david dubois race and gener influcenes on adjuntment in early adolescense

8/12/2019 David DuBois Race and Gener Influcenes on Adjuntment in Early Adolescense

1/21

Race and Gender Influences on Adjustment in Early Adolescence: Investigation of anIntegrative ModelAuthor(s): David L. DuBois, Carol Burk-Braxton, Lance P. Swenson, Heather D. Tevendale,Jennifer L. HardestyReviewed work(s):Source: Child Development, Vol. 73, No. 5 (Sep. - Oct., 2002), pp. 1573-1592Published by: Blackwell Publishing on behalf of the Society for Research in Child DevelopmentStable URL: http://www.jstor.org/stable/3696400 .

Accessed: 01/03/2012 13:10

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing and Society for Research in Child Development are collaborating with JSTOR to digitize,preserve and extend access to Child Development.
http://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/action/showPublisher?publisherCode=srcdhttp://www.jstor.org/stable/3696400?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/3696400?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=srcdhttp://www.jstor.org/action/showPublisher?publisherCode=black


2/21

Child Development, eptember/October 002,Volume 3,Number 5,Pages 1573-1592

Race and Gender Influences on Adjustment in Early Adolescence:

Investigation of an Integrative Model

David L. DuBois, Carol Burk-Braxton, Lance P Swenson, Heather D. Tevendale, and Jennifer L. Hardesty

This research nvestigated an integrative model of race- and gender-related nfluences on adjustment duringearly adolescence using a sample of 350Black and White youth. In the proposed model, prejudice/ discrimina-tion events, as well as race and gender daily hassles, contribute o a general stress context. The stress context,in turn, influences evels of emotional and behavioral problems n adjustment, with these associations medi-ated (in part) by intervening effects on self-esteem. Racial and gender identity similarly have positive effectson adjustment via their intermediary inkages with self-esteem. Structural quation modeling analyses pro-vided support for all of these aspects of the model. Findings also revealed theoretically predicted differences nmodel parameters across race by gender subgroups. These include a direct effect of prejudice/ discriminationevents on emotional problems specific to Black youth and an effect of gender identity on self-esteem specific togirls. Black girls appeared o be most vulnerable o exhibiting significant adjustment difficulties as a result ofthe processes under investigation.

INTRODUCTION

In recent years there has been increasing interest inrace- and gender-related influences on youth adjust-ment. There is growing appreciation that a betterunderstanding of concerns in these areas will be nec-essary not only for promoting optimal individual de-velopment, but also for meeting the nation's socialand economic needs (Cauce, Coronado, & Watson,1998; McLoyd, 1998; Sue, Bingham, Porche-Burk6, &

Vasquez, 1999). Contemporary theoryand research

reflects a shift from studying race and gender asstatic, relatively isolated indicators to examining theirmutual implications for youth adjustment withinmore process-oriented, integrative frameworks (Clark,Anderson, Clark, & Williams, 1999; Frable, 1997; N. G.Johnson, Roberts, & Worell, 1999; McLoyd & Stein-berg, 1998; Nettles & Pleck, 1996). Using this ap-proach, contextual and psychological experiences ofgroup members can be considered in relation to nor-mative processes of adaptation within specific peri-ods or stages of development (McLoyd, 1998). Severalfactors point to early adolescence as a period of par-ticular importance for issues pertaining to race (McLoyd& Steinberg, 1998; Phinney, 1990) and gender (De-partment of Health and Human Services, 2000; N. G.Johnson et al., 1999). These include increased expo-sure to others of differing backgrounds at this age(Petersen, 1988), stage-specific concerns such as pu-berty and dating (Compas, 1995; Compas, Hinden, &Gerhardt, 1995), and the emerging salience of groupidentity (e.g., racial/ethnic) in processes of self-understanding (Hill & Lynch, 1983; Quintana, 1998).The present research investigates a process-oriented,

integrative model of race- and gender-related influ-ences on the emotional and behavioral adjustment ofBlack and White young adolescents (see Figure 1).

Stressful Experiences

Exposure to prejudice or discrimination on thebasis of race or gender has been found to be associ-ated with less favorable levels of mental health andadjustment in several studies (for reviews, see Albee

& Perry, 1998; Clark et al., 1999; Contrada et al., 2000;Pugliesi, 1992; C. E. Thompson & Neville, 1999). Nu-merous important issues, however, are in need ofclarification regarding these associations. Prior re-search has focused nearly exclusively, for example, onolder adolescent (i.e., college student) and adult sam-ples. As a result, little is known about the adjustmentimplications of prejudice or discrimination experi-ences for younger age groups. Vulnerability of youngadolescents, in particular, would be consistent withtheir potential to experience significant levels of bothracism and sexism (Leadbeater & Way, 1996; McLoyd

& Steinberg, 1998), despite still-evolving skills for cop-ing with such experiences (Sandler & Twohey, 1998).A second concern is the prevailing tendency to in-

vestigate prejudice and discrimination as sources ofstress only for those groups expected to be most vul-nerable to their effects. This is reflected in a focus onracial or ethnic minorities in the case of stressors thathave a racial component, and a primary concern with

? 2002 by the Society for Research in Child Development, Inc.All rights reserved. 0009-3920/2002/ 7305-0017


3/21

1574 Child Development

RaceDaily / - Racial dentity

EmotionalProblems

Prejudice/ General Stress GlobalDiscrimination + ContextSelf-Esteem

Events Ae___

Gender ++ ++ Behavioral

Problems

Gender Daily Gender dentityHassles -/J ,-Gender Identity

Figure 1 Integrative model of race- and gender-related influences on adjustment during early adolescence.

females when considering negative experiences relat-ing to gender. These emphases are consistent with thestatus of minorities and females as oppressed or stig-matized groups within larger society. Still, it shouldnot be assumed that theoretically less vulnerablegroups are necessarily immune from the negative ef-fects of such stressors (Clark et al., 1999). In a recentinvestigation of adults (Williams, Yu, Jackson, &Anderson, 1997), for example, it was found that expo-sure to race-related stress was an equally strong pre-dictor of psychological distress for Whites as it wasfor Blacks. Especially when extending research tonew

age groups,it thus would seem

imperativeto con-

sider whether nonminorities are susceptible to nega-tive effects of race-related stressors and whethermales are affected by those with gender salience.

A third issue in need of greater understanding isthe nature and significance of interrelations amongdiffering types of stressors involving prejudice ordiscrimination. The distinction between major lifeevents and relatively more minor, but potentially re-current sources of stress (often referred to as "dailyhassles") is well-established in the broader stress lit-erature for children and adolescents (Compas, 1987,1995). Major events frequently result in increased lev-els of day-to-day stress, however, thus accounting, atleast in part, for the effects of such events on youthadjustment (Compas, 1995). This suggests that theadaptive consequences of larger incidents of preju-dice or discrimination may be mediated by subse-quent social and environmental experiences (e.g., re-sponses of others) that stem from these events (C. E.Thompson & Neville, 1999). The extent to which thelatter experiences constitute sources of stress can beexpected to be the product of a dynamic interplaybetween characteristics of the situations involved and

individual appraisal processes (Lazarus & Folkman,1984). Their linkage to relatively more major, acuteevents therefore represents an important mecha-nism through which differences in the meaning andsignificance that youth ascribe to such incidents andtheir aftereffects may be manifested (C. E. Thompson& Neville, 1999). Based on these considerations, theproposed model includes contributions of prejudice/discrimination events to daily hassles involving bothrace and gender (see Figure 1). It can be seen, fur-thermore, that these linkages mediate (in part) therelation of the former types of events to adjustmentoutcomes.

A final issue is the linkage of prejudice and dis-crimination experiences, in turn, to more general oroverall levels of stress experienced by youth. This isan important concern for at least two reasons. First,many types of events and situations involving preju-dice or discrimination may exacerbate levels of stressin other areas (Albee & Perry, 1998; C. E. Thompson &Neville, 1999). Incidents of unfair treatment on thebasis of race or gender in a particular setting (e.g.,school), for example, may create significant addi-tional obstacles to adaptation within that same context.Second, it is well-established that risk for negative ad-justment outcomes increases during developmentwith the accumulation of stressors across differingareas (Compas, 1995; Rutter, 1996). Based on these con-siderations, the confluence of prejudice discriminationexperiences with broader, more encompassing sourcesof stress seems a likely mechanism mediating theirconsequences for adaptation in early adolescence. Inthe proposed model, prejudice / discrimination events,as well as race and gender daily hassles, are thus as-sumed to influence adjustment via contributions tothe general stress context (see Figure 1).


4/21

DuBois et al. 1575

Identity and Self-Esteem

As conceptualized in the seminal theoretical workof Tajfel (1978, 1981), social identity is a central part ofthe self-concept based on identification with specificgroups. Through processes involving social compari-son and exposure to group norms in decision-makingand behavior, individuals have the potential to culti-vate a positive group identity and a sense of the dis-tinctiveness and value of the group(s) of which theyare members (Tajfel, 1981). Application of these con-cepts to race and ethnicity has emphasized the impor-tance of self-identification, exploration, learningabout group customs and behaviors, and a sense ofbelonging to and pride in one's group as the basis fora positive racial or ethnic identity (Phinney, 1990;Spencer & Markstrom-Adams, 1990). Components ofa strong gender identity similarly may include identi-

fication with other women (or men), personal andpublic gender esteem, and critical awareness of theinfluence of gender roles and stereotypes in larger so-ciety (Martin, 1999; Michaelieu, 1997).

A positive racial or ethnic identity has been foundto be associated with more favorable scores on indicesof adjustment in several studies of older adolescentsand adults (Bates, Beauvais, & Trimble, 1997; Brook,Whiteman, Balka, Win, & Gursen, 1998; Grossman,Wirt, & Davids, 1985; Munford, 1994; Pyant & Yanico,1991; Rotheram-Borus, 1990; Zimmerman, Ramirez,Washienko, Walter, & Dyer, 1998). Paralleling research

in the stress area, it is less clear whether racial or eth-nic identity retains a similar protective function atearlier stages of development. It seems reasonable toexpect, however, that benefits would extend at leastto early adolescence, when identity exploration andformation processes begin to intensify (Adams, 1992;Phinney, 1989; Quintana, 1998). A further limitationof available studies, again similar to the stress area, istheir nearly exclusively focus on members of racial orethnic minority groups (e.g., Black). Despite this omis-sion, it has been noted that a positive racial identitymay be of significant benefit not only to members ofminority groups, but also members of the White ma-jority (Helms, 1990; Phinney, 1996).

Investigations of gender identity and its implica-tions for adjustment involve primarily clinical popu-lations (i.e., those diagnosed with gender identity dis-orders; see Money & Lehne, 1999). These studiesnecessarily are of limited usefulness for understand-ing the adaptive significance of less extreme, morenormative variations in attitudes about gender groupmembership. The adjustment correlates of sex roleorientation also have received considerable attention(for a review, see Markstrom-Adams, 1989). Measures

used in this research, however, focused on the extentto which participants endorsed traditional masculineand feminine traits rather than gender identity as in-dicated by valuing of and commitment to one's gen-der group (Michaelieu, 1997). Positive adaptive corre-lates of a strong gender identity as assessed from thelatter perspective have been reported for older fe-male, primarily White adolescents (i.e., college stu-dents; Kennedy, 1993; Michaelieu, 1997). Theoreti-cally, there is a basis for expecting such benefits toextend to earlier periods of adolescence (N. G. Johnsonet al., 1999) as well as to a broader cross-section ofyouth, including members of minority groups (Lead-beater & Way, 1996) and possibly males (Carlock,1999; Good, Borst, & Wallace, 1994).

There has been little investigation of the combinedimplications of racial and gender identity for adjust-ment or how these linkages may be mediated byother self-system processes. Enhancement of overallfeelings of self-regard (i.e., global self-esteem), how-ever, appears to be one of the most consistent benefitsassociated with having a positive identity relating torace (Allen, Howard, & Grimes, 1997; Phinney &Rosenthal, 1992) or gender (Michaelieu, 1997). Theo-retically, as distinct and salient components of per-sonal identity (Frable, 1997), racial and gender identityeach may contribute to a broader, more encompass-ing sense of favorable self-regard. In the proposedmodel, the two types of identity thus are assumed toindependently and positively influence self-esteem.

Because overall feelings ofself-worth function as a

central organizational component of the self-system(Harter, 1999), these linkages with self-esteem are ex-pected to represent the primary mechanism throughwhich racial and gender identity influence youthemotional and behavioral adjustment (see Figure 1).

Influences of Stress on Identity and Self-Esteem

A process-oriented, integrative framework of race-and gender-related influences on adjustment in earlyadolescence also requires consideration of linkagesbetween stressful experiences and processes of iden-tity and self-esteem. Racial and gender group affilia-tions as "ego-extensions" (Rosenberg, 1979) may makeyouth susceptible to feelings of shame and hence dis-identification when they have negative experiencesrelating to group membership. Stress processes lead-ing to stronger sense of identity, however, also havebeen described. In Cross' (1995) theory of Nigres-cence, for example, an identity based on valuingBlack culture and history can be strengthened by un-favorable "encounters" relevant to race (e.g., experi-ences of discrimination), in part because these provide


5/21


opportunities for reconsideration of an identity inwhich being Black has low salience or is devalued.Studies conducted with college-age and adult popu-lations provide some support for both possibilities(Branscombe, 1998; Chambers et al., 1998; V. L. S.Thompson, 1996). Based on these considerations, theproposed model assumes that stress experiences re-lating to race and gender (i.e., race daily hassles andgender daily hassles) may exert either a primarilypositive or negative influence on corresponding as-pects of personal identity in early adolescence. A re-lated feature of the model is that effects of the generalstress context on emotional and behavioral adjust-ment are mediated (in part) by negative effects on glo-bal self-esteem (see Figure 1). Thus, to the extent thatexperiences of prejudice or discrimination contributeto an overall heightened exposure to stress, this is ex-pected to have negative consequences at a corre-

spondingly general level of the self-system (DuBois &Tevendale, 1999).

Present StudyIn the present study, structural equation modeling

(SEM) was used to investigate all features of the pro-posed model within an integrative framework ofanalysis. The sample included approximately equalnumbers of youth in each of four race by gender sub-groups (i.e., Black males and females as well as Whitemales and females). Accordingly, it was well-suited to

investigating both race and gender as moderators ofthe strength of hypothesized relations. This aspect of theresearch extends previous studies, furthermore, byallowing for consideration of race-related influencesamong nonminority, White youth and gender-relatedinfluences among males.

METHOD

School and Community Context

Youth participating in the research were attendinggrades 5 through 8 in the public school system of amedium-sized Midwestern city (population approxi-mately 70,000). Schools in the district included 18 ele-mentary schools (grades K-5), 3 middle schools(grades 6-7), and 3 junior high schools (grades 8-9),all of whom participated in the research. At the timethe research was conducted, racial composition of thestudent population at the targeted grade levels withinthe district was 80% White, 16% Black, and 4% othernon-Black minority backgrounds. The proportion ofminority students varied substantially by school forthose who were in grade 5 and thus still attending el-

ementary school (range = 0- 46% Black students), butwas relatively consistent across both middle schools(14-18% Black students) and junior high schools (15-21% Black students). Only approximately 5% of certi-fied teaching staff in the district were members of mi-nority groups, although nearly one third (31%) ofsupport staff had minority backgrounds; minoritystaff in both categories were predominantly Black. Interms of gender, approximately 90% of teachers werefemale at the elementary school level and between75% and 80% were female in middle and junior highschools.

The city served by the school district was sur-rounded by numerous smaller, predominantly White,rural communities that maintained separate schoolsystems. Neighborhoods within the city ranged fromlow income (e.g., public housing) to upper middle in-come; most had predominantly Black or White popu-lations, despite evidence of racial integration in someareas of the city. Historically, segregation had beenpracticed in the community as late as the 1960s. In re-cent years, race-related incidents and concerns hadcontinued to arise with some regularity. A commu-nity survey on race relations conducted shortly beforethe present research took place revealed significantdifferences in the perceptions of White and Black res-idents (Mayor's Columbia Race Relations Task Force,1996). Most Whites (81%), for example, reported thatWhite people generally treated Black people well inthe community, whereas just under one half of Blacks

(46%)felt this to be true. A

majorityof each

race,moreover, acknowledged the presence of at leastsome racial discrimination in the community (52% ofWhites; 82% of Blacks).

SampleThe sample for the present research consisted of 350

youth who participated in an initial, baseline assess-ment for a planned longitudinal study. Youth wereselected randomly for possible participation from thepublic school system described previously. Selectionwas stratified, however, in an effort to include equalnumbers of youth within relevant categories of eachof the following demographic and background charac-teristics: gender, race (White and Black/ African Amer-ican), developmental level (pre-adolescent-grades 5and 6, and early adolescent-grades 7 and 8), andfamily income level (low- and non-low-income sta-tus, as indicated by parent-reported annual family in-come of $20,000 or less). A further effort was made tobalance selection across differing combinations ofthese characteristics (e.g., preadolescent Black fe-males from low-income families). The participating


6/21

DuBois et al. 1577

school system provided the information necessary forselection on the basis of gender, race, and develop-mental level. Data on family income were not avail-able from the school system. As an alternative strat-egy for this selection factor, U.S. Census Bureau datawere used to

identify prevailinglevels of family in-

come within differing areas of the city (i.e., blockgroups). For purposes of sample selection, these datathen were combined with student home address in-formation to provide an indirect indication of likelyfamily income status for each youth. As noted, actualfamily income status for each study participant ulti-mately was determined directly on the basis of parent-reported data obtained at the initial time of assessment.

The resulting sample of 350 youth included 170boys and 180 girls, 160 White and 190 Black youth,189 preadolescents and 161 early adolescents, and 173and 177 youth from low-income and non-low-income

families, respectively. Limited proportions of bothWhite youth (n = 20; 12.5%) and Black youth (n = 25;13.2%), when provided with the opportunity to self-identify their race/ethnicity flexibly on the survey in-strument, indicated one or more additional designa-tions other than White/ European American or Black/African American, respectively (e.g., biracial, ethnicaffiliations, and so forth). Because their small num-bers precluded separate analysis, these youth werecombined with other White and Black youth for studyanalyses. With respect to race by gender subgroups,the sample included 89 Black males, 81 White males,

101 Black females, and 79 White females. Of furthernote, despite the effort to balance sample selectionacross differing combinations of demographic char-acteristics, there remained a significantly greater rep-resentation of low-income families among Blackyouth (n = 113 of 190; 59.5%) as compared with Whiteyouth (n 60 of 160; 37.5%), X2(1, N = 350) = 16.78,p < .001.

Procedure

Participation required both parent or guardianconsent and youth assent. Following an initial mail-ing that described the study, parents and youth werecontacted by phone and, when necessary, by homevisits to discuss possible participation. Transporta-tion to questionnaire sessions was offered to reduceobstacles to participation among youth from low-income families. To facilitate participation of minorityyouth, research team members from minority back-grounds were involved as much as possible in mak-ing initial contacts with these youth and their parents.Using these procedures, consent for participation wasobtained for approximately 60% of those youth who

were selected and able to be contacted. This overallrate of participation compares favorably with otherresearch that involved youth from similar demographicbackgrounds (e.g., Seidman, Allen, Aber, Mitchell, &Feinman, 1994). The consent rate, furthermore, wassimilar across differing race by gender subgroups inthe sample (range = 58% for White males-65% forBlack females) as well as preadolescent and early ad-olescent age groups (59% and 64%, respectively).

Youth completed their initial assessment for the re-search on a rolling basis as they were recruited intothe study during a 3-month period in the first half ofthe school year (i.e., late September through Decem-ber); as noted previously, it was data from this first,baseline assessment that were used in the present in-vestigation. Questionnaires were administered toyouth during 2-hr group sessions on Saturdays and inindividually scheduled sessions as necessary. In all

sessions, instructions and items were read aloud toyouth to ensure that reading level did not interferewith accurate completion of measures. Youth did notplace their names on questionnaires and were as-sured of the confidentiality of their responses withinethical guidelines.

Measures

Stress. Revised versions of two well-validated in-struments, the Daily Hassles Questionnaire (DHQ;Rowlison & Felner, 1988) and the Life Events Check-

list (LEC; J. H. Johnson, 1986), were used toderive

thefollowing measures of stress for use in primary studyanalyses: race daily hassles, gender daily hassles,prejudice/ discrimination events, and general stresscontext. The DHQ is designed for use with older chil-dren and adolescents and is patterned after the origi-nal Daily Hassles Scale (Kanner, Coyne, Schaefer, &Lazarus, 1981). Items represent typical day-to-dayconcerns of school-age children and adolescents.Children and adolescents are instructed to indicatewhether each event or situation occurred during thepast month and, if it did, to rate the extent to which itwas a hassle using a 4-point rating scale ranging from0 (not at all a hassle) to 3 (a very big hassle). The re-vised version of the measure reflects results of furthervalidation research and is designed to assess stressorsin each of five targeted domains (i.e., school, family, peerrelations, physical appearance, and sports / athletics;see DuBois, Felner, Brand, Phillips, & Lease, 1996;DuBois, Felner, Sherman, & Bull, 1994). For thepresent study, the revised DHQ was expanded to in-clude additional stressors pertaining specifically toboth race and gender. A total of 10 and 11 items rep-resenting race and gender stressors, respectively,


7/21


were added to the measure for this purpose. Theseitems included events and situations relating to eachof the five content domains noted previously: school(three items for race, e.g., "Were called names or in-sulted at school about your race / ethnicity"; and threeitems for gender, e.g., "Treated unfairly at school be-cause you are a girl/boy"), family (one item for race,"Family member was called racist names or insultedabout his /her race / ethnicity"; and two items for gen-der, e.g., "Were not allowed to do something at homebecause you are a girl/boy"), peers (three items forrace, e.g., "Were excluded by other kids because ofyour race/ ethnicity"; and three items for gender, e.g.,"Were called names or teased by other kids aboutsomething having to do with being a girl/boy"), ap-pearance (one item for race, "Experienced teasing orcomments about your appearance having to do withyour race / ethnicity"; and one item for gender, "Expe-rienced teasing about your appearance having to dowith being a girl/boy"), and sports/athletics (twoitems for race, e.g., "Were expected to be good orNOT be good at a sport just because of your race/ethnicity"; and two items for gender, e.g., "Were toldyou wouldn't be any good at a sport because you area girl/boy"). Separate scores were derived for raceand gender daily hassles. Consistent with establishedscoring procedures for the measure (DuBois et al.,1996; DuBois et al., 1994), these were obtained bysumming ratings for each type of stressor on the 0 to3 scale noted previously. Items not endorsed (i.e.,event or situation had not occurred in

past month)were assigned values of 0. The resulting scale scoresdemonstrated satisfactory internal reliability both forthe overall sample, coefficient a of .86 for race hasslesand .87 for gender hassles, and for each race by gen-der subgroup, as ranging from .74 to .91 for race has-sles and .76 to .88 for gender hassles. (Internal consis-tency reliability also was assessed across Family Low-Income Status x Developmental Level subgroups inthe sample and found to be adequate for all studymeasures, acsof .69 or greater.)

The LEC (J. H. Johnson, 1986) assesses the occur-rence of major life events experienced by older chil-dren and adolescents. The revised version of the LECused in the present study assesses the occurrence ofevents in the same content domains as the revisedDHQ (i.e., school, family, peer relations, physical ap-pearance, and sports / athletics; see DuBois et al.,1996; Dubois et al., 1994). The respondent is asked toindicate whether each event occurred in the previous6 months and, if it did, to appraise the event as goodor bad and rate its degree of impact on a 4-point scale.For the present research, respondents also were askedwithin each content domain (e.g., school) whether

they had experienced a major event or change relat-ing to prejudice/ discrimination during the desig-nated time frame. This allowed for computation of aprejudice/ discrimination events score for each par-ticipant (i.e., number of domains for which a negativeevent of this type was reported). Impact ratings werenot utilized because of potential confounding with re-ports of daily stressors that included a similar ap-praisal component and because, as in prior research(Compas, 1987), this type of weighting did little tochange the relation of the prejudice / discriminationevents score with other study measures. Fifty-nineyouth reported at least one prejudice/ discriminationevent, with Black youth doing so more than Whiteyouth (Black males: n = 17, 19.1%; White males: n =6, 7.4%; Black females: n = 27, 26.7%; White females:n = 9, 11.4%). Among those who reported at least oneprejudice/ discrimination event, the following percent-ages of youth within each Race X Gender subgroup re-ported an event in each targeted domain (values ineach domain represent Black males, White males,Black females, and White females, respectively): school(70.6%, n = 12;83.3%, n = 5;59.3%, n = 16;66.7%, n = 6),family (23.5%, n = 4; 16.7%, n = 1; 25.6%, n = 7; .0%,n = 0), peer relations (41.2%, n = 7; 66.7%, n = 4;25.9%, n 7; 44.4%, n = 4), physical appearance(17.6%, n = 3; 50%, n = 3; 14.8%, n = 4; 22.2%, n = 2),and sports/athletics (17.6%, n = 3; 16.7%, n = 1; 14.8%,n = 4; 0%, n = 0).

The final measure, general stress context, was de-

rived on the basis of responses to both of the preced-ing instruments. First, the remaining 104 items on theDHQ not included in the measures of race and genderhassles were summed using the scoring proceduresdescribed previously, a > .90 for full sample and eachRace x Gender subgroup. Next, a total negative lifeevents score was computed from the core set of 73events on the LEC (i.e., excluding prejudice / discrim-ination events). The two resulting indices, which ex-hibited a substantial positive association, r = .47, thenwere standardized (M = 0, SD = 1) and averaged toform a single composite measure.

Racial identity. Racial identity was assessed usingthe widely utilized Multigroup Ethnic Identity Mea-sure (MEIM; Phinney, 1992). Items on the MEIM arerated on a 4-point scale from strongly disagree tostrongly agree, with responses to 14 items summedto yield a total racial or ethnic identity score. Subse-quent validation research has indicated, however,that a slightly abbreviated, 12-item version of thescale is most appropriate for use with younger, pread-olescent and early adolescent youth (Roberts & Phin-ney, 1996).Accordingly, only this latter set of items wasadministered in the present research. As with those


8/21

DuBois et al. 1579

on the full measure, these items are designed to assessmultiple aspects of racial or ethnic identity. These in-clude affirmation and belonging (e.g., "I have a lot ofpride in my ethnic group and its accomplishments");ethnic identity achievement--both exploration (e.g.,"I have spent time trying to find out more about myown ethnic group, such as its history, traditions, andcustoms") and resolution (e.g., "I have a clear sense ofmy ethnic background and what it means to me");and ethnic behaviors and practices (e.g., "I participatein cultural practices of my own ethnic group, such asspecial food, music, and customs"). Items are scored 1to 4 in the direction of a stronger racial or ethnic iden-tity, with total scores thus having a possible range from12 to 48. As in prior research (Roberts & Phinney, 1996),the 12-item scale exhibited satisfactory internal reli-ability both for the overall sample, ac= .89,and for eachRace X Gender subgroup, aosranging from .86 to .89.

Gender identity. Gender identity was assessed usingan adaptation of a measure developed for use witholder adolescents (Michaelieu, 1997). The adaptedmeasure includes items that tap aspects of identitysimilar to those on the MEIM, including favorablefeelings about group membership (e.g., "I'm glad tobe a boy/ girl"), identity exploration and resolution(e.g., "I've given a lot of thought to how being a boy/girl affects my life"), and psychological centrality orsalience (e.g., "Being a boy/ girl has a lot to do withhow I feel about myself"). Items were rated on a 4-point agree / disagree scale similar to that used on theMEIM.

Preliminary factor analysesfor

the 10items

onthe measure indicated support for a one-factor solu-tion, with the initial factor extracted accounting forthe majority of the variance (88%) and all items load-ing positively on this factor; similar results were ob-tained in separate factor analyses for each Race xGender subgroup (e.g., initial factor accounting forbetween 71% and 83% of the variance). A total scoreon the measure thus was used in the present research,ot = .77 for full sample; as ranging from .73 to .80 forRace x Gender subgroups, with higher scores indi-cating reports of a stronger gender identity and hav-ing a possible range from 10 to 40.

Self-esteem. Self-esteem was assessed using theGlobal Self-Esteem Scale of the Self-Esteem Question-naire (SEQ; DuBois et al., 1996), a self-report measurefor older children and young adolescents. The SEQconsists of 42 items, each of which is rated on a 4-point scale ranging from strongly disagree to stronglyagree. In addition to global self-esteem, the measureyields scale scores for self-evaluations pertaining tofive different domains (e.g., peer relations). The Glo-bal Self-Esteem Scale consists of eight items that as-sess overall perceptions of self-worth (e.g., "I am

happy with myself as a person"). Each item is scoredfrom 1 to 4 in the direction of higher self-esteem andthen summed. Prior validation research providessupport for use of the SEQ with demographically di-verse samples of preadolescents and early adoles-cents, including both White and Black youth (DuBois,Bull, Sherman, & Roberts, 1998; DuBois, Felner, Brand,& George, 1999; DuBois et al., 1996; DuBois et al.,1994; DuBois, Tevendale, Burk-Braxton, Swenson, &Hardesty, 2000; Swenson et al., 1998). For the presentsample, scores on the Global Self-Esteem Scale exhib-ited satisfactory internal consistency, ao = .78 for fullsample and aos anging from .72 to .82 for Race x Gen-der subgroups.

Youth adjustment. Youth adjustment was assessedusing the Youth Self-Report (YSR), the self-report ver-sion of the Achenbach Behavior Checklist (Achen-bach, 1991). The YSR is a well-validated measure and

is used extensively in research with children and ad-olescents. For purposes of the present study, thebroad-band Internalizing and Externalizing Scales ofthe measure were utilized as indices of problems inemotional and behavioral adjustment, respectively.

In accordance with recommended procedure(Achenbach, 1991), raw scores on the Externalizingand Internalizing Scales of the YSR were used in pri-mary study analyses. It was of interest, however, toexamine the proportion of youth in the sample forwhom standardized scores on these scales, normedseparately by gender, were at or above levels suggest-

ingclinical

significance (i.e.,T score >

64; Achenbach,1991). Ratings on each scale were found to be in thisrange for between 15% and 20% of the overall sample(internalizing/emotional problems: 16.6%, n = 58;externalizing/behavioral problems: 20.9%, n = 73)and for similar proportions of youth within Race XGender subgroups (Black males: 18.0%, n = 16 and19.1%, n = 17, respectively; White males: 19.8%, n =16 and 18.5%, n = 15, respectively; Black females:16.8%, n = 17 and 22.8%, n = 23, respectively; Whitefemales: 11.4%, n = 9 and 22.8%, n = 18, respectively).The relatively high prevalence of elevated ratings ofadjustment problems in the sample may be, in part, areflection of the large proportion of youth from socio-economically disadvantaged backgrounds.

RESULTS

Preliminary Analyses

Zero-order correlations and descriptive statisticsfor study measures are shown in Table 1. Moderatepositive correlations were evident among the differ-ent measures of stress, for example, r = .51 between


9/21


Table 1 Zero-Order Correlations and Descriptive Statistics for Study Measures

Measure 1 2 3 4 5 6 7 8 9 n M SD

1. Prejudice/ discrimination events 346 .28 .742. Race daily hassles .40*** 350 1.41 3.673. Gender daily hassles .39*** .51*** 350 1.51 3.754. General stress context .43*** .53*** .53*** 350 .00 .875. Racial identity .11* .10 .06 .00 347 34.74 6.546. Gender identity -.05 -.06 -.11* -.13* .27*** 349 34.99 4.457. Global self-esteem -.13* -.18*** -.11* -.37*** .27*** .28*** 350 23.81 4.008. Emotional problems .31*** .30*** .31*** .50*** - .03 -.09 -.44*** 350 13.31 10.139. Behavioral problems .22*** .25*** .21**** .44*** -.05 -.07 -.31*** .72*** 350 12.91 9.64

Note: n ranges from 343 to 350 for correlations because of missing data on certain measures.*p < .05; ***p < .001.

race and gender daily hassles. Higher scores on eachof these measures, in turn, were associated with

greater reported levels of emotional and behavioralproblems. Measures of racial and gender identitysimilarly were associated positively with each other,r = .27, and with ratings of global self-esteem. Nei-ther identity measure exhibited a significant direct re-lation with indices of adjustment. More positive rat-ings of global self-esteem, however, were associatedwith lower reported levels of both emotional and be-havioral problems, rs of -.44 and -.31, respectively,ps < .001. The overall pattern of associations evidentthus allowed for a possible mediated linkage betweenmeasures of identity and adjustment via their mutualassociation with

ratingsof self-esteem. Stress and iden-

tity measures exhibited a pattern of weak and pri-marily nonsignificant relations with one another. Theassociations involved also varied in direction, withprejudice / discrimination events having a positive re-lation with racial identity, r = .11, p < .05, but genderdaily hassles having a negative relation with gen-der identity, r = -.11, p < .05. All stress measures did,however, demonstrate the expected negative associa-tion with ratings of global self-esteem. Finally, therewas a moderate to strong association between ratingsof emotional and behavioral problems, r = .72. Al-though noteworthy, this did not present a major concernwith regard to multicollinearity because the indiceswere to be utilized only as criterion or outcome mea-sures in primary analyses rather than as predictors.

Analysis of variance (ANOVA) was used to exam-ine differences on measures in association with youthdemographic and background characteristics. Foreach study measure, an ANOVA was performed totest for both main effects and all possible two-wayinteractions involving the four primary demo-graphic and background characteristics-that is,gender, race (White versus Black), developmental

level (preadolescent-grades 5 and 6 versus earlyadolescent-grades 7 and 8), and family income

level (low versus non-low-income status). All testsof significance controlled statistically for other maineffects and interactions given that the design was notcompletely balanced (Pedhazur, 1997).

Main effects for race were found for several mea-sures: prejudice/ discrimination events, F(1, 335) = 4.34,p < .05; race daily hassles, F(1, 339) = 5.56, p < .05;racial identity, F(1, 336) = 12.31, p < .001; and globalself-esteem, F(1, 339) = 5.45, p < .05. Black youthscored higher than did White youth on each of thesemeasures, thus reporting greater exposure to prejudicediscrimination events (M = .37, SD = .82 and M = .17,SD = .61 for Black and

White youth, respectively)and

race daily hassles (M = 1.89, SD = 4.37 and M = .85,SD = 2.51, respectively) as well as more favorable lev-els of both racial identity (M = 29.45, SD = 5.65 andM = 27.39, SD = 4.98, respectively) and self-esteem(M = 24.19, SD 4.06 and M = 23.36, SD = 3.89, re-spectively). The main effect found for the measure ofrace daily hassles was accompanied by a significantRace x Family Income Level interaction, F(1, 339) =6.81, p < .01. Simple effects analysis (Keppel, 1982) in-dicated a significant race difference among youth fromnon-low-income families, F(1, 339) = 13.33, p < .001,in the direction of greater reported race daily hasslesby Black youth (M = 2.49, SD = 2.65) in comparisonwith White youth (M = .48, SD = 1.60); the corre-sponding comparison for youth from low-income fam-ilies was not significant, F(1, 339) < 1, ns (M = 1.49,SD = 3.18 and M = 1.46, SD = 3.48, for Black andWhite youth, respectively). Similarly, the main effectfor the measure of racial identity was found in con-junction with a significant Race X DevelopmentalLevel interaction, F(1, 336) = 4.42, p < .05. On thismeasure, a race difference was found among early ado-lescent youth, F(1, 339) = 15.31,p < .001, in the direction


10/21

DuBois et al. 1581

of stronger racial identity for Black youth (M = 36.14,SD = 5.94) in comparison with White youth (M = 31.80,SD = 5.91); the corresponding comparison among pre-adolescent youth was not significant, F(1, 339) = 1.53,p > .10 (M = 35.74, SD = 7.42 and M = 34.62, SD = 5.76,for Black and White youth, respectively).

Significant main effects for gender also were foundfor gender daily hassles, F(1, 339) = 6.48, p < .05, andemotional problems, F(1, 339)= 4.04, p < .05. Girlsscored higher than did boys on each of these measures,thus reporting greater levels of both gender daily has-sles (M = 2.04, SD = 3.54 and M = .96, SD = 2.58 forgirls and boys, respectively) and emotional problems(M = 14.39, SD = 10.94 and M = 12.17, SD = 9.09, re-spectively). In addition, a significant Gender x De-velopmental Level interaction was found for genderidentity, F(1, 338) = 4.00, p < .05. On this measure, agender difference was found among early adolescent

youth, F(1, 338) = 5.38, p < .05, in the direction of lessstrong gender identity for girls (M = 34.19, SD = 4.20)in comparison with boys (M = 35.67, SD = 3.64); thecorresponding comparison among preadolescentyouth was not significant, F(1, 338) < 1, ns (M = 35.29,SD = 4.78 and M = 34.90, SD = 4.84, for girls andboys, respectively). No other main effects or interac-tions were significant in the ANOVA analyses.

Structural Equation Modeling Analysesfor Hypothesized Model

Primary analysesused SEM to

investigatethe

hy-pothesized conceptual model (see Figure 1). Theseanalyses were performed using Version 5.1 of the EQScomputer program (Bentler, 1996). A latent variablesapproach offers potential advantages in SEM analysesin terms of taking into account error in measurement(Bollen, 1989). Typically, multiple items or measuresare used to assess each latent variable. In the presentapplication, however, this approach would have re-sulted in an unacceptably high ratio of estimated pa-rameters to sample size and thus was not feasible.Therefore, an alternative procedure was used in whichestimates of the reliability of each measure were incor-porated as fixed parameters into all SEM analyses (seeBollen, 1989, pp. 168-171). This was accomplished bymodeling each construct as a latent variable with therelevant study measure as its single observed indica-tor; the error variance for each indicator then was setto a fixed value based on estimated reliability of themeasure for the present sample (i.e., coefficient oC).

Along with race and gender as shown in Figure 1,family low-income status and developmental level asdefined previously were included as exogenous vari-ables in SEM analyses. To control for the latter factors

(i.e., family low-income status and developmentallevel) as possible confounding influences, paths rep-resenting their direct effects on study measures wereincorporated into all analyses; the full set of covari-ances among these and the remaining exogenousvariables (i.e., race and gender) was modeled as well.In an analogous manner, correlation between error(i.e., residual) terms was modeled for each of the fol-lowing pairs of measures that were related conceptu-ally and occupied similar positions in the hypothe-sized model: race and gender daily hassles; racial andgender identity; and emotional and behavioral prob-lems. To facilitate interpretation of SEM results, allstudy measures used as input data were rescaled tohave a uniform mean of 0 and an SD equal to 1 at thelevel of the full sample. Unstandardized path coeffi-cients obtained in analyses using these rescaled mea-sures are reported.

Model fit first was evaluated for the full sample.Hypothesized paths were retained if they were eithersignificant (i.e., p < .05) or approached significance(i.e., p < .10). Given the exploratory nature of the re-search and the potential for even nonsignificant pathsto be important in overall fit, this relatively conserva-tive approach to model trimming was viewed as mostappropriate. Lagrange multiplier indices next wereexamined to identify any further, theoretically rele-vant paths that, if added, would enhance model fit(Bentler, 1989), again at a level that either reached orapproached significance. This final step in the model-

fitting processwas

important giventhe

potential forsignificant bias to result from model specification er-rors in SEM analyses (Bollen, 1989), especially whenexamining support for relatively unproven models, aswas the case in the present context (MacCallum, 1995).

Indices of overall fit generally were good for the hy-pothesized model, for example, Bentler-Bonett NormedFit Index = .95; furthermore, all hypothesized pathswere significant and in predicted directions with onlya few exceptions. The exceptions were nonsignifi-cant paths linking race to race daily hassles, gender toprejudice/ discrimination events and gender identity,and race daily hassles to racial identity. A significantoverall X2 tatistic, X2(31,N = 350) = 49.13, p < .05,suggested, however, that fit could be improved. Con-sistent with this possibility, after trimming the non-significant paths, iterative examination of Lagrangemultiplier indices resulted in the addition of three sig-nificant paths. One of the paths was a direct (i.e., un-mediated) effect of prejudice / discrimination eventson emotional problems, unstandardized coefficientb = .15, p < .05. The remaining two paths representeddirect linkages between race and global self-esteem(in the direction of higher reported self-esteem for


11/21


Black youth) and between gender and emotionalproblems (in the direction of greater reported emo-tional problems for girls). For this model, as shown inFigure 2, all hypothesized paths not trimmed in initialmodel fitting continued to reach significance and re-mained in expected directions. Also significant, butnot depicted in Figure 2, were paths representing ef-fects of family low-income status on gender hassles(in the direction of fewer reported gender hassles foryouth from low-income families), b = -.24, p < .05,and developmental level on general stress context (inthe direction of greater levels of reported stress forolder, early adolescent youth), b = .18, p < .05. Noother paths representing effects of family low-incomestatus or developmental level were significant in thefinal model. Overall, in accordance with the im-provements noted, fit indices for the model indi-cated a strong fit to the data, for example, Normed

Fit Index = .96, and now included a nonsignificant

X2,X2(32,N = 350) = 31.85, p > .20. As shown in Fig-

ure 2, the model accounted for nearly half of thevariance in reported levels of emotional problems,R2 = .43, and a smaller, but still noteworthy propor-tion of the variance in reported levels of behavioralproblems, R2 = .28.

The final model included several sets of pathsthrough which measures of race- and gender-relatedinfluences were linked indirectly to indices of youth

adjustment via intermediary effects on other mea-sures. Although this suggested indirect model effectsof race- and gender-related measures on adjustmentindices, it was important to test their statistical signif-icance (Bollen, 1989). Estimates of indirect effects ofrace- and

gender-relatedmeasures on indices of both

emotional and behavioral problems, respectively, andresults of tests of their significance are summarized inTable 2, along with corresponding information for di-rect model effects for purposes of comparison. As canbe seen in the table, significant indirect effects werefound linking each race- and gender-related measureto both criterion measures of youth adjustment, withall effect estimates in expected directions.

Race X Gender Subgroup AnalysesA multigroup analysis (Bentler, 1989) was con-

ducted next, using EQS to examine whether fit for thefinal model differed across the four Race x Gendersubgroups. This analysis first examined whether con-straining path coefficients from the model to be equalacross groups resulted in a significantly poorer over-all fit in comparison with allowing path coefficients tovary freely across groups. All other model parame-ters, including covariances among residual terms andpaths representing effects of developmental leveland low-income status, were free to vary across

Daily Hassles Racial Identity

.33s37*** Behavioral

i35 I

Prejudice/ General -.40*** Global .20**Discrimination Stress Context Self-Esteem

'""

, -.19"*Gender "

.30"**

.43**

.2"Behavioral

.33 -.22"

Problems

R2= .28

GenderIDaily Hassles -.15" - 1 Gender Identity

Figure 2 Final structural equation model for the full sample (N = 350). Path coefficients are unstandardized; to facilitate inter-pretation, all coefficient estimates were derived on the basis of study measures rescaled to have a mean of 0 and SD equal to 1. Sig-nificance levels were determined by critical ratios on the unstandardized coefficients, *p < .05; **p < .01; ***p < .001. Coefficientsof determination (R2) indicate the overall proportion of variance accounted for in each criterion measure of youth adjustment. Notdepicted in the figure are two additional exogenous variables (i.e., developmental level and family low-income status); the finalmodel included estimates of the covariance of these variables with each other and other exogenous variables (i.e., race and gender)as well as paths representing their possible effects on all remaining variables in the model. Also not depicted are correlations be-tween error (i.e., residual) terms for the following pairs of measures: race and gender daily hassles, racial and gender identity, andemotional and behavioral problems.


12/21

DuBois et al. 1583

Table 2 Direct and Indirect Effects of Race and Gender Measures on Indices of Youth Adjustment

Full Sample Black Males Black Females White Males White Females

Direct Indirect Direct Indirect Direct Indirect Direct Indirect Direct IndirectMeasure Effects Effects Effects Effects Effects Effects Effects Effects Effects Effects

Emotional problemsPrejudice / discrimination events .15* .30*** .25*** .21* .25*** .28** - .39** - .31**Race daily hassles .17*** - .19* - .21* - .23* - .18*Gender daily hassles .16*** - - - .09 - .18*" - .15*Racial identity - .08** - -.09 -.12+ - -.15 - -.07Gender identity - -.08** - - - -.13 - - - -.08

Behavioral problemsPrejudice / discrimination events .30*** - .23** - .27** - .35*** .35**Race daily hassles .17*** .20* - .20** - .20* - .20*Gender daily hassles - .16*** - - - .09 .16* .16*Racial identity -.04* - -.06 - -.06 -.06 -.06Gender identity -.04* - - - -.06 - - - -.06

Note: The effects reported are in unstandardized form. To facilitate interpretation, effect estimates were derived in structural equation mod-

eling analysesin which

studymeasures used as input data were rescaled to have a mean of 0 and an SD equal to 1 at the level of the full sam-

ple. This rescaling had no effect on tests for statistical significance, which were conducted using critical ratios on unstandardized effects.*p < .05; **p < .01; ***p < .001; + p < .10.

groups in both analyses (cf. Bentler, 1989).1 Estimatesof measure reliability again were incorporated asfixed parameters, but now with separate estimates(i.e., coefficient ae)used for each Race X Gender sub-group. This was designed to ensure that any differ-ences found in model fit could be attributed to differ-ences in the structural, as opposed to measurement,

portions of the model. Finally,it

shouldbe

notedthat

because the analysis was investigating model fit inthe context of considering Race x Gender subgroups

separately, it was no longer possible to include raceand gender as exogenous variables and estimate theirrelations to study measures as had been done in thefull-sample analysis. The resulting multigroup analy-sis revealed a significant difference (i.e., worsening)in model fit when constraining paths to be equalacross Race X Gender subgroups, X2(118,N = 350) =

171.44, as comparedwith when

theywere allowed to

vary freely, X2(76,N = 350) = 81.04;difference X2(42,N = 350) = 90.40,p < .001.

Based on this result, a theoretically guided ap-proach was used to identify specific paths that dif-fered across two or more subgroups of youth. Thegoal in doing so was to determine the most parsimo-nious approach to allowing model parameters to varyacross subgroups that would produce an acceptableoverall fit. A related aim was to avoid modeling ofrelatively minor, potentially chance variations inpath estimates across groups that might not reflect ac-tual or substantive differences (MacCallum, 1995).Available theory and research, it will be recalled, em-phasizes that race- and gender-related experiencesmay be especially salient in processes that affect theadjustment of ethnic minorities and females, respec-tively. The possible existence of subgroup differencesalong these lines thus was addressed as a first step inthe model refinement process. This involved examin-ing whether the fit of the constrained model was im-proved in any instances by allowing paths involv-ing race-related measures to differ across Black andWhite subgroups of youth or those involving gender-

1As in the full-sample SEM analysis, family low-income statusand developmental level were included primarily for purposesof statistical control in the multigroup analysis. Thus, significantpaths from these variables to study measures are not discussedin the text. Briefly, for the final multigroup analysis model thatwas fit, paths for family low-income status and developmentallevel indicated the following, all ps < .05: less favorable ratingsof gender identity for Black males from low-income families rel-ative to those from non-low-income families, b = -.49; lower re-ported global self-esteem for Black females from low-income fami-lies compared to those from non-low-income families, b = -.40,as well as higher scores on the general stress context measure forearly adolescent compared with preadolescent Black females,b = .34; among White males, fewer reported gender daily hasslesbut greater prejudice/ discrimination events foryouth from low-income families relative to those from non-low-income families,bs = -.24 and .47, respectively, as well as less favorable ratingsof racial identity for early adolescent compared with pre-adolescent youth, b = -.42; and, among White females, lessfavorable ratings of both racial and gender identity and greaterreported race daily hassles for early adolescent youth comparedwith preadolescent youth, bs = -.43, -.48, .27, respectively. Noother paths involving family low-income status or developmen-tal level were significant in the final multigroup analysis model.


13/21


related measures to differ across female and malesubgroups. Illustratively, for the path from racialidentity to self-esteem, improvement in model fit wasevaluated when this path was allowed to vary forBlack subgroups of youth (i.e., Black females andBlack males) in comparison with White subgroups(i.e., White females and White males), while still con-straining the path to be equal within each pairing ofrace subgroups that differed only on gender (e.g., Blackfemales and Black males). For purposes of these analy-ses, differences across both race and gender sub-groups were examined for the path representing aneffect of prejudice / discrimination events on emotionalproblems.

Lagrange multiplier indices identified three pathsthat, when allowed to vary in the preceding manner,yielded significant improvements in model fit bothindividually and in combination: difference X2(3,N =

350) = 19.28, p < .001. These included the paths rep-resenting effects of prejudice / discrimination eventson both race daily hassles and emotional problems.As expected, each of these paths, when allowed todiffer across race subgroups, revealed stronger esti-mated effects for Black youth (i.e., Black males andBlack females) in comparison with White youth (i.e.,White males and White females). The third path rep-resented the effect of gender identity on global self-esteem. Also as expected, when freed, this path wasrevealed to be stronger for female subgroups (i.e.,Black females and White females) in comparisonwith male

subgroups (i.e.,Black males and White

males).As noted, allowing the preceding paths to vary in

the manner indicated improved overall model fit. Theremaining constraints imposed, however, still re-sulted in a reliably less favorable fit in comparisonwith the fully unconstrained model: difference X2(39,N = 350) = 71.12, p < .01. Lagrange modifier indicesthus were examined to identify further paths that, ifallowed to vary across two or more subgroups,would significantly improve fit. Fit was revealed to beimproved significantly by allowing the paths fromgender daily hassles to gender identity and from glo-bal self-esteem to emotional problems to vary freelyacross all subgroups, and by allowing the path fromgender daily hassles to general stress context to varyfor White youth (i.e., White males and White females)in comparison with Black youth (i.e., Black males andBlack females). With these additional modifications,there was no longer evidence of a less favorable fit incomparison with a fully unconstrained model: differ-ence X2(32,N = 350) = 42.18, p > .10.

Two further steps in model fitting also were carriedout. The first explored whether any paths involving

race- and gender-related measures not included inthe final model derived for the full sample mightnonetheless be significant for specific Race X Gendersubgroups. These included paths from the originalhypothesized model that had been dropped due tononsignificance (e.g., effect of race daily hassles on ra-cial identity). Also examined were possible paths rep-resenting direct effects of race- and gender-relatedmeasures on indices of emotional and behavioralproblems (e.g., effect of gender daily hassles on emo-tional problems). These paths had not been hypothe-sized, but were nonetheless theoretically plausible. Inaccordance with considerations noted previously,tests focused on effects for race-related measures thatmight be specific to Black or White subgroups andthose for gender-related measures that might be spe-cific to females or males. These analyses failed to re-veal any additional paths that merited estimation for

either race or gender subgroups. The last step wasthen to trim from the multigroup model those pathsfor each Race x Gender subgroup that did not reachor approach significance. A total of eight paths weretrimmed on this basis, ranging from one path forBlack females to three for Black males.

The resulting multigroup model, shown in Figure3, provided a good overall fit to the data, as indicatedboth by a nonsignificant x2 goodness-of-fit statistic,X2(113,N = 350) = 129.11, p > .10, and relatively highlevels for other major indices of fit, for example,Normed Fit Index = .97. For each subgroup, modelvariables

accountedfor

noteworthy proportions ofthe variance in criterion measures of youth adjust-ment, that is, R2ranging from .36 to .48 for emotionalproblems and from .23 to .35 for behavioral problems.

Differences in paths for each Race x Gender sub-group were apparent consistent with findings re-ported previously. Illustratively, whereas the pathfrom prejudice / discrimination events to emotionalproblems was significant for both Black females andBlack males, b = .25, it was nonsignificant for Whitefemales and White males and thus not estimated (seeFigure 3). Similarly, the path from gender identity toglobal self-esteem was significant for female sub-groups, b = .31,but nonsignificant for male subgroups.The negative path from gender daily hassles to gen-der identity that had been significant in the model fitfor the full sample, furthermore, was revealed to besignificant only for White males.

As can be seen in Table 2, indirect effects of race-and gender-related measures on criterion indices ofyouth adjustment were significant within each sub-group. All indirect effects that reached significance in-volved measures of stress. These were found for bothemotional and behavioral problems in all subgroups.


14/21

ace acial dentityDaily Hassles

Emotional

A28 t Problems28***.25*** R2=.4540 ***d

.34"***k /.58-.32***

Prejudice/ General 41***h GlobalBlack Males Discrimination .22 Stress Context Self-EsteemEvents 20**mI-"20

.49***...42**.1

L.49..42*Behavioral Problems

R2 = .23

Dail HslesGender Identity

Race

RacialIdentityaily Hassles

EmotionalProblems

.28***g .25** R2 = .48

.40 ***d

k34***

*- .43***

Prejudice/1General

GlobalBlack Females Discrimination .22** Stress Context Self-EsteemEvents - .20*

.49***c .842***i .18* Behavioral31Problems

R2= .29

Dai le s Gender IdentityDaily Hassles

Race Racial dentityDaily Hassles

EmotionalProblems

.28** R2= .46

.40***d .34*k.56***

Prejudice/ General 41xh GlobalWhite Males Discrimination .22** Stress Context Self-EsteemEvents-.20

?49***c .33.**f

.42***1 BehavioralProblemsR2 = .27

3Genderai.

-

Daily Hassles -.83**Gender Identity

Race Racial dentityDaily Hassles

EmotionalProblems

.28***gR2= .36

.40***d 34**k1 -.25**

Prejudice nea -IGbWhite Females Discrimination.22*Se

GeneralCtlolvents Stress Context

Se f-EsteemE0 - .20.

.3 " " BehavioralProblemsR2 - .35

GenderDaily Hassles Gender Identity

Figure 3 Final multigroup structural equation model for Black males (n = 89), Black females (n = 101), White males (n = 81), andWhite females (n = 79). Paths that share the same lettered superscript for differing groups were constrained to be equal acrossthese groups. Path coefficients are unstandardized; to facilitate interpretation, all coefficient estimates were derived using studymeasures rescaled to have a mean of 0 and SD equal to 1 at the level of the full sample. Significance levels were determined by crit-ical ratios on the unstandardized coefficients, * p < .05; ** p < .01; ***p < .001. Coefficients of determination (R2) indicate the over-all proportion of variance accounted for in each criterion measure of youth adjustment. Not depicted in the models shown for eachsubgroup are two exogenous variables (i.e., developmental level and family low-income status); their covariance and possibleeffects on all remaining model variables were estimated separately for each group. Also not depicted are correlations betweenerror (i.e., residual) terms for the following pairs of measures: race and gender daily hassles, racial and gender identity, andemotional and behavioral problems.


15/21


DISCUSSION

Findings of this research indicate that socioenviron-mental experiences (i.e., stress) and psychologicalcharacteristics (i.e., identity) pertaining to race andgender are related to emotional and behavioral ad-justment in early adolescence. In analyses conductedwith the full sample, support was obtained for nearlyall aspects of the proposed model. Significant differ-ences in model fit across Race x Gender subgroupswithin the sample, furthermore, could be accountedfor (in part) by theoretically predicted variation inmodel parameters for both Black and White youthand males and females. Overall, findings extend pre-vious research by helping to delineate race- and gender-related influences on youth adjustment within a pro-cess-oriented, integrative framework and for groupsnot traditionally viewed as stigmatized or discrimi-

nated against on the basis of race or gender (i.e.,White and male youth, respectively).

As hypothesized, model pathways were foundthat linked race- and gender-related stress to greaterlevels of emotional and behavioral problems. Thesewere evident both at the level of the overall sampleand for each Race x Gender subgroup. Whereas pre-vious empirical studies that investigated effects ofprejudice and discrimination on mental health havefocused predominantly on adults and older adoles-cents (Albee & Perry, 1998; Clark et al., 1999), the cur-rent findings indicate a relation between such experi-

ences and adaptive difficulties at a substantiallyearlier point in development (i.e., early adolescence).This result is consistent with a limited amount ofother research that addressed the implications of prej-udice and discrimination for younger populations(Fisher, Wallace, & Fenton, 2000; Phinney, Madden, &Santos, 1998).

Within the stress portion of the model, reports ofmajor events involving prejudice or discriminationconsistently were linked to increased daily hassles re-lating to gender and race. This aspect of results is con-sistent with prior research in which negative lifeevents have been indicated to produce heightenedlevels of ongoing (i.e., day-to-day) stress in adoles-cence (Seidman et al., 1994; Wagner, Compas, & Howell,1988). Sensitivity to this type of mechanism likely wasenhanced in the present investigation by the sharedfocus of the measures involved on similar types of ex-periences (i.e., prejudice and discrimination) occur-ring within the same set of life domains (DuBois et al.,1994). The relatively strong linkage between preju-dice/ discrimination events and race daily hassles forBlack participants, furthermore, suggests that charac-teristics of youth (e.g., race or ethnicity) may interact

with events involving prejudice or discrimination toincrease susceptibility to particular types of dailystressors. As in prior studies (e.g., Wagner et al., 1988),daily stressors had a salient role in pathways thatlinked life events to adjustment outcomes. A note-worthy aspect of these results is that prejudice/dis-crimination events were related to poorer adjustmentthrough associations with heightened levels of day-to-day stressors pertaining to both race and gender.Multiple pathways of this nature could constitute amechanism through which certain types of incidents(e.g., events involving both sexism and racism)present greater risk for adverse effects on youth.

The relations of all the preceding measures with in-dices of adjustment were mediated, in turn, by theirassociations with a measure of the general stress con-text. It thus appears that prejudice and discriminationexperiences make a significant contribution to the

overall level of stress reported by young adolescentsand that this is influential in accounting for the nega-tive implications of such experiences for emotional orbehavioral functioning (Compas, 1995; Rutter, 1996).Prejudice / discrimination events and race and genderhassles each were indicated to make a distinct contri-bution to general stress level, despite the overlap amongthem. This pattern suggests another set of processesthrough which multiple forms of prejudice and dis-crimination may combine with one another to createsignificant risk for problems in adjustment. Findingsare nevertheless most consistent with a relatively

indirect linkage between prejudice / discriminationexperiences and the emotional and behavioral func-tioning of young adolescents that is dependent on asso-ciations with other variables. It will be recalled in thisregard that the relation of general stress context withadjustment problems was itself mediated (in part) byan intervening association with lower reported levels ofself-esteem (DuBois & Tevendale, 1999). Overall,then, results indicate that several linkages involvingnormative developmental processes may be impor-tant in accounting for effects of prejudice and dis-crimination on adjustment in early adolescence.

For minority youth in the sample (i.e., Black malesand Black females), exposure to events that involvedperceived prejudice or discrimination also was linkeddirectly to poorer mental health in the form of emo-tional problems. Research on stress and coping pro-cesses within the Black population (Anderson, 1991;Clark et al., 1999) suggests several mechanisms thatcould be important in accounting for this type of as-sociation. These include, for example, internalizedfeelings of anger stemming directly from incidents ofperceived racism. Specific characteristics of the con-text in which the research was conducted also are a


16/21

DuBois et al. 1587

relevant consideration. Most Black youth who re-ported events that involved prejudice or discrimina-tion had experienced at least one of these events in theschool setting. There was only a limited representa-tion of Black students in the population of the partic-ipating schools and their teaching and support stafflikewise was predominantly White. In this context,many Black youth may not have felt adequately sup-ported in efforts to deal with situations that involvedperceived victimization or unfair treatment on thebasis of race. Such concerns for minority students,furthermore, may be exacerbated by a relative scar-city of crossrace friendships at the secondary schoollevel (Hallinan & Williams, 1989).

Model findings for racial and gender identity mea-sures also largely conformed to prediction. Withinpath analyses for the full sample, results were con-sistent with stronger identity in each area contribut-

ing independently to higher self-esteem. Throughthese linkages, positive racial and gender identitywere implicated indirectly in fewer reported prob-lems in both emotional and behavioral adjustment(i.e., identity-eself-esteem-youth adjustment). Thesame type of mediational pathway was evidentwithin each of the models fit separately for Race XGender subgroups. In contrast, support for direct ef-fects of identity on youth adjustment was not found,despite relevant paths being eligible for inclusion inall models. Zero-order correlations between mea-sures of identity and adjustment within the overall

sample similarly were weak and nonsignificant.Find-

ings thus suggest that the protective benefits of posi-tive racial and gender identities for young adoles-cents are tied very closely to whatever advantagesaccrue from them in the form of an overall enhancedsense of self-worth (N. G. Johnson et al., 1999; Lead-beater & Way, 1996; Phinney, 1992). The mediationalpathways found that linked measures of identity andadjustment were relatively less strong than thosefound that linked measures of stress and adjustment.Accordingly, the associated estimates of indirect effectreached statistical significance only in analyses thattook advantage of the statistical power associated withthe full sample. One possible explanation for thistrend is that active identity exploration is typicallyonly in its first stages during early adolescence (Adams,1992), thus limiting the potential protective benefitsof a positive racial or gender identity at this age.

In multigroup SEM analyses, racial identity wasrelated positively to self-esteem among both maleand female Black youth. A corresponding relation be-tween gender identity and self-esteem was evidentfor both White and Black females. Those youth whowould be expected on the basis of broader societal

trends to receive more external negative feedbackconcerning a particular facet of their identity (i.e.,race or gender) thus were indicated to benefit signifi-cantly from being able to nonetheless construct a pos-itive view and understanding of that area of their self-concept. Minority and female youth differed, how-ever, in the extent to which they reported favorablelevels of the relevant components of their identities incomparison with other youth. Of particular note isthat in the early adolescent portion of the sample,Black youth reported significantly more positive lev-els of racial identity than did White youth, whereasfemales reported less favorable gender identity rela-tive to boys. Possible implications of findings in thisportion of the research include a contribution of iden-tity issues to observed race and gender differences inthe self-esteem of young adolescents. Black youth atthis age, for example, tend to report higher self-esteem

than do White youth (Gray-Little & Hafdahl, 2000).Present results suggest that this trend could be linkedto esteem-enhancing benefits of a relatively strongsense of racial or ethnic identity for these youth. Like-wise, girls report lower self-esteem than do boys inearly adolescence (Kling, Hyde, Showers, & Buswell,1999). This would be in accordance with the currentfindings that indicated a less positive gender identityfor girls at this age and its relevance to their feelingsof self-worth (N. G. Johnson et al., 1999).

Results for identity measures did not conform en-tirely to a theoretically predicted pattern of differen-tial

importancefor

minorityand female

youth.In

par-ticular, although the relation of gender identity toself-esteem was restricted to female subgroups, thecomparable association for racial identity was evi-dent not only for Black youth, but also for Whiteyouth. This result seems especially noteworthy withrespect to White males for at least two reasons. First,for these youth, further deviations from patterns ofassociation expected on the basis of prevailing theorywere apparent for gender-related measures. These in-clude a strong negative linkage between gender dailyhassles and gender identity as well as a contributionof gender hassles to overall stress context. Second,among the various subgroups in the sample, Whitemales were unique as the only youth whose back-grounds lacked a traditional basis (i.e., belonging to aminority group or being female) for expecting sensi-tivity to the types of influences under investigation.Although not necessarily anticipated fully, findingsfor this subgroup are nonetheless plausible from atheoretical perspective. By early adolescence, Whitemales can be expected to be aware of criticism of theadvantages they enjoy in contemporary society andto be exposed to situations in which others respond


17/21


negatively to them on this basis (Carlock, 1999). Theirreports of stress therefore often may include signifi-cant experiences of perceived prejudice or discrimi-nation that have the potential to affect them ad-versely. These considerations illustrate, furthermore,how achieving a realistic positive view and under-standing of what it means to be White (Helms, 1990)and male (Good et al., 1994) could be important forself-esteem and other aspects of psychological well-being. Compared with other youth, however, Whitemales appear to have experienced lower absolute lev-els of exposure to race- and gender-related stress (i.e.,fewer prejudice / discrimination events and race dailyhassles compared with Black youth and less exposureto gender daily hassles compared with girls). Alongwith Black males, they also reported a more favorablegender identity than did females. These differencessuggest that processes of influence for the constructs

involved, even if relevant to White males, are not aslikely to create vulnerability to significant adjustmentproblems for this group of youth. Process similaritiesfor this subgroup relative to minority and female youththus should be interpreted with this qualification.

Using the same reasoning, Black females would beexpected to be most susceptible to experiencing prob-lems in adjustment attributable to processes associ-ated with the constructs under investigation. As a re-sult of being minority and female, these youth wereincluded in all of the groups in the sample that re-ported relatively high degrees of race- and gender-related stress

(i.e.,Black

youth who reported greaterprejudice/ discrimination events and race daily has-sles and females who reported greater gender dailyhassles); Black females also were among the femaleswho reported less positive gender identity than didmales in the early adolescent portion of the sample.Black females did not, however, exhibit overall eleva-tions on measures of emotional and behavioral prob-lems relative to the remainder of the sample. Theymay have benefitted in this regard from a wide vari-ety of compensatory resources, including the rela-tively strong racial identity and high level of self-esteem evident for Black youth within the sample. Onthe other hand, there also were signs of added risk forsome Black females. In particular, those who werefrom low-income backgrounds and were older re-ported relatively high levels of stress in other parts oftheir lives (i.e., general stress context) as well as lessfavorable overall self-esteem. The presence of multi-ple sources of risk pertaining both to race and genderand to more general areas of development could havea cumulative effect in creating vulnerability to signif-icant adjustment difficulties (Rutter, 1996). As thesepossibilities illustrate, several differing types of pro-

cesses may interact with those under investigation toinfluence the levels of emotional and behavioral func-tioning that are exhibited by particular members ofgroups such as Black females.

In closing, several limitations and directions forfuture research should be noted. Because of thecross-sectional design, inferences concerning influ-ences among the constructs investigated are inher-ently tentative. In accordance with prevailing theory(Albee & Perry, 1998; Anderson, 1991; Cauce et al.,1998; Clark et al., 1999; Nettles & Pleck, 1996; C. E.Thompson & Neville, 1999), the proposed conceptualmodel assumed that race- and gender-related aspectsof stress and identity can have significant effects onyouth adjustment. Alternative processes, however,could account for the associations observed amongmeasures in these areas. It is possible, for example,that emotional problems experienced by youth (e.g.,anxiety) also have a role in reciprocally shaping thenature and extent of discrimination that they perceivein their environments (Phinney et al., 1998). A furthermethodological limitation is the relatively small sizeof the sample in comparison with what would havebeen ideal for conducting SEM analyses (Bollen,1989). This is an especially relevant issue with respectto the multigroup SEM analysis, given the substan-tially smaller numbers of participants within the sub-groups involved. Relatedly, sample size prohibitedinvestigation of interactions of race and gender witheither developmental level or socioeconomic status,

although these clearly could be important. To addressthe preceding concerns, longitudinal studies based onlarger samples should be a priority in future research.

Measures of stress and identity used in the currentinvestigation were differentiated with respect to raceand gender. A distinction also was made in assess-ments of stress between relatively acute, major eventsinvolving prejudice or discrimination and more minor,but potentially ongoing negative situations and expe-riences of this nature (i.e., daily hassles). Numerousadditional refinements are possible, however, andshould be pursued. These include distinguishingsources of race- or gender-related stress on the basisof such dimensions as domain (e.g., school) and theirintra- versus intergroup nature (Clark et al., 1999). Asimilar potential exists for multidimensional assess-ments of racial and gender identity, such as separateconsideration of their more affectively and cogni-tively oriented components (Sellers, Smith, Shelton,Rowley, & Chavous, 1998). This might prove usefulfor better delineating linkages between stresses andidentity, which were only partially elucidated withthe present measures. Because existing work hastended to lack a developmental focus, however, it will


18/21

DuBois et al. 1589

be important first to determine which types of dis-tinctions are most salient for populations as young asthose included in the present research.

Given the prominent role of coping responses inliterature that addresses effects of prejudice and dis-crimination stress among adults (Clark et al., 1999),these also should receive consideration. In the presentresearch, individual appraisal ratings were built intostress measures. It would be helpful in future work,however, to investigate their role in coping processesdistinct from objective event characteristics. Thismight help to identify cognitive tendencies or biasesof youth, for example, that contribute to either under-or overestimated degrees of perceived exposure todiscrimination (Feldman & Swim, 1998). The specifictypes of coping responses that youth employ to dealwith perceived experiences of discrimination or prej-udice are likely to be significant as well, such as with

respect to moderating effects of these types of stresson adjustment outcomes.

Qualitative sources of data could be of furthervalue in future investigations. These would allow formore in-depth interpretation and understanding offindings obtained with the types of structured, quan-titative measures used in this study and most previ-ous research. In the current investigation, for exam-ple, information was not available on the specificcontent of the major life events involving prejudiceand discrimination that youth reported. Data ad-dressing such concerns could be especially useful for

increasing understandingof the stress and

identityexperiences of groups not traditionally included in re-search of this nature (e.g., White males).

Finally, the design and scope of the present inves-tigation was restricted in several respects that arenoteworthy. This includes the focus on a specific stageof development (i.e., early adolescence) and hence arelatively limited range of ages. In addition, youthfrom only two racial or ethnic backgrounds (i.e.,White and Black) were sampled, all of whom wereliving in a single community and attending the sameschool system. Results might have differed, for example,if the research had not been conducted within a com-munity and school system in which there was a Whitemajority population. To address these concerns, com-parative investigation of youth of varying develop-mental levels; from multiple racial and ethnic minor-ity groups; and across contrasting types of schools,neighborhoods, communities, and regions of the coun-try will be necessary. In doing so, there also should beconsideration of the range of individual-, contextual-,and societal-level influences involved in perpetuatingprejudice and discrimination toward youth fromdiverse backgrounds. This type of multifaceted ap-

proach will lead to more comprehensive understand-ing of race- and gender-related influences on youthdevelopment and thus facilitate effective approachesto intervention.

ACKNOWLEDGMENTS

This research was supported by a grant to the first au-thor from the National Institute of Mental Health(DHHS 5 R29 MH55050). The authors would like tothank the students, staff, and parents of the ColumbiaPublic Schools, Columbia, MO, for their participationin and support of the research. The insightful com-ments of Helen Neville on an earlier draft of this arti-cle also were greatly appreciated.

ADDRESSES AND AFFILIATIONSCorresponding author: David L. DuBois, 210 Mc-Alester Hall, Department of Psychological Sciences,University of Missouri-Columbia, Columbia, MO65211; e-mail: [email protected]. Carol Burk-Braxton, Lance P. Swenson, Heather D. Tevendale,and Jennifer L. Hardesty are also at the University ofMissouri-Columbia.

REFERENCES

Achenbach, T. M. (1991). Manual for the Youth Self-Report nd1991Profile.Burlington: University of Vermont, Depart-ment of Psychiatry.

Adams, G. R. (1992).Introduction and overview. In G. R.Adams, T.P. Gullotta, &R.Montemayor Eds.),Adolescentidentity ormation pp. 1-8). Newbury Park, CA: Sage.

Albee, G.W., & Perry, M. (1998).Economic and social causesof sexism and of the exploitation of women. Journal fCommunity & Applied Social Psychology, 8, 145-160.

Allen, M., Howard, L., & Grimes, D. (1997). Racial grouporientation and self-concept: Examining he relationshipusing meta-analysis. The Howard Journal of Communica-tions, 8, 371-386.

Anderson, L. P. (1991).Acculturative tress: A theory of rel-evance to Black Americans. Clinical Psychology Review,11, 685-702.

Bates, S. C., Beauvais, F., & Trimble, .E. (1997).AmericanIndian adolescent alcohol involvement and ethnic iden-tification. Substance Use & Misuse, 32, 2013-2031.

Bentler, P. M. (1989). EQS structural equations program man-ual. Los Angeles: BMDP Statistical Software, nc.

Bentler, P. M. (1996).EQS:A structural quations program(Version 5.1) [Computer software]. Encino, CA: Multi-variate Software, nc.

Bollen, K. A. (1989). Structural equations with latent variables.New York: Wiley.


19/21

1590 Child D

david dubois race and gener influcenes on adjuntment in early adolescense

Documents