THE RELATIONSHIP BETWEEN MULTIRACIAL IDENTITY VARIANCE, SOCIAL
CONNECTEDNESS, FACILITATIVE SUPPORT, AND ADJUSTMENT IN
MULTIRACIAL COLLEGE STUDENTS
by
JAMES L. LYDA
A DISSERTATION
Presented to the Department of Counseling Psychology
and Human Services
and the Graduate School of the University of Oregon
in partial fulfillment of the requirements
for the degree of
Doctor of Philosophy
June 2008
11
University of Oregon Graduate School
Confirmation of Approval and Acceptance of Dissertation prepared by:
James Lyda
Title:
"The Relationship Between Multiracial Identity Variance, Social Connectedness, Facilitative
Support, and Adjustment in Multiracial College Students. "
This dissertation has been accepted and approved in partial fulfillment of the requirements for
the Doctor of Philosophy degree in the Department of Counseling Psychology and Human
Services by:
Benedict McWhirter, Chairperson, Counseling Psychology and Human Services
Krista Chronister, Member, Counseling Psychology and Human Services
Karyn Angell, Member, Counseling Psychology and Human Services
Naomi Zack, Outside Member, Philosophy
and Richard Linton, Vice President for Research and Graduate Studies/Dean of the Graduate
School for the University of Oregon.
June 14,2008
Original approv~l signatures are on file with the Graduate School and the University of Oregon
Libraries.
© 2008 JAMES L. LYDA
111
James L. Lyda
An Abstract of the Dissertation of
for the degree of
IV
Doctor of Philosophy
in the Department of Counseling Psychology and Human Services
to be taken June 2008
Title: THE RELATIONSHIP BETWEEN MULTIRACIAL IDENTITY VARIANCE,
SOCIAL CONNECTEDNESS, FACILITATIVE SUPPORT, AND
ADJUSTMENT IN MULTIRACIAL COLLEGE STUDENTS
Approved: _
Dr. Benedict T. McWhirter
Research has suggested that multiracial individuals may vary in how they racially
identify depending on the context in which they operate (Renn, 2004; Root, 1998, 2003).
To examine this assertion, multiracial identity and variance in multiracial identity were
examined in this exploratory study of a nationally representative sample of 199
multiracial college students. Additionally, the relationship ofmultiracial identity variance
with factors common to adult transitional development and to the college student
experience, including social connectedness, various forms of facilitative support, college
vadjustment, and depression, were also examined in this study. Sex differences among
these study variables were also explored.
The results of descriptive analyses revealed that this generally connected,
adjusted, and non-depressed sample consistently varied their racial identity depending on
their context. Results of Pearson product-moment correlations among study variables for
the whole sample demonstrated that this multiracial identity variance was not related to
adjustment, social connectedness, facilitative supports, or depression. But results differed
when breaking down the sample by sex. For males, increased variance in multiracial
identity across contexts was related to lower perceived availability of, support from, and
connectedness to student support groups. For females, increased multiracial identity
variance was related to lower participation in ethnic and cultural student support groups.
A series of subsequent simultaneous multiple regression analyses revealed that increased
involvement in one form of facilitative support in the college environment-
ethnic/cultural student support groups- actually predicted lower multiracial identity
variance for the sample.
Regarding connectedness, for the entire sample, higher social connectedness was
related to higher college adjustment but lower participation in ethnic and cultural student
support groups. Sex differences also emerged for connectedness. For males, social
connectedness was directly related to availability of student groups, adjustment, and
institutional attachment, and for females social connectedness was directly related to
college adjustment, but inversely related to participation in ethnic/cultural groups.
VI
Findings ofthis study are consistent with multiracial identity theory, social
connectedness theory, and with research related to college student and adult transitional
development, and confirm that multiracial individuals vary their identity based on social
context. Implications for future research and intervention are discussed.
vii
CURRICULUM VITAE
NAME OF AUTHOR: James L. Lyda
PLACE OF BIRTH: Iowa City, Iowa
DATE OF BIRTH: June 9, 1979
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon, Eugene, OR
Northwestern University, Evanston, IL
DEGREES AWARDED:
Doctor of Philosophy in Counseling Psychology, 2008, University of Oregon
Master of Science in Counseling Psychology Family and Human Services, 2004
University of Oregon
Bachelor of Arts in Psychology, 2001, Northwestern University
AREAS OF SPECIAL INTEREST:
Multiracial identity; multicultural counseling competence; late adolescent! young
adult development and mental health; college student development;
college counseling
PROFESSIONAL EXPERIENCE:
Post-Doctoral Fellow, Counseling and Psychological Services, University of
California- Berkeley, Berkeley, CA 2007-08
Pre-doctoral Intern, University Counseling and Consulting Services, University of
Minnesota, Minneapolis, MN 2006-07
Graduate Teaching Fellow, Counselor/ Outreach Specialist, University
Counseling & Testing Center University of Oregon, Eugene, OR 2004-06
Therapist, Volunteers in Medicine Clinic, Eugene, OR 2004-06
V111
Individual Therapist, Oregon Social Learning Center, Eugene, OR, 2002-04
Practicum Intern, University of Oregon Counseling and Testing Center, Eugene,
OR 2003-04
Practicum Intern, University of Oregon Child and Family Center, Eugene, OR
2002-03
Crisis Line Counselor University of Oregon Crisis Line, Eugene, OR 2002-03
Intern, Youth Resource Foundation, Inc. Warming House Youth Center,
Willamette, IL, 2000- 01
GRANTS, AWARDS AND HONORS:
American Psychological Association Minority Fellowship Program, Mental
Health and Substance Abuse Services Fellowship 2003-06
University of Oregon Center on Diversity and Community (CoDaC), Graduate
Summer Research Award, Summer 2005
ix
ACKNOWLEDGMENTS
This dissertation could not have been completed without the love, support,
patience and sacrifice of my partner, Angela Keller. I love you and thank you for
everything.
To my mother, Debra Lyda, thank you for nurturing my education, for being a
Supermom throughout my upbringing, and for all of your words of wisdom throughout
the years. To my father, Dr. Jack Lyda Sr., thank you for blazing the trail and setting the
bar high for me to strive for in my education. To my sister Elizabeth, thank you for being
the only person in the world with whom I share my unique biracial heritage and for being
my first friend. And to my brother Jack, Jr., thanks for being like a second father to me as
a child and a role model as a father to my nieces, Haley and Taylor.
To my grandparents, Leonard and Leona House, thank you for your love and
SUppOlt throughout the years. Your value for my education has been a driving force, and I
would not have made it this far without you. To Naomi Lyda, may you continue to rest
in peace and may your grace live on in me.
I would like to thank my dissertation Chair, Dr. Benedict McWhirter, for his
guidance, patience, and example. In addition I am thankful for the commitment,
dedication, and feedback of my dissertation committee members, Dr. Krista Chronister,
xDr. Karyn Angell, and Dr. Naomi Zack. Also, thank you to John Petrocelli for your
helpful consultation.
Thank you to my grad school friends and colleagues who made graduate school
bearable: Shari Lum, May Lim, Allison Lau, Keith Miller, Heather Marshall, Sarah
Wilson, Dave Miller, Amit Shahane, Gerald Gonzalez, Michelle Burt, Stephanie Shepard,
and Colleen Comeau.
Special and heartfelt thank you to Karen Paez, for being the big sister I never had.
Your support was invaluable, and your friendship and kindness (literally) nourished my
soul. And, George Hanawahine, thank you for your mentorship, brotherhood, and thank
you to you family for showing me the meaning of 'ohana.
Finally to my dear friends from internship, Denise Wyttenbach and Monica
Bigler, two strong women who are the embodiment of perseverance and grace.
This investigation was supported in part by APA Minority Fellowship Program
Mental Health Substance Abuse Fellowship (Grant # 3T06SM56564), as well as a grant
from the University of Oregon Center on Diversity and Community (CoDAC).
Finally, thank you to my participants and all those who assisted me in their
recruitment.
Dedicated to Leonard and Leona House,
who made my education possible.
Xl
XlI
TABLE OF CONTENTS
Chapter Page
I. RATIONALE 1
Historical, Political, and Social Implications of Mixed Race Identity.......... 2
Racial and Ethnic Identity 3
Multiracial Identity Models 5
Monoracial Identity Development Models 5
Biracial and Multiethnic Identity Development Models 6
Ecological Models of Multiracial Identity Development 7
Wardel and Cruz-jansen's Model..................................................... 8
Root's Model 11
Multiracial Identity Variance 13
Social Connectedness 17
Social Connectedness and Multiracial Identity: Influence of Sex 19
Facilitative Support 20
College Adjustment 21
Depression..................................................................................................... 22
Purpose of This Study 23
Research Questions........................................................................................ 24
II. METHODOLOGY 27
Participants 27
Measures 32
Demographics 33
Multiracial Identity Variance 33
Social Connectedness 35
Facilitative Supports 37
College Adjustment 38
Chapter
Depression .
Procedures .
Pmticipant Recruitment ..
Data Collection ..
Sample Size ..
III. RESULTS .
Overview .
Preliminary Analyses .
Descriptive Analyses .
Multiracial Identity Variance .
Sex Differences .
Social Connectedness .
Sex Differences .
Perceptions of Facilitative Student Supports .
Sex Differences .
College Adjustment .
Sex Differences .
Depression .
Sex Differences .
Correlation Analyses .
Multiracial Identity Variance .
Social Connectedness .
Facilitative Supports .
College Adjustment .
Depression .
Regression Analyses .
Explaining Multiracial Identity Variance ..
Explaining Social Connectedness ..
Explaining Depression .
Summary of Results .
Xlll
Page
40
41
41
43
44
45
45
45
48
48
49
49
49
49
50
51
51
51
51
55
56
56
57
57
57
61
61
62
62
65
xiv
Chapter Page
IV. DISCUSSION 68
Main Findings: Relationship Among Variables 68
Demographics , 68
Multiracial Identity Variance 71
Social Connectedness 75
Facilitative Supports 77
College Adjustment 78
Depression 79
Sex Differences...... 80
Implications of the Findings 82
Study Limitations 87
Future Research and Intervention............................................................. 89
Conclusion 90
APPENDICES 91
A. REVIEW OF THE LITERATURE 91
B. INFORMED CONSENT STATEMENT 109
C. MEASUREMENT INSTRUMENTS 112
REFERENCES 127
xv
LIST OF FIGURES
Figure Page
1. Ecological Framework for Understanding Racial Identity........................... 12
2. Multiracial Identity Options Frequencies 53
XVI
LIST OF TABLES
Table Page
1. Participant Demographics 28
2. Multiracial Demographics 29
3. Combined Multiracial Demographics 31
4. Constructs and their Corresponding Measures 32
5. Alpha Reliability and Number of Items for Instrument Scales 46
6. Percentages of Multiracial Identification by Context 52
7. Descriptive Statistics of all Study Variables by Sex and One-Way
ANOVA Results 54
8. SACQ Mean Scale Scores and T-Scores 55
9. Sample Means, Standard Deviations and Intercorrelations of all
Measured Variables 59
10. Intercorrelations of all Measured Variables by Sex 60
11. Simultaneous Regression Results Summary of Multiracial Identity
Variance onto Sex and Four Components of Facilitative Support
Variables 63
12. Simultaneous Regression Results Summary of Social Connectedness
Regressed onto Sex, Facilitative Support Variables and Student
Adaptation to College Variables 64
13. Simultaneous Regression Results Summary of Depression Regressed
onto Sex, Support and Connectedness from Student Groups, and
Student adaptation to College Variables 65
1CHAPTER I
RATIONALE
The "biracial l baby boom" is upon us (Root, 1992). Changes in immigration laws
in 1965 that allowed for an increase in immigration particularly from Asia, and the
abolition of the last anti-miscegenation laws in 1967 (Loving vs. State of Virginia) have
yielded the emergence of a new generation of multiracial individuals with increased
possibilities, range, and flexibility to identify their racial and ethnic identity (Root, 1998,
2003a). The 2000 U.S. Census marked the first time in history that respondents were given
the option to check off multiple categories of the race question, resulting in approximately
6.8 million individuals in the U.S. identifying as two or more races (U.S. Census Bureau,
2000). The number of mixed-race births is increasing at a faster rate than the number of
1 In this proposal I use the terms biracial, multiracial and mixed to represent people whose
acknowledged identity includes two or more of the U.S. government Census racial and ethnic
categories. These categories are White (non-Latino), Latino/Spanish/Hispanic, Black!African
American (not Hispanic), American Indian or Alaska Native, and Asian Indian, Chinese, Filipino,
Japanese, Korean, , Vietnamese, Native Hawaiian, Guamanian or Chomorro, Samoan, Other
Asian, Other Pacific Islander, and some other race (US Bureau of the Census, 2000). While the
terms multiracial and multiethnic are not synonymous, they are often used (or misused)
interchangeably in the literature. Race is a socially constructed way of grouping people, which
differs from society to society and over time. It has no biological basis, yet has been historically
borrowed from the systematic classification of plants and animals to give this concept scientific
validity (Root, 1996). Therefore, multiracial (biracial) refers to an individual of two or more
racial heritages; the offspring of parents with two or more racial designations. Ethnicity refers to
self-identified membership in a group because of shared emotions, attitudes, and identification
with values and history (Root & Kelley, 2003). It represents a shared cultural worldview and/or
geographic origin. Ethnicity is a social identity that is not passed on by genetic lines, but rather
the teachings of a family or community. Consequently, multiethnic means relating to or including
several ethnic groups.
2single-race births (U.S. Census Bureau of the Census, 1992). The impact is being felt all
over the U.S. where in cities like Seattle, Sacramento, and San Antonio one in every six
newborns is multiracial (Kelley & Root, 2003). The multiracial population is young,
diverse and rapidly growing (Jones & Smith, 2003). According to Jones and Smith (2003),
nearly 42 percent of the population is under the age of 18, and 68 percent are under 35.
Undoubtedly the college and university campuses of the United States are already
reflecting the upsurge in this population. In 2000, the US Census reported approximately
500,000 multiracial students enrolled in colleges or universities.
Historical, Political, and Social Implications of Mixed Race Identity
Zack (2006) describes mixed-race American's as instances of microdiversity, in
that their racial diversity exits on the individual level, unlike the diversity that is
acknowledged to exist between or among the four major racial groups (Asian, Black,
White, Native AmericanlIndian). According to Zack (2001):
The right to distinctive race-based group cultural identities has come to be generally
accepted as a necessary component of democratic pluralism, and it is usually
referred to as "multiculturalism." In ordinary usage, except when the term
"ethnicity" is used as a synonym for "race," the cultural aspect of group identity is
called "ethnic" if the group in question is racially white (e.g. Italian, German,
Jewish ethnic identities as developed in the United States). And "racial" if the group
is African American, Asian American, or American Indian (e.g. black, Asian, and
Indian racial identities). ( p. 39)
Accordingly, we have developed a labeling system for racial and ethnic identification that
3differentiates race and ethnicity based on majority/minority lines.
There is a difference between mixed race people who are aware of their mixed race
heritage from grandparents and great-grandparents who were not of the same race, and
mixed race people who are only aware of their parents differing racially. The latter, first-
generation mixed race individuals are more likely to more acutely experience the
tribulations of being mixed in a monoracial system. This is embodied in the differences
between the experience of mixed race individuals in the continental U.S. versus those in
Hawaii, where there is an historical and cultural tradition of public recognition of mixed
race/ethnicity, as well as more social acceptance of mixed race individuals.
For African Americans in particular, the significance of the "one-drop rule" has had
significant implications on the historical, political and social implications of mixed race
identity. The "one drop rule," otherwise known as racial assignment by hypodecent, is an
historical colloquial term in the U.S. that states to be considered Black, one only needs to
have one drop of Black blood in one's ancestry (M. Harris, 1964; Omi & Winant, 1986; as
cited in Shih & Sanchez, 2005).
Racial and Ethnic Identity
Race and ethnicity are both primary sources of identity and significant bases for
polarization and stratification of groups. According to Erikson (1963), establishing a stable
sense of personal identity in U.S. society is an integral aspect of the developmental process.
Accordingly, for the traditional aged college student, the college years represent a critical
and formative transition from adolescence to adulthood (Chickering, 1969). Racial identity
is a psychological construct reflecting aspects of our membership in, and identification
4with, a specific racial group (Cuellar, Nyberg, Maldonado, & Roberts, 1997 as cited in
Nishimura, 1998). According to Phinney (1992), racial identity development peaks during
late adolescence and early adulthood making the traditional college years a critical period.
Internal conflict associated with mixed racial identity and the ways in which it
contextualizes experiences in college settings are often overlooked or ignored (Nishimura,
1998; Wallace, 2003).
As multicultural awareness on campuses increases, mixed race identity is being
acknowledged more freely now than ever before (Hart-Web, 1999). Consequently, college
campuses will be confronted with serving this newly emerging population. Yet, despite
proliferation in the numbers of multiethnic students, not much is understood about their
development and interactions in the college context (Renn, 2000).
In the ongoing battle over access, equity, and affirmative action policy in higher
education, ethnic and racial demographics matter. As the numbers of multiracial births
increase, the need for addressing the psychosocial development of multiracial individuals is
critical (Root, 1996). Yet little is done to address the unique needs of this population
(Benedetto & Olisky, 2001; Herring, 1992). Specific concerns regarding ecological factors
contributing to the psychosocial and identity development of multiracial individuals have
demonstrated the need to better understand this rapidly increasing population (Herring,
1992; McRoy & Freeman, 1986; Sebring, 1985). Literature on the psychosocial experience
of multiracial individuals expounds a number of psychosocial stressors. Examples include:
marginalization (Root, 1990); difficulty finding others who will accept them as one of
"their own" (Herring, 1992); ethnic identity conflicts (Gibbs & Moskowitz-Sweet, 1991;
5Overmier, 1990); oppression from multiple ethnicities/races of heritage (Root, 1990);
exclusion (Paez & Lyda, 2002); pressure to choose one aspect of their heritage over another
(Paez & Lyda, 2002); and dating/relational difficulties (Root & Kelley, 2003). Experienced
separately or in combination, these stressors are a threat to the healthy psychological
development and adjustment of multiracial young people. Certainly, being born multiracial
in the United States does not summarily predestine an individual to a life of hardship. A
multiracial background does not only possess challenges, but may also provide resources
that contribute to resilience in meeting these challenges. For example, researchers have
suggested multiracial individuals (a) may have an increased ability to move between racial
groups (Basu, 2007), (b) may experience acceptance by monoracial peers who respect
multiracial backgrounds (Renn, 2000), and (c) have the potential for access and support
from a variety of cultural communities (Shih, & Sanchez, 2005). Even so, it is important to
consider that certain contextual influences shape the developmental experience of this
population.
Multiracial Identity Models
Monoracial Identity Development Models
During the past three decades, several models of monoracial identity formation
have been established and applied to college students in the U.S. (e.g. Atkinson & Sue,
1993; Cross, 1991, 1995; Helms, 1990, 1995). These models generally rely on a
progression from conformity with majority (white) culture through stages (or "statuses") of
dissonance and resistance to an immersion in a monoracial/monoethnic culture, ending by
integrating racial/ethnic identity with other aspects of the person' s self-definition. Each of
6these models attempts to uncover the psychosocial processes that different racial/ethnic
groups experience as they interface with other racial/ethnic groups and cultures, and each
has served as an important stepping-stone to the current literature on the development of
multiracial identity. Nonetheless, according to Root (2003a):
Contemporary thinking on racial identity development is derived from the
intersection of psychology and the racial pride movements of the 1960s and 1970s.
The racial pride movements catalyzed solidarity within race and further reinforced
notions that one must identify with a single race ... The history of race in the United
States and specifically the "one drop rule" ... provides the key to understanding
how psychological models evolved in a way that has excluded the reality of many
mixed race persons. (p. 34)
Monoracial identity models are unable to accommodate the full complexity, non-linearity,
and heterogeneity of the multiracial identity processes (Root, 2003a). In light of the
historical foundations behind racial and ethnic identity models, a paradigm shift in the way
that we construct racial identity is necessary to account for the unique experience of
multiracial individuals, as well as the contextual factors that influence their experience.
Biracial and Multiethnic Identity Development Models
The majority of research on the multiracial population in the United States has
focused on children with mixed black and white heritage. Poston (1990), Jacobs (1992),
and Kerwin and Ponterotto (1995) have each developed models of biracial identity
development in children of black and white parentage (see Appendix A for detailed review
of these articles). This presents some difficulties with generalizability to all multiracial
7individuals. According to Wardle and Cruz-Janzen (2004), there are two reasons for this
lack of external validity: (a) there is simply not enough research on other multiracial groups
to form any body of knowledge, and (b) research on the identity development of single race
children suggests different patterns for children whose physical features are more
ambiguous with regard to racial and ethnic background (see Aboud, 1987 in Appendix A).
These models provided a valuable foundation for modeling multiracial identity, but in and
of themselves are not adequate for understanding the diverse multiracial population.
Phinney's multiethnic identity model (1993) applies to single-race minority,
multiracial, or multiethnic adolescents and has three distinct stages through which the
adolescent progresses (See Appendix A for more detailed review). While this model
attempts to capture the common experience in identity development of all people of
color, it does not fully account for the contextual influences that contribute to identity
formation, and adheres to a more linear stage model. For multiethnic individuals,
confidently choosing and accepting all aspect of an ethnic identity is more complex if for
no other reason than they must choose from a broader selection of racial and ethnic
possibilities. As mentioned above, these choices may also vary depending on the
individual's context.
Ecological Models ofMultiracial Identity Development
Because racial identity is constructed in the context of social relationships (Renn,
2004), and in the ongoing interactions between individuals and their environments, a
theoretical model that encompasses those interactions is useful in exploring racial identities
and multiracial identification. The ecological model, first introduced by Bronfenbrenner
8(1977, 1979, 1989, 1993), represents such an alternative to stage-based models of identity
development. The ecological model accounts for interactions among and between the
various subenvironments an individual encounters while providing a means to examine the
dynamic, fluid nature of college life. A life where students move from one setting to
another constructing and reconstructing identities both in interpersonal relationships, and in
reaction to the messages they receive from interacting environments.
The Ecological Model of Human Development articulated by Bronfenbrenner
(1977, 1979, 1989, 1993) accounts for both individual factors and the individual's
interactions within his or her dynamic environment as accounting for development. In this
model, human development does not occur in isolation, but within multiple, embedded
ecological systems (Bronfenbrenner, 1979, 1989). There are three assumptions of the
ecological model: (a) the individual and his or her environment are continually interacting
and exerting mutual influence-and as a result constantly changing, (b) the individual is an
active patticipant in his or her development, and (c) changes in one ecological system may
influence changes in systems that are more proximal and distal to the individual (i. e.
bidirectionality).
Over the past two decades two multiracial identity development models have
emerged that reflect constructs of Bronfenbrenner's ecological model (e.g Root, 2003;
Wardle & Cruz-Janzen, 2004). As such, I describe both models in the sections that
follow.
Wardle and Cruz-jansen's model. In Wardle's developmental model (Wardle
1992; Wardle & Cruz-Janzen, 2003), the central task of identity development for the
9multiracial child is the achievement of healthy multiracial identity. This is marked by
successful completion of the three developmental stages incorporated into the model, as
well as the child's interaction with the components that make up the ecological model:
family, group antagonism, minority/lower status context, majority/ higher status context,
and community.
The three developmental stages described by Wardle and Cruz-Jansen's model
(Wardle 1992; Wardle & Cruz-Janzen, 2004) are Early childhood (3 to 7 years),
Transition Period (6 to 12 years), and Adolescence. Each of these stages incorporates the
developmental theories of both Piaget, and Erickson. According to this model in the
Early Childhood stage, the child becomes aware of physical features, and the similarities
and differences between themselves and their parents and peers. At this stage they are
often asked to explain their physical and racial ambiguity and they need a label to
proactively do so. The Transition Period is marked by the multiracial child's increased
awareness of sexuality. At this age group belonging is important, and parallels an interest
in competencies and the concept of race. The Adolescent stage incorporates Erikson's
(1963) identity crisis. At this stage multiracial youth learn to separate out race, ethnicity,
abilities, likes, dislikes and career choices. Their family and school may increasingly
support non-race specific groups, and they become more comfortable with multiracial
identity.
The interactional components of the ecological model described by Wardle and
Cruz-Jansen's model (Wardle 1992; Wardle & Cruz-Janzen, 2004) determine the success
or failure of the healthy multiracial or multiethnic identity process. Family includes
10
biological, adoptive, foster, teen, extended, and blended families on all sides. The
family's impact on the child's multiracial identity depends on the attitude of the family
towards such an identity, discussion of the topics such as the child's identity and racism,
as well as the means by which the family supports the child's overall identity
development.
Wardle and Cruz-Jansen's (Wardle 1992; Wardle & Cruz-Janzen, 2004) explains
that multiracial children must deal with two kinds of group antagonism: the traditional
and institutionalized racism experienced by all people of color in the United States, and
the antagonism of all single-race and ethnic groups toward people of mixed heritage. The
level of racism multiracial children experience largely depends on phenotype, community
context, and school environment.
Wardle and Cruz-Jansen's model (Wardle 1992; Wardle & Cruz-Janzen, 2004)
breaks down cultural contexts into two separate categories, minority/ lower status
context, and majority/ higher status context. Wardle and Cruz-Janzen (2004) explain:
Multi[racial] children with some mainstream White heritage have a distinctive
cultural context that includes a minority and majority status; children whose
parents are from two minority groups also have a lower status and higher status
context based on the position of their parents' race and ethnicity within the strict
racial hierarchy. (p.123)
The community represents the most important ecological impact on a child after
the family (Bronfenbrenner, 1989), and according to Wardle and Cruz-Jansen's model
(Wardle 1992; Wardle & Cruz-Janzen, 2004) includes school, church groups, immediate
11
peers, and neighborhood groups. According to this model there are three factors that play
heavily on the impact of the community on healthy multiracial identity: (a) Families'
feeling of belongingness; (b) Community's acceptance of a range of diversity; and (c)
Presence of minority representation and multiracial/multiethnic children in the
community groups the child attends (2004).
As each child integrates all of his or her experiences within his or her own unique
context, they will more toward or away from healthy multiracial identity. Wardle's
model attempts to convey the developmental and ecological complexity of this process,
yet, still incorporates a stage model that implies a linear racial identity progression.
Root's Model. Informed by Bronfenbrenner, Root's (2000, 2003a) Ecological
Frameworkfor Understanding Racial Identity provides a model for understanding the
complexity of racial identity development, that is both inclusive of multiracial identity,
and non-linear. Based on nearly a decade of research, Root's framework illustrates the
bidirectional effects of common ecological influences (e.g. Family functioning,
racial/ethnic identities, community attitudes, racial socialization, and physical
appearance) and invisible factors (e.g. regional and generational history of race and
ethnic relations, gender, sexual orientation, and class; see figure 1) on multiracial
individuals. Root's (1990, 1997, 2000, 2003) model is grounded in the understanding that
a multiracial person's identity can and will change, depending on their ecological context.
Further Root posits the central task for the multiracial child is to achieve a positive
resolution between a sense of identity and his or her environment (1990, 1997, 2003a).
Figure 1. Ecological Framework for Understanding Racial Identity (adapted from
Root, 2003a).
Background "Invisible" Context
• Regional and generational history of race and ethnic relations
• Gender and sexual orientation
• Class
~
Family Functioning Identities Community
Attitudes &
Family
... Racial
Socialization Ethnicities Socialization
Child!Adolescent's
Traits & Aptitudes Races
~
Physical Appearance or Phenotype
12
13
Root (2003a) identifies five types of identities that emerge from research on
mixed race persons: assignment by hypodescent, monoracial fit/self assignment, blended
identity, bi-or multiracial, and White with symbolic identity. Generational norms often
influence the identity adopted by these individuals. For example, for older generations
single-race identity is often the result of either assuming an assignment according to a
"one-drop rule" and hypodescent. Younger generations now have the option of publicly
assuming a blended or multiracial identity. "Symbolic Whiteness" appears to reflect
identification with a class lifestyle and values or a lack of exposure to an ethnic
background with which one identifies (p. 34).
Root (1990, 1997, 2003a) postulates four possible healthy identity resolutions: (a)
the individual accepts society's definition of his or her race or ethnicity, (b) the individual
identifies with both ethnic or racial groups, (c) the individual identifies with a single
racial/ethnic group, (d) the individual identifies with the mixed-race or multiethnic/
multiracial group that his or her background represents. Each of these resolutions are
positive if the individual is accepted in their chosen group, does not feel pressure to
change, does not deny or put down any aspect of their heritage, and is supported in their
identity by their immediate environment. In the case of the last resolution, the identity is
positive if the new identity embraces all aspects of the person's identity.
Multiracial Identity Variance
The ecological nature of multiracial identity elucidates that there is more than one
way to for mixed-race individuals to identify (hereafter referred to as multiracial identity
variance). The potential mixed race categories enabled by the numerous possible racial
14
combinations under current census standards do not yet exist as recognized groups.
Accordingly, Zack (2006) has identified the following mixed race identity options:
fractional, inclusive, traditional nonwhite, white, generic, and aracial. Fractional
identification occurs when a multiracial person chooses to identify his or herself in rough
fractional terms. To illustrate, take the hypothetical case of Angela who has a White
mother and a father who is Black and Asian and chooses to label herself as one third White,
one-third Black, and one-third Asian (Or she may choose to apply fractions more literally
and identify as one-half white, one-quarter black, and one-quarter Asian). Inclusive
identification occurs when an individual identifies as all or most of hislher racial identities
without choosing to "divide" them into parts. Angela may see herself as Asian and Black
and White. Traditional nonwhite identification takes place when an individual chooses to
identify with a singular, nonwhite aspect of their racial heritage. Angela may choose to
identify as Black, because she grew up in an urban neighborhood in the southeastern U.S.
surrounded by friends and family who all identified as black and accepted her as Black.
Nonwhite identification may be due to family and/or geographical context, generational
factors, pressure from others to chose one racial identification, as well as the extent to
which an individual subscribes to societal norms about racial categorization (one drop rule,
blood quantum, phenotypic characteristics, etc.). White identification is self-explanatory,
but is conceptually complex because of the challenges put forth by societal norms based in
the one-drop rule of racial classification. Suppose Angela's parents divorce when she is
young and she grows up with her mother and her stepfather, who is also White, in a
predominantly White neighborhood. She has no contact with her father and therefore no
15
experience of Black or Asian ancestry from him. Angela is fair skinned, with long straight
hair and can pass as white. Despite Angela's mixed heritage, she chooses to identify as
White. If race has no base in biology, this should be as acceptable as Angela (or anyone
with similar racial heritage) identifying as singularly Black.
Suppose Angela decided that she will identify simply as "mixed" and finds no need
to specify, insisting that it is distinctly American to have the racial heritage that she does. In
this case Angela would be choosing a generic identity option. Suppose Angela decided to
raise her kids believing that they are simply human beings and citizens of the world. Her
children readily adopt this identity, thereby adopting an aracial identity.
Zack's mixed race identity options parallel the identity patterns Renn (2004)
identified in a qualitative study of 54 multiracial college students from a diverse selection
of universities across the U.S. The identity patterns include: Monoracial identity, multiple
monoracial identities, multiracial identity, extraracial identity, and situational identity.
Monoracial identity occurred when students identified with one of their monoracial
heritage groups either some or all of the time. Multiple Monoracial Identities occurred
when students held two or more monoracial identities. Multiracial Identity occurred when
students did not conform to the monoracial categories that exist in the U.S. Instead, they
expressed a new construction of identity such as hapa, biracial, multiracial, mixed etc.
Extraracial Identity occurred when students chose not to identify themselves by any
racial classification and do not accept to the construction of racial identities. Finally
Situational Identity was marked by students' identification with two or more of the four
patterns described above. This was either a conscious, or an unconscious shift based on
16
the situation. For them, identity was fluid and contextual.
The idea of racial identity varying based on context represents an important shift in
the evolution of racial identity models. Whereas previous models have accounted for the
non-linearity of racial identity development (e.g. Cross, 1995; Helms, 1995; Poston, 1990),
they have not acknowledged a paradigm that allows for an individual to carry multiple
racial identities simultaneously in a manner that is adaptive and psychosocially healthy.
Research by Renn (2000, 2004), and the conceptual work of Root (2000, 2003) and Zack
(2006) have suggested that multiracial identity variance exists, is shaped by various
ecological factors, and can substantiate healthy development in multiracial individuals.
Root's ecological model (2000, 2003a) serves as the basis for understanding the overall
development of multiethnic adolescents. In concert with Zack' s (2006) racial identity
options, and Renn's (2004) identity patterns, a language now exists whereby the many
contextual choices that influence racial identification can be articulated and understood.
To date, there is little research exploring the relationships between multiracial
identity development, multiracial identity variance and other important developmental
factors that occur during the same period in young persons' lives. As traditional college age
represents a significant period of development and transition into young adulthood
(Chickering, 1974), constructs that have been shown in recent years to playa critical role in
similar developmental events include: social connectedness (Lee & Robbins, 2000), sex
differences in connectedness (Lee, Keough, & Sexton, 2002), environmental factors (Renn,
2000; Renn 2004; Nishimura, 1998, Wallace 2003), college adjustment factors, and
affective states (depression being the most critical of these; see Shih & Sanchez, 2005).
17
Because of the centrality of these constructs during the same developmental period in
which racial identity evolves, research that focuses on multiracial identity development
would be greatly informed by commensurately focusing on constructs that may contribute
in one way or another to the multiracial identity development process. As such, I discuss
the constructs of social connectedness, sex differences in connectedness, facilitative
supports, college adjustment, and depression in the following sections. For a more detailed
review of the literature, please refer to Appendix A.
Social Connectedness
According to Townsend and McWhirter (2005), connectedness is gaining
increased recognition as an important factor in human development and psychology.
Connectedness has been described as a multidimensional construct reflecting "both the
breadth (quantity) and depth (quantity) of human relationships" (Townsend &
McWhirter, 2005, p. 193). The multidimensionality of connectedness stems from the
evolving definition(s) of the construct, and from inter- and intrapersonal dimensions.
Townsend and McWhirter (2005) give a comprehensive description of the types and
forms of connectedness, that include interpersonal (Newcomb, 1990), social (Timpone,
1998), family (Troll, 1994), school (Karcher, 2001; Neumark-Szainer, Story, French,
&Resnick, 1997), and cultural (Daneshpour, 1998) connectedness, and well as
community (Marton, Hrabowski, & Greif, 1998), affective (Rosen, 1999), emotional
(Phares, 1993), and empowered (Kearney, 1998) connectedness. Among these, the type
of connectedness that has been empirically studied most among college students is social
connectedness.
18
Psychosocial distress often occurs when people feel lonely, isolated, excluded,
marginalized or alienated (McWhirter, 1990). These experiences are also related to low
levels of social connectedness. Lee and Robins (2000) define social connectedness as "an
aspect of the self that reflects subjective awareness of interpersonal closeness with the
social world in toto" (p. 78). This construct has significant implications for adolescent
development, particularly within the multiracial population. In their review of the literature,
Karcher and Lee (2002) indicate that greater connectedness is related to psychological
happiness, increased physical health, and better coping skills among adolescents. And that
lower connectedness is related to more psychological difficulties and poorer physical health
outcomes. Incorporating the social connectedness theory of Lee and Robbins (2000),
Karcher's ecological theory of connectedness (Karcher, 2001 as cited in Karcher & Lee,
2002) states that adolescents seek to strengthen peer-mediated connectedness to friends,
while simultaneously maintain their adult-mediated connectedness to school, teachers, and
family. Consequently, as adolescents individuate over time, their experience shifts towards
connectedness that is mediated by peer-influenced norms. The college context may
catalyze this process, as traditional-age college students begin to spend less time with their
family of origin and more time surrounded by peers. For multiracial adolescents
specifically, the various ecological factors that contribute to connectedness may be
uniquely influenced by existence with multiple heritages.
According to Wallace (2003), in the college context, interethniclinterracial group
relations take on a heightened salience and intensity. There is a higher emphasis on group
solidarity and mixed heritage students may encounter biases within their heritage
19
communities due to minority, majority, or multiple ancestries. Identity politics based on
perceived racial authenticity may direct their opportunities for identification! participation
in a community (Wallace, 2003). This provides a logical bridge to the potential
relationship of social connectedness to multiracial identity, in that both inter- and intra-
personal relationships are influenced by racial and ethnic identity. As explained by Root
(2004):
When [multiracial students] leave high school and move to a new community for
college.. .they must negotiate their identity on a daily basis... they live in an
environment in which race may matter very much and for which they may be
unprepared (Root, 1999)... Stress may precipitate acute homesickness, anxiety, or
depression. (p. 28)
This speaks to the importance of understanding the impact of multiple ecological factors,
such as connectedness, on the development of multiracial college students, especially given
the transitional nature of this time period.
Social Connectedness and Multiracial Identity: Influence of Sex
With regard to social connectedness, a number of researchers have suggested that
men and women differ in their interpersonal experiences (Lee & Robbins, 2000; Caffarela
& Olson, 1993; Miller & Stiver, 1993; Nelson, 1996). Lee and Robbins (2000, 2002) have
shown sex differences in the construction of social connectedness. Other researchers have
suggested that women in western culture tend to define themselves in more socially
connected terms, while men's self definitions emphasize the quality of separateness
(Caffarela & Olson, 1993; Miller & Stiver, 1993). Meanwhile, although some authors have
20
found clear sex differences in the nature of connectedness-implying that women tend to
value connectedness more that men, the work of Lee and colleagues (e.g. Lee, Keough, &
Sexton, 2002; Lee & Robbins, 2000) has suggested that social connectedness is equally
salient in both women's and men's lives. The finding of these researchers suggest that
college women and men differ in the types of relationships that constitute their experiences
of social connectedness and also differ in terms of the kinds of social provisions that
contribute to interpersonal closeness.
Research on multiracial identity suggested sex differences in identification. Root
(2003) has posited gender differences in the experience of affirming ethnic or racial
minority heritage. For example, research by Twine (1996) suggests that affirmation of
minority status appears more difficult for multiracial women in relationships with white
men, than multiracial men with white women. This suggests that multiracial males and
females may experience differences in legitimizing, reinforcing, or demonstrating their
racial heritage or authenticity. In essence, multiracial men and women may be faced with
different means to "prove" their racial make up. Additionally, Root (1994) categorized six
developmental challenges faced by multiracial individuals. Two of these, physical
appearance and sexuality, emerged as significantly impactedby sex differences, with
women experiencing more challenges than their male counterparts in both areas.
Facilitative Support
An additional factor specifically related to the experiences of college men and
women has to do with the potential support they do or do not experience in the college
environment. In a qualitative study, Renn (2000) explored how multiracial students'
21
interactions with peers, involvement in activities, and academic work influenced the kinds
of identity-based spaces (e.g. student groups, student support services, and campus
activities) they chose to occupy. The influence of identity-based space was found to be
impOltant to students' private construction of multiracial identity. Factors such as common
cultural knowledge, similar physical appearance, and group involvements shaped the peer
culture on the college campuses studied and ultimately defined who would occupy which
identity-based space. Many multiracial students struggled to find a sense of belongingness
based on these factors because they were based on monoracial norms. Many multiracial
individuals experience pressure to make a single choice regarding racial self-labeling
(Nishimura & Bol, 1997; Paez & Lyda, 2002), and this pressure may be reinforced in
identity based space. According to Nishimura (1998), the existence of established student
groups, particularly minority student groups, does not automatically meet the needs of a
diverse student population. Therefore the relationship between various campus supports
and multiracial racial identity, social connectedness, and college adjustment remains to be
understood and examined in the literature. The relationship of level of involvement and
level of supportiveness of facilitative supports on the psychological adjustment of
multiracial college students also merits further study.
College Adjustment
Shih and Sanchez (2005) recently conducted the first comprehensive review of both
the theoretical and empirical work on the psychological adjustment of multiracial
individuals. This review addressed two questions posed by theories of multiracial identity
development: (a) do multiracial individuals experience difficulties in the process of
22
forming a racial identity, and, (b) does this impact psychological adjustment? The authors
could only find support for the presence of difficult identity-related experiences (e.g.
rejections by others, identity confusion) in qualitative studies using clinical samples. As a
result, they concluded that, generally, multiracial individuals do not devalue their
multiracial identity. The authors found evidence in the literature of positive and negative
trends for psychological adjustment of multiracial individuals. These trends were based on
the type of sample being investigated (i.e. clinical or nonclinical), the outcome being
considered, and the specific monoracial population to which the multiracial population was
being compared. In sum, though scholars have proposed a number of theories to explain
multiracial identity development and its consequences (e.g. Gordon, 1964; Park, 1928,
1931; Root, 1996; Stonequist, 1937; Wardle, 1987, 1999), researchers have conducted few
empirical studies to test these theories. Shih and Sanchez (2005) suggest that because this
area of research is still in the beginning stages, expansion of the research should explore the
factors that impact the psychological adjustment of multiracial individuals.
Depression
Literature on the relationship between multiracial identity and depression has
yielded inconsistent and in some cases discrepant conclusions (see Shih & Sanchez, 2005
for detailed review). In their review of the qualitative literature on multiracial identity and
depression for example, Shih and Sanchez (2005), were only able to identify patterns in
which the prevalence of depression is related to the type of sample (clinical vs. non-
clinical). They observed low depression, and happiness, in nonclinical samples but not in
clinical samples, and observed depression more often in studies with clinical samples
23
than in nonclinical samples. Shih and Sanchez's (2005) survey of the quantitative
literature, suggested a preliminary trend for multiracial individuals to show higher levels
of depression when compared with their monoracial majority peers but not when
compared with their monoracial minority peers. However, these results were based on
examination of only four quantitative studies, and the samples of many of the qualitative
and quantitative studies were often limited in the multiracial combinations represented.
Therefore, corroborating the reliability of this trend calls for further investigation.
Purpose of This Study
A comprehensive review of the literature revealed no quantitative studies
examining specific elements of Root's ecological model (2000, 2003). Models of
multiracial identity development extant in the literature are conceptual and were almost
exclusively based on qualitative research methods (see Phinney, 1993; Root, 2003; Wardle
& Cruz-Janzen, 2004). The role of multiracial identity variance in the experience of
multiracial individual has not been examined in quantitative studies, and few qualitative
studies have been conducted (see Renn, 2000, 2004; Root, 1996) to observe this
phenomenon. Moreover, there were no studies exploring the relationship between
multiracial identity variance, college adjustment, social connectedness, and facilitative
supports on campus, and depression in multiracial college students.
Given the status of the empirical literature just described, in this study I sought to
enhance existing knowledge of the experiences of multiracial college students in a manner
that was theoretically related to models of racial identity development that are inclusive of,
and relevant to multiracial individuals. This research will help college student personnel
24
(i.e. Counseling center staff, advisors, student life staff, etc.) and researchers better
understand the contextual influences that may impact multiracial/multiethnic college
students' development. This research will provide data that will assist colleges and
universities to be better equipped to address the unique needs of this population.
Using Root's ecological framework for understanding racial identity as a theoretical
guide (Root, 2000, 2003), I examined a number contextual, interpersonal, and intrapersonal
factors that are potentially related to the racial identity development of college of
multiracial college students. In particular, I focused on the following factors present in
Root's model: racial identity, sex, and connection to community (as measured by social
connectedness and one's experience of campus support). I investigated the relationship of,
racial identity variance, social connectedness, facilitative supports, depression, and college
adjustment supports in multiracial college students, with consideration given to sex
differences.
Research Questions
In this investigation I asked the following exploratory research questions:
1) When presented with Zack's (2006) multiracial identity options how do
multiracial college students choose to identify?
a. What are the participants' frequencies of identification with each option?
b. Does participants' multiracial identity vary across the following contexts:
family, friends, classroom, new settings, with authority?
c. Does participants' reflection of others' perceptions of themselves vary?
d. Does participants' multiracial identity vary by sex?
25
e. If sex variations in multiracial identity are present, how do they vary by
context (e.g. family, friends, classroom, new settings, with authority
figures)?
2) What is sample's overall level of social connectedness?
a. Are there sex differences in social connectedness?
3) What is the level of involvement in and subsequent level of support provided by
campus facultative supports?
a. Are their sex differences in involvement in and support from facilitative
supports?
4) What is the sample's overall level of college adjustment?
a. Are there sex differences in college adjustment?
5) What is the sample's overall level of depression?
a. Are there sex differences in depression?
6) Is there a correlation among multiracial identity variance, social connectedness,
facilitative supports, college adjustment, and depression in multiracial college
students?
a. What is the nature of these relationships?
b. Are there sex differences in these relationships?
c. Is there a relationship between involvement in campus facilitative supports
and multiracial identity variance, social connectedness, college
adjustment, or depression?
26
d. Is there a relationship between involvement in campus supports and social
connectedness?
e. Is there a relationship between adjustment and mood?
b. Is multiracial identity variance related to adjustment, mood?
c. Is social connectedness related to adjustment, or mood?
d. Is involvement in facilitative supports related to adjustment, or mood?
e. Is level of support experienced by facilitative supports related to
adjustment, or mood?
7) How much is the variance in multiracial identity variance, social connectedness,
and depression explained by sex, support from student groups and connectedness
to student groups?
To examine the questions above, I used a passive observational survey design, initializing
a web-delivered questionnaire to assess the relationships among racial identity variance,
facilitative supports, social connectedness, college adjustment, and depression for
multiracial college students. Simple relationships among constructs were assessed by
correlation analyses to address above research questions. Simple differences were
assessed by analysis of variance to address the above research questions. Simultaneous
regression models were used to assess the amount of variance shared by identified
gender, support and connectedness to student groups on the following criterion variables:
multiracial identity variance, social connectedness, and depression.
27
CHAPTER II
METHODOLOGY
Participants
A total of 231 students completed the online survey. Based on criteria for
inclusion ([a] identified as having two or more racial heritages based U.S. government
Census racial and ethnic categories, [b] were age 18 to 23, and [c] were enrolled in a
college or university) data from 199 participants was used for data analysis. The majority
of participants whose responses were completely eliminated from inclusion had endorsed
only one race on the demographic questionnaire (n = 29, 12.6%). The remaining
participants (n= 3, 9.4%) whose responses were eliminated did not meet age criteria (all
were over age 23). This resulted in a final total N of 199 participants who were included
in subsequent study analyses. Tables 1 - 2 provide demographic information about the
participant groups. As presented in Table 1, 154 (77.4%) of the participants were female
and 45 were male. None of the participants identified as Transsexual or Transgender. All
students were in the spring semester or winter/spring quarter (January through May) of
the academic year when they completed the survey. There were 39 freshmen (19.6%),41
sophomores (20.6%),49 juniors (24.6%), 50 seniors (25.1 %), and 19 graduate students
(9.5%). Participants were on average 20.36 years old (SD = 1.51), and carried an average
Table 1
Participant Demographics (N=199)
Frequency Percentage Frequency Percentage
Sex Region of Upbringing
Female 155 77.9 Northeast 51 25.6
Male 44 22.1 Mid-Atlantic 10 5.0
Southeast 19 9.5
Year in College Midwest 36 18.1
Freshman 39 19.6 Southwest 13 6.5
Sophomore 41 20.6 Pacific Northwest 24 12.1
Junior 49 24.6 West 38 19.1
Senior 50 25.1 Alaska 2 1.0
Grad. Student 19 9.5 Hawaii 2 1.0
US Territory 2 1.0
Age
18 25 12.6 College Region
19 44 22.1 Northeast 60 30.2
20 34 17.1 Mid-Atlantic 12 6.0
21 49 24.6 Southeast 18 9.0
22 29 14.6 Midwest 44 22.1
23 19 9.5 Southwest 8 4.0
Pacific Northwest 21 10.6
Major West 31 15.6
AfAm Stud 3 1.5 Alaska 1 0.5
Anthropology 6 3.0 Hawaii 2 1.0
Psychology 8 4.0 US Territory
Sociology 10 5.0
Ed. Studies 7 3.5 Father Mother
Political Sci. 7 3.5 Education Level
Freq % Freq %
Biology 8 4.0 Did not graduate 8 4.0 11 5.5
H.S.
Finance 4 2.0 High Schoo1/ G.E.D. 22 11.1 20 10.1
Neuroscience 6 3.0 Some College 29 14.6 24 12.1
Engineering 8 4.0 Associates Degree 10 5.0 23 11.6
English 9 4.5 Bachelors Degree 36 18.1 51 25.6
Other mqiors 123 61.8 Some Grad. School 5 2.5 12 6.0
Graduate Degree 80 40.2 57 28.6
Don't Know 9 4.5 1 .5
Note. AfAm = African American; Ed.= Education; Sci. = Science; Freq = Frequency; H.S. = High
School; Grad = Graduate.
28
Table 2
Multiracial Demographics (N = 199)
Frequency % Frequency %
Black, White 34 17.0 NA, Black, White, SOR 1 .5
Japanese, White 21 10.5 Filipino, Chinese, White 1 .5
Chinese, White 20 10.0 Black, Japanese, White 1 .5
Filipino, White 13 6.5 Mexican, NA 1 .5
AI, White 9 4.5 AI,OL 1 .5
NA, Black, White 8 4.0 Filipino, White, Mexican 1 .5
PR, White 8 4.0 Chinese, OA, White 1 .5
OL, White 7 3.5 NA, Black, Chinese, White 1 .5
Korean, White 6 3.0 OL, NA, Black, AI, Chinese, 1 .5
White
OL, Latino 4 2.0 PR, NA, Filipino 1 .5
OA, White 4 2.0 Chinese, Filipino 1 .5
White, SOR 4 2.0 Vietnamese, SOR 1 .5
Cuban, White 4 2.0 Chinese, OA, White, SOR 1 .5
Mexican, White 3 1.5 OL, Chinese, Filipino, White 1 .5
Black, Chinese 3 1.5 Korean, Vietnamese 1 .5
PR, Black 3 1.5 Filipino, Japanese 1 .5
TwoOL 2 1.0 Black, Japanese 1 .5
OL, Chinese, White 2 1.0 Black, Filipino 1 .5
Mexican, NA, White 2 1.0 Chinese, Filipino, Hawaiian, 1 .5
White
Black, White, PR 2 1.0 Mexican, SOR 1 .5
Mexican, Japanese 2 1.0 Chinese, SOR 1 .5
Black, Filipino, 2 1.0 PR, NA, Black, White 1 .5
White
Filipino, OA, White 2 1.0 Vietnamese, White 1 .5
Black, SOR 2 1.0 NA, White 1 .5
Black, Asian, OL 1 .5 Black, Japanese, Korean 1 .5
Chinese, Vietnamese 1 .5 Mexican, Filipino 1 .5
Chinese, Korean, 1 .5 AI, Chinese, Vietnamese, White 1 .5
White
Latino, Japanese 1 .5 Filipino, White, SOR 1 .5
Black, AI, Samoan, 1 .5
SOR
Note. AI =Asian Indian; NA =Native American; PR =Puerto Rican; OL = Other Latino; OA =Other
Asian; SOR =Some other race.
29
30
GPA of 3.40 (SD =0.38). Participants reported their region of upbringing as follows:
25.6% Northeast, 5.0% Mid-Atlantic, 9.5% Southeast, 18.1 % Midwest, 19.1 % West, 6.5
% Southwest, 12.1 % Pacific Northwest, and 1.0% each for Alaska, Hawaii, and US
Territory. Participants also reported the region of their college or university, yielding
30.2%,22.1 %, 15.6% attending in the Northeast, Midwest, and West respectively.
Participants provided educational information for their parents, with over 60% of mothers
and fathers having earned a bachelors degree or higher (graduate or professional degree).
All participants had declared a major.
The 199 participants represented 57 different unique racial combinations (see
Table 2) indicating a racially diverse sample of multiracial college students. Of the 57
racial combinations, 53 (93%) appeared 9 or fewer times in the data, and 33 (57.9%)
appeared only once. Accordingly, racial combinations were combined into larger
categories for the following reasons: (a) for the sake of reporting the data, and (b)
because of the lack of statistical power in the number of less represented combinations, it
did not make logical sense to analyze multiracial group differences, but identifying the
frequencies of identification was of interest. For this reason, categories "Asian/ Pacific
Islander" and "Latino/a" were'created based on combined data (see Table 3). Chinese,
Filipino, Japanese, Korean, Vietnamese, Other Asian, Guamanian or Chomorro, Samoan,
and Other Pacific Islander were combined into the category "Asian/ Pacific Islander."
Mexican, Mexican American, Chicano, Puerto Rican, Cuban, and other Latino were
combined into the category "Latino/a." Asian Indian, which represents participants with
31
heritage from the country of India, was kept as a unique category because of the high
number of participants (13; 6.5%) identifying with this racial group.
Table 3
Combined Multiracial Demographics (N = 199)
Frequency Percentage
White and Asian/Pacific Islander 65 32.5
Black and White 34 17
3 or More Races (including White) 32 16
White and Latino/a 22 11
All Other Racial Pairs 12 4.5
White and Asian Indian 9 3
2 different Latino/a 6 2.5
Black and Asian/Pacific Islander 5 2
3 or More Races (Non-White) 4 2
2 different Asian/Pacific Islander 4 2
Asian/ Pacific Islander and Latino/a 4 2
Black and Latino/a 3 1.5
Note. Latino/a includes Mexican, Mexican American, Chicano, Puerto Rican, Cuban, and other Latino;
Asian/ Pacific Islander includes Chinese, Filipino, Japanese, Korean, Vietnamese, Other Asian, Guamanian
or Chomorro, Samoan, and Other Pacific Islander.
32
Measures
A list of all measures used in this study is provided in Table 4 (For complete
measures see Appendix C).
Table 4
Constructs and their Corresponding Measures
CONSTRUCT MEASURE
Demographic Variables 0 Demographic Questionnaire (Lyda, 2007)
0 Demographic Questionnaire (Lyda, 2007)
Racial Identity
0 Multiracial Identity Options Survey (MIaS;
Lyda,2007)
Connectedness 0 Social Connectedness Scale-Revised (SCS-R;
Lee et al., 2001)
Facilitative support
0 College Supports Questionnaire (Lyda, 2007)
College Adjustment 0 Student Adaptation to College Questionnaire
(SACQ; Baker & Siryk, 1984)
Mood 0 Patient Health Questionnaire (PHQ-9; Spitzer,
Williams, Kroenke, et al., 1999)
33
Demographics
Background information was collected via a demographic questionnaire created
specifically for this study (see Appendix C). This questionnaire included questions about
age, sex, race/ethnicity, national citizenship, region of residence, parents' level of
education, and G.P.A.
Multiracial Identity Variance
Multiracial identity variance was measured by the Multiracial Identity Options
Survey (MIOS; Lyda, 2005). The Multiracial Identity Options Survey was developed for
the purpose of this research to assess each participant's self-reported identification with
one of Zack's (2006) six theorized identity options for mixed race people (fractional,
inclusive, traditional nonwhite, white, generic, and aracial) as well as to determine
variance in identification based on interpersonal or situational context. The survey sought
to answer the following questions: (a) When a multiracial college student is presented
with theoretically defined multiracial identity options and a context, what identity option
does he or she choose? And, (b) how does his or her chosen identity option vary across
contexts? The measure consisted of 11 items, each requiring the participants to chose
from six or seven statements, each corresponding to a specific multiracial identity option.
For example, the statement "I identify myself by describing fractions/percentages of my
racial breakdown (e.g. I'm Ih white, J.A black, and 14 Native American)" corresponded
with the fractional identity option. Participants were first asked which multiracial identity
option statement best fit their answer to the following question: "How do you identify
yourself in terms of race generally?"
34
The next ten items were formulated based on the influence of context in racial
identity development emphasized by Root's (2003) ecological model of racial identity
development, and Renn's (2000, 2004) work on both the significance of public and
private space for identity, and the phenomena of situational identity. These ten questions
were divided into two parts: participants' self identity (first five questions), and
participants' reflection on other's perception of them (second five questions). The first
five questions asked participants to endorse how they identified in the following contexts:
new or unfamiliar settings, with close friends, with immediate family, in class, and with
authority figures (e.g professors, advisors). Participants were then prompted to chose one
of the six identity options described above as well as a seventh option "It depends." This
option was included to accommodate for participants who may feel as if the question
requires a more nuanced answer, thus preventing participants from leaving the question
blank, or answering in a way that was not congruent with their experience. From a
theoretical standpoint, the "It depends" response, accommodated for context based
identity options described by Renn (2004), such as situational and multiple monoracial
identity.
The second set of five questions asked "How do you think the following groups
identify you?" The groups listed were: society as a whole, close friends, immediate
family, classmates, and authority figures. The participant was then given the same set of
multiracial identity options to chose from as above. The purpose of this question was to
assess for the participants' reflection of others' racial perception of them. In other words,
35
how do they believe others perceive them racially? For the same reason as the previous
questions in the survey, the close friends, immediate family, classmates, and authority
figures prompts, include the seventh option "It depends." "Society as a whole" did not
include this option because the global (macrosystemic) nature of society as a construct at
a given point in time (i.e. the time the participant took the survey) required a more
specified answer. Thus, eliminating the "It depends" response option, forced participants
to choose their best reflection of society's views of them at that time.
Multiracial identity variance was calculated by first computing a standard deviation
score for each participant across each of the eleven items of the identification scale.
Greater scores indicated greater inconsistency, or variance, in identity across situations
and contexts. Responses on the survey created categorical variables for self-
identification, identification based on the various contexts, and participants' reflection of
others' perception of multiracial identity. Responses were used in analysis for two
purposes: (a) to assess frequency of multiracial identification per context, and (b) to
obtain a standard deviation score across contexts for each participant, that would provide
a measure of variance in identification across contexts. The latter analysis provided a
quantitative means to measure multiracial identity variance, a central construct in the
current study that has been observed in phenomenological, ethnographic, and case
studies, but is to date unmeasured quantitatively.
Social Connectedness
Social connectedness was measured by the Social Connectedness Scale-Revised
36
(SCS-R; Lee et al., 2001), a 20-item, 6-point Likert-type scale (l = strongly agree to 6 =
strongly disagree) that measured the degree of interpersonal closeness that was
experienced between a participant and his or her social world (i.e., friends, peers,
society), as well as the degree of difficulty in maintaining a sense of closeness with
others. Sample items included "I am able to connect with other people," "I am able to
relate to my peers," and "I feel comfortable in the presence of strangers." The negatively
worded items were reverse scored, and the measure scores had a possible range of 20 to
120. An item mean score with a possible range from 1 to 6 was also calculated by
dividing the total scale score by 20 (or number of scale items). Higher scores on the SCS-
R reflected a stronger sense of social connectedness. A mean item score greater than 3.5
suggests greater tendency to be socially connected vs. disconnected (Lee, Draper & Lee,
2001).
Lee, Draper and Lee (2001) revised the original Social Connectedness Scale (Lee &
Robbins, 1995) for use with a college age sample. A validation study yielded a mean
scale score (89.84; SD = 15.44) and mean item score of 4.49 (SD = .77), with a reliability
alpha of .92. In the validation study, Lee et al. found no significant sex or race differences
on the SCS-R. However, a sub-study using the SCS-R to test a mediator model between
social connectedness, dysfunctional interpersonal behaviors, and psychological distress,
found differences by sex, but no differences by race (Lee et al., 2001). In tests of
convergent validity, the SCS-R was positively correlated with a measure of independent
self-construal (r = .37), and collective self-esteem (membership r = -.57, private r = .42,
37
public r = .39). The SCS-R was negatively correlated with measures of loneliness (r = -
.80), social distress ( r = -.55), avoidance (r = -.57), depression (r = -.45), hostility (r = -
.24), and social discomfort (r = -.28). Tests of discriminant validity failed to yield
significant correlations with measures of interdependent self-construal, collective
identity, somatization, obsessive-compulsiveness, phobic anxiety, and excessive
interpersonal responsibility and controlling behaviors. SCS-R validation study findings
were consistent with previous research by Lee and Robbins (1995, 1998,2000) on the
original SCS. As a measure of internal consistency, a Chronbach's alpha coefficient of
.93 was found on the SCS-R for the current sample.
Facilitative Supports
The College Supports Questionnaire (CSQ; Lyda, 2005) was created for the
purposes of this study to measure participants' utilization of campus support services and
groups/organizations (hereafter referred to as facilitative supports), as well as perceived
connectedness to and support received from these facilitative supports. The CSQ
consisted of 7 items assessing for the following: (a) the presence of facilitative supports
available to participants' on their campus (e.g. Ethnic student groups/ Ethnic Student,
Fraternities and Sororities with a traditionally racial minority membership, multicultural
student group/ multicultural student unions, minority mentorship program, university
initiated minority student programs, Biracial or Multiracial students group), (b)
participants' level of participation in groups/ organizations at college or in the community
(four point Likert-type scale,; l=extremely active to 4 = not-active), (c) participation in
38
racially/ethnically based groups/organizations, (d) the level of activity in
racially/ethnically based groups/organizations (four point Likert-type scale; l=extremely
active to 4 = not-active), (e) the level of support received from campus supports (four-
point Likert-type scale; 1= Not supported at all to 4 = Very supported), (f) level of
connectedness to others involved in campus supports (four point Likert-type scale, 1 =
Not Connected to 4= Very Connected), and (g) the usefulness of a number of support
services present on campus (e.g. advising, orientation activities, and career services; five-
point Likert-type scale, from 1 = very useful to 2= am not aware of such a service being
available). Items from the College Supports survey were used to establish the following
variables used for analysis in the current study: activity in student Groups (nominal
variable), student group availability (nominal variable), participation in student groups
(nominal variable), participation in ethnic/cultural groups (nominal variable), support
from groups (interval variable), connectedness to groups (interval variable), usefulness of
student services (interval variable).
College Adjustment
College adjustment was measured using scores from the Student Adaptation to
College Questionnaire (SACQ; Baker & Siryk, 1984). Specific to the college context,
Baker and Siryk (1984, 1999) defined college adjustment as including four dimensions:
Academic Adjustment (acclimation to various educational demands characteristic of the
college experience), Social Adjustment (students' reconciliation of interpersonal-societal
demands inherent in adjustment to college), Personal-Emotional Adjustment (how the
student is feeling psychologically and physically-experience of general psychological
39
distress and/or any associated somatic problems), and Goal Commitment/Institutional
Attachment (students' feeling about being in college, in general, and the college he or she
is attending, in particular). Their work led to the development of a measure of the
SACQ. The SACQ is a self-report measure with 67 Likert-scale items, ranging from 1 =
"Applies to me very closely") to 9 = "Doesn't apply to me at all." By summing the scores
of the 67 items, the instrument yields a total raw score for adjustment to college. Raw
scores are converted into T-scores, based on conversion tables appropriate for
participants' sex and student standing listed in the SACQ manual (Baker & Siryk, 1999).
Per the SACQ manual, all participants' raw scores were converted to T-scores using the
"second semester norms," because all students were in at least their second semester in
college. T-score conversion tables are based on the normative sample stratified by sex
and semester, and have a mean of 50 and a standard deviation of 10.
There are also four subscales that assess various forms of adjustment. On the basis
of two independent samples, Baker and Siryk (1986) reported alpha coefficients for the
full scale (.91 and .92) , and the following alphas for the four subscales: academic
adjustment (SACQ AC; .82 and .87), social adjustment (SACQ SOC; .88 and .88),
personal-emotional adjustment (SACQ PERS; .82 and .79), and attachment-goal
commitment (SACQ AGC; .89 and .86), reflecting an adequate degree of internal
consistency for each subscale. In this study I employed all subscales. Full scale and
subscale alphas for this measure on the current sample were: Full Scale SACQ alpha =
.94; personal-emotional adjustment alpha = .85; social adjustment alpha= .88, academic
adjustment alpha = .89, attachment-goal commitment alpha = .83. Sample items from the
40
SACQ include: "I feel that I fit in well as part of the college environment" (social
adjustment subscale); "I have been keeping up to date on my academic work." (academic
adjustment subscale); "I have been feeling tense or nervous lately." (personal-emotional
adjustment subscale) "I feel that I am very different from other students at college in
ways that I don't like."; "I wish I were at another college or university." (attachment-goal
commitment subscale).
Depression
A more specific assessment of mood will be employed as an adjunct to an overall
assessment of college adjustment. The Patient Health Questionnaire (PHQ-9; Spitzer,
Williams, Kroenke, et aI., 1999, Kroenke, Spitzer, & Williams, 2001) is a brief, 10-
question survey of depression, which scores each of the 9 DSM-IV criteria for
Depression as "0" (not at all) to "3" (every day). The PHQ-9 was used to assess
participants' mood. The PHQ-9 was validated and normed on a sample of 6,000 patients
in 8 primary care clinics and 7 OB-GYN clinics. Construct validity was assessed using
the 20-item Short-Form General Health Survey, self reported sick days and clinic visits,
and symptom related difficulty. Criterion validity was assessed against an independent
structured mental health professional interview in a sample of 580 patients. PHQ-9 scores
greater than or equal to 10 had a sensitivity of 88% and a specificity of 88 % for major
depression as determined by the structured interview, indicating strong criterion validity.
Scores of 5, 10, 15, and 20 represented mild, moderate, moderately severe, and severe
depression respectively. In addition to making criteria based diagnoses of depressive
disorders, the PHQ-9 is also a reliable and valid measure of depression severity, thus
41
making it useful for this research study as a means to asses participants mood state on a
continuum (no depression to severe depression). The internal reliability of the PHQ-9 is
excellent with Chronbach's alpha of 0.89 in the primary care study, and 0.86 in the OB-
GYN study. Test-retest reliability, as measured by 48 hour follow up administration
yielded a correlation of 0.84 between the two administrations.
The reasons for incorporating a more specific assessment of mood, beyond that
measure by the SACQ are threefold: (a) An assessment of depression provided important
mental health related data that will increase the understanding of the relationship between
the constructs and intrapersonal processes, (b) it was important to understand the
relationship of the constructs of interest and high levels of depression as a means to assess
for potential risk factors in multiracial college students experience at college (c) it was
important to understand the relationship between low depression and the constructs of
interest as a means to assess potential protective factors in multiracial college students
experience at college. In the current study, PHQ-9 item responses were summed and
averaged, resulting in scores ranging from zero to three to obtain a standard score by which
results could be compared to a midpoint (1.5). The Chronbach's alpha for the current study
was .89.
Procedures
Participant Recruitment
Participants were recruited via the Internet, which has several benefits (Gosling,
Vazire, Srivastava, & John 2004). First, this method provided access to samples beyond
42
the reach of traditional methods of psychological research. Second, Internet methods
allowed for relatively inexpensive and practical methods of data entry. Finally, the web
allowed me to access a much greater number of individuals who identify as multiracial
from across the nation as opposed to a limited geographical area, resulting in a larger and
more diverse sample. This improved the external validity of the current study so that
findings may better generalize to multiracial college students nationwide.
I worked with PsychData to create and maintain my questionnaires. PsychData is
an online data collection service that was created specifically for the social science
community to conduct secure Internet-based research. PsychData provided an online
research manager that allowed me to control all aspects of my data collection including
building and editing surveys, changing survey response options, tracking numbers of
participants, and downloading data at any time during the data collection process. Each
survey that I created through PsychData was hosted in a secure environment that was
designed to protect participants' privacy. In addition, the service stored informed consent
forms or any other identifying participant information on a separate, secure server. I also
gathered contact information from participants who wanted to participate in an incentive
raffle to win a $200 gift certificate to Amazon.com, by creating two separate, linked
surveys.
Participants in this study were recruited in three ways: (a) through local and
national emaillistservs targeting ethnic minority students, specifically multiracial college
students; (b) establishing contacts with ethnic student group members, academic
departments and student support services, and requesting that these contacts distribute
43
information about the survey and/or allow permission to contact individuals with whom
they are affiliated directly; (c) through a snowball sampling technique (Gall, Gall, &
Borg, 2003). Snowball sampling is a non-probability sampling technique that relies on
previously identified group members to identify other members of the population. In this
study, I conducted a snowball sampling technique in which multiracial college students,
faculty, staff, colleagues, and co-workers encouraged other multiracial college students to
participate in this investigation via email and word of mouth. I provided monetary
incentives to increase recruitment success by providing each participant with one
opportunity to win a $200 gift certificate to Amazon.com.
Participants consisted of multiracial individuals, age 18 to 23 (SD = 1.51), who
were enrolled in various colleges and universities. The sample included students
attending numerous universities across the United States. In an effort to minimize a
possible acculturation and ethnic identity bias associated with sampling exclusively from
ethnic and cultural student organizations, recruitment efforts included sampling from
numerous other types organizations and academic programs (i.e. non-culturally and non-
ethnically focused student groups).
Data Collection
Data collection took place between January and May 2007 (spring semester, or
winter and spring quarters depending on participant's academic system). I initially
established contact with potential participants through an informational email requesting
their participation in this study. This email included: (a) a brief description of the study,
44
specifying that 18-23 multiracial students were sought; (b) the estimated time to complete
the survey, (c) a statement of participants' chances to win a $200 gift certificate to
Amazon.com and, (d) a web address hyper1ink to the survey web pages.
The web address included in this email first lead participants to the informed
consent form (see Appendix B). After reading the informed consent statement, those
who chose to participate were directed to instructions for completing the questionnaire
battery. I provided participants with time prompts so that they knew how long it took to
complete the questionnaire battery. While individuals who are not multiracial, as defined
by the US Census criteria for racial categorization, were able complete the survey, only
participants who endorsed multiple races on the demographic form (Appendix C) were
included in analysis.
Sample Size
The power of a statistical test is the probability that the test will find a statistically
significant effect in a sample of size N, at a pre-specified level of alpha, given that an
effect of a particular size exists in the population (Cohen, 1992). In this study, an alpha
level of .05 and a medium population effect size (ES =.15) were desired. A candidate set
of 14 independent variables was obtained. Based on examination of Cohen's (1992)
indexes for determining sample sizes necessary for .80 power to detect effects, the above
criteria for a multiple regression/correlational analysis required a sample of 125
participants for this study.
45
CHAPTER III
RESULTS
Overview
In this chapter I discuss the results of this study using the following organizational
framework: First, I present a summary of the preliminary analyses as they relate to
missing data management, reliability, and preparation of data for statistical analyses.
Next, I outline the results of descriptive analyses used to address exploratory research
questions 1 - 5. Third, I discuss correlational analyses used to examine the relationships
among constructs specified in the exploratory research questions 6a- e. Fourth, results of
simultaneous regression analyses employed to answer exploratory research question 7 are
presented and explained. Finally, I conclude this chapter by summarizing all results..
Preliminary Analyses
Table 5 describes the alpha reliability coefficients for each of the variables used in
the study. All scales had alpha reliability coefficients above .83. All variables had an N of
199, with one exception. Data from five participants on the Patient Health Questionnaire-
9 were not included in reliability analysis because they met criteria for exclusion (missing
more than 1 item response; see Table 5).
46
Table 5
Alpha Reliability and Number ofItems for Instrument Scales (N= 199)
Scale Alpha Number of Items
Social Connectedness Scale Revised (SCS-R) .93 20
SACQ (Full Scale) .94 67
SACQ Personal-Emotional Adjustment .85 15
SACQ Social Adjustment .88 20
SACQ Academic Adjustment .89 24
SACQ Attachment-Goal Commitment .83 15
Patient Health Questionnaire - 9 .89* 9
Note. SACQ = Student Adaptation to College Questionnaire;
* N= 195.
Data were first transformed and calculated for the purposes of data analysis
(Field, 2005). It is important to note that the degrees of freedom vary across analyses.
This variance is due to missing data resulting from participants missing a significant
number of responses on entire measures. The SACQ, SCS-R, and PHQ-9 were eligible
for missing data replacement if the number of missing items were equal to no more than
10% of the total number of items. This means that the number of missing items needed to
be less than or equal to six (6) for the Student Adaptation to College Questionnaire
(SACQ) Full Scale, one (1) for the SACQ Personal-Emotional Adjustment and
Attachment-Goal Commitment Subscales, two (2) for the SACQ Social Adjustment
Subscale, SACQ Academic Adjustment Subscale, and SCS-R, and one (1) for the PHQ-9
47
to meet criteria for data replacement. When there were sufficient data for a particular
measure or subscale to maintain a participant's responses for analysis, a multiple
imputation procedure for missing data replacement (see Schafer & Graham, 2002) was
conducted for the missing points on the Student Supports Questionnaire, Multiracial
Identity Options Scale, the Social Connectedness Scale- Revised, and the Patient Health
Questionnaire- 9. This procedure was conducted using NORM software (Schafer, 1999),
a process that uses the expectation maximization algorithm. Unlike mean substitution or
other simple regression missing data replacement techniques, multiple imputation
maintains similar means, standard deviations, normality estimates, and covariances with
related variables. In a comprehensive exploration of missing data replacement techniques,
Shafer and Graham (2002) found that multiple imputation data replacement was the most
accurate method for every type of data they examined including actual and synthesized,
normal to non-normal, highly correlated to uncorrelated, and varied to restricted ranges.
In this study, I conducted missing data replacement on either the item or scale level for
various measures, depending on the particular scale. Replacement is viable on either
level, according to Schafer and Graham (2002). Thus, all surveys were usable because the
multiple imputation procedure estimated and replaced the missing data. For the SACQ,
missing data were replaced per the instructions provided in the SACQ manual (see Baker
& Siryk, 1999). The Demographics, College Supports Questionnaire, and Multiracial
Identity Options Scale were not eligible for missing data replacement because I was
interested in responses on this measure on an item-by-item basis. Reponses on each item
of these surveys was independent of responses on the previous items.
48
Descriptive Analyses
The following section outlines results of the analyses used to answer exploratory
research questions one through five.
Multiracial Identity Variance
The current study examined frequency of identification with Zack's (2006)
multiracial identity options and the incidence of variance in personal and perceived
identity options across different contexts. Frequency and variance in multiracial
identification was examined by sex. The frequencies of multiracial identification on the
MIOS are listed in Table 6. Sample size was 199 for each of contexts listed except for
self-identification in the presence of family (N = 197), and participant's reflections of
classmates' perception of them racially (N = 198). In the latter two cases, participants'
did not give responses to these items, thus causing the N to drop below 199 for each item.
According to Table 6, when asked to endorse how they chose to self-identify "in
general," the majority of participants (78.4%) endorsed "Fractional" (44.7%) or
"Inclusive" (33.7%) multiracial identity options. Only.5 percent of participants endorsed
"White," this despite the large majority of participants (87.4%) indicated at least some
white racial heritage. By contrast, when asked to endorse the multiracial identity option
by which they believed "Society," in general perceived them, 47.5% endorsed
"Traditional-Non White," while another 17.6 % endorsed "White." "Fractional" and
"Inclusive" identity options remained the most endorsed multiracial identity options for
participants self-identification in various contexts (see Table 6 and Figure 2). This was
not the case for participants' reflections of others' perception of their identity, which saw
49
an increase in endorsement of "Traditional Non-White," "White," "Generic," and "It
Depends"(see Table 6 and Figure 2). This suggests a discrepancy between participants'
personal sense of multiracial identity and their reflections on others' perception of their
multiracial self-identity.
Figure 2 illustrates the frequency (in number) of multiracial identification by
context (both by self and the participants' reflections of others' perceptions of them).
Figure 2 is not a representation of interaction effects, but simply frequency and
descriptive information graphed in such a way to illustrate differences and patterns.
Sex differences. As displayed in Table 7, males and females did not differ
significantly in their multiracial identity variance.
Social Connectedness
On average, participants reported a greater tendency to be socially connected vs.
disconnected (SCS-R mean item;::: 3.5; Lee, Draper, & Lee, 2001). On average, social
connectedness was significantly greater than the scale midpoint (3.50), t(198) = 12.64, p
< .001 (See Table 7).
Sex differences. As displayed in Table 7, males and females did not differ
significantly in their report of social connectedness. Both males and females reported a
greater tendency to be socially connected vs. disconnected (SCS-R mean item;::: 3.5;
Lee, Draper & Lee, 2001).
Perceptions ofFacilitative Student Supports
Data from the second section of the questionnaire (i.e., College Supports
Questionnaire) were analyzed using single-sample t-tests. That is, mean levels on each of
50
the seven items were tested against their respective scale midpoints (Field, 2005). The
mean level of active participation in student support groups was statistically greater than
the midpoint of 2.50 (M = 2.94), t(198) = 6.01, p < .001.
With regard to perceived availability of student support groups and affiliation or
participation in student support groups, the data were calculated such that for each
participant the total values were equal to the number of options they selected. As shown
by the mean for perceived availability of student support groups (M = 4.17), the sample
reported a moderate level of support group availability; this value did not differ
significantly from the midpoint (4.00), t(199) = 1.50, ns. However, few participants did
appear to belong to or participate in student support groups, as evidenced by a
significantly lower mean (M = 1.35) than the midpoint (4.00), t(199) = -27.57, p < .001.
Participation in ethnic and cultural student support groups appeared to be more
encouraging, as participants reported a significantly greater participation mean (M =
2.82) than the midpoint of the response scale, t( 199) =4.18, p < .001. However, the
degrees to which participants perceived support from student support groups and felt
connectedness with student support groups was only moderate (t(187) = .98, ns; t(187) =
.70, ns respectively). Participants did report that the usefulness of student support
services was significantly greater than moderate, t(190) = 4.86, p < .001.
Sex differences. With regard to sex, facilitative supports did differ in two respects.
As displayed in Table 7, females reported significantly greater active participation in
student support groups than males, F(1, 185) =4.7, p < .05. The same was true for
perceived usefulness of student support services, F(1, 185) = 4.34, p < .05.
51
College Adjustment
Mean SACQ scores and standard deviations, and T-scores are reported in Table 8.
Overall, scores for men, and women were within one standard deviation (10) of the mean
T score (50), suggesting average adjustment to college compared to their respective sex
specific normative sample (Baker & Siryk, 1999).
SACQ items responses were summed and averaged, resulting in full-scale and
subscale scores ranging from one to nine. Similar to other study variables, means of the
five SACQ were tested against the response scale midpoint (5.00) to interpret the levels
of adaptation to college (Field, 2005). The SACQ-full scale mean reached a level
significantly greater than the midpoint, t(198) = 13.63, p < .001. Likewise, each of the
four subscales reached a level significantly greater than the midpoint: academic: t( 198) =
1O.38,p < .001; social: t(198) = l1.73,p < .001; personal-emotional: t(198) = 3.61,p <
.001; and institutional attachment: t(198) = 15.25, P < .001. Thus, on average,
participants reported a considerably high level of adaptation to college.
Sex differences. As shown in Table 7, males reported significantly greater
institutional attachment than did females, F(1, 185) =6.00, p< .02. No other significant
sex differences were found on the SACQ full scale or subscales.
Depression
The sample mean for depression was significantly lower than the midpoint (1.5),
t(198) = -9.50, p < .001, indicating low levels of depression across the sample.
Sex differences. As displayed in Table 7, males and females did not differ
significantly in their reports of depression symptoms.
Table 6
Percentages ofMultiracial Identification by Context (N=199)
Fractional Inclusive Traditional N.W. White Generic Aracial It Depends
GID 44.7 33.7 5.5 0.5 1.0 14.6
(89) (67) (11) (1) (2) (29) (--)
Unfamiliar 25.1 25.6 10.1 2.0 2.5 13.1 21.6
4-< (50) (51) (20) (4) (5) (26) (43)
;>...0
Friends 37.2 30.2 7.0 0.5 3.5 15.6 6.0.~ (1)
...... U
I:: I:: (74) (60) (14) (1) (7) (31) (12)<l.l <l.l
-0 (/J
Family* 27.6 25.6 8.0 14.6 13.6 9.5...... <l.l --4-< ....
"Q)o.. (55) (51) (16) (--) (29) (27) (19)\/J I::
...... Classmates 26.6 26.6 9.5 2.0 2.0 18.6 14.6
(53) (53) (19) (4) (4) (37) (29)
Authority 27.1 29.1 6.5 1.0 5.0 16.6 14.6
(54) (58) (13) (2) (10) (33) (29)
Society 11.6 8.5 47.2 17.6 5.5 9.5
(/J (23) (17) (94) (35) (11) (19) (--)
.... 0 Friends 23.6 19.6 8.0 2.0 20.6 13.6 10.1,2 ......
...... 4-< (47) (39) (16) (4) (41) (27) (20)o 0
'b ~ Family 26.1 19.6 8.0 2.0 20.6 13.6 10.1
I:: .S (52) (39) (16) (4) (41) (27) (20)o ......
...... 0..
t) ~ Classmates** 9.0 9.0 32.7 11.6 5.0 14.6 17.6
<l.l ....
<;::: <l.l (18) (18) (65) (23) (10) (29) (35)<l.lo..
~ Authority 7.5 9.5 33.7 8.5 7.5 10.6 22.6
(15) (19) (66) (17) (15) (21) (45)
Note. GID =General Identity; Unfamiliar =Unfamiliar People; Frequencies in parentheses.
* N =197
** N= 198.
VI
N
Figure 2. Multiracial Identity Options Frequencies (N =199)
MIOS Frequencies
100
90
*
80
70
60
50
40
30
20
10
o
Fractional
--- - Inclusive
/ - Trad.-NW
-----1----- - -- - -----j- .'>.'White
--" -----G,"'",
Aracial
-+-- Depends
G>() 0'0
V..,~
'?)/~.
~<:i'/-
0'0
~./~
.,~
S>()
~
'?)/1-
0'0 0'0 '¢o '¢O '¢O '¢O '¢O
~ -'1 '% ~. ~ ~ -'1v0 0~ /~ '?)/1- 'l'S' v0~ .,~ t 0";-1-** S''';> 0";-1- ~
<:i'6~S'
Context
Note. GID =General Identity; SID =Self Identity; Unfamiliar =Unfamiliar Settings; PO =Reflected Perceptions of Others.
* In number
** N = 197
t N =198
Vl
W
54
Table 7
Descriptive Statistics ofall Study Variables by Sex and One- Way ANOVA Results
Male (n = 42) Female (n = 145)
Variable M SD M SD F(l, 185)
Multiracial identity variance 1.77 .82 1.80 .67 .07
Social connectedness item score 4.31 .90 4.23 .82 .29
Activity in student groups 2.64 1.07 3.02 1.01 4.67*
Student group availability 4.27 1.48 4.14 1.58 .24
Participation in student groups 1.09 1.22 1.42 1.40 2.09
Participation in E/C SG 3.00 1.02 2.76 1.08 1.77
Support from groups 2.36 1.01 2.62 .94 2.47
Connectedness to groups 2.36 1.01 2.57 .90 1.66
Usefulness of student services 3.04 .52 3.22 .53 4.34*
SACQ-full scale 6.32 1.01 5.97 1.11 3.52
SACQ-academic 6.15 1.11 5.82 1.24 2.59
SACQ-social 6.42 1.44 6.05 1.35 2.44
SACQ-personal/emotional 5.66 1.46 5.30 1.46 2.06
SACQ-institute. attachment 6.62 .98 6.14 1.18 6.00**
Depression .83 .62 1.01 .80 2.01
Note. N = 187. E/C SG = Ethnic Cultural Student Groups
*p < .05. **p < .02.
55
Table 8
SACQ Mean Scale Scores and T-Scores
Male (n =43) Female (n =154)
Variable M SD T-Score M SD T- Score
Full Scale 424.21 68.37 49 400.28 74.13 45
Acad. Adjust. 148.49 26.46 50 139.65 29.87 48
Social Adjust. 128.05 29.03 50 121.08 26.96 44
Pers-Em. Adjust. 85.02 22.11 45 79.55 21.93 43
Inst. Attach. 99.33 14.89 49 92.06 17.74 43
Note. SACQ normative sample T-Score M = 50, SD = 10.
Correlation Analyses
In the current study I also explored the relationships among the primary constructs
of interest: multiracial identity variance, social connectedness, facilitative supports
(specifically involvement in campus facilitative supports), college adjustment, and
depression. I employed correlational analyses to identify the nature of these relationships
as specified in exploratory research question six. In this section I describe the results of
these analyses for the entire sample, as well as with respect to sex. Correlation results are
displayed in Tables 9 and 10.
It is important to note, none of these correlations differed significantly from their
corresponding correlations according to Fisher's z transformation comparison of
independent correlations tests. Thus, significant correlation coefficients are significant at
the 95 percent confidence level or higher.
56
Multiracial Identity Variance
When viewed in light of the entire sample, identity variance was not significantly
correlated with any of the other study variables (see Table 9). However, when
intercorrelations were examined by sex (see Table 10) different findings emerged. For
males, multiracial identity variance was significantly and negatively correlated with
perceived availability of student support groups, perceived support from student support
groups, and connectedness to these groups. For females, however, multiracial identity
variance was significantly and negatively correlated only with participation in ethnic and
cultural student support groups. This suggests that for men the more their multiracial
identity varied across contexts: (a) the less they were aware of student support groups as
available to them, (b) the less they perceived student support groups as being supportive,
and (c) the less connected they felt to student support groups. While for women, the more
their multiracial identity varied across contexts, the less they participated in ethnic and
cultural student support groups.
Social Connectedness
From the view of the entire sample, social connectedness was significantly and
negatively correlated with participation in ethnic and cultural student support groups, but
positively with the SACQ-full scale, and the four SACQ subscales. Thus, increases in
social connectedness were associated with less activity in ethnic and cultural student
support groups, but greater adjustment and adaptation to the college setting according to
the SACQ (see Table 9).
57
With regard to the correlation analysis by sex, social connectedness was
significantly correlated with the perceived availability of student groups, overall
adjustment, social adjustment, personal-emotional adjustment, and institutional
attachment, for males. For females, social connectedness was significantly correlated
with participation in ethnic/cultural groups, overall adjustment, social-emotional
adjustment, academic adjustment, personal-emotional adjustment, and institutional
attachment (see Table 10).
Facilitative Supports
With the exception of perceived support from student support groups, the pattern
of relationships with depression for females was similar to the overall sample's pattern of
relationships. However, for males, the pattern was a bit more discrepant (see Table 9).
College Adjustment
The four SACQ subscales were significantly and positively correlated with each
other (see Table 9). These four subscales were significantly and positively correlated
with social connectedness. It is also interesting to note that each of the four subscales
was significantly and negatively correlated with depression. The pattern of relationships
between the four SACQ subscales and other study variables for males was similar to that
for females (see Table 10).
Depression
With respect to the entire sample, PHQ-9 depression scores were significantly and
negatively correlated with perceived support from student support groups, connectedness
to student support groups, and each of the SACQ scales (as noted above; see Table 9).
58
Regarding sex, the pattern of relationships between PHQ-9 depression scores and other
study variables were similar. Of note, for females depression was significantly negatively
correlated with each of the full scale and four subscales of the SACQ, suggesting an
inverse relationship between depression and overall college adjustment, academic
adjustment, social adjustment, personal-emotional adjustment, and institutional
attachment (see Table 9). For males, depression scores were only significantly and
negatively correlated with the SACQ-Full Scale and SACQ- Institutional Attachment
subscale (see Table 10), suggesting an inverse relationship between depression and both
overall adjustment, and institutional attachment.
Table 9
Sample Means, Standard Deviations and Intercorrelations ofall Measured Variables
Variable M SD I 2 3 4 5 6 7 8 9 10 I I 12 13 14
I. Multiracial ID-VAR 1.79 .71
2. Social connectedness 4.25 .84 -. I I
3. Activity in SG 2.94 1.03 -.04 .08
4. SG availability 4.17 1.55 -.12 .13 .22**
5. Participation in SG 1.35 1.36 .07 .06 .41 ** .38**
6. Participation in E/C SG 2.82 1.07 -.12 -.15* -.53** -.18* -.62**
7. Support from groups 2.57 .97 -.13 .11 .42** .21** .24** -.42**
8. Connectedness to groups2.53 .93 -.12 .02 .47** .17* .35** -.44** .74**
9. Usefulness student servo 3. I 9 .53 -.12 .09 .39** .19** .29** -.28** .25** .25**
10. SACQ-full scale 6.05 1.09 -.03 .60** .00 .00 .01 -.06 .07 .04 -.04
11. SACQ-academic 5.90 1.22 -.04 .37** -.10 -.11 -.02 .06 .04 .05 -.03 .85**
12. SACQ-social 6.14 1.37 -.07 .71** .14 .15* .10 -.22** .17* .05 .09 .80** .45**
13. SACQ-personallemot. 5.38 1.47 -.02 .43** -.05 -.06 -.06 .02 .00 .02 -. I 3 .84** .64** .54**
14. SACQ-institute. attach. 6.25 I.I6 .04 .58** .07 .1 I .05 -.12 .07 .02 -.01 .84** .57** .83** .61 **
15. Depression .98 .77 .07 -.13 -.12 -.07 -.02 .06 -.16* -.20** .02 -.35** -.24** -.25** -.42** -.27**
Note. N = 188. ID-VAR= Identity Variance; SG = Student Groups; E/C = Ethnic and Cultural; SACQ = Student Adaptation to College Questionnaire.
*p < .05.
**p < .01.
VI
\0
Table 10
lntercorrelations orall Measured Variables by Sex
Variable I 2 3 4 5 6 7 8 9 10 II 12 13 14 15
1. Multiracial ID-VAR - -.12 .03 -.06 .13 -.16* -.04 -.03 -.14 -.05 -.06 -.05 -.10 .08 .09
2. Social connectedness -.06 - .14 .07 .08 -.20* .16 .11 .15 .62** .40** .72** .48** .58** -.15
3. Activity in SO -.21 -.08 .27** .35** -.51 ** .36** .45** .31 ** .08 -.04 .22** .03 .14 -.13
4. SO availability -.31 * .34* .08 .43** -.23** .21* .22** .25** -.05 -.12 .08 -.10 .06 -.06
5. Participation in SO -.13 .04 .58** .20 -.64** .22** .34** .24** .03 -.01 .13 -.07 .08 -.02
6. Participation in E/C SO .03 .03 -.60** .00 -.51 ** -.41 ** -.50** -.24** -.11 .00 -.25** -.01 -.15 .08
7. Support from groups -.34* .00 .55** .21 .31 * -.47** .77** .14 .10 .09 .17* .05 .07 -.16
8. Connected. to groups -.32* -.24 .48** .05 .37* -.24 .62** .21 * .11 .13 .13 .05 .06 -.19*
9. Usefulness student servo -.06 -.10 .57 -.02 .46** -.39* .58** .31 * .00 .01 .15 -.12 .03 .03
10. SACQ-full scale .06 .54** -.20 .19 .01 .07 .02 -.14 -.11 .85** .80** .84** .84** -.35**
II. SACQ-academic .10 .25 -.25 -.09 -.01 .27 -.12 -.20 -.08 .81 ** - .47** .63** .58** -.24**
12. SACQ-social -.10 .71 ** -.06 .40** .04 -.19 .22 -.16 -.03 .78** .34* - .56** .83** -.25**
13. SACQ--personal/emot. .18 .30* -.20 .10 .02 .07 -.07 .02 -.12 .85** .69** .47** - .63** -.44**
14. SACQ-institute. attach. -.02 .63** -.06 .36* .06 -.09 .14 -.05 -.02 .81** .44** .87** .56** -.27**
15. Depression .06 -.07 -.22 -.13 -.11 .01 -.28 -.43** -.13 -.31 * -.23 -.24 -.28 -.31 *
Note. N =188. ID-VAR =Identity Variance; SO =Student Oroups; E/C =Ethnic and Cultural; SACQ =Student Adaptation to College Questionnaire. servo =
services; emot. =emotional; Correlations for males (n =44) are below the diagonal and correlations for females (n =144) are above the diagonal.
*p < .05.
**p< .01.
0\
0
61
Regression Analyses
Regression analysis should be based on theoretically supported hypotheses about
the connection between variables (Field, 2005). Based on Root's ecological model (2003)
there was theoretical basis to examine how sex, perceptions of facilitative supports, and
college adjustment scores explain multiracial identity variance, social connectedness, and
depression. Thus sex, student perception of student group availability, participation in
ethnic and cultural student groups, perceived support from student groups, and
connectedness to student groups were employed as predictors of multiracial identity
variance, social connectedness, and depression, per exploratory research question seven
(see p. 26). The following section enumerates the results of three simultaneous regression
analyses used to examine the relative contribution and amount of variance explained by
these variables (see Tables 11-13). Because the criterion variables (multiracial identity
variance, social connectedness, and depression) were not significantly correlated with
each other, they were not included as predictors in the regression.
Explaining Multiracial Identity Variance
To explain statistical variance in the construct "multiracial identity variance" a
simultaneous multiple regression model was computed using sex, and four facilitative
support variables (student group availability, participation in ethnic/cultural student
groups, support from groups, connectedness to groups) as five different predictor
variables (see Table 11). Participation in ethnic and cultural student support groups was
the only variable to significantly contribute to explaining approximately 7% of the
variance in the construct "multiracial identity variance," R2 =.07, F(5, 180) =2.56, p <
62
.05. The equation indicates increased participation in ethnic and cultural student support
groups explains lower multiracial variance across contexts.
Explaining Social Connectedness
To explain variance in social connectedness a simultaneous multiple regression
analysis was computed using the seven different predictor variables: sex, support from
groups, connectedness to groups, academic adjustment, social adjustment,
personal/emotional adjustment, and institutional attachment (see Table 12). This test
revealed that the model of predictors accounted for 52% of the variance in social
connectedness, R2 = .52, F(7, 190) = 29.46,p < .001. However, social adjustment
appeared to be the only significant contributing variable in the model, thus indicating a
direct relationship between the two, whereby higher social adjustment explains higher
social connectedness.
Explaining Depression
To explain variance in depression a simultaneous multiple regression model was
computed using seven different predictor variables: sex, support group groups,
connectedness to groups, academic adjustment, social adjustment, personal/emotional
adjustment, and institutional attachment (see Table 13). This test revealed that the model
of predictors accounted for 23% of the variance in depression, R2 = .23, F(7, 178) = 7.66,
P < .001. Connectedness to groups appeared to contribute to the significance of the
model at a marginal level, whereas personal and emotional adjustment appeared to
explain the model most significantly. This suggests both connectedness to student
groups, and personal-emotional adjustment moderately explain level of depression.
Table 11
Simultaneous Regression Results Summary ofMultiracial Identity
Variance Regressed onto Sex and Four Components ofFacilitative Support
Group Variables
Predictor B SE ~ t
Sex .05 .12 .03 .41
Student group availability -.04 .04 -.09 -1.18
Participation in E/C SG -.16 .05 -.24 -2.90*
Support from groups -.09 .08 -.12 -1.11
Connectedness to groups -.08 .08 -.11 -1.01
Note. For gender, males were coded as zero and females were coded
as one. E/C SG = Ethnic and Cultural Student Groups.
*p < .01.
63
Table 12
Simultaneous Regression Results Summary ofSocial Connectedness
Regressed onto Sex, Facilitative Support Variables and Student
Adaptation to College Variables
Predictor B SE ~ t
Sex .08 .10 .04 .81
Student group availability .02 .03 .04 .84
Participation in E/C SG .01 .04 .01 .22
SACQ-academic .05 .05 .07 .96
SACQ-social .45 .06 .74 7.84*
SACQ-personal/emotional .04 .04 .06 .86
SACQ-institutional attachment -.07 .07 -.10 -1.00
Note. For gender, males were coded as zero and females were coded as one. E/C=
Ethnic and Cultural Student Groups. SACQ = Student Adaptation to
College Questionnaire.
*p < .001.
64
Table 13
Simultaneous Regression Results Summary ofDepression Regressed
onto Sex, Support and Connectedness from Student Groups, and
Student Adaptation to College Variables
Predictor B SE ~ t
Sex .12 .12 .07 1.04
Support from groups -.04 .08 -.05 -.49
Connectedness to groups -.14 .08 -.18 -1.79t
SACQ-academic -.01 .06 -.02 -.21
SACQ-social .02 .07 .03 .27
SACQ-personal/emotional -.22 .05 -.43 -4.50*
SACQ-institutional attachment .01 .08 .02 .12
Note. For gender, males were coded as zero and females were coded
as one. SACQ = Student Adaptation to College Questionnaire.
tp < .08.
*p < .001.
Summary of Results
The results demonstrate that, when presented with Zack's (2006) multiracial
identity options, participants tended to self-identify with Fractional (44.7 %) and
Inclusive (33.7) identity, while only one participant identified as White (see Table 6).
When participants were asked how they would self-identify in different interpersonal
contexts, they again tended to identify with Fractional and Inclusive identity, but
65
66
endorsement of these identities decreased in favor of other multiracial identity options.
This suggests multiracial students choose different multiracial identity options depending
on their context. There were differences in multiracial identity option endorsement based
on whether the participants were asked to endorse multiracial self-identity vs. their
reflections of others' perceptions of their multiracial self-identity. Within the multiracial
college student sample, males and females demonstrated similar levels of multiracial
identity variance across contexts, suggesting: (a) males and females both vary in their
chosen multiracial identity across contexts, and (b) males and females do not differ in the
way they vary their chosen multiracial identity across interpersonal contexts.
Overall, participants in this study demonstrated a greater tendency to feel socially
connected vs. disconnected (Lee, Draper, & Lee, 2001) and a greater tendency to be
involved in facilitative supports than not. Participants also demonstrated average levels of
adjustment when compared to the normative sample stratified by sex (Baker & Siryk,
1999), and the sample expressed low levels of depression.
Females reported significantly greater active participation in student support
groups and greater perceived usefulness of student support services than males. Males
reported significantly higher levels of institutional attachment than females.
Overall, multiracial identity variance was not significantly correlated with any of
the other study variables. Male multiracial identity variance was significantly and
negatively correlated with perceived availability of student support groups, perceived
support from student support groups, and connectedness to these groups. Female
67
multiracial identity variance was only significantly and negatively correlated with
participation in ethnic and cultural student support groups.
Social connectedness was significantly and negatively correlated with
participation in ethnic and cultural student support groups, but positively with college
adjustment. Male social connectedness was significantly correlated with the perceived
availability of student groups, adjustment, and institutional attachment. Female social
connectedness was significantly negatively correlated with participation in ethnic/cultural
groups, and positively correlated with all measure of college adjustment.
Depression was significantly and negatively correlated with perceived support
from student support groups, connectedness to student support groups, and adjustment.
Increased participation in ethnic and cultural student support groups significantly
explained decreased multiracial identity variance. Higher social adjustment explained
greater social connectedness. And finally, higher personal-emotional adjustment
explained lower depression.
68
CHAPTER IV
DISCUSSION
In this chapter I discuss the outcome of this study using the following
organizational framework. First, I present a summary of the results as they relate to the
proposed exploratory research questions, while discussing overall implications as they
relate to theory, previous research, and interventions for traditional-age multiracial
college populations. I then outline more specific suggestions for intervention, prevention,
and college and university programming for the problems identified in this study. Finally,
I present some of the limitations of this particular study, highlight ideas for improving
research in this area, and articulate ways to enhance and extend this study's findings.
Main Findings: Relationship Among Variables
In the current study I was interested in exploring if relationships existed among
multiracial identity variance, college adjustment, social connectedness, facilitative
supports, and depression in multiracial college students. Overall, the sample tended to be
socially connected, was in the average range of adjustment to college, and exhibited low
levels of depression.
Demographics
Overall the sample was racially diverse, predominantly female, and came from
highly educated parents. The sample was racially diverse in the sense that multiple
combinations of multiracial heritage were represented. This is important in validating the
69
sample as a cross section of the multiracial population, which distinguishes the current
study from previous multiracial identity research that has focused specifically on a
limited representation of specific bi- or multiracial sub-groups, such as black/white
biracial individuals (Shih & Sanchez, 2005; Wardle & Cruz-Jansen, 2004). The sample
tended to consist of participants with highly educated parents and as a result were likely
to be of higher socioeconomic status. It is unknown if the general socioeconomic status
of the sample is representative of the multiracial college student population as a whole.
Also, women outnumbered men three to one. These factors are important when
considering the generalizabiity of these results.
One striking finding from this study was the noted difference between a
participant's self-identity, and their perception of society's view of their self-identity. On
the whole, Fractional identity and Inclusive identity were either alternatively the first or
second most frequently endorsed self-identity options endorsed by study participants,
depending on the context. Across the sample, general self-identity was 44.7% Fractional
and 33.7% Inclusive, with only 5.5 and 0.5% of the sample choosing the monoracial
options of Traditional Non-white and White, respectively. This finding suggests that there
was a discrepancy between how participants were identifying themselves racially, and
how they perceive others categorize them racially. Based on these findings, participants
believed that "society" most often categorizes them in a monoracial category, with 47.2%
of the sample endorsing Traditional Non-white, and 17.6% endorsing White. As such,
there appears to be a gap between multiracial persons' awareness of their own identity
options, and what they believe others recognize as their identity options. That is,
70
multiracial college students perceive that strangers, peers, classmates, professors, and
society in general have a narrow concept of the spectrum of multiracial identity. Across
all contexts, very few participants reported White self-identity with frequency ranging
from 0.5% to 2%. When others' perceptions were considered, the range of White
frequency increased from 2% to 17.6%.
These results may most likely be explained by examining the role of (a)
phenotype, and (b) "language barrier" in multiracial nomenclature (e.g Root, 2003a,
Zack, 2001). First, taking into account the significance of phenotype in the multiracial
self-identity, students who appear phenotypically more white (based on skin color, hair
texture, eye shape, etc) may not identify as such, but others' may identify them that way.
Root's (2003a) Ecological Frameworkfor Understanding Racial Identity highlights the
role of phenotype in multiracial identity development, noting that phenotype can be both
a social facilitator and/or barrier to racial group acceptance and affiliation. Phenotype can
influence a multiracial students' self-identity, as well as the identity others' (particularly
monoracial others') project onto them, causing a racial socialization dialectic that can
shape multiracial students' identity. Racial socialization, is also highlighted in Root's
framework (2003a), and is defined as the process by which multiracial individuals
develop a healthy multiracial identity.
Secondly, this discrepancy in self- and others' perceptions of multiracial identities
may also be the result of a "language barrier" in the multiracial identity nomenclature, as
the concept of multiracial identity "options" (Renn, 2004; Zack, 2006) challenges
conventional wisdom of the "check one box" paradigm (Gaskins, 1999). The results of
71
this study suggest that multiracial college students believe others place them into
monoracial categories, without consideration of the multiple aspects of their heritages.
This supports and extends the writings of Wallace (2003), who posits that the imposition
of group norms of identification (i.e. monoracial identity) and solidarity are often
employed, reinforced, and challenged on college campuses. This discrepancy may
influence developmental trajectories of students' racial identities, as well as adjustment
and, therefore, warrants further study to enumerate the multidimensional nature of
multiracial identity (Renn, 2004; Root, 2003a). The tension caused by discrepancy in self
and others' perceptions of multiracial identity is summarized by a young multiracial
woman quoted in Gaskins (1999) who states, '''Being biracial isn't hard because we're
confused about our racial identity. It's hard because everyone else is confused. The
problem isn't us-it's everyone else'" (p. 15). The language barrier issue along with
other study findings made evident in the current study, and supported by previous
research, may indicate that there is a need to expand overall awareness of what
multiracial identity is, and that multiracial identity options exist for a variety of reasons
and in different contexts.
Multiracial Identity Variance
Examining the sample as a whole, multiracial identity variance was not
significantly correlated with any of the other study variables. Hence, changes in
multiracial identity across various contexts were not related to adjustment, social
connectedness, depression, or the influence of facilitative supports. In and of itself,
evidence of multiracial identity variance as a phenomenon allows for the possibility
72
fluidity in multiracial identification and lends support to Zack's (1998, 2006) notion of
microdiversity, whereby multiracial individuals are understood to be "diverse within
themselves and not merely diverse as members of groups that are believed, in often
erroneous ways, to be racially different from other groups." (Zack, 1998, p.82). These
findings suggest that variance in multiracial identification, and therefore microdiversity,
does not have a significant relationship with the psychosocial adjustment and mental
health of multiracial college students. While these results neither confirm nor deny
whether multiracial identity variance protective in maintaining psychosocial health,
previous research has demonstrated that multiracial youth generally exhibit positive
adjustment and well-being (Field, 1996; Cauce, Hiraga, Mason, Aguilar, Ordonez, &
Gonzales, 1992; R. C. Johnson, 1992; Johnson & Nagoshi, 1986; Stephan, 1992). In
previous studies, multiracial youth have demonstrated (a) positive self-concepts, (b) no
difference in life distress, general distress, behavior problems, or self-worth than
monoracial peers; (c) no differences on aspects of personality when compared to
monoracial peers, and (d) no difference in ability to experience positive peer relationship
compared to monoracial peers. Shih and Sanchez's (2005) comprehensive review of the
literature on multiracial identity and adjustment revealed that among studies sampling
from non-clinical populations (as in the current study) there was little evidence that
multiracial individuals in the general population were dissatisfied, unhappy, or
uncomfortable with their racial identity. The current study's failure to refute these
previous findings, as well as the lack of a demonstrated relationship between multiracial
identity variance other study variables is important contribution to understanding the role
73
racial identity plays in the psychosocial health and development of multiracial college
students.
Root (1990, 1997, 1998,2003) has proposed that multiracial identity options are
each considered healthy if the multiracial individual is: (a) accepted in their chosen
group, (b) does not feel pressure to change, and (c) does not deny or put down one aspect
of their heritage, and instead embraces all aspects of their identity, and (d) is supported in
their identity by their immediate environment. Though this is beyond the scope of the
current study, it is possible that these factors influenced the outcomes across all variables.
Strikingly, level of participation in ethnic and cultural student groups was the only
predictor to significantly explain multiracial identity variance, with higher participation
explaining lower multiracial identity variance. Thus, multiracial students who participate
in ethnic and cultural student groups appear to less demonstrate less fluidity in their
multiracial identity across contexts and vice versa.
This inverse relationship between multiracial variance and participation in ethnic
and cultural student groups (a facilitative support variable), suggests that multiracial
students who were more fluid in their identity were less likely to receive support from the
types of student supports those in the higher education community traditionally expect to
foster healthy identity and adjustment for students of color, such as ethnic and cultural
student groups (Nishimura, 1998; Wallace, 2003). This reflects Wallace's (2003) view
that in the college context, interethniclinterracial group relations often take on a
heightened salience and intensity. In these groups there may be a higher emphasis on
group solidarity, and mixed heritage students may encounter biases within their heritage
~---------
74
communities due to minority, majority, or multiple ancestries. Racial identity politics-
coalition and organization around shared experiences of injustice of members of
marginalized racial groups with the aim to assert or reclaim greater self-determination
(Heyes, 2007)-- based on perceived racial authenticity may influence student
opportunities for identification and participation in a community.
In Renn' s (2000) qualitative study of situational identity (i.e. fluid or context
based identity) among multiracial college students, participants spoke of finding space-
both physical and psychological-to fit in. Space was both a public and a private concept.
The former, were spaces where peer culture was present: residence halls, student
organizations, classrooms, and social events. The latter, were spaces used to define
students own identities where they, as individuals, sorted through the meanings of peer
culture, family background, and personally to derive notions of culture, race, and self.
Public spaces shaped students' sense-making in private spaces, but students also brought
their privately held ideas about race, culture, and identity into the public spaces on
campus.
This public-private dichotomy mirrors aspects of Root's (2003a) Ecological
Frameworkfor Understanding Racial Identity, which suggests that there are visible and
invisible factors that contribute to multiracial identity development. According to Renn
(2000), the three main elements of public space-making are shared culture, physical
appearance, and participation in legitimizing activities. Therefore, common cultural
knowledge, similar physical appearance, and group involvements, define who can occupy
which public spaces. For example, in Renn's (2000) study a cultural knowledge deficit
75
was just as powerful in keeping a student out of a certain space, as cultural knowledge
proficiency was at keeping them in. Physical appearance also emerged as a marker of
racial/ethnic authenticity, and determined who could belong and who could not. This may
provide some explanation for the inverse relationship between ethnic and cultural student
group participation and multiracial identity variance found in the current study.
Accordingly, multiracial participants in the current study who evidenced fluid racial
identity may have been less likely to participate in ethnic and cultural student groups
which can ostensibly reinforce static, monoracial identity.
Wallace (2003) points out that college can be considered a criticalleaming
period, as multiracial students leave their families to enter a social context in which the
expectations and roles related to group membership are especially concentrated. In
college contexts, cultural legitimacy and loyalty can become increasingly important in
such spaces, where these students' physical appearance (as well as interactional style or
ancestry) can stigmatize them as culturally suspect, ambiguous, or even invisible to other
group members. Additional research is required to better understand the impact these
sociocultural dynamics have on multiracial college students' development and is further
discussed later in the chapter.
Social Connectedness
The current study's incorporation of the construct of "social connectedness"
represents an extension of both the multiracial identity and social connectedness
literature. On average, study pmticipants appeared to feel socially connected and males
and females did not differ in their level of social connectedness. This study extends
76
current research on social connectedness, by identifying a relationship between social
connectedness and college adjustment for multiracial college students. Social adjustment,
which measured participants' success in coping with the interpersonal societal demands
inherent in the college experience (Baker & Siryk, 1999), was directly associated with
social connectedness. This makes sense, as social connectedness is defined as an aspect
of the self that reflects subjective awareness of the interpersonal closeness with the social
world as a whole (Lee & Robbins, 1995, 1998). This finding therefore supports the utility
of the Social Connectedness Scale-Revised and the Social Adjustment subscale of the
SACQ for cross validation of social connectedness and social adjustment as constructs.
Accordingly, further confirmatory factor analysis, and concurrent validity studies are
warranted in this respect.
The finding that higher tendency to feel socially connected was related to higher
levels of college adjustment across all domains supports findings of previous research, by
Karcher and Lee (2002), which suggested that greater connectedness is related to
psychological happiness, increased physical health, and better coping skills among older
adolescents and young adults.
On the other hand, an unusual finding of this study was that for the entire sample,
higher levels of participation in ethnic and cultural student groups was related to lower
social connectedness. Social connectedness theory would suggest that participation in
groups would have a buffering effect on students who are not well connected with others,
and so participation would seem to be related to greater social connectedness (Lee &
Robbins, 2000). The current study's findings, may imply that perhaps the students
77
involved in groups were trying to enhance their interactions with, and supports from,
similar others in order to increase their social connectedness, but had not yet achieved
this. Alternatively, these findings may also indicate that those who were more socially
connected, in fact, had less reliance on these groups for facilitative supports.
At face value this appears counterintuitive, and indeed for monoracial ethnic and
racial minorities on college campuses it is often essential that they have a space in which
to connect with others who look like them and/or share similar cultural experiences
(Nishimura, 1998; Wallace, 2003). Considering the complex interface of ecological
factors involved in multiracial identity development, as well as the dialectic between
public and private expression of multiracial identity outlined by Renn (2000), future
research should further explore the manner in which multiracial individuals achieve
social connectedness and how it may compared to their monoracial minority and majority
peers. Indeed, Townsend and McWhirter (2005) point out that there are cultural
differences in the means by which people achieve connectedness. Greater interpretation
of these findings would be beyond the scope of this study. I did not ask participants how
long they had been involved in or the quality of their experience with student support
groups. Further research is needed to identify any potential mediational pathways
contributing to social connectedness in multiracial college students to better understand
the experience of social connectedness among this population.
Facilitative Supports
In the current study, I was interested in the role of perception of and participation
in facilitative supports on college campuses (e.g. student groups, various campus support
78
services, etc.), and their relationship with other variables studied. The findings indicate
that multiracial college students perceive a moderate level of facilitative support
availability in their college campus environments. However, few participants appeared to
participate significantly in general student support groups that were available to them.
Participation specifically in ethnic and cultural student support groups was more
encouraging, as the sample reported a significantly greater level participation in these
groups than general student groups. However, participants only perceived moderate
support from student support groups and only felt moderately connected within student
support groups. Overall, the findings suggest that when participants did participate in
student support groups, they were more actively involved in ethnic and cultural student
groups than in other types of campus groups. This makes sense vis-a-vis a participants'
likely status as a student of color, and the need to share identity based space with like-
others. Interestingly however, increased involvement in ethnic and cultural student
groups explained lower multiracial identity variance, implying a direct connection
between multiracial student's participation in ethnic and cultural groups and a more static
racial identity.
College Adjustment
Each of the four subscales on the SACQ-- academic adjustment, social
adjustment, personal-emotional adjustment, and institutional attachment-- were
significantly positively correlated with social connectedness. As mentioned previously,
these findings support research that has demonstrated a connection between social
connectedness and psychosocial wellness (Karcher & Lee, 2002). Of additional interest is
79
Institutional Attachment, which measures students' feelings about being in college in
general, their feelings about the particular college they are attending, as well as their goal
commitment (Baker & Siryk, 1999). This is an important construct when considering
issues of multiracial college student retention. Low retention and graduation rates among
underrepresented minority students are reported to be the result of many factors,
including academic unpreparedness (Lee, 1991; Priest & McPhee, 2000; Thomason &
Thurber, 1999; Simon, 1993), and lack of adequate social and academic support (Lee,
1991). Understanding the relationship between factors such as social connectedness and
institutional attachment will assist college support personnel with retention efforts and
warrants further study.
Depression
In the current study I explored depression because I wanted to understand the
relationship of this risk factor with the other primary study constructs in multiracial
college students. Overall, the findings indicated that the multiracial college student
participants in this study experienced low levels of depression. Previous literature on the
relationship between multiracial identity and depression, has yielded inconsistent, and in
some cases, discrepant conclusions (e.g. Shih & Sanchez, 2005). In their review of the
qualitative literature on multiracial identity and depression Shih and Sanchez (2005)
surmised that similar to the population in general, non-clinical multiracial samples
experienced lower levels of depression than multiracial clinical samples. In their survey
of the quantitative literature, Shin and Sanchez (2005) suggested the following trend:
Multiracial individuals tended to show higher levels of depression when compared with
80
their monoracial majority peers, but not when compared with their monoracial minority
peers. As a cautionary note, their findings were based on the examination of only four
quantitative studies, and the samples of many of the qualitative and quantitative studies
were limited in the type and number of multiracial member combinations represented.
While the current study did not compare multiracial college students to their monoracial
majority and minority peers, the findings support Shin and Sanchez's observation that
non-clinical multiracial populations exhibit low levels of depression in general, and lower
levels of depression than their monoracial minority peers.
With regard to depression and college adjustment, each of the four SACQ
subscales was significantly and negatively correlated with depression, meaning higher
levels of adjustment in each of the four domains were related to lower levels of
depression. The results of the current study indicated that only personal-emotional
adjustment to college significantly explained depression. This result is congruent with the
literature in that personal-emotional adjustment measures how a college student is feeling
psychologically and physically, as well as their experience of general psychological
distress and/or any associated somatic problems (Baker & Siryk, 1999), and is therefore
tantamount to depression. It then makes logical sense that one might explain the other.
Sex Differences
In the current study I was interested in exploring potential sex differences in the
relationships between study variables among multiracial college students. Overall there
appears to be few differences between the male and female participants across study
variables. I found two significant sex differences: Women perceived higher usefulness of
81
student services, and men endorsed higher levels of activity in student groups and higher
institutional attachment. While the current literature on the multiracial population has
discussed sex differences influencing the multiracial experience, explanations for these
differences are conceptual in nature. The following discussion of the influence of sex on
the variables studied extends this literature, and generates hypotheses that can guide
future research.
I found that the more a male participant varied in his multiracial identity choice
across contexts, (a) the less he perceived student groups as available to him, (b) the less
he felt supported by them, and (c) the less he felt connected to these groups. Also, when
male multiracial college students perceive a lack of facilitative support on campus, they
were more likely to express a range of multiracial identity options across contexts. There
are number of possible explanations for this trend. It is possible that males with less
facilitative support affiliations were freer to vary in their identification because their
racial self-identity choices were not constricted by specific groups. Alternately,
multiracial males with higher identity variance may not have been drawn to participate in
student groups where they perceived a static identity (especially monoracial) is desired,
or expected. Also, these findings may reflect a dearth of services available to male
multiracial students and/or their underutilization of available services supporting research
that students of color (particularly males) underutilize campus support services (Ross-
Gordon, 1998). Though the current study does not demonstrate a causal relationship
between these variables, it is possible that male students who participate in student
groups either find less need to vary their multiracial identification, or are may influenced
82
by frequent contact with peers who are less apt to vary identification and with whom they
identify. These peers may identify as monoracial, or multiracial with static identity.
For females, findings suggest that those who had a higher variation in their
multiracial identity across contexts were less likely to participate in ethnic and cultural
student support groups. With respect to females, the results suggest low participation with
ethnic and cultural student groups were related to increased fluidity in multiracial
identification. Similar to the findings of Renn (2000) and observations of Wallace (2003),
it is possible that female students who do not conform to static, monoracial identity do
not participate in groups where monoracial authenticity is perceived as desired or
required for membership. Further research is warranted in this respect.
In sum, important trends emerge when taking sex differences into consideration: (a) an
association between afflation with student groups and a tendency to have a more
consistent multiracial identity across contexts exists for multiracial students, and (b) the
nature of this association differs by sex.
Implications of the Findings
The current study has many implications regarding the psychosocial factors
contributing to multiracial identity and its expression in different contexts. As a
correlational study, causal relationships between variables cannot be inferred. The
decision to include all of the participants into a general "multiracial" category was made
so that analyses could be effectively and reliably conducted to answer the principal
exploratory questions of this study. Looking at multiracial college students collectively,
however, as opposed to examining specific racial combinations or smaller groups, is not
83
intended to imply universality in the experiences of multiracial people. In fact, there are
certainly specific and important ecological influences that impact the who, what, where,
when, why, and how of microdiversity and the effects of microdiversity on people in their
daily lives. The current study, in fact, reinforces the notion that race, as a social
construction, is not a viable construct for defining a single group of people, nor for using
that group definition to then try to predict predisposition to psychosocial stressors, poor
adjustment, and other experiences. Nevertheless, until larger samples of more specific
groupings of students are assessed, the findings of this study shed a great deal of light on
some aspects of the experiences of people of multiple racial and ethnic backgrounds.
The aforementioned ecological influences likely to contribute to multiracial
identity development have to do with constructs such as sex, and environmental factors
such a facilitative supports available to and used by multiracial students. The current
study demonstrates that flexibility in racial identification is related to a decreased need
for student facilitative supports services. Taking this into consideration, student support
personnel in higher education (advisors, counselors, residence assistants, administrators,
etc.) may want to reconsider referring multiracial students to support groups based on
assumption that they will benefit from connection with other students of color who
identify as monoracial. For example, referring a biracial black/white student to the
campus black student union, may not address the unique influence multiracial identity
has on a students sense of connectedness to peers along racial, ethnic, and cultural lines.
In Renn's (2000) qualitative research on identity-based space and multiracial
college students, she observed that while appearances quickly marked someone as not
84
belonging in a particular monoracial group, most students also looked "ethnic" enough to
be perceived as "not white." This perception created space for them to belong to a general
community of students of color. Through participation in certain activities, clubs or
classes (e.g. social, education, or political events) on each campus, students negotiated
the boundaries of various communities. Also, Renn found that joining a group
specifically for multiracial students, marked students as not fitting in. Among the
participants interviewed, belonging to a group was considered antithetical to fully
belonging to the black student organization or to one of the nationality-specific Asian
student groups (Korean Students Association, Japan Club, etc.), for example. The
multiracial students felt that they could fit into monoracial groups only if they did not
claim membership in a multiracial student group, or otherwise assert their multiple-
heritage background. In light of Renn's work, the current study's findings regarding
multiracial identity variance may be interpreted as assertion of mixed race heritage,
therefore underpinning the current study's finding that a significant inverse relationship
exits between multiracial identity variance and participation in student groups-- and
specifically for women, participation in ethnic and cultural student groups.
Renn (2000) posits that the creation of a public multiracial space requires a
critical mass of students who are willing to identify as multiracial and who feel a need for
a separate multiracial space. The combination of demographic factors (e.g. the number of
multiracial students) and peer culture highly influence the experience of multiracial
students. It is important for higher education support staff to consider that, (a) extant
student groups/supports (particularly ethnic and cultural supports) may not be effective in
85
addressing the needs of multiracial college students, (b) multiracial college students may
benefit from a group specifically for multiracial students, (c) joining such a group may
have social and identity based consequences for multiracial students, and (d) a critical
mass of students is required to sustain such a group. According to Renn (2000):
If it is true that students benefit from maximum freedom to experience and
participate in different identity-based spaces, then information about how students
move in and out of communities on campus can be used to consider whether an
individual institution or higher education as a whole is meeting the needs of
multiracial students and others. (p. 415)
Accordingly, both Renn' s work and the findings of the current study highlight the
importance of colleges' and universities' efforts to decide how best to support the needs
of multiracial students by assessing their social, racial, cultural climates. This includes
assessment for a critical mass of students (and subsequent recruitment and retention
efforts if a deficit is identified), and providing appropriate identity-based space to foster
the overall growth and development of these students.
It is important for college support personnel, and counselors in particular, to
understand that expression of multiracial variance in multiracial youth can be
misunderstood and misinterpreted as signs of poor adjustment (Daniel, 1992; Stephan,
1992). Situational racial identity is often thought to be confusion rather than flexibility or
adaptability (Renn, 2004), thus diminishing the validity of microdiversity. The current
study demonstrates that variance in multiracial identity does not equate to identity
confusion, and is not related to the psychosocial well being of multiracial college
86
students. When studies have identified negative adjustment in multiracial youth, the
cause is typically conflict arising from their family or environment, or lack of guidance in
resolving developmental crises (Dimas, 1995; Field, 1996; Gibbs & Hines, 1992; Root,
1998; Root, 2001; Tomishima, 1999). Adolescents who had a white reference group but
had negative feelings about black persons had poorer adjustment. Thus, psychological
problems can arise when a mixed race adolescent internalizes stereotypes or negative
attitudes expressed toward them. Adjustment difficulties have been correlated with lack
of family conversation about, living in a single-parent family and having less contact
with non-custodial parents and relatives, and family strife (e.g. divorces, addicted parents
or siblings, absent parents, abuse, or adoptions). Therefore counselors and other SUpp0l1
staff should be careful to accurately assess psychological problems associated with
multiracial identity, in order to understand all of the ecological factors that may
contribute to a student's hardship.
Multiracial youth may need to develop a multiracial vocabulary to fully explore
their racial identity, such as Zack's (2006) multiracial identity options and Renn's (2004)
multiracial identity patterns. Indeed, the current study suggests that multiracial youth can
adopt a multiracial vocabulary to describe themselves, but a discrepancy exits between
their knowledge of this lexicon and society's knowledge of it. Talking about race can
give multiracial youth an outlet to discuss hurtful comments and recognize racial
stereotypes. According to Renn (2004):
When racial identity is associated with poor adjustment, it is often because family
dysfunction and traumas were color-coded at a young age. This color coding of
87
the dysfunction attaches to external negative stereotypes associated with being a
person of color. (p. 114).
It is crucial that college campuses do not repeat this dynamic, by fostering open dialogue
about race, and acknowledging multiracial identity as valid, healthy, and unique.
Study Limitations
There were a number of limitations to the current study. First, the sample is based
on the responses of traditional-age mixed-race college students, limiting the
generalizability of results beyond this group. Second, Root (2003b) outlines a number of
common flaws in research of mixed race people. These include: non-random distribution
of population, self-selection limits, generational changes in the meaning of mixed race,
heterogeneity versus homogeneity, the impact of nationality, immigration status, and
acculturation on racial/ethnic identity. While I sought to remedy many of these flaws in
the current study, self-selection bias, and heterogeneity versus homogeneity of the sample
emerged as limitations.
First, Mixed race people are not randomly geographically distributed in the
United States (U.S. Census Bureau, 2001). Despite my attempts to obtain a sample with a
broad geographical scope, self-selection into the study may have been a factor in this
sample (see Table 1). Second, the current study examined a heterogeneous mixed race
sample and therefore differences between individuals who identify with specific
combinations of mixed race were not explored. This sample is very representative of
most samples of mixed race heritage individuals in the U.S. in that the combination of
racial backgrounds was extremely diverse. It is unlikely that any valid between-group
88
comparisons could be made given the changing nature of the U.S. population without
recruiting extremely large samples.
As noted in the results section, the data from 29 pmticipants were eliminated from
inclusion in data analysis because they did not select more than one racial category on the
demographic survey. It is important to note that these participants endorsement of only
one racial category does not preclude them from having mixed race heritage. The
demographic survey used in this study, like others that ask participants to "choose all that
apply," did in fact establish a forced choice regarding racial identification. Students self-
selecting a monoracial racial category may have been expressing their chosen multiracial
identity option, as opposed to objectively delineating the full extent of their racial
heritage. While the aim of the race question on the demographic survey was the latter, it
is possible that this question may have served to recapitulate the very societal and
psychosocial dynamics that create neglect the presence of options for racial identification
for mixed race persons. The Multiracial Identity Options Scale (MIOS) which
accommodates options in identification, required that participants be self-reflective, and
results suggest that this has an impact on identification.
Finally, the MIOS was comprised of two categories: self and other perception in
various contexts. Because Multiracial Identity Variance was measured by examining the
variance in self-perception and other perception responses, it is possible that these results
were skewed by the difference between participant responses to self-perception questions
and other perceptions context. If multiracial identity variance results were separated into
89
self vs. other perception there might have been less variance in the data and therefore a
more flat overall profile of multiracial identity variance might have emerged.
Future Research and Intervention
Future research should explore the causal relationships between multiracial
identity variance and facilitative supports, social connectedness, and college adjustment.
The influence of facilitative supports in the current study raises a number of important
questions: (a) Are programs and services available for monoracial groups equally
available [and approachable] to multiracial students? (b) Do multiracial students have
equal access to resources, student organizations, and support? (c) How knowledgeable
are administrative and student support personnel (e.g. advisors, counselors, residence life
coordinators, etc) of the unique experiences of their multiracial college students? (d) Do
university communities support, or invalidate variance in racial identity? and (e) In what
ways do academic outcomes (e.g., retention, academic achievement, graduate rates, other
measure of student learning and development) vary for multiracial students compared to
the population as a whole, and to monoracial students?
To Improve the experiences and general well-being of multiracial college
students, Renn (2004) has also suggested a number of steps be taken on college
campuses. Colleges and universities: (a) can ask about racial heritage and racial identity,
(b) sponsor speakers on mixed race topics, cultural performances/displays, including
opportunities to participate, (c) challenge assumptions of monoraciality among faculty,
staff, and students, (d) encourage the incorporation of the complexity of race into
undergraduate learning which could aid in students' identity development, and (e)
90
increase awareness on campus of the national policies that impact mixed race students.
It is important to note, that the sheer number of possible combinations and
permutations of mixed race heritage makes it impossible to detect differences between
multiracial subgroups. The sample size required to obtain enough power to accommodate
all possible racial mixtures is impossible to obtain. This point is significant, as it
highlights the complexity of the mixed race, particularly as a subject of quantitative
study. It is unlikely future research will be able to successfully explore the aspects and
correlates of all mixed race identities.
Conclusion
This research study provides descriptive evidence that multiracial individuals
exercise variation in their chosen racial identity depending on different interpersonal and
social contexts. The results of this study shed important light on the concept of a racial
identity development process that is context based rather than stage based, and requires a
unique nomenclature that reflects the microdiversity of this population. Multiracial
identity variance does not appear to be related to psychosocial adjustment and wellbeing.
Yet, important, sex specific relationships among multiracial identity variance, social
connectedness, facilitative supports, college adjustment, and depression emerged, which
will inform current practice in regard to multiracial student support, challenge
conventional wisdom regarding healthy racial identity development, and guide future
research with this ever-increasing U.S. population.
91
APPENDIX A
REVIEW OF THE LITERATURE
In this section I review the literature regarding historical and contemporary
perspectives on human development and psychology, with specific attention to
multiethnic/multiracial identity development, and social connectedness. Because
literature reviews require clear parameters that are inclusive of the literature base of
interest (Jackson, 1980), I outline the review's search parameters and procedures.
Manuscripts selected for this review's initial content pool met the following
criteria: (a) published since 1967; (b) written in English; and (c) included the key words
"multiracial," "multiethnic" "biracial" "Interracial offspring," "mixed race," "ethnic
identity" "racial identity," "social connectedness" and "adolescence," "social
connectedness" and "college students," and "psychological adjustment". My
computerized literature search queried the PsychINFO and ERIC databases, yielding a
total of 43, 757 manuscripts, both quantitative and qualitative in nature. I then cross-
referenced these keyword searches to focus the search on the population and constructs of
interest. Next I examined the abstracts and ultimately selected, 15 books, 21 chapter
articles, and 32 journal articles for review, based on their relevance to the historic mission
of counseling psychology with an emphasis on prevention, intervention, person and
environment interactions, and multicultural competence, and (Brabeck, Walsh, Kenny,
and Comilang, 1997; Heppner, Casas Carter, & Stone, 2000; Gelso & Fretz, 2001). Other
92
writings listed in the references section were referred to for theoretical, methodological,
and statistical purposes, but were not included in the original review because they were
not within the parameters of the electronic search.
In the next section I discuss historical and contemporary perspectives on
multiracial/multiethnic identity development. I then discuss social connectedness as a
construct grounded in theoretical and empirical study. Through this review I discuss the
relationship between these relational/developmental factors and psychological health and
adjustment.
Multiracial Identity Development
On the whole, there is a dearth of research specific to the multiracial population.
According to Root (2003):
Contemporary thinking on racial identity development is derived from the
intersection of psychology and the racial pride movements of the 1960s and
1970s. The racial pride movements catalyzed solidarity within race and further
reinforced notions that one must identify with a single race The history of race
in the United States, and specifically the "one drop rule" provides the key to
understanding how psychological models evolved in a way that has excluded the
reality of many mixed race persons. (p. 34)
While scholarship in this area is increasing, the focus of research on multiethnic
and multiracial individuals has been focused on their racial and ethnic identity
development (Wardle & Cruz-Janzen, 2004). Each of the multiraciallmultiethnic identity
93
models is founded on the idea that these individuals face a unique challenge in
developing their sense of racial and ethnic identity.
But before examining issues unique to multiracial/multiethnic identity
development, it is important to examine the common processes involved in the early
childhood racial and ethnic identity development of all children.
Early Childhood Development ofRacial and Ethnic Identity
Learning a sense of racial and ethnic identity is a complex developmental process.
In order to understand the unique aspects of multiethnic identity development in the
United states it is important to understand how all individuals develop a sense of racial
and ethnic identity in early childhood. Aboud (1987) describes the following stage model
for normative development from infancy to age ten.
Infancy. During this period children can discriminate between dark and light
stimuli. Consequently, they can tell the difference between light and dark faces.
Ages 3 to 4. At this stage children can recognize (and are very interested in) the
physical racial differences between Black and White children, but are unable to discern
their own racial or ethnic identity.
Ages 5 to 9. During this stage of development, children learn to recognize their
own racial identity label and develop an initial awareness of group affiliation. Awareness
of group belonging occurs after the child learns about group similarities.
Age 7. At this stage, White and Black children are able to recognize each other by
racial distinction.
94
Age 8. By this stage children of other ethnic groups (Native American, Asian,
Hispanic, etc.) are able to recognize their peers who are from other ethnic groups.
Apparently, the salient features of these groups are less clear to children of these groups
than those of White and Black children.
Age 8 to iO. Children at this stage of development form a stronger sense of group
belonging and develop a sense of racial or ethnic constancy.
This model provides a general sense of how all children in the US begin to
incorporate the social construction of race, and the concept of ethnicity into a cognitive
schema. This occurs through a developmental process that includes normative
development and contextually influenced learning. While Aboud's (1987) model is a
cogent description of the early development of ethnic identity, it is based on a monoracial
paradigm. All children progress through various psychosocial, physical, and emotional
stages. Unique to multiracial and multiethnic children, however, is their development
toward a healthy multiracial and multiethnic identity (Wardle & Cruz-Janzen, 2004).
BlackIWhite identity Models
The majority of research on the multiracial population in the United States has
focused on African AmericanlEuropean American children. Poston (1990), Jacobs
(1992), and Kerwin and Ponterotto (1995) have each developed models of biracial
identity development in children of Black and White parentage. This presents some
difficulties with generalizability to all multiracial individuals. According to Wardle and
Cruz-Janzen (2004), there are two reasons for this lack of external validity: (a) there is
simply not enough research on other multiracial groups to form any body of knowledge,
95
and (b) research on the identity development of single race children suggests different
patterns for children whose physical features are more ambiguous with regard to racial
and ethnic background (see Aboud above). These models provided a valuable foundation
for modeling multiracial/multiethnic identity, but in and of themselves are not adequate
for understanding the diverse multiethnic population.
Poston's model. Poston's developmental model (1990) has five levels, or developmental
stages. At the time this model was developed the terminology for children of mixed
heritage was biracial and biethnic; today's terminology is multiracial and multiethnic. In
Stage 1, the Personal Identity stage, the child's sense of self, or identity, does not include
the concept of ethnic or racial belonging. In Stage 2, the Choice ofGroup Categorization
stage, biracial children are pressured to choose one background for their identity_ In this
case, biracial identity will not be the choice due to limited cognitive development. Most
often the race/ethnicity of choice is that of the parent of color.
In Stage 3, the Enmeshment/Denial stage, the individual experiences confusion,
self-dislike, and guilt from choosing the identity of one parent and denying that of the
other. Resolution of this sense of confusion and dislike must occur before the child can
move on to Stage 4, Appreciation. In this stage the individual may still identify with the
single group selected in Stage 2, but the child learns about and appreciates more of the
background of his or her other parent. Finally, in Stage 5, Integration, the child fully
integrates both sides of his or her heritage, viewing him- or herself with an identity that
includes the ethnic/ racial, cultural and family heritages of both parents and their
extended families.
96
Jacobs' model. Jacobs' model (1992) is based on data collected from studies
using dolls with biracial children of Black and White parentage. There are three
developmental stages in this model. At Stage 1, Pre-Color Constancy: Play and
Experimentation with Color (0-4 liz years old), children are developing the knowledge of
the physical color of their skin, but are not concerned with it. In this stage they cannot
match a doll's skin color to the skin color of another family member. At Stage 2, Post-
Color Constancy: Biracial Label and Racial Awareness (4 V2 to 8 years old), biracial
children learn that skin color is permanent. This results in ambivalence regarding their
own skin color. In order for the child to positively progress to the final stage, he or she
must resolve this ambivalence. At this stage, the child realizes his or her identity is
different from their parents, and the child in becoming increasingly aware of racial
discrimination.
At Stage 3, Biracial Identity (8-12 years old), a child has successfully developed
three conclusions: a) the color of his or her skin is related to his or her mixed heritage-
based on parent's genotypes; b) the color of his or her skin does not determine his or her
race; and c) the child is biracial because of the different racial heritages of both parents.
Kerwin-Ponterotto's model. This model is a six-stage lifespan model starting at
childhood and progressing to adulthood (Kerwin & Ponterotto, 1995). At Stage 1,
Preschool (up to age 5), biracial children become compare and contrast their appearance
and the appearance of other children. At stage 2, Entry to School, biracial children are
asked by their peers to explain their physical, and racial ambiguity ("what are you?").
97
Parents need to give them a term such as "mixed," or "biracial" to help them respond
proactively.
By stage 3, Preadolescence, biracial children are increasingly aware of physical
differences, including skin color and hair texture, and are beginning to understand
cultural determinants of group membership. These children reach the knowledge that
their parents belong to different racial groups. Often, a precipitating event, such as
experiencing racism, seems to force the child into full salience of his of her multiracial
identity. In stage 4, Adolescence, multiracial youth are pressured to identify with only one
ethnic or racial group. This stems from the natural identity crisis of this age and the need
for adolescents to associate with similar people. By stage 5, College! Young Adulthood,
young people may associate with friends mainly from one of their racial/ethnic
backgrounds, but feel less pressure to choose sides, feel comfortable in group of different
racial-ethnic people, and can generally see things from more than one point of view.
Finally, by stage 6, Adulthood, multiracial adults integrate aspects from every part of
their heritage, thus functioning comfortably in a variety of contexts. This is a conceptual
model with a basis in previous qualitative research by Kerwin, Ponterotto, Jackson, and
Harris (1993).
Inclusive Model-Phinney's Model
Phinney's model (1993) applies to single-race minority, multiracial, or
multiethnic adolescents and has three distinct stages through which the adolescent
progresses. In stage 1, Unexamined Ethnic Identity, the child has not examined his or her
ethnic identity and either a) has no interest in ethnic identity at all, or (b) automatically
98
accepts the view of the ethnic or racial identity that is assigned by others. At stage 2,
Ethnic Identity Search/ Moratorium, individual children search for information about
their ethnic or racial heritage and try to determine its relevance to them and their personal
identity. At this point, however, they are not yet ready to select an ethnic or racial
identity. By stage 3, Achieved Ethnic Identity, the individual chooses a racial or ethnic
identity and is totally confident and accepting of all aspects of that identity.
While this model attempts to capture the common experience in identity
development of all people of color, it does not fully account for the contextual influences
that contribute to identity formation. For multiethnic individuals, confidently choosing
and accepting all aspect of an ethnic identity is more complex if for no other reason than
they must choose from a broader selection of racial and ethnic possibilities. These
choices may also vary depending on the individual's context.
Multiracial/Multiethnic Identity Models
The two multiracial/ multiethnic specific identity models extant in the literature
are not singularly stage models. Root's Ecological framework for Understanding Racial
Identity (2000, 2003) is concerned with the impact of ecological factors-family, social
ability, school, sexual orientation, gender, class, community, peers, and physical
appearance-on the identity of multiracial individuals. Wardle's Model (Wardle, 1992;
Wardle & Cruz-Janzen, 2004) attempts to integrate all of the models previously
mentioned and expand on them to apply to all multiracial children. This model has both
an ecological and developmental component.
99
Root's model. Root (1990, 1997, 1998, 2003a) believes that a multiracial person's
identity can, and will change, depending on the ecological context. Further she believes
the central task for the multiracial child is to achieve a positive resolution between a
sense of identity and his or her environment (1990, 1997,2003).
Root (2003a) identifies five types of identities that emerge from research on
mixed race persons: assignment to hypodescent, monoracial fit/self assignment, blended
identity, bi-or multiracial, and White with symbolic identity. Generational norms often
influence the identity adopted by these individuals. For example, for older generations
single-race identity is often the result of either assuming an assignment according to a
"one-drop rule" and hypodescent. Younger generations now have the option of publicly
assuming a blended or multiracial identity. "Symbolic Whiteness" appears to reflect
identification with a class lifestyle and values or a lack of exposure to an ethnic
background with which one identifies (p. 34).
Root (1990, 1997, 2003a) believes there are four possible healthy identity
resolutions: (a) the individual accepts society's definition of his or her race or ethnicity,
(b) the individual identifies with both ethnic or racial groups, (c) the individual identifies
with a single racial/ethnic group, (d) the individual identifies with the mixed-race or
multiethnicl multiracial group that his or her background represents. Each of these
resolutions are positive if the individual is accepted in their chosen group, does not feel
pressure to change, does not deny or put down one aspect of their heritage, and is
supported in their identity by their immediate environment. In the case of the last
100
resolution, the identity is positive if the new identity embraces all aspects of the person's
identity.
Wardle's model. In Wardle's model (Wardle 1992; Wardle & Cruz-Janzen, 2003),
the central task of identity development for the child is the achievement of healthy
multiethnic/ multiracial identity. This is marked by successful completion of the three
developmental stages incorporated into the model, as well as the child's interaction with
the components that make up the ecological model: family, group antagonism,
minorityllower status context, majority/ higher status context, and community.
The three developmental stages described by Wardle's model (Wardle 1992;
Wardle & Cruz-Janzen, 2004) are Early childhood (3 to 7 years), Transition Period (6 to
12 years), and Adolescence. Each of these stages incorporates the developmental theories
of both Piaget, and Erickson. According to Wardle in the Early Childhood stage, the
child becomes aware of physical features, and the similarities and differences between
themselves and their parents and peers. At this stage they are often asked to explain their
physical and racial ambiguity and they need a label to proactively do so. The Transition
Period is marked by the multiracial child's increased awareness of sexuality. At this age
group belonging is important, and parallels an interest in competencies and the concept of
race. The Adolescent stage incorporates Erikson's (1963) identity crisis. At this stage
multiracial youth learn to separate out race, ethnicity, abilities, likes, dislikes and career
choices. Their family and school may increasingly support non-race specific groups, and
they become more comfortable with multiracial identity.
101
The interactions of components of the ecological model described by Wardle
(Wardle 1992; Wardle & Cruz-Janzen, 2004) determine the success or failure of the
healthy multiracial or multiethnic identity process. Family includes biological adoptive,
foster, teen, extended, and blended families on all sides. The family's impact of the
child's multiracial identity depends on the attitude of the family towards a multiracial
identity, discussion of the topics such as the child's identity and racism, as well as the
means by which the family SUppOltS the child's overall identity development.
Wardle (Wardle 1992; Wardle & Cruz-Janzen, 2004) explains that multiethnic
and multiracial children must deal with two kinds of group antagonism: the traditional
and institutionalized racism experienced by all people of color in the United States, and
the antagonism of all single-race and ethnic groups toward people of mixed heritage. The
level of racism multiracial children experience largely depends on phenotype, community
context, and school environment.
Wardle's model (Wardle 1992; Wardle & Cruz-Janzen, 2004) breaks down
cultural contexts into two separate categories, minority/ lower status context, and
majority/ higher status context. Wardle and Cruz-Janzen (2004) explain:
Multiethnic children with some mainstream White heritage have a distinctive
cultural context that includes a minority and majority status; children whose
parents are from two minority groups also have a lower status and higher status
context based on the position of their parents' race and ethnicity within the strict
racial hierarchy. (p.123)
102
The community represents the most important ecological impact on a child after
the family (Bronfenbrenner, 1989). And, according to Wardle (Wardle 1992; Wardle &
Cruz-Janzen, 2004) includes school, church groups, immediate peers, and neighborhood
groups. According to this model there are three factors that play heavily on the impact of
the community on healthy multiracial identity: (a) Does the family feel it belongs?; (b) Is
the community accepting of a range of diversity?; and (c) Is there minority representation
and multiracial/multiethnic children in the community groups the child attends? (2004)
As each child integrates all of his or her experiences within his or her own unique
context, they will more toward or away from healthy multiracial identity. Wardle's
model attempts to convey the developmental and ecological complexity of this process.
Social Connectedness
Lee and Robbins (1995) developed two measures of belongingness-the Social
Connectedness Scale and the Social Assurance Scale-based on Kohut's (1984) self-
psychology theory. Kohut proposed that as a manner of avoiding feelings of loneliness
and alienation people seek to confirm a subjective sense of belongingness or "being part
of." The authors proposed that belongingness is composed of companionship, affiliation
and connectedness and sought out to develop self-report measures to evaluate these
aspects of belongingness.
A random split-sample procedure was used with 626 participants from a large,
metropolitan university. Researchers gathered data for 204 students for the first split
sample, 198 students for the second split sample, and a separate sample of 18 students for
test-retest reliability. Each participant was given two weeks to complete a questionnaire
103
packet composed of demographic information, the Marlowe-Crowne Social Desirability
Index (Marlowe-Crown, 1998) and 45 randomized items measuring belongingness.
Analysis revealed two main factors for the construct of belongingness. The top eight
items from factor one were selected to compose the Social Connectedness Scale and eight
of the nine items from factor two were selected to compose the Social Assurance Scale.
The Social Connectedness Scale focuses on the emotional distance or connectedness
between oneself and others while the Social Assurance Scale focuses on need for
reassurance from others in social situations to sustain a sense of belonging. The authors
stress need for additional research to address the multiple dimensions of belongingness.
Lee and Robbins (1998) presented (a) a detailed description of social
connectedness as a theoretical construct and (b) findings from two studies examining
effects of social connectedness. Study one explored the effects of connectedness on
anxiety; study two explored the role of connectedness in social situations. Social
connectedness was defined as "the subjective awareness of being in close relationship
with the social world". The authors hypothesized that individuals with strong
connectedness are better able to (a) manage their emotions and needs via cognitive
processes and (b) develop interpersonal trust with the ability to construct and maintain
connectedness in social situations.
Study one: 185 undergraduate women were used to examine the following
question: 'Are individuals with higher connectedness better able to manage anxiety than
individuals with lower connectedness'? The mean age of the group was 24.15 years and
the ethnic breakdown was 37 African Americans, 13 Asian Americans, 129 European
104
Americans and 6 unidentified. Researchers measured connectedness using the Social
Connectedness Scale (Lee & Robbins, 1995) and measured anxiety using the State-Trait
Anxiety Inventory-Trait Form Y (Spielberger, 1983). The contributing effects of
collective self-esteem and perceived social support on anxiety were also explored.
Collective self-esteem was measured using the Collective Self-Esteem Scale (Luhtanen &
Crocker. 1992) and perceived social support was measured using the Social Support
Questionnaire-short form (Sarason, Sarason, Shearin, & Pierce, 1987). Social
connectedness was uniquely related to lower scores of anxiety, more so than collective
self-esteem and perceived social support. Researchers concluded that social
connectedness is more relevant to lower anxiety than temporary sources of belonging.
Study two: qualified participants from study one were asked to participate in
study two to explore whether (a) women with high connectedness would experience
lower state anxiety and higher self-esteem across experimental conditions and (b) would
report greater social identification than women with low connectedness. 44
undergraduate women agreed to participate. Mean age of the participants was 24.93 and
ethnic breakdown was 4 African Americans, 2 Asian Americans, 37 European
Americans, and 1 unidentified. Researchers asked participants to complete a minimal task
for 30 minutes as members of either the 'high cohesion' condition or the 'low cohesion'
condition. In the 'high cohesion' condition participants were allowed to communicate
with one another every six minutes via written messages. In the 'low cohesion' condition
participants were allowed to rest every six minutes with no communication. Women in
the 'high cohesion' group reported a more positive group attitude and higher levels of
105
social identity than women in the 'low cohesion' group. Women with high connectedness
reported higher levels of social identity and overall state self-esteem than women with
low connectedness.
Lee & Robbins (2000) studied social connectedness in college women and men
by examining the types of relationships that uniquely contribute to social connectedness.
It was hypothesized that men would feel more connected when relationships were based
on social comparison, and that women would feel more connected when relationships
were based on forms of intimacy and physical proximity. 387 undergraduate students
with a mean age of 20.96 were the participants of this study. Ethnic breakdown was 80
African Americans, 25 Asian Americans, 259 European Americans, 6 Latino/Hispanics,
and 17 unidentified. The Social Connectedness Scale (Lee & Robbins, 1995) was used to
measure social connectedness, the Social Provisions Scale (SPS) (Russell & Cutrona,
1984) was used to measure different forms of perceived social support. Subscales on the
SPS include attachment, social integration, reassurance of worth, reliable alliance,
guidance, and opportunity for nurturance. The UCLA Loneliness Scale-Revised was
used to measure general feelings of loneliness.
One-way analyses of variance with listwise deletion were first performed on each
measure with sex as the independent variable. Alpha level for analyses was adjusted to
.006 using the Bonferonni procedure. Results found no significant difference between
women and men on levels of social connectedness, while women reported significantly
less loneliness than men. On measures of social provision, women reported more
attachment, guidance, opportunity for nurturance, reassurance of worth and social
106
integration than men. Social connectedness was highly correlated with loneliness for
women, but only moderately related for men. Correlations between social connectedness
and social provision were also greater for women than men.
Hierarchical multiple regression analyses were used to measure the unique
contribution of each social provision on social connectedness after controlling for
loneliness. Analyses were performed separately on women and men. Results found that
for women, reliable alliance positively related to social connectedness while guidance
negatively related to social connectedness. For men, reassurance of worth positively
related to social connectedness while reliable alliance and opportunity for nurturance
negatively related to social connectedness. This study found social connectedness is
equally salient in the lives of women and men. Authors' recommend future studies of
social connectedness include factors of race and ethnicity, social class, gender,
adjustment, prediction and intervention evaluation.
Lee, Keough, and Sexton (2002) applied self-verification theory to investigate
how social appraisal mediated the effect of social connectedness on perceived stress in
the college context. The authors expected gender differences in the proposed relationship
with women presenting an interdependent self-construal based on intimacy and physical
proximity of relationships and men presenting an independent self-construal based on
issues of power and status. 214 students participated in this study with a mean age of
20.27 years. Ethnic breakdown was 65% White, 4% African American, 13% Asian
American, 14% Latino American, 1% Arabic/Middle Eastern and 3% unidentified.
Measures used were the Social Connectedness Scale (modified for college context; Lee &
107
Robbins, 1995), Appraised Status of Social Groups Scale (ASSGS) (Sexton, Lee, &
Keough, 1999), and Perceived Stress Scale (Cohen, Kamarck, & Mermelstein, 1983).
The ASSGS specifically measures students' perceptions about the degree to which
university student community holds various groups in high versus low esteem. No
statistically significant gender differences were found on social connectedness or
perceived stress. Social connectedness was significantly correlated with social appraisal
and perceived stress for both women and men. Racial status and year in school were not
related to social connectedness or social appraisal.
Yeh, and Inose (2003) investigated the influence of reported English fluency,
social support satisfaction, and social connectedness in predicting acculturative stress
among international students from various regions. Authors discussed how international
students, with an interdependent self-concept, may experience difficulty developing
perceived social support and social connectedness in an American, individualistic culture.
Surveys were distributed at international student organizations and clubs. Packets
included a demographic questionnaire, Acculturative Stress Scale for International
Students (ASSIS) (Sandhu and Asrabadi, 1994), Social Connectedness Scale (SCS) (Lee
& Robbins, 1995) and Social Support Questionnaire-Short Form (SSQSR) (Sarason et
al., 1987). A stepwise regression model was performed using age, gender, region, English
fluency, social connectedness, and social support network satisfaction as predictor
variables and acculturative stress as the criterion variable. The overall regression model
was significant, accounting for 34% of the variance for acculturative stress. English
fluency, social connectedness and social support network satisfaction were significant
predictors of acculturative stress. Specifically, higher levels of English fluency, social
connectedness and social support network satisfaction predicted lower levels of
acculturative stress. Researchers believe feelings of social connectedness were strong
predictors of acculturative stress due to the significance placed on relationships for
individuals from interdependent societies. Quality of social connections is an essential
aspect concerning self-identity, values and ways of 'being' with others.
108
APPENDIXB
INFORMED CONSENT STATEMENT
109
110
Informed Consent Form
(You may print a copy of this form for your records.)
Research Study: MultiraciaV Multiethnic College Students' Identity and
Mental Health
Invitation to Participate: You are invited to participate in a dissertation research study
conducted by James Lyda, M.S., a doctoral candidate in Counseling Psychology at the
University of Oregon. The following information is provided to help you make an
informed decision about whether or not you want to participate in this study. Please feel
free to email me with questions (jlyda@uoregon.edu).
Eligibility: You are eligible to participate in this study if you identify as having two more
more racial or ethnic groups in your heritage, are 18 to 23 years of age, currently enrolled
in a college or university, and you are able to write and speak English.
Purpose of the Study: The purpose of this study is to examine how your racial and
ethnic identity affects your emotional and psychological well being.
Explanation of Procedures: If you decide to participate, you will fill out a web-based
group of questionnaires, which should take about 45 minutes to complete. Once you have
completed the questionnaires, they will be electronically sent back to me over a secure
server. The information that you provide (e.g., name, contact information, and survey
answers) will remain completely confidential and will be stored in a secure, password-
protected server on the Internet. Your identifying information will not be linked in any
way with your survey answers.
Potential Risks and Discomforts: On occasion some people may experience some
distress when completing psychological questionnaires. If you should feel distressed at
any time during the study, you may email me and I will email you a list of community
resources and referrals. You may also withdraw from the study at any time without any
consequences.
Potential Benefits to Participant: The benefits you may personally receive from
participating in this study include the oppOltunity to increase self-awareness about your
racial/ethnic identity and your emotional and psychological well being. You will be given
the chance to explore your thoughts and feelings about certain aspects of yourself and
you will have the opportunity to receive community resources and referrals from the
primary investigator. You will have the opportunity to participate in a confidential raffle
in which you may enter to win a $200 Amazon.com gift certificate. If you choose to
participate in this raffle, you will be required to submit your contact information (phone
number and email address) in addition to the answers on your survey battery. Your
111
identifying and contact information will not be linked in any way to your answers in the
survey battery. Upon completion of my participant recruitment process, I will randomly
select 5 participants who will win a $200 Amazon.com gift certificate.
Potential Benefits to Society: Your participation in this study will add to the research on
Multiracial individuals' emotional and psychological well being. Your participation will
increase the knowledge of how to assist Multiracial persons in addressing mental health
concerns. Results from this study may help counselors and other social service providers
more effectively give Multiracial college students the culturally appropriate assistance,
support, and encouragement they need.
Voluntary Participation and Withdrawal: Your participation is voluntary. You may
withdraw your consent and discontinue participation at anytime without penalty.
Assurance of Confidentiality: Your identity will be kept confidential because your
name or other identifying information will not be attached to the answers that you
provide. I will store confidential survey data on a secure computer server, protected by a
login ill and password available only to me. Signed informed consent forms or any other
identifying participant information will be kept on a separate, secure server. Contact
information will be used only for those who want to participate in the raffle to win a gift
certificate. Participation in the raffle drawing is optional and answers on your
questionnaires will not be connected to requested identifying contact information. All
downloaded data will be stored on a password-protected computer and/or in a locked file
cabinet to further ensure confidentiality. The information obtained in this study may be
published in a scientific journal or presented at scientific meetings for which only
aggregate data will be presented.
Rights of Research Participants: Your rights as a research participant have been
explained to you. If you have any additional questions about this study, please contact the
primary researcher or research advisor:
James Lyda, M.S.
Doctoral Candidate
(541) 515-2313
jlyda@uoregon.edu
Benedict McWhirter, Ph.D.
Associate Professor, Counseling Psychology
(541) 346-2410
benmcw@uoregon.edu
If you have any questions concerning your rights as a research participant, contact
Human Subject Compliance, 5219 University of Oregon, Eugene, OR 97403, (541) 346-
2510. You may print a copy of this Informed Consent Form for your records.
Click "I agree" to fill out the survey.
I I Agree I
Decline
APPENDIX C
MEASUREMENT INSTRUMENTS
112
113
Demographic Questionnaire
Thank you for your participation in this study. Please fill out this questionnaire as
completely as possible. Provide additional comments as needed. It should take
approximately 5 minutes to complete this survey.
1. Age: _
2. What is your sex? (check one)
o Female
o Male
o Transgender
o Transsexual
3. Are you Spanish/Hispanic/Latino?
o No, not Spanish/Hispanic/Latino
o Yes, Mexican, Mexican American, Chicano
o Yes, Puerto Rican
o Yes, Cuban
o Yes, other Spanish/ Hispanic/Latino (please print group)
4. What is your race? (Mark one or more of the races you consider yourself to be)?
o American Indian or Alaska Native
o Black, African American
o Asian Indian
o Chinese
o Filipino
o Japanese
o Korean
114
o Vietnamese
o Other Asian (please type) _
o Native Hawaiian
o Guamanian or Chomorro
o Samoan
o Other Pacific Islander (please type): _
o Some other race (please type): _
5. What comes closest to describing your year in college?
o Freshman
o Sophomore
o Junior
o Senior
o Graduate Student (Masters)
o Graduate Student (Doctoral)
6. Have you declared a major?
o No
DYes
If yes, please specify: _
7. To date, which of the following courses have you taken? (Select all that apply):
o Ethnic studies courses
o Women's studies courses
o Race Relations courses
o Other multicultural content courses
(please type in): _
9. What is your current G.P.A.? (please type in) _
10. (a) Are you a U.S. Citizen? A. Yes B. No
(b) If not, please indicate your national citizenship
115
11. How old were you when you began living in the U.S.?
o I was born in the U.S.
o Six years of age or younger
o Between age 7 and 17
o After age 18
12. In which of the following regions of the U.S. do you currently reside?
0 Northeast
0 Mid-Atlantic
0 Southeast
0 Midwest
0 Southwest
0 Pacific Northwest
0 West
0 Alaska
0 Hawaii
0 US Territory (please type):
13. In which of the following regions of the U.S. did you spent the majority of your
upbringing?
0 Northeast
0 Mid-Atlantic
0 Southeast
0 Midwest
0 Southwest
116
o Pacific Northwest
o West
o Alaska
o Hawaii
o US Territory (please list): _
14. What is your father's highest level of education?
o Did not graduate High School
o High School/ G.E.D.
o Some college
o Completed 2-year college (Associates degree)
o Completed 4 year-college (Bachelors degree)
o Some graduate school
o Completed graduate degree (e.g. Masters, Ph.D., M.D., Law School)
o Don't know
15. What is your mother's highest level of education?
o Did not graduate High School
o High School/G.E.D.
o Some College
o Completed 2-year college (Associates degree)
o Completed 4-year college (Bachelors degree)
o Some graduate school
o Completed graduate degree (e.g. Masters, Ph.D., M.D., Law School)
117
o Don't know
16. How are you paying for your college education? (Select all that apply):
o Parents/Guardians are helping pay
o I am employed full time
o I am employed part time
o I receive a financial scholarship/grant
o I receive a financial loan
17. How are you paying for your Cost of Living (e.g., rent, bills)? (Select all that
apply):
o Parents/Guardians are helping pay
o I am employed full time
o I am employed part time
o I receive a financial scholarship/grant
o I receive a financial loan
18. How did you hear about this research study (e.g. e-mail, listerv, class, friend told
me, "word of mouth," etc.)
118
College Supports Questionnaire
1. How actively do you participate in student groups or community organizations on
campus or in the surrounding community?
A. Extremely active (e.g., attend weekly meetings; engage in social activism;
organize activities; participate in social events)
B. Very active (e.g., attend weekly meetings but do not actively participate in
events or social activism; or participate in events but do no attend weekly
meetings)
C. Fairly active (e.g., attend meetings, events, socials once in a while)
D. Not active (no involvement)
2. Are any of the following types of groups available on your campus? (check all that
apply)
o Ethnic student groups/ Ethnic Student Unions (e.g Black Student Union,
Asian Student Association, MEChA, etc.)
o Fraternities and Sororities with a traditionally racial minority membership
o Multicultural student group/ Multicultural student unions
o Minority Mentorship Program
o Minority Student Programs initiated by your College/ University (e.g.
activities geared toward minority students during orientation)
o A group specifically for Biracial or Multiracial Students
o Other (please type): _
o None
3. Do you belong to or participate in any of the following types of groups? (check all
that apply)
o Ethnic student groups/ Ethnic Student Unions (e.g Black Student Union,
Asian Student Association, MEChA, etc.)
o Fraternities and Sororities with a traditionally racial minority membership
o Multicultural student group/ Multicultural student unions
o Minority Mentorship Program
o Minority Student Programs initiated by your College/ University (e.g.
activities geared toward minority students during orientation)
o A group specifically for Biracial or Multiracial Students
o Other (please type): _
119
o None
4. How actively do you participate in specifically ethnic or cultural student groups
and organizations on campus, or in the off campus community?
1. Not active (no involvement)
2. Fairly active (e.g., attend meetings, events, socials once in a while)
3. Active (e.g., attend weekly meetings but do not actively participate in
events or social activism; or participate in events but do no attend weekly
meetings)
4. Highly active (e.g., attend weekly meetings; engage in social activism;
organize activities; participate in social events)
5. How supported do you feel in the groups you are involved
with?
1) Not supported at all
2) Somewhat supported
3) Supported
4) Very supported
6. How connected to others do you feel in the groups your are involved with?
1) Not connected
2) Somewhat connected
3) Connected
4) Closely connected
120
7. Some colleges and universities have the following student support services. How
f lh £ d h f 11' t' f . ft f ?use u ave you oun teo owmg suppor serVIces rom your ms I u Ion.
Very Useful Useful Not Useful Have not used Am not aware
this service, of such a
although it is service being
available available.
5 4 3 2 1
o Introduction or Orientation
o Study Skills support class
o College/University advising services
o Students' Union advice services
o Social activities organized by college or university
o Student unions, clubs, societies, fraternities, sororities
o Religious or Spiritual centers/groups
o Health Center
o Career Services
121
Multiracial Identity Options
Biracial or multiracial people chose to racially identify in different ways and this may
depend on their environment or context. For example, how you define yourself with
family may differ from when you are in a classroom at college with people who may not
know you very well. Below are questions followed by a series of brief statements. Please
read through each of them and choose which one you believe best fits for you.
Which statement best fits with how you identify yourself in terms of race
generally? (choose one)
1. I choose to identify myself by describing fractions/percentages of my racial breakdown (e.g. I'm
Y2 White, JA Black, and JA Native American).
2. I choose to identify as all/most of my racial identities without fractions (e.g. I am Black and White
and Native American)
3. I choose to identify as one Non-White race (e.g Black, Asian, Native American, etc.).
4. I choose to identify as White.
5. I choose not to identify with any race.
6. I identify simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain how
I'm mixed (unlike choices 1 & 2).
Now, how do you identify yourself in the following contexts? (choose one)
When in new or unfamiliar settings:
1. I choose to identify myself by describing fractions/percentages of my racial breakdown (e.g. I'm
Y2 White, JA Black, and 1,4 Native American).
2. I choose to identify as all/most of my racial identities without fractions (e.g. I am Black and White
and Native American)
3. I choose to identify as one Non-White race (e.g Black, Asian, Native American, etc.).
4. I choose to identify as White.
5. I choose not to identify with any race.
6. I identify simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain how
I'm mixed (unlike choices 1 & 2).
7. It depends.
When with close friends:
122
1. I choose to identify myself by describing fractions/percentages of my racial breakdown (e.g. I'm
V2 White, 1/.1 Black, and 1/.1 Native American).
2. I choose to identify as al1/most of my racial identities without fractions (e.g. I am Black and White
and Native American)
3. I choose to identify as one Non-White race (e.g Black, Asian, Native American, etc.).
4. I choose to identify as White.
5. I choose not to identify with any race.
6. I identify simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain how
I'm mixed (unlike choices I & 2).
7. It depends.
When with immediate family:
1. I choose to identify myself by describing fractions/percentages of my racial breakdown (e.g. I'm
V2 White, If.I Black, and 1/.1 Native American).
2. I choose to identify as al1/most of my racial identities without fractions (e.g. I am Black and White
and Native American)
3. I choose to identify as one Non-White race (e.g Black, Asian, Native American, etc.).
4. I choose to identify as White.
5. I choose not to identify with any race.
6. I identify simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain how
I'm mixed (unlike choices 1 & 2).
7. It depends.
When in class:
1. I choose to identify myself by describing fractions/percentages of my racial breakdown (e.g. I'm
V2 White, If.I Black, and 1/.1 Native American).
2. I choose to identify as all/most of my racial identities without fractions (e.g. I am Black and White
and Native American)
3. I choose to identify as one Non-White race (e.g Black, Asian, Native American, etc.).
4. I choose to identify as White.
5. I choose not to identify with any race.
6. I identify simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain how
I'm mixed (unlike choices 1 & 2).
123
7. It depends.
When with authority figures (i.e. Professors, advisors, bosses, supervisors, etc.)
]. I choose to identify myself by describing fractions/percentages of my racial breakdown (e.g. I'm
Y2 White, 'l4 Black, and 'l4 Native American).
2. I choose to identify as all/most of my racial identities without fractions (e.g. I am Black and White
and Native American)
3. I choose to identify as one Non-White race (e.g Black, Asian, Native American, etc.).
4. I choose to identify as White.
5. I choose not to identify with any race.
6. I identify simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain how
I'm mixed (unlike choices] & 2).
7. It depends.
Now, how do you think the following groups identify you? (choose one)
Society as a whole:
]. Identifies me by describing fractions/percentages of my racial breakdown (e.g. I'm Y2 White, 'l4
Black, and 'l4 Native American).
2. Identifies me as all/most of my racial identities without fractions (e.g. I am Black and White and
Native American)
3. Identifies me as one Non-White race (e.g Black, Asian, Native American, etc.).
4. Identifies me as White.
5. Does not identify me with any race.
6. Identifies me simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain
how I'm mixed (unlike choices I & 2).
Close friends:
1. Identify me by describing fractions/percentages of my racial breakdown (e.g. I'm Y2 White, 11<1
Black, and 11<1 Native American).
2. Identify me as all/most of my racial identities without fractions (e.g. I am Black and White and
Native American)
3. Identify me as one Non-White race (e.g Black, Asian, Native American, etc.).
124
4. Identify me as White.
5. Do not identify me with any race.
6. Identify me simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain
how I'm mixed (unlike choices I & 2).
7. It depends.
Immediate family:
1. Identify me by describing fractions/percentages of my racial breakdown (e.g. I'm 1J2 White, IA
Black, and IA Native American).
2. Identify me as all/most of my racial identities without fractions (e.g. I am Black and White and
Native American)
3. Identify me as one Non-White race (e.g Black, Asian, Native American, etc.).
4. Identify me as White.
5. Do not identify with any race.
6. Identify me simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain
how I'm mixed (unlike choices 1 & 2).
7. It depends.
Fellow students in your classes:
1. Identify me by describing fractions/percentages of my racial breakdown (e.g. I'm 1J2 White, IA
Black, and IA Native American).
2. Identify me as all/most of my racial identities without fractions (e.g. I am Black and White and
Native American)
3. Identify me as one Non-White race (e.g Black, Asian, Native American, etc.).
4. Identify me as White.
5. Do not to identify with any race.
6. Identify me simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain
how I'm mixed (unlike choices 1 & 2).
7. It depends
Authority Figures (e.g. Professors, advisors, bosses, supervisors etc.):
125
1. Identify me by describing fractions/percentages of my racial breakdown (e.g. I'm Y:z White, 1,4
Black, and 1,4 Native American).
2. Identify me as all!most of my racial identities without fractions (e.g. I am Black and White and
Native American)
3. Identify me as one Non-White race (e.g Black, Asian, Native American, etc.).
4. Identify me as White.
5. Do not identify me with any race.
6. Identify me simply as "mixed" (or "biracial", "multiracial") without feeling the need to explain
how I'm mixed (unlike choices 1 & 2).
7. It depends.
126
Reference Information for Other Instruments
The additional instruments used in this study were used with permission by he instrument
developers and or purchased through the companies holding the copyrights.
Social Connectedness Scale-Revised (SCS-R)
The request approval to use this instrument, please contact Richard M. Lee, Ph.D. at
m:l2J.~~s~~£(j_~llllJi1.£QJ~. To access the items of this instrument please refer to Lee, Draper, and
Lee's article "Social connectedness, dysfunctional interpersonal behaviors and
psychological distress: Testing a mediator model" in the Journal of Counseling
Psychology, 48 (3), July 2001,310-318.
Student Adaptation to College Questionnaire (SACQ)
To review and/or purchase this copyrighted instrument, please contact Western
Psychological Services, 12031 Wilshire Boulevard, Los Angeles, CA 90025-1251.
Patient Health Questionnaire (PHQ-9)
To review this copyrighted instrument, please contact Pfizer, Inc. 235 East 42nd Street
New York, NY 10017.
127
REFERENCES
Aboud, F. E. (1987). The development of ethnic self-determintation and attitudes. In J.S.
Phinney & M. J. Rotheram (Eds.), Children's Ethnic Socialization: Pluralism
and Development (pp. 29-55). Newbury Park, CA: Sage.
Atkinson, D. R., & Sue, D. W. (1993). Counseling American minorities: A cross cultural
perspective (4th ed.). Madison, WI: W. C. Brown & Benchmark.
Baker, R. W., & Siryk, B. (1984). Measuring adjustment to college. Journal of
Counseling Psychology, 31,179-189.
Baker, R. W., & Siryk, B. (1986). Exploratory intervention with a scale measuring
adjustment to college. Journal of Counseling Psychology. 33, 31-38.
Baker, R. W., & Siryk, B. (1999). SACQ: Student adaptation to college questionnaire,
manual. (2nd ed.). Los Angeles: Western Psychological Services.
Barron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal ofPersonality and Social Psychology, 51, 1173-1182.
Basu, A. (2007). Negotiating social contexts: identities of biracial college women.
Charlotte, NC: lAP.
Benedetto, A. E., & Olisky, T. (2001). Biracial youth: The role of the school counselor
in racial identity development. Professional School Counseling, 5,66-69.
Brabeck, M., Walsh, M. E., Kenny, M., & Comilang, K. (1997). Interprofessional
collaboration for children and families: Opportunities for counseling psychology in
the 21 st century. The Counseling Psychologist, 25, 615-636.
Bronfenbrenner, U. (1977). Toward an experimental ecology of human development.
American Psychologist, 32, 513-31.
Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature
and design. Cambridge: Harvard University Press.
128
Bronfenbrenner, U. (1989). Ecological systems theory. In R. Vasta (Ed.), Six theories of
development (pp. 187-249). Greenwich, CT: JAI press.
Bronfenbrenner, U. (1993). The ecology of cognitive development: research models and
fugitive findings. In R. H. Wozniak & K. W. Fischer, (Eds.), Development in
context: Acting and thinking in specific environments (pp. 3-44). Hillsdale, NJ:
Erlbaum.
Caffarella, R. S., & Olson, S. K. (1993). Psychosocial development of women: A critical
review of the literature. Adult Education Quarterly, 43, 125-151.
Cauce, A. M., Hirango, Y. Mason, C., Aguilar, T., Ordonez, N., & Gonzales, N. (1992).
Between a rock and a hard place: social adjustment of biracial youth. In M. P. P.
Root (Ed.), Racially mixed people in America (pp. 207-22) Newbury Park, CA:
Sage.
Chickering, A. W., (1969). Education and identity. San Francisco: Jossey-Bass.
Cohen,1. (1992.) A power primer. Psychological Bulletin, 112, 155-159.
Cross, W. E., Jr. (1995). The psychology of nigrescence: Revising the Cross model. In J.
G. Ponterotto, 1. M. Casas, L. A. Suzuki, & C. M. Alexander (Eds.), Handbook of
Multicultural Counseling (pp. 93 -122). Thousand Oaks, CA: Sage.
Daneshpour, M. (1998). Muslim families and family therapy. Journal ofMarriage and
Family Counseling, 24, 355-368.
Daniel, G. R. (1992). Beyond black and white: The new multiracial consciousness. In M.
P. P. Root (Ed.), Racially mixed people in America (pp. 333-41). Newbury Park,
CA: Sage.
Erikson, E. (1963). Childhood and society (2nd ed.) New York: Norton.
Field, L. D. (1996). Piecing together the puzzle: Self-concept and group identity in
biracial black/white youth. IN M.P.P. Root (Ed.), The multiracial experience:
Racial borders as the new frontier (pp. 211-26). Thousand Oaks, CA: Sage.
Gall, M. D., Gall, J. P., & Borg, W. R. (2003). Educational Research: An Introduction
(7th ed.). Boston: Allyn & Bacon.
Gaskins, P. (1999). What are you? Voices ofmixed-race young people. New York: Holt.
129
Gelso, C. J., & Fretz, B.R. (1992). Counseling Psychology. Forth Worth, TX: Harcourt
Brace Jovanovich.
Gibbs, J. T., & Hines, AM. (1992). Negotiating ethnic identity: issues for black-white
biracial adolescents. In M. P. P. Root (Ed.), Racially mixed people in America
(pp. 223-38). Newbury Park, CA: Sage.
Gibbs, J. T, & Moskowitz-Sweet, G. (1991). Clinical and cultural issues in the treatment
of biracial and bicultural adolescents. Families in Society, 72, 579-592.
Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust Web-
based studies? A comparative analysis of six preconceptions about Internet
questionnaires. American Psychologist, 59, 93-104.
Hart-Webb, D. (1999). The biracial bind. In Y. M. Jenkins (Ed.), Diversity in college
settings (pp. 15-25). New York: Routledge.
Helms, J. E. (1990). Black and white racial identity: Theory, research, and practice. New
York: Greenwood Press.
Helms, J. E. (1995). An update of Helms's white and people of color racial identity
models. In J. G. Ponterotto, J. M. Casas, L. A Suzuki, & C. M. Alexander (Eds.),
Handbook ofMulticultural Counseling (pp. 181-198). Thousand Oaks, CA: Sage.
Heppner, P. P., Casas, J. M., Carter, 1. & Stone, G. 1. (2000). The maturation of
counseling psychology: Multifaceted perspectives. In S. D. Brown, & R. W. Lent
Handbook ofcounseling psychology (3rd ed) (pp. 3-49). New York: Wiley.
Herring, R. D. (1992). Developing biracial identity: A review of the increasing dilemma.
Journal ofMulticultural Counseling and Development, 23, 29-38.
Heyes, C. (2007). Identity Politics. Retrieved Jan 30, 2008, from Stanford Encycolpedia
of Philosophy Web site: http://plato.stanford.edu/entries/identity-politics/
Jacobs, J. H. (1992). Identity development in biracial children. In M. P. P. Root (Ed.),
Racially mixed people in America (pp. 190-207). Newbury Park, CA: Sage.
Jones, N. A, & Smith, A S. (2003). A statistical portrait of children of two or more races
in Census 2000. In M. P. P. Root & M. Kelley, (Eds.), Multiracial Child Resource
Book (pp. xiv-xvi). Seattle, WA: Mavin Foundation.
130
Karcher, M. J. (2001, August). Measuring adolescent connectedness: Four validation
studies. Poster session presented at the annual convention of the American
Psychological Association, San Francisco, CA.
Karcher, M. J. & Lee, Y. (2002). Connectedness among Taiwanese middle school
students: A validation study of the Hemingway measure of adolescent
connectedness. Asia Pacific Education Review, 3, 92-114.
Kearney, M.H. (1998). Truthful self-nurturing: A grounded formal theory of women's
addiction. Quantitative Health Research, 8, 495-512.
Kelley, M. & Root, M. P. P. (2003). Introduction. In M. P. P. Root & M. Kelley,
(Eds.), Multiracial Child Resource Book (pp. xiv-xvi). Seattle, WA: Mavin
Foundation.
Kerwin, C. & Ponterotto, 1. G. (1995). Biracial identity development: Theory and research.
In 1. G. Ponterotto, J . M. Casas, L. A. Suzuki, & C. M. Alexander (Eds.),
Handbook ofmulticultural counseling (pp. 199-217). Thousand Oaks, CA: Sage.
Kohut, H. (1984). How does analysis cure? New York: International Universities Press.
Kroenke K., Spitzer R. L., Williams J. B. (2001). The PHQ-9: Validity of a brief
depression severity measure. Journal ofGeneral Internal Medicine, 16,
606-613.
Lee, R. M., Bowen, N. A., & Keough, K. A. (1999). The moderating influence of social
connectedness in the college adjustment ofprovisional students. Unpublished
manuscript, University of Texas at Austin.
Lee, R. M., Draper, M., & Lee, S. (2001). Social connectedness, dysfunctional
interpersonal behaviors, and psychological distress: Testing a mediator model.
Journal of Counseling Psychology, 48,310-318.
Lee, R. M., Keough, K. A., & Sexton, J. D. (2002). Social connectedness, social
appraisal, and perceived stress in college women and men. Journal ofCounseling
and Development, 80, (255-361).
Lee, R. M., & Robbins, S. B. (1995). Measuring belongingness: The Social
Connectedness and the Social Assurance Scales. Journal of Counseling
Psychology, 42,232-241.
13]
Lee, R. M., & Robbins, S. B. (1998). The relationship between social connectedness and
anxiety, self-esteem, and social identity. Journal of Counseling Psychology, 45,
(338-345).
Lee, R. M. & Robbins, S. B. (2000). Understanding social connectedness in college
women and men. Journal of Counseling and Development, 78, 484-491.
Maton, K. 1., Hrabowski, F. A., III, & Greif, G. L. (1998). Preparing the way: A
qualitative study of high-achieving African American males and the role of the
family. American Journal of Community Psychology, 26, 639-668.
McRoy, R., & Freeman, E. (1986). Racial identity issues among mixed race children.
Social Work in Education, 8, 164-174.
McWhirter, B. T. (1990). Loneliness: A current review of the literature with implications
for counseling and research. Journal of Counseling and Development, 68, 417-
422.
Milan, S., Keiley, M. K. (2000). Biracial youth and families in therapy: Issues and
interventions. Journal of Marital and Family Therapy. 26, 305-315.
Miller, J. B., & Stiver, 1. P. (1997). The healing connection: How womenform
relationships in therapy and in life. Boston: Beacon Press.
Neumark-Sztainer, D., Story, M., French, S. A., & Resnick, M. D. (1997). Psychosocial
correlates of health compromising behaviors among adolescents. Health Education
Research, 12, 37-52.
Newcomb, M. D. (1990). Social support by many other names: Toward a unified
conceptualization. Journal ofSocial & Personal Relationships, 7,479--494.
Nishimura, N. J. (1998, Spring). Assessing the issues of multiracial students on college
campuses. Journal of College Counseling, 1,45-53.
Nishimura, N., & Bol, L. (1997). School counselors' perceptions of biracial children's
counseling needs in an urban educational setting. Research in the Schools, 4(3),
17-23.
Overmier, K. (1990). Biracial adolescents: Areas of conflict in identity formation. The
Journal ofApplied Social Sciences, 14, 157-177.
132
Paez, K., & Lyda, J. (2002, August). Integrating Identities: Understanding common
themes in biracial individuals. Poster session presented at the annual meeting of
the American Psychological Association, Chicago, IL.
Phares, V. (1993). Father absence, mother love, and other family issues that need to be
questioned: Comment on Silverstein (1993). Journal ofFamily Psychology, 7,293-
300.
Phinney, J. (1992). The Multigroup Ethnic Identity Measure: A new scale for use with
adolescents and young adults from diverse groups. Journal ofAdolescent
Research,7,156-176.
Phinney, J. S. (1993). A three stage model of ethnic identity development in adolescence.
In M.E. Bernal & G. P. Knight (Eds.). Ethnic Ientity: Formation and
transmission, among Hispanic and other minorities (pp. 61-79). Abany, NY:
SUNY Press.
Poston, W. S. C. (1990). The biracial identity development model: A needed addition.
Journal ofCounseling and Development, 69, 152-55.
Renn, K. R. (2000). Patterns of situational identity among biracial and multiracial college
students. The Review ofHigher Education, 23, 399-420.
Renn, K. (2004). Mixed race students in college: The ecology of race, identity, and
Community. Albany: SUNY Press.
Root, M. P. P. (1990). Resolving "other" status: Identity development of biracial
individuals. In L. S. Brown & M.P.P. Root (Eds.), Diversity and complexity in
feminist therapy (pp.185-205). New York: Haworth.
Root, M. P. P. (1992). Within, between, and beyond race. In M. P. P. Root (Ed.),
Racially Mixed People in America (pp. 3-11).
Root, M. P. P. (1994). Mixed Race Women. In L. Comas Diaz and B. Green (Eds.),
Women ofcolor and mental health: The healing tapestry. New York: Guilford.
Root, M. P. P. (Ed.) (1996). The Multiracial Experience: Racial Borders as the New
Frontier. Thousand Oaks, CA: Sage.
Root, M. P. P. (1997). Multiracial Asians: Models of ethnic identity. Amerasian Journal,
23 (1), 29-41.
133
Root, M. P. P. (1998a). Preliminary findings from the biracial sibling project. Cultural
Diversity and Mental Health, 4, 237-247.
Root, M. P.P. (1998b). Multiracial Americans: Changing the face of Asian America. In L.
C. Lee & N. W. Zane (Eds.), Handbook ofAsian American Psychology (pp. 261-
281). Thousand Oaks, CA: Sage.
Root, M. P. P. (2000). Rethinking Racial Identity Development: An Ecological
Framework. In P. Spickard and J. Burroughs (Eds.) We are a people: Narrative
in the construction and deconstruction ofethnic identity. Philadelphia, PA:
Temple University Press.
Root, M. P. P. (2003a). Racial identity development and persons of mixed race heritage.
In M. P. P. Root & M. Kelley, (Eds.), Multiracial Child Resource Book (pp. 34-
41) Seattle, WA: Mavin Foundation.
Root, M. P. P. (2003b). Multiracial families and children: Implications for educational
research and practice. In J. A. Banks and C. A. McGee Banks (Eds.), Handbook
of research on multicultural education (second edition), pp. 121-122. San
Francisco: Jossey-Bass.
Root, M. P. P. (2004). From exotic to a dime a dozen. In A. Gillem and C. Thompson
(Eds.), Biracial Women in Therapy: Between the Rock ofGender and the Hard
Place ofRace (pp. 19-31). New York: The Haworth Press.
Root, M. P. P. & Kelley, M. (2003). Multiracial Child Resource Book. Seattle, WA:
Mavin Foundation.
Rosen, W. B. (1999). Moments of truth: Notes from a lesbian therapist. Smith College
Studies in Social Work, 69, 293-308.
Sebring, D. L. (1985). Considerations in counseling interracial children. Journal ofNon-
White Concerns in Personnel and Guidance, 13, 3-9.
Schafer, J. L. (1999). NORM: Multiple imputation of incomplete multivariate data
under a normal model [Computer software]. University Park: Pennsylvania State
University, Department of Statistics.
Schafer, J. L. & Graham, J. W. (2002). Missing data: Our view of the state of the art.
Psychological Methods, 7, 147-177.
134
Shih, M., & Sanchez, D. T., (2005). Perspectives and research on the positive and
negative implications of having multiple racial identities. Psychological
Bulletin, 131, 569-591.
Stephan, C. W. (1992). Mixed-heritage individuals: Ethnic identity and trait
characteristics. In M. P. P. Root (Ed.) Racially mixed people in America (pp. 50-
63). Newbury Park, CA: Sage.
Timpone, R. J. (1998). Ties that bind: Measurement, demographics, and social
connectedness. Political Behavior, 20, 53-77.
Townsend, K.c., & McWhirter, B. T., (2005). Connectedness: A review of the literature
with implications for counseling, assessment, and research. Journal of Counseling
and Development, 83, 191-201.
Troll, L. E. (1994). Family connectedness of old women: Attachment in later life. In B. F.
Turner & L. E. Troll (Eds.), Women growing older: Psychological perspectives (pp.
169-201). Thousand Oaks, CA: Sage.
Twine, F.W. (1996) Brown skinned white girls: Class, culture and the construction of
white identity in suburban communities. Gender, Place, and Culture 3(2): 205-
224.
U.S. Bureau of the Census. (1992, March). Marital status and living arrangements.
Current Population Reports, Population Characteristics, Series P20-468,
December. Washington, DC: U.S. Government Printing Office.
U.S. Census Bureau. (2000). Racial and ethnic classifications used in Census 2000 and
beyond. Retrieved May 10,2005 from the World Wide Web:
http://www.census.gov/population/www/socdemo/race/racefactcb.html
Wallace, K. R. (2003). Contextual factors affecting identity among mixed heritage
college students. In M. P. P. Root & M. Kelley, (Eds.), Multiracial Child
Resource Book (pp. 87-93). Seattle, WA: Mavin Foundation.
Wardle, F. (1992). Biracial Identity: An ecological and developmental model. Denver,
CO: Center for the Study of Biracial Children.
Wardle, F., & Cruz-Janzen, M. 1. (2004). Meeting the needs ofmultiracial and
multiethnic children in schools. Boston: Pearson Education, Inc.
Wolf, P. (1988). Treating the self: Elements of clinical selfpsychology. New York:
Guilford Press.
135
Zack. N. (1998). Mixed black and white race and public policy. In N. Zack, L. Shrage, &
C. Sartwell (Eds.), Race, class, gender, and sexuality: The big questions (pp. 77-
85). Boston: Blackwell.
Zack, N. (2001). American mixed race: The U.S. 2000 Census and Related Issues.
Harvard BlackLetter Law Journal, 17, pp. 33-46.
Zack, N. (2006). Thinking about race (2nd ed.). Belmont, CA: Thomson-Wadsworth.