This is to certify that the

thesis entitled

Racial Identity and Psychological Well-being
Among African American Freshmen at a
Predaninantly White University

presented by

Mercedes E. Carswell

has been accepted towards fulfillment
of the requirements for

MA- degree in __Es¥chology

 

 

Date /2//2/0 3
/ /

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

' ‘ ‘fi' w

v -.— ~— ‘

IJBRARY y

MiChigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65—p.15

RACIAL IDENTITY AND PSYCHOLOGICAL WELL-BEIN G AMONG AFRICAN
AMERICAN FRESHEMEN AT A PREDOMINANTLY WHITE UNVIVERSITY

By

Mercedes E. Carswell

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTERS OF ARTS
Department of Psychology

2003

ABSTRACT

RACIAL IDENTITY AND PSYCHOLOGICAL WELL-BEING AMONG AFRICAN
AMERICAN FRESHEMEN AT A PREDOMINANTLY WHITE UNVIVERSITY

By

Mercedes E. Carswell

The present study examined racial identity and psychological distress (i.e. depression and
anxiety) among African American fieshmen at a predominantly White university.
Participants were 98 African American undergraduate freshmen (ages 17 — 19) at a
Midwestern university who completed measures at the beginning of and after six weeks
into their first semester. Measures included the Racial Identity Attitudes Scale, STAI, the
BDI-II, and a self-report estimate of their high school racial composition. Although no
main effects for Intemalization attitudes, Immersion attitudes, or high school racial
composition were found, results indicated two significant interaction effects on
depression. Among students with relatively stronger Immersion attitudes, those with
weaker Intemalization attitudes showed the greatest increase in depressive symptoms and
those with stronger Intemalization attitudes showed the least. Among students with
weaker Immersion attitudes, the risk for increased levels of depression was also greater
among those lower Intemalization attitudes. [Beta = .21, t_= 2.57, p<.05]. Additionally,
among participants who had attended a predominantly Black high school, those with
relatively strong Immersion attitudes experienced a greater increase'in depressive
symptoms than those with relatively weak Immersion attitudes. Students who attended
predominantly White high schools showed relatively little change in levels of depression,

regardless of their Immersion attitude strength [Beta = —.16, _t= -2.05, p_<.05].

ACKNOWLEDGEMENTS

I would like to first thank God for his continuous blessings. I would like to
express my deepest gratitude to my advisor, Dr. Susan Frank for her wisdom, patience,
and guidance. I would like to thank my committee members, Dr. Richard P. DeShon and
Dr. Dozier W. Thornton for their insight and thoughtful suggestions. I would also like to
acknowledge Dr. Isis Settles for her substantial assistance during the earlier phases of this
project.

I would like to give special thanks to my family. Thank you Mom, Dad and John
for your love, support and belief in my abilities. I want to thank my friends Melissa and
Jonathan for embodying true friendship every day. Lastly, I would like to thank my

Michigan family, Shallimar, Mark, Gillian, Carolyn and Janet. Thank you for everything.

iii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................... vi
LIST OF FIGURES .......................................................................... vii
INTRODUCTION
Conceptualizing Racial Identity ......................................................... 2
Nigrescence Theory ................................................................. 2
Multidimensional Model of Racial Identity ...................................... 5
Ego Identity Theory .................................................................. 6
Racial Identity Revisited ............................................................. 7
Identity and Psychological Well-being .................................................. 10
Prior Interracial Experiences and Well-being .......................................... 14
Summary and Hypotheses ............................................................... 16
METHOD
Sample ........................................................................................ 19
Procedure .................................................................................... 19
Measures
Racial Identity ......................................................................... 21
Psychological Distress
Beck Depression Inventory-II ............................................. 22
State-Trait Anxiety Inventory .............................................. 23
Demographic Questionnaire ........................................................ 24
RESULTS
Preliminary Analysis ..................................................................... 25
Tests of Predictions ..................................................................... 32
DISCUSSION .................................................................................. 40
APPENDICIES
Appendix A. Racial Identity Attitudes Scale (Immersion and Intemalization). . .48
Appendix B. Demographic Questionnaire ............................................. 50
REFERENCES ................................................................................ 51

iv

Table
Table
Table
Table
Table

Table

LIST OF TABLES

. Racial Identity Attitudes Scale Items ........................................... 26
. Properties of Study Measures: Means and Standard Deviations .......... 30
. Intercorrelations Among Study Variables ..................................... 31
. Comparison of Means on Immersion and Intemalization .................. 33
. Hierarchical Regression of Anxiety ............................................ 38
. Hierarchical Regression of Depression ........................................ 39

LIST OF FIGURES

Figure 1. Measurement Model of Immersion and Intemalization Scales .................. 28
Figure 2. Interaction effect of Immersion and Intemalization on Depression ............ 36
Figure 3. Interaction effect of Immersion and High School on Depression............... . .37

vi

INTRODUCTION

The freshman experience can be distressing without factoring in race, with a
substantial number of freshmen experiencing depression (Kashani & Priesmeyer, 1983)
and low self-esteem (Marron & Kayson, 1984). Race is likely to complicate the problem.
Research suggests that during high school, minority students experience greater
psychological well-being when attending a school in which they are in the racial majority
rather than a distinct minority (Gray-Little & Carels, 1997, Hudley, 1997). The pattern
often continues into college. D’Augelli and Hershberger (1993) found that the Afiican
American students in a predominantly White university reported lower energy levels, life
satisfaction and overall well-being than White students. Further, James (1998) found that
around 25% of Afiican American students surveyed in his study showed a decline in self-
concept and self-image when attending a predominantly Caucasian university.

Though not specifically mentioned, racial identity is likely to contribute to
African American students’ psychological well-being as well. Adolescents who are not a
part of the White majority generally find identity development more complex with efforts
to integrate ethnicity into their self-concept as an added complication (Steinberg, 1999).
African Americans must decide if and to what degree their race is a defining factor in
their overall experience of self. Many researchers have observed that racial minority
teenagers find their search for an integrated, positive sense of self significantly more
difficult than their White peers (for a recent review see Steinberg, 1999). Moreover,
studies have found a relationship between difficulties consolidating a positive racial
identity and both anxiety (Carter, 1991; Parham & Helms, 1985; Taub & McEwen,

1992), and depression (Chamber et al., 1998; Munford, 1994; Siegel, Yancy &

McCarthy, 2000). In addition, racial identity problems may operate as a risk factor that
increases the likelihood that minority students entering a mostly Caucasian college
environment will succumb to increases in psychological distress.

Congptualizing Racial Identig

Just how racial identity is likely to influence psychological well-being among
African American freshmen at White universities is difficult to articulate without a clear
definition of the racial identity concept. Racial identity researchers have used the term in
different ways, with a major debate surrounding whether to conceptualize and measure
Afiican American racial identity as a categorical construct (as in Cross’ Nigrescence
model) or dimensional construct (as in Seller’ Multidimensional Model of Racial
Identity) construct.

Nigrescence Theory.

The development of a racial identity for African-Americans has been modeled
and theorized about extensively over the last 40 years. Arguably the most well-known
and widely used model of Black identity development is the Cross model of Nigrescence.
According to this model, a Black person matures through five stages of identity: Pre-
encounter, Encounter, Immersion/Emersion, Intemalization, and Intemalization-
Commitment (Cross, 1991), though only the first four are commonly used in research.

During the Pre-encounter stage, the Black individual incorporates a pro-
White/Anti-Black dichotomous view towards life. All materials and representations of
“Blackness” are devalued and a “White” frame of reference is adopted. Cross (1991) also
identifies a subgroup of Pre-encounter Blacks who do not necessarily have anti-Black

views, but instead do not feel that race is a significant element of their self-concept. The

second stage, Encounter, results from a distinct event in which the individual’s frame of
reference is challenged. The “White” norm is no longer adequate and previously held
conceptions associated with the Pre-encounter stage are discarded. The sense of self is
fragmented and the individual begins to build a new identity. Immersion-Emersion
follows as individuals explore their “Blackness” and attempts are made to construct a
new frame of reference. African Americans in this stage may become anxious because of
a lack of identity stability, as they attempt to organize their new identity (Parham, 1989).
The individual may live in complete contrast to the former Pre-encounter stage, now
devaluing “Whiteness.” The attitudes may manifest themselves in racial exclusivity in all
social interactions and persistent dismissal of anything deemed “not Black enough.” Anti-
White feelings are often present. In the final stage, Intemalization, the positive
“Blackness” sought out and sculpted in the Immersion-Emersion stage is internalized for
the first time. Security is the key concept in this stage in that the individual is finally
secure with his or her “Blackness.” Security allows for reconnection with other racial
groups, including Whites. There is less of an idealization of “Blackness,” and in its place,
a full appreciation of what it means to be Black. The final stage, Intemalization-
Commitment, is quite similar to Intemalization. The distinction is in the continuing
concern for and active involvement in Black issues (Cross, 1991). Blacks in the
Intemalization-Connnitment stage maintain their newfound security and invest their time
and efforts towards educating and helping other Blacks.

In an attempt to create a measure that would reflect the stages of Cross’
developrhental theory, Helms and Parham (1985) constructed the Racial Identity

Attitudes Scale (RIAS). Using four subscales to reflect Cross’ first four stages, they

developed items representing attitudes hypothesized to be found in the corresponding
stages. Helms and Parham (1996) report the scale has internal consistency reliabilities of
.76 (Pm-encounter), .51 (Encounter), .69 (Immersion/Emersion), and .80
(Intemalization), based on their work with African American university students. The
RIAS also shows concurrent validity, in that the scale is positively associated with a scale
of etlmic identity (Lemon & Waehler, 1996).

Though widely used, some researchers and theorists have been critical of the
RIAS. The Pre-encounter scale has displayed a significant negative correlation with the
Intemalization scale (range = -.28 to -.48), which implies that the two scales, to some
extent, are measuring opposite poles of the same construct (Lemon & Waehler, 1996;
Munford, 1994; Yancio, Swanson, & Tokar, 1994). In examining the distribution of
RIAS scale scores, Yancio, Swanson, and Tokar (1994) found that Pre-encounter scores
tended to be positively skewed with most peOple choosing not to endorse Pre-encounter
statements. This may be a result of social desirability or an individual taking offense to
some Pre-encounter items (e. g. “I believe that Black people came from a strange, dark,
and uncivilized continent”).

Racial identity researchers also question whether the RIAS can actually measure
Encounter attitudes given the stage’s inherently transitory nature. Cross (1990) theorized
the Encounter stage develops from an encounter event and then quickly leads into the
Immersion/Emersion stage. Such a fleeting period would be difficult to capture.
Additionally, past research has found little statistical support for the Encounter dimension
when using the RIAS (Ponterotto & Wise, 1987; Yancio, Swanson, & Tokar, 1994). The

RIAS has also been criticized because, consistent with the theory on which it is based, the

scores are interpreted categorically. Categorical scoring does not allow for individuals to
have equally strong attitudes associated with two or more stages. However, the stages
often share characteristics. For example, strong Intemalization and Immersion attitudes
both imply that race is central to the individuals’ self-concept and suggest active
involvement in black causes. In addition, the RIAS assumes that Intemalization is the
ideal and does not allow for values associated with other stages to be considered
beneficial (Sellers, Chavous, & Cooke, 1998). For instance, because the Intemalization
stage is considered the ideal or “highest” identity status, activist characteristics associated
with the Immersion/Emersion stage are seen as developmentally lower or inferior.
Multidimensional Model of Racial Identity.

Aware of the limitations of the Cross model of Nigerscence and its accompanying
instrumentation, the RIAS, Sellers, Smith, Shelton, Rowley, and Chavous (1998)
developed a model of racial identity that considers processes common to all minority
identities while attending to the distinctive experience of being Afiican American in
America and the identity formation process that is exclusive to that minority group. This
Multidimensional Model of Racial Identity (MMRI, Sellers et. a1, 1998) consists of four
components: salience, centrality, regard, and ideology. For the purposes of this study, the
notion of centrality is most pertinent because it helps to bridge the racial identity work of
Cross and that of other identity theorists who conceptualized ego identity development as
a universal process. Centrality is the degree to which race is significant in an Afiican
American individual’s general self-concept. The concept of identity centrality is inherent
in all developmental models. Identity development involves designating greater or lesser

importance to beliefs, values and behavior patterns experienced as self-defining.

However, the centrality construct does not capture the possibility of varying levels of
exploration and consolidation emphasized in the general ego identity development
literature, and a major aspect of Cross’ theory as well.
Ego Identity Theory.
Erikson’s (1968) well-known ego identity development theory, articulating identity
development as a universal process, is based on two defining features; the amount of
effort spent searching for an identity and the extent to which the individual is dedicated
or committed to the selected identity. An achieved identity, presumably the ideal,
develops from a period of exploration followed by commitment to a particular identity.
Using Erikson’s work, Marcia (1966) developed an identity model composed of four
statuses based on the extent to which the individuals have been involved in searching for
a new identity and the degree of their commitment to the present identity. Individuals in
the Moratorium stage are not yet committed to an identity, but rather are in crisis,
exploring new identity options. Foreclosure occurs when the individual has settled on an
identity without exploration, usually as the result of identification with significant others.
Identity Diffirsion refers to the experience of individuals who are indifferent to identity
issues, neither exploring nor committing themselves to a coherent identity. Identity
Achievement is reached after a period of exploration followed by commitment to an
identity.

More recently, theorists have included the notion of “regression” (falling back
into an earlier developmental stage) and “progression” (bypassing an identity status) in
their conceptualizations of identity development. For example, life events may force

individuals in Achievement to reassess how they construe their identity resulting in a

cycling process to Moratorium (or even Diffusion), then back towards a new version of
Achievement. Researchers have found that some adolescents cycle before and after
reaching Achievement (Adams & Fitch, 1982; Meeus, Iedema, Helen, & Volleberg,
1999). As Erickson (1968) theorized, identity formation seems to be a life long process.
Identity may change continuously, but at a slower rate after adolescence (Honess &
Yardly, 1987).

Racial Identity Revisited.

Major racial identity work has also taken place under the rubric of ethnic identity.
Phinney (1996) saw race as one aspect ethnicity, a construct based on physical
appearance and societal beliefs. She conceptualized ethnicity as a categorization of
individuals on the basis of race and culture, which could be associated with country of
origin. In her work, Marcia’s ego identity stages are reformulated to describe ethnic
identity development. Phinney (1989) argues that minority individuals experiencing
identity Diffusion have made little attempt to explore their ethnic identity and have no
clear grasp of ethnic identity issues. Alternatively, those in identity Foreclosure are clear
about their own ethnicity, but have not yet really explored identity alternatives. In the
Foreclosure status, feelings about ethnicity can be positive or negative depending on
socialization (i.e. unconditionally accepting parental views). In the Moratorium phase,
there is evidence of ethnic identity exploration, but individuals are still somewhat
confused about the meaning of their own ethnicity and its importance in their overall
experience of self. Achievement occurs when minorities have explored their various
identity options and have come to own, appreciate and accept their ethnicity. Using

Marcia’s model, while focusing on ethnicity, Phinney examined the ethnic identities of

91 American-bom high school students. Black, Hispanic, Asian, and White students
participate in the study. Based on her results, Phinney proposed a three-stage model of
ethnic identity. The Foreclosure and Diffusion identity statuses could not be reliably
distinguished in the study and therefore make up the first stage characterized by a lack of
ethnic identity exploration. The second stage, Moratorium, involves exploration of one’s
ethnicity, but with little or no commitment. In the final stage, an individual is in an
Achieved ethnic identity state that, like the Achieved states in other theories, results from
a period of exploration and a firm commitment to the identity. Phinney (1992) created a
l4-item scale to assess the theorized identity statuses. However, with items such as “I
participate in cultural practices of my own group, such as special food, much, or
customs” and “I have spent time trying to find out more about my own ethnic group, such
as its history, traditions and customs,” the scale seems too general to accurately reflect
the totality of experiences and attitudes associated with African American identity.

It is conceptually easier to translate Cross’ model as compared to the MMRI
model into a general theory of identity that emphasizes exploration and commitment. The
Pre-encounter stage equates well with the Identity Diffusion/Foreclosure stage, in which
there is a lack of exploration and identity is based on prior socialization. Immersion-
Emersion shares aspects with the Moratorium stage, most importantly the notion of
exploration. Afiican Americans in Immersion/Emersion are actively searching for a sense
of what it means to be Black and have not consolidated their racial identity. The
Intemalization stage is fimdamentally the same as the Achievement stage; race has been
explored and secured into the individual’s overall identity structure. The Immersion-

Emersion stage depicts the experience of searching and experience of change in identity

whereas the Intemalization stage describes the “habituation and internalization of the new
identity” (Cross, 1991 , p. 190). Only the Encounter stage does not fit into the
transformation; the Encounter stage represents an immediate crisis (which may
precipitate regression as well as progression) in contrast to the general feeling of
indifference found in Diffusion. To summarize, the Immersion-Emersion and
Intemalization subscales of the Racial Identity Attitudes Scale can be viewed as
dimensional measures of exploration with and without and commitment (respectively).
The Cross model also allows for cycling among stages across the lifespan, as in the ego
identity literature, with an Encounter experience as the catalysis for a shift in racial
identity (Parham, 1989). However the cycling theory in racial identity still awaits
empirical validation.

The MMRI is less adaptable than the RIAS given that, individuals can score high
on a centrality measures, but engage in varying degrees of exploration and consolidation.
The meaning of centrality for individual functioning is likely to vary as a function of
whether it is experienced in the context of exploration or as the result of exploration. For
example, African Americans with strong Immersion attitudes and high centrality may
limit their social networks to only other Blacks, whereas African Americans with strong
Intemalization attitudes and high centrality will likely associate with people of different
ethnicities. Additionally, scoring high on either scale appears indicative of having high
racial centrality. Although the MMRI allows for greater complexity in defining Black
racial identity, it may be bypassing the earlier theoretical contributions, losing essential

distinctions associated with Cross’ model and general ego identity theory.

Together, the Immersion/Emersion and Intemalization concepts may capture the
exploration and commitment ideas essential to the identity construct and articulated by
Phinney (1989) and other identity theorists (Erikson, 1968; Marcia, 1966). Individuals
low on both measures presumably have not explored identity issues (i.e. they are in
identity diffusion or foreclosure) and are in the equivalent of Phinney’s first stage of
ethnic identity development. African Americans with high scores on the Immersion
measure and low scores on Intemalization can be viewed as in Moratorium; immersing
themselves in a Black identity, but not having reached an understanding and
consolidation of aspects of the self found in the Achieved status because they are
unwilling (and perhaps too insecure) to explore the relation of their Blackness to other
identities based in the larger society. Alternatively, high scores on the Intemalization
measure, when combined with low scores on Immersion, may be indicative of an
Achieved identity, what Phinney (1989) conceptualizes as the third and final identity
status. It may appear that high scores on both scales should be associated with the
achieved state. However, it seems unlikely that an individual would score high on
Immersion and Intemalization based on the developmental nature of the Achieved state.
Marcia’s (1966) theory, on which Phinney’s work is based, assumes that an individual
has, for the most part, completed the exploration period before entering identity
Achievement; although more minor reorganizations are likely to occur over the lifespan
(i.e. spouse identity, parental identity, occupational identity).

Identng and Psychological Well-being
Studies relating identity to psychological well-being include several focusing on

identity in general and a smaller number focusing on ethnic identity in particular. In

10

terms of general identity research, Cross and Allen (1970) completed one of the earliest
studies examining identity status and mood. Using the Zuckerrnan Mood Affect
Adjective Checklist, they tested 81 male undergraduates and found no difference among
the identity status groups in undergraduates’ depression and anxiety scores. However, the
racial and gender homogeneity of the sample (i.e. White males) limit generalizability of
these results to the current study.

While Cross and Allen did not find differences in mental health among identity
statuses, other research in ego identity work has linked identity statuses to psychological
well-being. In a study completed during the same time period, Marcia and Friedman
(1970) examined the ego identity status and anxiety level of 49 college women. The
results showed that college women in the foreclosure identity status obtained the lowest
Welsh Anxiety Scale scores, whereas women in identity diffusion attained the highest
anxiety scores. However, like the Cross and Allen (1970) study, the homogeneity (i.e.
White females) of the sample limits the generaliziablity to the present research. Further,
these results are dated and obviously require future replication. In a later review of ego
identity studies concerning psychological well-being, Waterman (1992) noted that across
studies, moratorium individuals reported the highest levels of anxiety, whereas
individuals in the more committed statuses (foreclosure and achievement) reported the
least. The results imply that higher levels of commitment to an identity are associated
with lower anxiety whereas high levels of exploration are associated with high anxiety.

A similar pattern has been discovered with depression. Meeus (1996) drew 2557
Dutch adolescents, 648 of which were between the ages of 18 and 20, fi'om a national

sample to investigate developmental identity and psychological well-being. Meeus found

11

that Moratorimn adolescents were more depressed than adolescents in the remaining
three statuses. An achieved identity was associated with the least amount of depression
and highest level of general well-being. Further, diffusion and moratorium identities
were related to lower general well-being than the identity statuses high in commitment.
The cultural differences between Dutch and American adolescents limit the applicability
of results. Yet the pattern of higher identity commitment and lower exploration relating to
lower psychological distress is in accord with studies done with American adolescents.
In the general identity literature, the common thread is for greater identity
exploration/moratorium to be associated with higher levels of psychological distress
whereas higher identity commitment/achievement is associated with highest levels of
psychological well-being. If this pattern holds for ethnic identity status as well, high
Immersion attitudes should be related to high levels of anxiety and depression, whereas
high Intemalization attitudes should be associated with low levels of anxiety and
depression. Some findings show this pattern. For example, Parham and Helms (1985)
examined the relationship between racial identity attitudes and affect states in Black
college students. The researchers gave 116 Afiican American students fiom four
predominantly White universities the Racial Identity Attitudes Scale to assess their racial
identity status. The Symptom-90 Checklist (SCL—90), a measure of psychological
symptoms, was given to assess the students’ affective state. Using multiple regression
analyses, the researchers found that higher Immersion attitudes were associated with
higher scores of anxiety. The Immersion — high anxiety relationship was not among
Parham and Helms’ initial hypotheses and they did not offer any possible explanations

for the connection. Yet, borrowing form the ego identity theorist, the lack of commitment

12

and high level of exploration associated with high Immersion scores could contribute to
the increase in anxiety (Waterman, 1992).

Higher Intemalization attitudes are theorized to be related to greater psychological
well-being (Cross, 1991; Helms, 1990). One study that supports this premise examined
the relationship between depression and racial identity. Munford (1994) gave 146 Afiican
American university students the RIAS, to evaluate racial identity, and the Beck
Depression Inventory, a highly-regarded measure of depression. Regression analyses of
the results demonstrated that racial identity attitudes were a significant predictor of
depression. Although the associations were not extremely strong, Intemalization scores
were negatively correlated with depression scores (r = -.23) whereas Immersion scores
were positively correlated with depression (r = .14).

Other researchers have not been able to relate psychological well-being to racial
identity. Saylor and Aries (1999) found that first year minority college students with
strong ethnic identity did not differ in levels of psychological well—being from those
students with weaker ethnic identities. However, the sample in the study consisted of six
ethnic minority groups and the number of Afiican American participants was too small to
make a separate analysis. In a study with similar non-significant results, Pyant and
Yancio (1991) found that there was no association between depression or well-being of
African American college students and their Immersion and Intemalization scores on the
RIAS. However, all of the participants were drawn from students at a predominantly
Black university. Phinney (1990) noted that in racially homogenous environments, racial
identity is likely to be less relevant to the self-concept of an individual of that race.

Although not specifically stated, racial identity may be less relevant to psychological

l3

well-being. If racial identity is more or less irrelevant at a predominantly Black
university, it should not influence the psychological well-being of the students that attend
the university. There is some support for this belief. As noted earlier, research suggests
that during high school, minority students experience greater psychological well-being
when attending a school in which they are in the racial majority rather than a distinct
minority (Gray-Little & Carels, 1997, Hudley, 1997). However, in a more heterogeneous
environment, racial identity should matter, and perhaps even more so in a predominantly
White environment. Consequently, racial identity should bear some influence on
psychological well-being in the proposed study.
Prior Interracial Experience

More generally, the lack of consistently strong, direct relationships across studies
between racial identity and psychological well-being in African American college
students, even at predominantly White institutions, suggest that other moderational
factors may be involved. Previous interracial experience in an academic setting may be
one such factor moderating the relationship between racial identity and psychological
distress. For example, Graham, Baker, and Wapner (1985) examined whether the racial
composition of an African American freshman’s former high school also contributed
their experience of psychological well-being during their first semester at a
predorrrinantly White university. The researchers assessed college adjustment in 42 Black
fi'eshmen on a predominantly White campus at three points during the Fall and Winter
academic semesters. Using AN OVA for repeated measures to analyze their results, the
researchers found that African American students from predominantly White high

schools adjusted better emotionally to the predominantly White college than those who

14

had attended integrated high schools. African American students who attended
predominantly Afiican American high schools adjusted most poorly. However, D’Augelli
and Hershberger (1993) did not find an association between high school racial
composition and well-being. While examining the social networks of African American
students at predominantly White colleges, the researchers noticed that the African
American students reported lower total well-being, measured by the General Well-being
Schedule, than the White students. In an attempt to clarify the discrepancy, the authors
compared African American students’ high school racial composition and well-being
scores. Graham, Baker, and Wapner’s study (1985) would suggest that the percentage of
Blacks at the students’ former high school would be a factor. However, D’Augelli and
Hershberger found no significant differences between the well-being scores of Afiican
American students who attended predominantly White high schools and those who
attended predominantly Afiican American high schools. Yet, the analysis did not take the
students’ levels of racial identity into account. In addition, the extent to which African
American students identify with their race may interact with their high school
composition to influence their level of well-being. Certain patterns of racial identity
attitudes could be an individual difference among Afiican American students that
determines that relationship between high school racial composition and psychological
well-being. The psychological implications of identity exploration may intensify in a
racial environment inconsistent with past experiences. African American students may be
driven to greater Immersion attitudes at a predominantly White university if they attended
a predominantly Black high school, leading to significant changes in psychological

distress.

15

Summg and Hypothesis

Summary

The present study examined the predictors of psychological distress, specifically
depression and anxiety, for incoming Afiican American students at a predominantly
White university. Arr integrative model of Black racial identity that consolidated the work
of Cross and Phinney, by using Immersion and Intemalization attitudes as dimensions of
exploration/moratorium and intemalization/achievement respectively, was used to assess
the influence of racial identity on psychological well-being. Additional analyses were
completed to determine if the percentage of Blacks at the students’ former high school
interacted with their racial identity patterns in explaining change in anxiety and
depression.

As noted earlier, African American students frequently experience considerable
difficulties consolidating race into their global identity (Steinberg, 1999) which is
associated with increased psychological distress (Carter, 1991; Chamber et al., 1998;
Munford, 1994; Parham & Hehns, 1985; Siegel, Yancy & McCarthy, 2000; Taub &
McEwen, 1992). These findings suggest that during a period of racial identity exploration
(stronger Immersion attitudes), Afiican American students may be more vulnerable to
psychological distress. A predominantly White campus may pose an additional threat to
their identity formulation, amplifying their risk for change in distress; especially for
African American students without prior interracial experiences. Students who are
already actively exploring their racial identity and who have also attended predominantly
Black high schools are now forced to deal with the “race card” on a daily basis. They are

no longer just “students” as they were in high school, they are “Black students” and must

16

reorganize their racial identity accordingly. The added complication of a new racial
environment could increase the risk for change in psychological distress. However,
stronger Intemalization attitudes, indicative of a stronger appreciation of self and
commitment to racial identity, may protect Afiican American students; diminishing these
risk factors (i.e. Immersion attitudes and racial high school composition) through a
confident sense of self. While these students still may be concerned about what the
environment means “to them,” stronger Intemalization attitudes allow students to be less
concerned with what being Afiican American on a predominantly White campus means
“about them,” thereby decreasing the risk of change in identity related psychological
distress.

In this study, Intemalization attitudes, Immersion attitudes, and high school racial
composition were expected to interact in predicting change in the level of psychological
distress during the first six weeks of the students’ first semester. Higher Intemalization
attitudes were expected to serve as a protective factor, given the implications of the
attitudes for inner security with race and self-concept. Regardless of context, the
intemalized/achieved students were expected to show consistently low levels of
psychological distress between time periods. In accord with higher levels of anxiety and
depression related to a Moratorium identity status, higher Immersion attitudes were
expected to be a risk factor for anxiety and depression. However, Intemalization attitudes
and high school racial composition were expected to moderate the relationship between
Immersion attitudes and changes in psychological distress; with distress evident when
Intemalization attitudes are relatively weak and even more so when the students’ former

high school was also predominantly Black.

17

Hypotheses

Hypothesis 1. Intemalization attitudes will be directly related to change in
psychological distress. Afiican American students with high Intemalization scores will
experience significantly smaller increases in psychological distress between Time 1 and
Time 2 than those reporting weaker Intemalization attitudes, regardless of the level of
Immersion and/or the percentage of Blacks at their high school.

Hyppthesis 2. The change in psychological distress between time points will be
highest when individuals’ scores on Intemalization dimension are relatively low,
Immersion scores are relatively high (i.e. the individual is in moratorium) and these same

individuals come fiom a predominantly Black high school.

18

METHOD

Participants

Initially, 149 Afiican American undergraduate fieshmen at Michigan State
University, entering the University during the Fall semester of 2002, completed part of or
the entire study questionnaire during the first data wave. Participants missing more than
10% of the data (N = 8, 5%) or failing to complete a questionnaire during the second data
wave (N = 43, 28%) were removed fi'om further analysis. The final sample of 98
participants (65% retention rate) included 25 men and 73 women. Participants were
drawn from the Michigan State University Maximizing Academic Growth in College
(MAGIC, N = 27) program targeting entering minority students, the College
Achievement Admissions Program (CAAP, N = 17), a program directed toward first
generation and low-income students, and from the introductory psychology subject pool
(N = 9). Participants were also recruited in the MSU residence halls through posted fliers
and information handed out by research assistance on site (N = 45). The mean age of the
participants was 17.91 (_S_l_)_ = .38), while the mean high school GPA was 3.30 (g = .38).
Procedure
This study included data from the first two waves of a larger study over five time points,
that examined the psychological well-being and college adjustment of African American
freshmen at a predominantly White college during their first year. Participants drawn
from the MAGIC and CAAP programs listened to a brief presentation about the study
and received a flier with a tear-off section to provide contact information if they are
interested in participating. Students indicating interest were contacted and scheduled for a

testing session. Participants recruited from the department subject pool signed-up for the

19

study using the departmental subject pool web-site after reading a short description on the
study. Participants drawn from the residence halls were invited to attend a scheduled
testing session, by means of direct contact with a research assistant or an informational
flier posted in the vicinity. At Time 2, all participants received a questionnaire packet by
mail three months afler the initial testing date and were asked to complete it and/or return
it during one of the scheduled pick-up sessions on-campus the following week. At each
time point, students participating through the subject pool received course credit and
those recruited in other ways received $5 for completing the surveys. As further
compensation, the names of all participants, regardless of the recruitment site, who
completed questionnaire packets at a specified number of data points (inclusive of Time 1
and Time 2) were entered into a drawing for a $25 cash award.

Testing sessions for the present study took place a) during the first two weeks of
the Fall 2002 semester and b) at the end of the 6th week of classes. Participants attended
one of six 2-hour group-testing blocks were scheduled for each data point. Participants
unable to take part in the group sessions were contacted in order to set up alternative
testing times. At the beginning of the first test administration, participants were asked to
sign two copies of the consent form (one copy to turn in and one to keep for personal
records) before receiving their first questionnaire packet. All questionnaire packets
contained the Racial Identity Attitudes Scale Immersion/Emersion and Intemalization
subscales (Appendix A), Beck Depression Inventory-II, and State-Trait Anxiety
Inventory. Participants were also asked to fill out a demographic questionnaire (Appendix
B) during the first testing session. Instruments were counterbalanced to account for order

effects.

20

Measures
Black Racial Identity

The Racial Identity Attitudes Scale - longflrr‘m (RIAS). Two subscales for the RIAS
(Helms & Parham, 1985) were used to assess the levels of Immersion/Emersion and
Intemalization. The RIAS is a self-report measure that consists of 50 items and uses a 5-
point Likert scale (“strongly agree to “strongly disagree”). The Immersion/Emersion
subscale contains 9 items, such as “I feel unable to involve myself in White experiences
and am increasing my involvement in Black experiences” and “The most important thing
about me is that I am Black.” The Intemalization subscale is made up of 12 items which
include “People, regardless of their race, have strength and limitations” and “I know
through experience what being Black means.” The scores for each item in a particular
subscale are smnmed and then divided by the number of items in the subscale. Omitted
items are scored as the rounded average of the other items in the subscale. In one
interpretation, racial identity is defined by the attitude subscale with the highest mean
score. This assumes that a racial identity attitude is synonymous with its corresponding
racial identity stage. As an alternative, Helms and Parham (1985) encourage researchers
to interpret the results for all four subscales rather than assigning individuals to a single
group based on their highest mean score. Helms and Parham (1996) report internal
reliability estimates, based on a study of university students, as follows: Pre-encounter =
.76; Encounter = .51; Immersion/Emersion = .67; Intemalization = .80. However, due to
prior criticisms surrounding the validity of the RIAS (Lemon & Waehler, 1996; Munford,

1994; Yancio, Swanson, & Tokar, 1994), a content analysis and confirmatory factor

21

analysis (CF A) were employed, to confirm the underlying structure of the Immersion and
Intemalization scales.
Psychological Distress

Anxiety and depression, found to be relatively prevalent among college students
(Whatley & Clopton,l992). It is well-known that anxiety is often comorbid with
depression. In an extensive review of this comorbidity, Brady and Kendell (1992) found
comorbidity estimates among ranging from 15.9 to 61.9%. Though they are comorbid,
anxiety and depression are conceptually distinct. Although both are saturated with high
negative affectivity (negative emotional feelings), low positive affectivity (positive-
affective states) is unique to depression (Watson, 1984; Watson, Clark, & Carey, 1988).
Therefore, depression and anxiety were analyzed as separate dependent variables.

Beck Dgpression Inventopy — II (EDI-II). The BDI -II (Beck, Steer & Brown, 1996) is
made up of 21 items regarding symptoms considered consistent with unipolar depression.
Each symptom is rated on a 4-point scale, 0 — 3, in terms of severity. A total score is
calculated by summing the ratings of all 21 items with a maximum score of 63. If an
individual chooses more than one answer, the response with the highest rating is used in
the computation. Scores from O — 13 represent minimal depression, 14 — 19 are indicative
of mild depression and 20 -— 28 are associated with moderate depression. Severe
depression is indicated by scores of 29 and above. The BDI-II is considered an
improvement over the previous version of the scale due to changes in content and
sensitivity (Beck, Steer & Brown, 1996; Steer & Clark, 1997). The scale assesses

symptomtology over two weeks to correspond with DSM-IV guidelines. In addition,

Beck and colleagues (1996) report that the scale’s reliability (Coefficient or = .92) is

22

higher than that of the original BDI (Coefficient or = .86). No studies validating the BDI-
II in Afiican American samples were identified by the author, but a few studies of scale
properties when using college students. The BDI-II has shown strong internal consistency
(range = .89 - .92) with undergraduate samples and is considered a strong measure of
depression for college students (Beck, Steer & Brown, 1996; Dozois, Dobson, &
Ahnberg, 1998; Steer & Clark, 1997; Whisman, Perez, & Ramel, 2000).

State-Trait Anxieg Inventory (STAI). The STAI (Spielberger, Gorsuch, & Lushene,
1970) is a self-report measure that consists of two scales differentiating between state
(fluctuating contextual condition) and trait (personality characteristic) anxiety and will be
used to asses students’ anxiety in the study. Each item is answered using a 4-point Likert
scale, from 1 (not at all) to 4 (very much so). The State anxiety scale contains 20 items,
such as “I feel calm” and “I am jittery,” that assess how an individual at a particular
moment. The Trait anxiety scale includes 20 items, such as “I feel like a failure” and “I
am a steady person,” that measure how the same individual normally feels. Scores for
both scales can range fi'om 20 (indicating low anxiety) to 80 (indicating maximum
anxiety). Only the trait anxiety scale was used in the present study because the experience
of distress was expected to be relatively stable, not just situational. In this study, the
predominantly White university context was viewed as a constant, not situational, factor
in the fi'eshmen’s lives.

The STAI has been validated using university students and has been used extensively
with that population (Spielberger, 1983). The STAI shows strong reliability, with alpha
coefficients from .91 to .93 for the State anxiety scale and .90 to .91 for the Trait anxiety

scale (Spielberger, 1983); and concurrent validity with other measures of anxiety. Less

23

work has been done examining the psychometric properties of the STAI with African
Americans. Brown and Duren (1988) tested the construct validity of the STAI with Black
members of the Pentecostal Church of God. The STAI differentiated well between state
and trait anxiety in this population. In a later and more encompassing study, Novy and
colleagues (1993) found that the STAI reliably assessed state and trait anxiety in literate
Whites, Black and Latinos. Therefore, it was feasible that the STAI provided a reliable
and valid measure of trait anxiety in for the study.
Demographic Information.

A demographic questionnaire was administered to attain participants’ age, sex,
high school racial composition, and parental total household income. Participants were

also asked to indicate their high school GPA.

24

RESULTS
Preliminm Analyses
Measurement Models for the Immersion and Intemalization scales.
A content analysis of items in the RIAS Immersion and Intemalization subscales
originally identified by Helms and Parham (1996) (Appendix A) led to the elimination of
four Immersion items and seven Intemalization items that did not reflect the constructs as
defined in the study. Three of the original nine items in the Immersion scale (items 6, 22
and 25) reflected distain for the White race, not an exploration of Black identity, and
were deleted; as was one item measuring the extent to which individual respondents are
vocal about their general, rather then their racial beliefs (item 18). Of the twelve
Intemalization items in the original scale, three items (items 17, 26, and 27) seemed
related to an individual’s level of tolerance, not necessarily associated with acceptance of
self, while three additional items appeared to measure an appreciation for Blackness
likely to be equally characteristic in individuals with Immersion and Intemalization
attitudes (items 4, 7 and 20). Examination of frequency distributions led to the
elimination of another Intemalization item (item 7) that had minimum variability (80% of
the respondents had extreme scores). Final items retained for further analysis are shown
in Table 1.

A confirmatory factor analysis (CF A) using AMOS confirmed the underlying
structure of the final 5-item Immersion and 5-item Intemalization scales. Data from
participants who responded to all 21 items of the RIAS (n=132 or 89% of the sample)
were used for these analyses. Model fit was evaluated employing the Chi-square

goodness of fit test, the Tucker-Lewis Index (TCI), Comparative Fit Index (CPI),

25

Table l

Racial Identity Attitudes Scale Items

 

 

Latent Scale Item
Construct Path Coefficients
Immersion 3. I am Increasing my involvement in Black activities
because I don’t feel comfortable in White
.70 (.49)
environments.
13. I believe that the world should be interpreted from
.64 (.64)
a Black or Afrocentric perspective.
14. I am changing my style of life to fit my beliefs about
.71 (.70)
Black people.
19. 1 limit myself to Black activities as much as I can. .36 (.34)
23. The most important thing about me is that I am
.52 (.51)
Black.
Intemalization l. I believe being Black is a positive experience. .67 (.67)
2. I know through experience what being Black in
.76 (.75)
America means.
24. Being black just feels natural to me. .56 (.57)
29. I am satisfied with myself. .59 (.58)
30. I have a positive attitude about myself because I am
.72 (.74)

Black.

 

 

Note. ( ) — coefficients derived when both constructs where run simultaneously

Cronbach’s alphas for final Immersion and Intemalization scales were a. = .70 and .79 respectively.

Immersion and Intemalization unit weight scale, r’2 = .24, rho = .81

26

Goodness of Fit Index (GFI) and the Root Mean Square Error of Approximation
(RMSEA), as well as cut-off values recommended by Hu and Bentler (1998, 1999) and
J oreskog and Sdrbom, (1993). Statistics for the 5-item Immersion scale indicated
adequate fit between the predicted and observed measurement models ([x2 (df 5, N =
132) = 3.56, p = .62]; RMSEA = .00, TLI = 1.04, CF I = 1.00, GFI = .99). Similarly, fit
statistics for the 5-itern Intemalization scale were adequate

([x2 (df 5, N = 132) = 7.47, p = .19]; RMSEA = .06, TLI = .97, CF I = .99, GFI = .98).

Differences in the Chi-squares of a model using the original Helrn’s items and a
model using the final five items were computed for each construct. For the Immersion
scale, the difference in the Chi-squares [x2 = (42.45(df= 27) vs. 3.56 (df= 5)] was
statistically significant, thereby favoring the use of the 5-item rather than the original 9-
iterrr scale as a measure of the Immersion construct. Likewise, a comparison of the
original 12-item Intemalization model against the 5-item model of the same construct
resulted in a significant Chi-square difference If (189.75[df= 49] vs. 7.47 [df= 51)],
favoring the use of the 5-item scale as a measure of the Intemalization construct.

A final CFA confirmed the relative distinctiveness between the Immersion and
Intemalization constructs as measured by their respective final 5-item scales. In the
model, Immersion and Intemalization were allowed to covary (rho = .81), given that both
share a high valuation of “Blackness” (Figure 1). Regression weights derived fi'om the
analysis are shown in parenthesis in Table 1. Consistently good fit results across the

various indicators led to the measurement model being judged as a good fitting model of

27

 

 

 

 

 

 

 

 

 

 

 

 

a...

 

Intemalization

Figure 1. Measurement Model of Immersion and Intemalization Scales

28

itemi

 

 

item2

 

 

 

 

item24

 

 

 

 

item29

 

 

item30

e e 6 e 8

 

the data [(x2 (df 34, N = 132) = 45.01, p = .10); RMSEA = .05, TLI = .95, CFI = .96, GF I
= .94. The correlation between the unit-weight scales was .24. Scale reliabilities for the
resulting Immersion and Intemalization scales were acceptable (Coefficient O. = .70 and
.79 respectively).

Descriptive statistics.

Means and standard deviations for the variables in this study are presented in Table 2 and
correlations among these variables in Table 3. A small number of participants had
missing data on the BDI and/or STAI. A total of 24 participants were missing one or two
on BDI or STAI at one of the two time points. Mean substitution was used to replace
missing items for these participants. For this sample Cronbach’s alphas for the BDI and
the STAI-Trait were .87 and .88, respectively.

A comparison of means for the normative college samples for the BDI (M = 7.65,
M2 = 5.9) and the STAI (M = 38.30 S_D = 9.18) with those for this sample showed that
participants in this study generally experienced non-clinical levels of depression (at Time
1 and Time 2, M_’s = 7.81 and 8.75 respectively) and average levels of anxiety, (M’s =
37.69 at Time 1 and 37.95 at Time 2). Before coming to the predominantly White
university, most students in the study had attended a high school with a higher percentage
of African American students than students of other races (M = .63).

T -tests comparing scores on each of the variables showed a significance
difference in the Immersion scores of males (M = 2.60, S_D = .96 ) and females (M =
2.14, _SQ = .62); t(31.20) = 2.22, p = .03; eta squared = .05). Male participants had
somewhat stronger Immersion attitudes than female participants. Additionally, there was

a significant sex difference in depression scores at Time 1. Male participants initially

29

Table 2

Properties of Study Measures: Means and Standard Deviations

 

 

Misure Mean SD
RIAS - Immersion 2.26 .75
RIAS — Intemalization 4.26 .59
High School Racial Comp. .63 .35
ED] — 11 (Time 1) 7.81 6.24
BDI — 11 (Time 2) 8.75 8.27
STAI - Trait (Time 1) 37.69 8.96
9.84

STAI — Trait (Time 1) 37.95

 

30

Table 3

Intercorrelations Among Study Variables

 

 

Variable l 2 3 4 5 6 7
Participants (n = 98)
l. Immersion — .138 .042 .024 -.105 .2728 .077
2. Intemalization — -.1 11 -.314a -.320a -.273al -.301a
3. High School — .118 .130 .060 .175
4. Depression Time 1 — .590“ .594a .459a
5. Depression Time 2 — .507a .6698
6. Anxiety Time 1 — .6191'

7. Anxiety Time 2

 

I'p<.01

31

expressed fewer depressive symptoms than female participants (Ms = 5.46 (S_D = 6.69)
vs. 8.61, (SD = 5.91); t(96) = -2.22, p = .03; eta squared = .03). T-test analyses showed
no significant sex differences in depression at Time 2, anxiety at Time 1 and Time 2,
Intemalization attitudes, or high school racial composition (Table 4).

As can be seen in Table 3, correlations among the variables in this study were
relatively low, minimizing concerns of multicolliearity. Skewness among variables
ranged fi'om -O.57 to 1.94, while kurtosis ranged from -1.35 to 4.59 in the sample.
Depression accounted for the higher instances of skewness and kurtosis (1.94 and 4.59
respectively at Time 2) consistent with the low prevalence of depression among African
Americans in the general population (Brown, Ahmed, Gary & Milburn, 1995). Skewness
and kurtosis values for the remaining variables generally did not significantly deviate
from a normal distribution (West, Finch & Curran, 1995).

Tests of the research predictions

A hierarchical multiple regression analysis was employed to test the predicted
relationships among racial identity, high school racial composition and “changes” in
students’ psychological distress (i.e., trait anxiety and depression) over the first six weeks
at a predominantly White university. Separate analyses regressed changes in anxiety and
depression on other variables in the model. Change was measured by residual scores
derived by regressing Time 2 anxiety or depression on Time 1 scores for the same
construct. Variables were standardized with interactions computed as cross-products.
Time 1 psychological distress scores were entered in on the first step. Main effects were
entered in the second block. Three 2-way interactions (the cross-products of

Intemalization and Immersion, Immersion and high school racial composition, and

32

Table 4

Comparison of Means on Immersion and Intemalization: T-test

 

 

 

 

Means (SD)

Malis Females df t
Immersion 2.60 (.96) 2.14 (.62) 96 2.72“
Intemalization 4.35 (.61) 4.22 (.58) 96 .93
High School .73 (.32) .60 (.36) 96 1.60
Depression Time 1 5.46 (6.69) 8.61 (5.91) 96 -2.22“
Depression Time 2 6.08 (4.86) 9.66 (9.00) 96 -l.89
Anxiety Time 1 37.53 (10.06) 37.74 (8.62) 96 -.11
Anxiety Time 2 36.12 (9.74) 38.58 (9.86) 96 1.60
a p < .05

33

Intemalization and high school racial composition) were entered in on the third step. The
3-way interaction was entered on the fourth step.

Hypothesis 1. Intemalization attitudes were hypothesized to be a negatively
related to changes in levels of psychological distress between Time 1 and Time 2.
Although Intemalization attitudes were significantly negatively correlated with
depression and anxiety at Time 1 and Time 2 (see Table 3), results did not detect any
main effects for Intemalization, either for changes in depression [Beta = -. l 3, t: -1.47, p
=.15] or changes in anxiety [Beta = -.11, _t_= -l .33, p =.l9] between the two time points.
Hence African American freshmen who reported stronger levels of Intemalization
attitudes at Time 1 did not experience lower levels of psychological distress, when
controlling for initial distress, compared with African American freshmen reporting
lower Intemalization attitudes.

Hypothesis 2. Changes in depressive and anxiety symptoms were hypothesized to
be most evident among participants with low Intemalization attitudes and high
Immersion attitudes coming from predominantly Black high schools. However, there
were no significant three-way interaction effects for either depression [E (8, 89) = 9.83; p
= .20] or anxiety [E (8, 89) = 9.35; p = .29]. Alternatively, two significant two-way
interactions where found for depression. An interaction between Immersion and
Intemalization attitudes [Beta = .21, 1= 2.57, p<.05] was associated with changes in
depression scores. Among students with relatively stronger Immersion attitudes, those
with weaker Intemalization attitudes showed the greatest increase in depressive
symptoms and those with stronger Intemalization attitudes showed the least. Among

students with weaker Immersion attitudes, the risk for increased levels of depression was

34

also greater among those lower Intemalization attitudes. This relationship was less strong
than for those with stronger Immersion attitudes (Figure 2).

A significant interaction between Immersion attitudes and high school racial
composition was also associated with changes in levels of depression between Time 1
and Time 2 [Beta = -.16, t: -2.05, p<.05]. Among participants who had attended a
predominantly Black high school, those with relatively strong Immersion attitudes
experienced a greater increase in depressive symptoms than those with relatively weak
Immersion attitudes (Figure 3). Students who attended predominantly White high schools
showed relatively little change in levels of depression, regardless of their Immersion
attitude strength. Additionally, while no significant two-way interaction was found when
regressing the predictors on anxiety, there was a trend for the same
Immersion/Intemalization interaction for anxiety that was found for depression [Beta =
.15, t: 1.82, p =.07]. Statistics fiom the regression analyses can be found in Table 5 and

Table 6.

35

 

High
Residual Depression

Low

 

High
Immersion

 

 

 

Low Intemalization High

Figure 2. Immersion and Intemalization Interaction on Residual Depression

36

High

Residual Depression

Low

 

Predominantly
Black
Hthdwd

  

Predominantly
White
High School

 

 

LOW Immersion ”‘9“

Figure 3. Immersion and High School Racial Composition Interaction on
Residual Depression

37

 

Table 5

Hierarchical Regression Analysis for Predictors of Time 2 Anxiety

 

 

(N = 98)

Predictors B p/In t-test AR2
Step 1

Anxiety Time 1 .64 .62 7.73° .3 83
Step 2

Immersion -.O8 -.08 -.91

Intemalization -.14 -.1 1 -1 .33

HS Racial Composition .13 .13 1.63 .04
Step 3

Immersion x Intemalization .18 .15 1.17b

Immersion x High School -.09 -.09 1.82

Intemalization x High School .02 .02 -1.08 .03
Step 4

3—way Interaction -.10 -.09 -1.08 .01

 

'p<.05, l’p=.07

38

Table 6

Hierarchical Regression Analysis for Predictors of Time 2 Depression
(N = 98)

 

 

 

Predictors B B/In t-test AR2
Step 1

Depression Time 1 .63 .55 6.31“ .35
Step 2

Immersion -.15 -.15 -1.76

Intemalization -.l6 -.13 -1.52

HS Racial Composition .02 .02 .26 .03
Step 3

Immersion x Intemalization .25 .21 2.57“l

Immersion x High School -.18 -.16 -2.053

Intemalization x High School -.13 -.11 -l .36 .08
Step 4

3—way Interaction .12 .11 1.31 .01

' p < .05

39

DISCUSSION

This study was conducted in order to evaluate the predictive nature of racial
identity for increased psychological distress, operationalized as increased anxiety and
depression, among African American freshmen at a predominantly White university.
Black racial identity was conceptualized by an innovative approach that consolidated the
work of Cross (1991) and Phinney (1990); using varying magnitudes of Immersion and
Intemalization attitudes as dimensions of racial identity. African Americans with high
scores on the Immersion measure were regarded as being in an exploration/moratorium
phase whereas high scores on the Intemalization measure were thought to be indicative of
intemalization/achievement. Strong Intemalization attitudes were hypothesized to serve
as a protective factor; African American students with high Intemalization scores were
expected to experience less dramatic increases in psychological distress over the first six
weeks of the first semester than students with weaker Intemalization attitudes, regardless
of Immersion attitude strength or high school racial composition. Further, Intemalization
scores, Immersion scores, and high school racial composition were hypothesized to
interact in predicting psychological distress. Strong Immersion attitudes and a
predominantly Black high school racial composition were expected to be a risk factors
for changes in psychological distress and Intemalization attitudes was expected to
moderate these relationships. Psychological distress was hypothesized to be highest when
the student’s Intemalization attitudes were relatively low, Immersion attitudes were high
(identity exploration/moratorium) and the student had attended a predominantly Black

high school.

40

Contrary to the first hypothesis, strong Intemalization attitudes were not directly
related to changes in psychological distress. Although correlations at each of the two time
points were in the expected direction (i.e. negative), African American students who held
strong Intemalization attitudes were no more likely to report an increase in anxiety or
depression than students with relatively weaker Intemalization beliefs. Prior research
using cross sectional data are not entirely consistent about the relationship between
Intemalization beliefs and psychological distress, with Cross (1991), Helms (1990) and
Munford (1994) showing the expected negative relationship and Pyant and Yancio
(1991), as well as Saylor and Aries (1999) finding no relationship. Failure in this study to
link Intemalization to changes in psychological distress may be a due to negative skew in
Intemalization scores. It is also conceivable that in a self-selection bias, students with
stronger Intemalization beliefs are more likely to attend a predominantly White
university, diminishing variability. African American students with significantly strong
Immersion attitudes may be less likely to consider attending a predominantly White
university, favoring the “comfort” of Historically Black Colleges and Universities
instead.

Study results partially supported the second hypothesis proposing a three-way
interaction. While no three-way interaction effects were discovered, findings supported
the notion of a relationship between an exploration/moratorium Black identity and
depression, consistent with research in the general identity literature (Meeus, 1996;
Waterman, 1992). Results identified two separate two-way interactions: one between
Intemalization and Immersion and another between high school racial composition and

Immersion. Students with relatively strong Immersion attitudes an_d relatively weak

41

Intemalization attitudes experienced a greater increase in depressive symptoms than those
with both strong Immersion arid strong Intemalization attitudes. Only six weeks into their
first semester, participants in an exploration/moratorium phase experienced a significant
increase in depression. Desiring an opportunity for self-reflection and introspection
around race, individuals with exploration/moratorium characteristics may have found
themselves frustrated in a predominantly White campus. Additionally, students with
relatively weaker Immersion attitudes and relatively higher Intemalization did not
experience a sizable increase. in depressive symptom in comparison to participants with
both relatively weak Immersion and Intemalization attitudes. Notably, students with a
High Immersion/Low Intemalization pattern of identity experienced a greater increase in
depressive symptoms than students with both relatively strong Immersion and
Intemalization attitudes. Immersion, in the context of varying degrees of Intemalization,
may be beneficial to well-being. These findings are noteworthy because they suggest that
a High Immersion/High Intemalization status, originally not considered meaningful in
this study, could be considered a preferable identity status for psychological well-being,
not what was perceived as an achieved Intemalization state (commitment after
exploration). This pattern of racial identity, quantifiable with this scale, may be what
Cross (1991) considered Intemalization-Commitment. The Immersion scale used in this
study assessed a degree of “activeness” that Cross characterized in the Intemalization-
Imrnersion phase. Intemalization-Commitment, not assessed in any racial identity
measure at this time, could have far reaching implications for student well-being. Overall,
these findings suggest that “exploration” and “commitrnen ” are better defined by a

composite of Immersion and Intemalization attitude strength.

42

Interpretation of the interaction between Immersion and high school racial
composition showed that among participants who had attended a predominantly Black
high school, those with stronger Immersion attitudes were more at risk for increases in
depression during their first semester at a predominantly White university than those with
relatively weak Immersion attitudes. The mental health implications of Immersion
attitudes are not moderated only by the strength of Intemalization attitudes, but also by
environment. An Afiican American student may be in the process of defining themselves
racially, but be less distressed if the enviromnent remains supportive of their emerging
racial identity. However, because this study did not examine the implications of racial
identity among fieshmen at a predominantly Black university, this possibility remains to
be confirmed. Notably, students who attended a predonrinantly White high school had
relatively no change depression regardless of the strength of their Immersion attitudes,
suggesting that prior interracial experience may be a buffer for depression in this
population.

Limitations

An important strength of this research was its longitudinal design. However the
study was limited by a smaller sample size than originally intended, which may have
reduced the possibility of finding a three-way interaction.

Another major limitation was possible recruitment biases. A substantially larger
number of women participated in the study than men. It is unclear how sex may have
acted as a selection bias (i.e. were women more open to help) or why men were more
difficult to recruit. Given that males reported higher levels of Immersion than females in

this study, it is conceivable that racial identity may have contributed to the difficulty in

43

recruiting men. While there is no information regarding the role racial identity may have
played in the likelihood of participation, stronger Immersion attitudes may have been
associated with a suspiciousness of a “Black” study on a predominantly White campus. It
is also unclear if having an African American primary investigator on the study
minimized the effect for some participants that were initially reluctant.

Although the study design called for the recruitment of “at risk” students
(identified by their involvement in a campus program geared toward that population), a
minimal number of those students participated. There were not enough students enrolled
in the “at risk” program to do separate analyses, but it is possible that “at risk” students
were also more at risk for psychological distress,"given the number of stressors they may
presented with (e. g. financial difficulties, poor college preparation). Future researcher
may wish to look a level of “risk” as a moderating construct.

As previously discussed, other investigators have pointed out problems in the
RIAS. Many of the items appear too ambiguous to differentiate between the dimensions
they are theorized to reflect. While a CPA led to development of “tighter” measures of
lrnmersion and Intemalization, as conceptualized in this study, the measurement
constructs may prove sample specific. Future research is needed to test whether these
particular solutions can be found elsewhere. It is also unclear to what degree social
desirability played a role in item response on the two scales.

Implications and Future Resegch

Given that the results tend to support the construct of Black racial identity

theorized in this study; using levels of Immersion and Intemalization as dimensions in

which to identify Black identity statuses (i.e. moratorium and achieved) researchers in the

area of racial identity can utilize the RIAS in a new way, providing a way to appreciate
the dimensional nature of the Immersion and Intemalization constructs while employing
them in a way consistent with generally identity “stages.” Since individuals, particularly
adolescents, have been hypothesized to cycle through identity stages (Adams & Fitch,
1982; Meeus, Iedema, Helen, & Volleberg, 1999), it may be interesting to explore how
varying levels of Immersion and Intemalization may influence how Afiican Americans
cycle during their search for racial identity.

The study had a limited time frame for a change in distress to occur and it would
be worthwhile to examine racial identity and psychological distress over a longer period
of time. Stronger Intemalization may have a protective, buffering effect over time instead
of an immediate effect as theorized in this study. Alternatively, stronger Immersion
attitudes may be associated with a trajectory of initially low to increasingly high
psychological distress over the freshman year; which could explain the absence of a main
effect for Immersion in this study. Analyzing individual differences in variable
trajectories will help determine which variables have a significant influence on the
development of depression and anxiety over time, thereby facilitating the creation of
preventive treatment plans for student that may show little to no pathological
symptomotogly earlier in the year, but experience greater distress as the semester
progresses. Additionally, the study only looked at changes in psychological well-being as
a firnction of racial identity measured at one point in time. Given the brief amount of time
between data points, the likeliness of finding appreciable changes in racial identity was
thought to be minimal. Therefore, utilizing only the initial racial identity scores seemed

appropriate. However, firrther research examining changes in racial identity and the

45

associated changes in psychological well-being are needed for a more profound
understanding of that relationship.

It is unclear why there were no significant changes in anxiety. One possible
explanation is that changes in trait anxiety, considered a generally stable emotion, may
take longer to manifest. Another possibility is that the changes in distress were limited
changes in low positive affectivity, thereby inaccessible to an anxiety measure. Potential
researchers may want to consider state and trait differences, as well as positive and
negative affectivity, in their psychological distress outcome measures.

Another area of future interest may lie in students’ perceptions of the university.
Although this study viewed a predominantly White university as a new stressor in the
students’ lives, necessitating some type of reaction, students’ perceptions of the
environment were not assessed and hence not considered as a moderating variable. The
way in which Afiican American students’ perceive their new environment may have
significant implications on their well-being. For instance, an African American student
coming from a predominantly Afiican American high school with strong Immersion
beliefs could perceive the campus as negative (discriminating, racist, and alienating) or
positive (opportunity for upward mobility, career opportunities) leading to different
levels of distress.

Factors that compromised psychological well-being among Afiican American
students attending predominantly White universities are still inadequately understood.
The findings fi'om this study can be utilized in designing and enhancing services and
interventions targeting psychological well-being among African American university

students. These findings may also serve as a basis for interventions in high schools

46

designed to prepare Afiican American students for the transition to college. Programs
that prevent psychological morbidity among African American freshmen can help ensure
that these students have equal access and opportunities to enjoy the educational and life

benefits of these institutions.

47

Appendix A
Racial Identity Attitudes Scale (Helms & Parham, 1985)

Directions: This questionnaire is designed to measure people’s attitudes about social and
political issues. There are no right or wrong answers. Different people have different
viewpoints; so try to be as honest as you can. Beside each statement, circle the number
that best describes how you feel using the scale below.

 

 

l 2 3 4 5
Strongly disagree Disagree Uncertain Agree Strongly agree
1. I believe being Black is a positive experience. 1 2 3 4
2. I know through experience what being Black in America means. 1 2 3 4

3. I am increasing my involvement in Black activities because I don’t

feel comfortable in White environments. 1 2 3 4
4. I feel an overwhelming attachment to Black people. 1 2 3 4
5. I involve myself in causes that will help all oppressed people. 1 2 3 4
6. People’s race do not influence how comfortable I feel when I am with 1 2 3 4
them.
1 2 3 4
7. I feel good about being Black, but do not limit myself to Black
activities.
1 2 3 4
8. I often find myself referring to Whites as “honkies,” “crackers,”
“devils,” ect. 1 2 3 4

9. I believe certain aspects of “the Black experience” apply to me, and 1 2 3 4
others do not.

10. I frequently confront the system and the (White) man.

 

1 2 3 4
11. I constantly involve myself in Black political and social activities
(such as art shows, political meetings, Black theatre, and so forth).
1 2 3 4
12. I involve myself in social and political groups even if there are no
other Blacks involved.
. 1 2 3 4
13. I believe that the world should be interpreted from a Black or
Afrocentric perspective. 1 2 3 4
14. I am changing my style of life to fit my beliefs about Black people. 1 2 3 4
15. I feel excitement and joy in Black surroundings. 1 2 3 4

 

48

 

 

 

l6.

l7.

l8.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

I am determined to find my Black identity.
People, regardless of their race, have strengths and limitations.

I speak my mind regardless of the consequences (e.g. being kicked out
of school, disappointing my parents, being exposed to danger).

I limit myself to Black activities as much as I can.

I believe that I have many strengths because I am Black.

I believe that White people should feel guilty about the way they have
treated Blacks in the past.

White people can’t be trusted.
The most important thing about me is that I am Black.
Being black just feels natural to me.

Black people who have any White people’s blood should feel ashamed
of it.

A person’s race usually is not important to me.

A person’s race has little to do with whether or not he/she is a good
person.

I believe that a Black person can be close friends with a White person.
I am satisfied with myself.

I have a positive attitude about myself because I am Black.

 

49

 

 

Appendix B
Demogpaphic Qpestionnaire

Name (last, first)
Sex: M F
MSU PID A
Date of Birth Age

 

 

 

Local Address

 

 

Local Phone

MSU e-mail address
secondary e-mail address
(please place a star by the account you would like us to use first)

 

 

 

Home Address

 

 

Home Phone

 

Mother’s Name
Occupation

 

 

Father’s Name
Occupation

 

 

What is the total income of your family household per year?

 

 

 

below $19,999 $120,000 to $139,999
$20,000 to $39,999 $140,000 to $159,999
$40,000 to $59,999 $160,000 to $179,999
$60,000 to $79,999 $180,000 to $199,999
$80,000 to $99,999 $200,000 to $219,999
$100,000 to $119,999 $220,000 and above

High School

Address

What was your high school GPA?

Please estimate the racial composition of your high school to the best of your ability (using

percentages)

White % Asian % Other %

Black % Latino % A

How long (in years or months) were you at this school?

 

50

REFERENCES

Adams, G. R. & Fitch, S. A. (1982). Ego stage and identity status development: A cross-
sequential analysis. Journal of Personality and Social Psychology, 43(3),
574-583.

Allen, W. R. (985). Black student, White campus: Structural, interpersonal, and
psychological correlates of success. Journal of Negro Education, 54(2),
134 — 147.

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck Depression Inventory manual
(2“1 ed.). San Antonio, TX: Psychological Corporation.

Brady, E. U. & Kendall, P. C. (1992). Comorbidity of anxiety in children and
adolescents. Psychological Bulletin, 111(2), 244 — 255.

Brown, M. T. & Duran, P. S. (1988). Construct validity for Blacks of the State-Trait
Anxiety Inventory. Measurement and Evaluation in Counseling and
Development, 21(1), 25 — 33.

Brown, D. R., Ahmed, F ., Gary, L. E., & Milburn, N. G. (1995). Major depression in a
community sample of African Americans. American Journal of Psychiatry,
152(3), 373-378.

Carter, R. T. (1991). Racial identity attitudes and psychological functioning. Journal of
Multicultural Counseling and Development, 19, 105 — 114.

Carter, R. T., DeSole, L., Sicalides, E. 1., Glass, K., & Tyler, F. B. (1997). Black racial
identity and psychosocial competence: A preliminary study. Journal of Black
Psychology, 23(1), 58 — 73.

Chambers, J .W., Kambon, K., Birdsong, B.D., Brown, J ., Dixon, P., & Robbins-Brinson,
L. (1998). Afi'ocentric Cultural Identity and the Stress Experience of African
American College Students. Journal of Black Psychology, 24(3), 368-396.

Cross, H. J. & Allen, J. G. (1970). Ego identity status, adjustment, and academic
achievement. Journal of Consulting and Clinical Psychology, 34(2), 288.

Cross, W. E. (1991). Shades of black: diversity in African-American identity.
Philadelphia, PA: Temple University Press.

D’Augelli, A. R. & Hershberger, S. L. (1993). African American undergraduates on a
predominantly White campus: Academic factors, social networks, and campus
climate. Journal of Negro Education, 62(1), 67 — 81.

Dozois, D. J., Dobson, K. S., & Ahnberg, J. L. (1998). A psychometric evaluation of the
Beck Depression Inventory-II. Psychological Assessments, 10(2), 83 — 89.

51

Erikson, E. (1968). Identity: Youth and crisis. New York: W. W. Norton & Company.

Graham, C., Baker, R. W., & Wapner, S. (1985). Prior interracial experience and Black
student transition into predominantly White colleges. Journal of Personality and
Social Psychology, 47 (5), 1146 — 1154.

Gray-Little, B. & Carels, R. (1997). The effect of racial dissonance on academic
achievement in elementary, junior high, and high school students. Journal of
Research on Adolescence, 7, 109 — 131.

Helms, J. E. (1990). Black and White racial identity: Theory, research, and practice.
Westport, CT: Greenwood Press.

Helms, J. E. & Parham, T. A. (1996). The racial identity attitudes scale. In R. L. Jones
(Ed.) Handbook of Test and Measurements for Black Populations Volume 2.
Hampton, VA: Cobb & Henry.

Hershberger, S. L. & D’Augelli, A. R. (1992). The relationship of academic performance
and social support to graduation among African-American and White university
students: A path-analytic model. Journal of Community Psychology, 2, 188 — 199.

Honess, T. & Yardly, K. (1987). Self and identity: Perspectives across the lifespan.
London: Routledge & Kegan Paul.

Hu, L. Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to
underparameterized model misspecification. Psychological Methods, 3(4),
424 — 453.

Hu, L. Bentler, P. M. (1998). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation
Modeling, 6(1), 1 —55.

Hudley, C. (1995). Assessing the impact of separate schooling for Afiican American
male adolescent. Journal of Early Adolescence, 15, 38 — 57.

James, R. (1998). The perceived effects of social alienation on black students enrolled at
a Caucasian southern university. College Student Journal, 32 (2), 228 - 239.

Jdreskog, K.G., & Sdrbom, D. (1993). LISREL®8 User ’s Reference Guide. Chicago:
Scientific Software International.

Kashani, J. H. & Priesmeyer, M. (1983). Differences in depressive symptoms and

depression among college students. American Journal of Psychiatry, 140 (8),
1081 - 1082.

52

Lemon, R. L. & Waehler, C. A. (1996). A test of stability and construct validity of the
Black Racial Identity Scale, F orm B (RIAS-B) and the White Racial Identity
Attitude Scale (WRIAS). Measurement and Evaluation in Counseling and
Development, 29(2), 77 — 85.

Marcia, J .E. (1966). Development and validation of ego-identity status. Journal of
Personality & Social Psychology, 3(5), 551-558.

Marcia, J. E. & Friedman, M. L. (1970). Ego identity status in college women. Journal of
Personality, 38(2), 249 — 263.

Marron, J. A. & Kayson, W. A. (1984). Effects of living status, gender, and year in
college on college students' self-esteem and life-change experiences.
Psychological Reports, 55 (3), 811 — 814.

Meeus, W. (1996). Studies in identity deve10pment in adolescents: An overview of
research and some new data. Journal of Youth and Adolescence, 25(5), 569 — 598.

Meeus, W., Iedema, J ., Helsen, M., & Vollebergh, W. (1999). Patterns of adolescent
identity development: Review of literature and longitudinal analysis.
Developmental Review, 19(4), 419—461 .

Munford, M. B. (1994). Relationship of gender, self-esteem, social class, and racial
identity to depression in Blacks. Journal of Black Psychology, 20(2), 157 — 174.

Novy, D. M., Nelson, D. V., Goodwin, J ., & Rowzee, R. D. (1993). Psychometric
comparability of the State-Trait Anxiety Inventory for different ethnic
subpopulations. Psychological Assessments, 5(3), 343 — 349.

Parham, TA. (1989). Cycles of Psychological Nigrescence. The Counseling
Psychologist, 17, 187-226.

Parham, T. A. & Helms, J. E. (1985). Relation of racial identity to self-actualization and
affective states of Black students. Journal of Counseling Psychology, 32 (3),
431 - 440.

Phinney, J. S. (1989). Stages of ethnic identity development in minority group
adolescents. Journal of Early Adolescence, 9(1-2), 34—49.

Phinney, J. S. (1990). Ethnic identity in adolescents and adults: Review of research.
Psychological Bulletin, 108(3), 499 — 514.

Phinney, J. S. (1996). When we talk about American ethnic groups, what do we mean?
American Psychologist, 51(9), 918-927.

53

Ponterotto, J. G. & Wise, S. L. (1987). Construct validity study of the Racial Identity
Attitude Scale. Journal of Counseling Psychology, 34(2), 218 — 223.

Pyant, C. T. & Yancio, B. J. (1991). Relationship of racial identity and gender-role
attitudes to Black women’s psychological well-being. Journal of Counseling
Psychology, 38(3), 315 — 322.

Saylor, E. S. & Aries, E. (1999). Ethnic identity and change in social context. The
Journal of Social Psychology, 139(5), 549 — 566.

Sellers, R. M., Chavous, T. M., & Cooke, D. Y. (1998). Racial identity and racial
centrality as predictors of African American college students’ academic
performance. Journal of Black Psychology, 24 (1), 8 — 27.

Sellers, R. M., Smith, M. A., Shelton, J. N., Rowley, S. A., Chavous, T. M. (1998).
Multidimensional model of racial identity: A reconceptualization of Afiican

American racial identity. Personality and Social Psychology Review, 2(1),
18 -— 39.

Siegel, J. M., Yancey, A. K & McCarthy, W. J. (2000). Overweight Symptoms among
African-American Women. Preventive Medicine. 31, 232-240.

Spielberger, C. D., Gorsuch, R. L, & Lushene, R. E. (1970). Manual for the State-Trait
Anxiety Inventory (Self-Evaluation Questionnaire). Palo Alto, CA: Consulting
Psychologists Press.

Spielberger, C. D. (1983). State-Trait Anxiety Inventory. Redwood City, CA: Mind
Garden.

Steer, R. A. & Clark, D. A. (1997). Psychometric characteristics of the Beck Depression
Inventory-II with college students. Measurements and Evaluation in Counseling
and Development, 30(3), 128 - 136.

Steinberg, L. (1999). Adolescence, 5‘h Edition. Boston: McGraw-Hill.

Taub, D. J. & McEwen, M. K. (1992). The relationship of racial identity attitudes to
autonomy and mature interpersonal relationships in Black and White
undergraduate women. Journal of College Student Development, 33 (5),

439 — 446.

Waterman, A. S. (1992). Identity as an aspect of optimal functioning. In G. R. Adams, T.
P. Gullotta & R. Montemayor (Eds.) Adolescent Identity Formation. Newbury
Park, CA: SAGE.

Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience
aversive emotional states. Psychological Bulletin, 96, 465 - 490.

54

Watson, D., Clark, L. A., & Carey, G. (1988). Positive and negative affectivity and their
relation to anxiety and depressive disorders. Journal of Abnormal Psychology, 97,
346 — 353.

West, S.G., Finch, J. F., Curran, P. J. (1995). Structural equation models with nonnormal
variables: Problems and remedies. In: R. H. Hoyle (Ed), Structural equation
modeling: Concepts, issues, and applications (pp. 56-75). Thousand Oaks, CA:
Sage Publications.

Whatley, S. & Clopton, J. (1992). Social support and suicidal ideation in college students.
Psychological Reports, 71 (3/2), 1123 — 1128.

Whisman, M. A., Perez, J. E., & Ramel, W. (2000). Factor structure of the Beck
Depression Inventory-II in a student sample. Journal of Clinical Psychology,
56(4), 545 — 551.

Yancio, B. J ., Swanson, J. L., & Tokar, D. M. (1994). A psychometric investigation of

the Black Racial Identity Attitude Scale — Form B. Journal of Vocational
Behavior, 44(2), 218 - 234.

55

 

llllllllllglllrlllillll11111111

 

// .A-._—-——
_l -_-‘.- m