation orientations. Mehan et al. investigated the effects of a
program to prepare underachieving African American and Hispanic adolescent students
for college admission by placing them in classrooms with high achieving peers. Most of
the students adopted what Gibson (1988) called accommodation without assimilation
(similar to Berry's, 1980, bicultural orientation) toward academic tasks. That is, "they
affirm[ed] their cultural identities while at the same time recognize[d] the need to
develop certain cultural practices, notably achieving academically, that are acceptable to
the mainstream" (Mehan et al., 1994, p. 105). By doing so these students resolved the two
conditions set forth by Phinney (1990) for the development of an identity: (a) negotiate
the different expectations of the cultures and (b) deal with the negative perceptions the
majority culture has toward the minority culture. Also, as Gonzalez and Cauce (1995)
and Peretti and Wilson (1995) argued, resolving the expectations of both the minority and


52
Fathers' parenting style. Inspection of Table 15 reveals the following significant
relationships between fathers parenting style and ideological identity status with
ethnicity controlled: authoritative style with foreclosure, F (2, 265) = 14.87, p = .0001,
permissive style with diffusion, F (2, 234) = 11.31, p = .0009. Each relationship was
positive. The same general findings were obtained for the relationship between
interpersonal identity status and father's parenting style (see Table 16). That is, analyses
revealed the following significant relationships between father's parenting and
interpersonal identity status: authoritative style with foreclosure, F (2, 169) = 15.07, p =
.0001, permissive style with diffusion, F (2, 232) = 30.02, p = .0001 Each relationship
was positive. As with the ideological identity statuses, no relationship was found between
father's parenting style and interpersonal achievement or moratorium. The interaction of
father's parenting style and ethnicity was not significant.
Question #5
The purpose of this question was to measure the relationship between identity
status and forms of acculturation orientation. The results of the GLM analyses are
presented in Table 17 (ideological identity) and Table 18 (interpersonal identity).
A significant, positive relationship was found between acculturation orientation
and ideological identity status with ethnicity controlled-marginalized acculturation with
diffusion, F (1, 170) = 12.27, p = .0006. The following positive significant relationships
between acculturation orientation and interpersonal Identity were found: marginalized
acculturation with diffusion, F(l, 173)=18.32, p = .0001, and separated acculturation
foreclosure, F(l, 168) = 24.92, p = .0001. Analyses were conducted to determine whether


54
Table 16. Summary of GLM Analyses for Relationship between Interpersonal Identity
Status and Father's Parenting Style Controlling for Ethnicity
Source
ss
df
Mean Square
F
E
Achieved
Authoritative
108.05
1
108.05
3.45
0.06
Authoritarian
196.54
1
196.54
6.37
0.01
Permissive
149.61
1
149.61
4.87
0.03
Moratorium
Authoritative
6.43
1
6.43
0.21
0.65
Authoritarian
0.21
1
0.21
0.01
0.93
Permissive
100.28
1
100.28
3.33
0.07
Foreclosure
Authoritative
482.96
1
482.96
15.07
0.0001
Authoritarian
9.47
1
9.47
0.28
0.60
Permissive
22.00
1
22.00
0.68
0.41
Diffusion
Authoritative
5.25
1
5.25
0.19
0.66
Authoritarian
45.52
1
45.52
1.74
0.19
Permissive
691.72
1
691.72
30.02
0.0001


4
unclear. Given these weaknesses in Hauser's study, the validity of his conclusion that the
African American students were more foreclosed than their Caucasian counterparts must
be questioned. Furthermore, the study was conducted over 30 years ago (between 1962
and 1967). Hauser referred to students' comments during the interview that suggested that
racial restrictions on the African American students' daily lives (e g., restrictions in jobs,
social life, recreation, and housing), limitations in work opportunities, and the lack of role
models who had succeeded in the majority culture may have contributed to their sense of
foreclosure. The greater social and economic opportunities available to African
Americans today may have reduced that sense of foreclosure.
Despite the weaknesses in Hauser's study, a more recent study by Abraham
(1986) supported Hauser's findings regarding the relationship between ethnic-minority
status and identity foreclosure. Abraham investigated the ideological and interpersonal
identity of 841 Mexican American and Caucasian students in grades 9-12 at a high school
in the southwest United States. Respondents completed the Extended Objective Measure
of Ego Identity Status (EOM-EIS) (Grotevant & Adams, 1984), a 64-item measure of
identity in four areas of ideological identity (occupation, religion, politics, and lifestyle)
and interpersonal identity (gender roles, dating, friendship, and recreation). In a
multivariate analysis of covariance, Abraham used mothers' and fathers' education as
covariates to control for differences in SES in the two groups. The results indicated
Caucasians and Mexican Americans differed in ideological identity. A greater percentage
of Mexican American respondents were foreclosed than their Caucasian counterparts.
Abraham concluded, therefore, that the Mexican American adolescents seemed to be
more receptive to their parents' views and ideals on politics, religion, occupation, and


3
were more reflective and less impulsive than those in foreclosure and diffusion, and
Adams, Abraham, and Markstrom (1987) found that individuals in the higher statuses
were less self-conscious than those in foreclosure and diffusion.
Statement of the Problem
As individuals move through adolescence, most tend toward the progressive
statuses of achievement and moratorium rather than the regressive statuses of diffusion
and foreclosure. Several studies, however, have suggested that progressive identity
development is more common for Caucasian adolescents than for their African American
and Hispanic peers.
In a longitudinal study of ideological identity in 22 low socioeconomic status
(SES) African American and Caucasian high school males, Hauser (1972) administered a
Q-sort annually for 3 years and obtained correlations within each annual Q-sort and
between each annual Q-sort to estimate the consistency within aspects of the students'
self-image and change over time in their self-image. From these correlations, Hauser
concluded that, compared to the Caucasian students, the African American students were
more foreclosed than the Caucasian students.
Several problems in the Hauser (1972) study raise questions about the validity of
his findings. First, the study consisted of a convenience sample of 22 matched African
American and Caucasian students. The small size of the sample and the inability to
determine the adequacy of the matching contribute to doubts about the validity of the
findings. Second, the author used a Q-sort of undetermined reliability rather than the
traditional interview or questionnaire measures of identity. Third, the reliability of the
differences between the correlations in the African American and Caucasian groups is


CHAPTER 3
RESULTS
This study was designed to investigate the relation between ethnic identification
and identity status before and after controlling for the effect of theoretically significant
demographic variables-gender, SES, and family status. It was also of interest to examine
the relationships among parenting style, familism, and identity status classification. The
final question of interest was whether acculturation orientation is related to identity status
classification. This chapter presents a summary of the statistical findings of this study
including (a) the summary statistics of demographic variables (age, gender, SES, family
status), parenting style, familism, acculturation orientation, and identity status
classification by ethnic group and (b) results of the analyses of the relationships among
the variables of interest.
Summary Statistics
Demographic Variables
Of the 300 respondents who participated in this study, 82 identified themselves as
African American, ¡20 as Hispanic, and 98 as Caucasian. All participants in this study
were between the ages of 18 and 22. The mean age of African American respondents was
20.8 years. The mean ages of Hispanic and Caucasian respondents were 20.8 and 20.6
years, respectively. Forty percent of the respondents were men (n = 120); 60% of the
respondents were women (n = 180). Of African American respondents, 62 (76%)
36