• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Race, segregation and diversity in college: How precollege diversity experiences influence students' college preferences, behaviors, and perceptions

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Donnell J Butler
Abstract:
Existing research suggests that racial segregation in neighborhoods and schools might lead students to develop negative predispositions regarding multiethnic settings and situations. However, very little research exists regarding how the presence or absence of precollege diversity experiences shapes students' future multiethnic college experiences. This dissertation consists of three related empirical analyses based on conceptual frameworks that I develop from research evidence and theories from the fields of sociology, psychology, and education. Using two recent multi-institution student surveys, I evaluate how students' ethnoracial origin and precollege experiences with racial and ethnic diversity influence their college preferences as well as their interracial interactions and perceptions of comfort in college. This dissertation contributes to a growing literature on the enduring consequences of racial and ethnic segregation, particularly how experiences of segregation influence individual attitudes and behavior. I find that both opportunities for and direct experiences with diversity prior to college influence students' future engagement with and interpretation of racial and ethnic experiences. In the concluding chapter, I discuss important areas for research and action related to minimizing the reproduction of social stratification and maximizing the preparation of students for a global community where experience with diversity is becoming a form of capital.

Table of Contents

Abstract.............................................................................................................................iii Table of Contents..............................................................................................................iv Acknowledgements............................................................................................................v Chapter 1. Introduction......................................................................................................1 Chapter 2. Ethnoracial Composition and College Preference: Revisiting the Perpetuation of Segregation Hypothesis.................................................................................................8 Chapter 3. Precollege Ethnoracial Diversity Experiences and Interracial Interactions in College.............................................................................................................................45 Chapter 4. Ethnoracial Group Membership and Comfort on Campus.............................84 Chapter 5. Conclusion....................................................................................................127 Appendix Tables............................................................................................................137 References......................................................................................................................178

iv

Acknowledgements I am eternally grateful to my dissertation chair, Marta Tienda, for the time and energy that she has invested in me. She has been the consummate teacher: providing the most constructive criticism, imparting lessons in scholarship, and challenging me to produce the best work possible. This moment would not be possible without the unwavering and generous support of Thomas Espenshade. He was instrumental in convincing me to come back to finish a dissertation, as well as selfless in providing me opportunities and flexibility to complete this research. There is no adequate way for me to thank him. Thus, my intention is to pay it forward with as much integrity and humility as he has shown. Thanks to Doug Massey for thoughtful comments, encouragement, and research that has inspired my sociological imagination since my undergraduate days. My appreciation extends to Paul DiMaggio and Scott Lynch for providing such thoughtful and expert suggestions for my continuing research. Special thanks also to Katherine McClelland, my undergraduate adviser, who started me out on this journey. My sincerest gratitude extends to the staff of the Department of Sociology and the Office of Population Research who have made my experiences as a graduate student and doctoral candidate most comfortable. I am particularly grateful to Mary Lou Delaney, Nancy Canulli, Dawn Koffman, Jennifer Flath, Wayne Appleton, and Joyce Lopuh. Much love and respect to those with whom I have shared in the highs and lows of graduate study (although not necessarily as classmates) and academia. Thank you for sharing your knowledge, constructive feedback, and invaluable support. It is your indomitable spirits that often kept me going. Shout outs to: Kate Jaeger, Anita Adhitya,

v

Margarita Mooney, Michelle Bart-Plange, Cristina Mora, Rebecca Casciano, Hilary Levey, Christine Percheski, Sofya Aptekar, Sharon Bzostek, Nick Ehrmann, Chin Jou, Tehama Lopez, Anna Münch, Stefanie Brodmann, Audrey Beck, Karen-Jackson Weaver, David Redman, Elaine Wiley, Leah Targon, Howard Taylor, Patricia Fernandez-Kelly, Ralph Piper, Brenda Joyce, Jason Klugman and Jim Floyd. I have been blessed with many families in my life who have been essential to my journey. Special thanks to the Levins, Saslows, Ziemers, Shaifers, and Hernandezes. Since there are no guarantees that I will have this opportunity again, I have a few more sentimental thanks to particularl family members. To Sylvia, who patiently read my ideas when they were only beginning to form, thank you for allowing me to love you and for loving me back. To James and Gail, thank you for my life. To Nancy Levin, thank you for welcoming me into your family like no one had before. To Thelma Seward, every heart needs a home, thank you for giving me a home when I needed it most; I now live in that home everywhere I go. Because of Nancy and Thelma, I am reminded that life is precious and needs to be lived fully. There are so many more colleagues, teachers, peers, friends and loved ones with whom I have engaged in a multitude of topics both scholarly and otherwise from my first breath through the moment I finish this sentence. However, the full set of people I wish to thank for my intellectual and personal development will not fit on this page. So, whether a person was mentioned here by name or not, I am thankful to all who have helped me to become the person I am.

vi

Chapter 1. Introduction Background Efforts to diversify American higher education, which date back to the later half of the 19th century, have endured largely based on the idea that students from different social, economic, religious, and racial backgrounds obtain cognitive and social benefits from interacting together on a diverse campus (Rudenstine 2001). Several studies claim that interracial interactions are associated with students' self-reported improvements in critical thinking, perspective-taking ability, intercultural competence, and civic engagement (Chang, Astin and Kim 2004; Gurin, Dey, Hurtado and Gurin 2004b; Zhao 2002). Moreover, benefits from interracial interactions extend to the larger society and the workplace where, because of demographic changes and economic globalization, people are increasingly expected to work and interact with people from different backgrounds (Harrison, Price and Bell 1998; Milem 2003). Despite the potential benefits of racial and ethnic diversity, when white and non- white students arrive on ethnoracially-diverse campuses, positive interracial interactions are not assured, as evidenced by reports of self-segregation and ethnoracial clustering on college campuses (Archibald 2006; Chua 2004; Editorial Board 2007b; Tatum 1997). Some researchers argue that integration and affirmative action might contribute to racial tension on campuses (Rothman, Lipset and Nevitte 2003). Historically, racial incidents on college campuses in the form of implied threats through vandalism, heated encounters, and ethnoviolence were not unusual (Ehrlich 1990; Fisher and Hartmann 1995; Hurtado 1992). Despite decades of integrated campuses, racial incidents persist. In a 1998 study of hate crimes on college campuses, the FBI found that almost half (222 of

450) of responding postsecondary institutions reported a hate crime, of which 57 percent were racially-motivated (Wessler and Moss 2001). Campus racial tension became newsworthy again in recent years when colleges were faced with a sudden proliferation of noose incidents in 2007 (Haines 2007; Nealy 2007) and hung effigies of the black Democratic presidential candidate in 2008 (Huckabee 2008). One reason for why positive interracial interactions in college are not always assured is that American students lack experiences with ethnoracial diversity prior to arriving on campus. Using 2005-06 academic year data from the U.S. Department of Education, Orfield and Lee (2007) note that many students still attend ethnically or racially homogeneous elementary or high schools. For example, the average percentage of same-group classmates was 77 percent for white students, 55 percent for Hispanic students, 52 percent for black students, and 23 percent for Asians students in the 2005- 06 academic year. Such racially homogeneous elementary and secondary school composition limits students’ opportunity to interact with peers from different ethnoracial backgrounds. Within-school racial stratification via high school curriculum tracking also limits students’ interactions with diverse classmates (Lucas and Berends 2002; Southworth and Mickelson 2007). Previous studies suggest that racial segregation in neighborhoods and schools might lead students to develop negative predispositions regarding multiethnic settings and situations. Drawing on Allport’s (1954) intergroup contact theory, existing research suggests that students who lack diversity experiences develop negative stereotypes and prejudices regarding other ethnoracial groups that result in avoidance of or anxiety during interracial encounters (Braddock 1980; Braddock and McPartland 1989;

Pettigrew and Tropp 2006; Plant and Butz 2006; Plant and Devine 2003). Thus, the presence or absence of precollege diversity experiences may shape students’ future multiethnic college experiences in particular ways. The consequences of student engagement with diversity is important because by 2036, non-Hispanic whites will comprise less than half of the American population of 18- to 24-year olds (U.S. Census Bureau 2004). As a result, postsecondary institutions could have applicant pools and students that are more ethnically and racially diverse than ever before. Nevertheless, these students, institutions, and society will not realize the aforementioned benefits of interracial interactions if the common American experience of racial segregation inhibits students' attendance, interactions, and comfort level on ethnoracially-diverse college campuses. This dissertation examines the consequences of racial and ethnic segregation in high school on students’ preferences for and experiences with racial and ethnic diversity in college. The dissertation contributes to a growing body of literature on the enduring consequences of racial and ethnic segregation, particularly as it relates to the college experience. The dissertation consists of three related papers presented here as empirical chapters. The analyses in these chapters are based on conceptual frameworks that I develop from research evidence and theories from the fields of sociology, psychology, and education. Using two recent multi-institution student surveys, I evaluate how students' ethnoracial origin and precollege experiences with racial diversity influence their college preferences as well as their interracial interactions and comfort levels in college.

Overview of Analytical Chapters In Chapter 2, I evaluate Braddock’s perpetuation hypothesis, which argues that racial segregation is self-perpetuated over the life cycle and across institutional settings (Braddock 1980). Studies that examine the relationship between segregation and college choice consistently show that black students who attend segregated schools are more likely to choose predominantly black colleges over predominantly white colleges (Wells and Crain 2005). Using data from the 2002 Texas Higher Education Opportunity Project (THEOP) survey of a representative sample of graduating high school seniors in Texas, I examine the association between the ethnoracial composition of students’ high schools and their first college preference. I measure ethnoracial composition as the population share that is the same race or ethnicity as the respondent. My analysis extends previous research by considering not only blacks, but also Hispanics, Asians, and whites. Consistent with Braddock’s claim, my results show a positive association between the ethnoracial composition of schools and first college preference for all demographic groups considered. However, this association disappears for blacks and Hispanics after controlling for the ethnoracial composition of the nearest 2- and 4-year colleges. These findings indicate that geographic context better explains the perpetuation of segregation rather than same-group preferences. In the remaining two empirical chapters, I further examine the consequences of high school ethnoracial composition by evaluating the degree to which students’ precollege diversity experiences shape their students’ collegiate diversity experiences. For these analyses, I use data from the 2004 and 2006 Campus Life in America Student Survey (CLASS), a panel survey of college students from six universities. With these

data, I can evaluate students’ precollege diversity experiences using not only high school and neighborhood ethnoracial composition, but also students’ self-reported frequency and quality of interracial interactions. In Chapter 3, I draw on intergroup contact research, which suggests that individuals who lack positive experiences with other racial and ethnic groups develop negative stereotypes and prejudices regarding other groups that, in turn, inhibit future interracial interactions (Pettigrew and Tropp 2006; Plant and Butz 2006; Stephan, Boniecki, Ybarra, Bettencourt, Ervin, Jackson, McNatt and Renfro 2002). From this research, I test four hypotheses that might explain how precollege diversity experiences are associated with students’ interactions with diverse classmates in college. Until now, previous studies lacked the data to disentangle environmental factors (i.e., ethnoracial composition of environment) from behavioral factors (i.e., interactions with diverse peers) when measuring students’ precollege diversity experiences. Using the CLASS data, I find that high school interracial interactions are a better predictor of students’ collegiate interracial interactions than ethnoracial composition of precollege environment, perceptions of diversity experiences, or attitudes toward other ethnoracial groups. In Chapter 4, I draw on previous studies suggesting that racial segregation in neighborhoods and schools might lead students to develop negative expectations of how they are perceived in integrated settings (Plant and Devine 2003; Wells and Crain 1994). Further, these negative expectations can contribute to feelings of not “belonging” or “fitting in”, which are associated with dropping out of college (Hausmann, Schofield and Woods 2007; Tinto 1975). Until now, previous studies that describe ethnoracial

differences in students’ level of acceptance or comfort on campus lacked data to distinguish race and ethnicity from other personal and institutional characteristics that shape students’ comfort level. Using the CLASS data, I find that students of color are more likely than whites to believe that their ethnoracial group membership impedes their ability to be comfortable in college. Results suggest that attending ethnoracially-diverse high schools and growing up in ethnoracially-diverse neighborhoods lessen students’ perceptions that their ethnoracial background impairs their comfort level on college campuses. However, the association only reaches statistical significance for whites. Diversity Capital In the concluding chapter, I summarize my most significant findings, recommend future research directions, and offer practical and sociological implications based on my research. Further, the conclusion highlights important areas for research related to minimizing the reproduction of social stratification and maximizing the preparation of students for a global community where experience with diversity is becoming a form of capital. Drawing on the concept of social capital, I contend that diversity can be used to secure benefits (Portes 1998). In the same way that education (human capital) and social networks (social capital) increase individual and group productivity, diversity influences the productivity of individuals and groups (Coleman 1988; Portes 1998; Putnam 1995). In addition to the previously mentioned individual cognitive and social benefits from frequent collegiate interracial interactions, research in organizations show that both racial and gender diversity are positively associated with profits and customer size (Herring 2009; Williams and O'Reilly 1998).

Using two recent multi-institution student surveys, I find that students who lack precollege diversity experiences tend to avoid racially-diverse colleges, refrain from interracial interactions in college and feel less comfortable on their college campuses. Considering the persistent high levels of ethnoracial segregation in American schools, the present study has the potential to contribute to our understanding of the reproduction of social stratification. My findings suggest that the absence or presence of diversity experiences could shape where students attend college, the social networks that they develop in college, and their sense of belonging—which could influence their academic success and achievement. Further analysis is needed to establish when and under what conditions those outcomes might occur. Nevertheless, this dissertation provides evidence that one can obtain diversity capital via diverse environments as well as actual interactions with diverse peers. My findings also show how precollege diversity capital can shape students’ collegiate diversity capital. Note on Terminology Throughout this dissertation, unless otherwise stated, I use the term diversity to describe racial and ethnic diversity. Whether my hypotheses or findings hold for other forms of diversity (e.g., socioeconomic, gender, sexuality, religious, political) is beyond the scope of this dissertation. In addition, I use the term race, racial, and ethnoracial interchangeably to denote both Census-defined racial groups (Asians, blacks, and whites) and the Census-defined Hispanic ethnic group. Further, I use the terms interracial, cross-racial, and intergroup interchangeably to denote interactions across the following four ethnoracial groups—Asians, blacks, Hispanics, and whites.

Chapter 2. Ethnoracial Composition and College Preference: Revisiting the Perpetuation of Segregation Hypothesis 1 Background The Supreme Court declared in 1954 that “separate educational facilities were inherently unequal” in its landmark Brown v. the Board of Education decision. The Court reasoned that segregated schools engendered a sense of inferiority among minority students that reduced their motivation to learn and thus deprived them of equal educational opportunities (U.S. Supreme Court 1954). Partly owing to court ordered desegregation, the percentage of black students who attended a majority nonwhite public school fell from 76 to 63 percent between 1968 and 1980, but rose again by 2005 to 73 percent (Orfield and Lee 2007). School segregation has proven resistant to change due to parental aversion to integrated schools (Mickelson, Bottia and Southworth 2008; Renzulli and Evans 2005; Saporito and Lareau 1999), the rise of ethnic-themed charter schools (Gootman 2009; Rimer 2009), and federal court decisions overturning school integration programs (Editorial Board 2007a; Orfield, Bachmeier, James and Eitle 1997). Early studies of racial segregation focused solely on blacks who were denied access to property and space through legal and social institutions for much of American history. Today, segregation is no longer an issue only for blacks. Hispanics surpassed

1 An abridged version of Chapter 2 is forthcoming in ANNALS of the American Academy of Political and Social Science. I am greatly indebted to Marta Tienda, Mark Long, and Sunny Niu for their suggestions as well as Dawn Koffman for her assistance in modifying the database that I use for the analysis. This research was supported by grants from the Ford, Mellon, Hewlett and Spencer Foundations and NSF (GRANT # SES-0350990), with institutional support from the Office of Population Research, Princeton University (NICHD Grant # R24 H0047879). Some of the data used for this study are restricted and therefore not available from the author.

non-Hispanic blacks as the largest minority population in 2003, and concurrently they became the most segregated minority population in schools: the percentage of Hispanics in majority nonwhite public schools rose from 55 percent in 1968 to 78 percent in 2005 (Orfield and Lee 2007; U.S. Census Bureau 2005a). Although high school students are a generation removed from de jure segregation and America is more ethnoracially diverse now than ever before, the schools are as segregated as ever. Braddock (1980) argued that racial segregation is self-perpetuated across institutional settings and life cycle stages both because spatial distribution of resources constrain choices and because individuals raised in segregated settings develop preferences for segregated environments. Such preferences reflect a general uneasiness in unfamiliar settings and a desire to avoid anticipated hostile reactions that might occur during interracial contact (Braddock 1980; Braddock and McPartland 1989). Presumably, young people who attend segregated high schools will choose to attend a segregated college and subsequently work and live in segregated environments as adults. Studies that invoke Braddock’s perpetuation hypothesis to understand college choice consistently find a strong association between segregated high school attendance and preferences for postsecondary institutions with large minority populations, such as Historically Black Colleges and Universities (HBCUs) (Braddock 1980; Wells 1995). Existing studies are based on students who graduated from high school in the 1970s and 1980s. There is less evidence about whether and how high school segregation is associated with college preferences for the growing Hispanic population or students graduating from high schools in the current millennium. This study investigates whether there is a positive association between students’ high school ethnoracial composition and

the ethnoracial composition of their first college preference. Stated as a question, are students who attend high schools with a large proportion of classmates who share their race or ethnicity more likely to prefer colleges with a large proportion of classmates who share their race or ethnicity? Three circumstances warrant revisiting Braddock’s perpetuation of segregation hypothesis at this time. First, the change from de jure to de facto school segregation and the “re-segregation” of blacks in schools presents a different social context from the one originally theorized by Braddock. In addition, changing demographics complicate Braddock’s concept of segregation as a black-white phenomenon. Braddock identified a segregated school as one where blacks comprised 50 percent of the student body or greater. This operational measure of segregation does not have the same meaning in a modern multi-ethnic context. Second, although blacks and Hispanics share similarities, such as high poverty rates, low levels of wealth, and social ills associated with both (Massey and Denton 1993; Oliver and Shapiro 1997), the two groups are very different with regard to their experiences as ethnoracial groups in America. 2 The dissimilarity in arrival history and government classification contribute to differences between blacks and Hispanics with regard to identity formation and stigmatization from the white majority (De Genova and Ramos-Zayas 2003; Omi 2001). Third, recent research on college choice options and geography suggests that characteristics of nearby colleges might explain self-perpetuated segregation in college choice better than same-group preference (Cohen and Brawer 2002; Tienda and Niu 2006a; Turley 2009).

2 The government classifies blacks as a race, categorically distinct from whites, and Hispanics as an ethnicity, which can be a subgroup of any race (Hirschman, Alba and Farley 2000)

College represents a unique opportunity for minority students raised in segregated settings to break the cycle of segregation by selecting a diverse college and gaining access to a wider variety of experiences, knowledge, and social networking opportunities. For many years, HBCUs provided essential postsecondary opportunities for black students denied admission to many postsecondary institutions (Pascarella, Smart and Stoecker 1989). Hispanic-Serving Institutions (HSIs) were not founded as such, but emerged because of regional demographic growth and migrant settlement patterns (Benítez 1998). Consequently, unlike HBCUs, HSIs often lack the rationale, mission, and practices designed to serve the specific educational needs of Hispanic students (Benítez 1998; Hubbard and Stage 2009). Evidence about benefits to attending Minority-Serving Institutions (MSIs) is mixed, partly due to the lack of systematic evidence for HSIs comparable to that available for HBCUs. Allen (1992) finds that black students in the 1980s achieved higher grade point averages when attending black colleges versus predominantly white colleges. Notwithstanding the potential benefits of attending minority-serving institutions, according to Bowen and Bok (1998), blacks who attended predominantly white selective colleges are more likely to graduate than blacks who attended less selective colleges. Furthermore, several studies claim that diverse college campuses enhance students’ critical thinking, perspective-taking ability, intercultural competence, and civic engagement (Chang et al. 2004; Gurin et al. 2004b; Zhao 2002). Students who attend diverse colleges also have access to a wider variety of job opportunities through social networks that they would not acquire at a MSI (Wells and Crain 1994).

Presumably, diverse colleges prepare students for leadership in a globalized workplace and increasingly diverse society (Harrison et al. 1998; Milem 2003). The following section presents a conceptual framework that uses previous research to develop three hypotheses to test and expand Braddock’s perpetuation of segregation hypothesis. After discussing the survey data, I present an analysis plan and report the findings. Results offer limited support for Braddock’s perpetuation of segregation hypothesis for all groups considered. The association between ethnoracial composition of high school and first college preference disappears for both blacks and Hispanics after controlling for the ethnoracial composition of the nearest 2- and 4-year colleges. In light of these findings and rising high school segregation levels, the conclusion discusses the implications of the findings for future research and policymaking. Conceptual Framework Originally proposed by Hossler and Gallagher (1987), the three-stage college choice model provides a framework for understanding how students make the decision to enroll in a college. Cabrera and La Nasa (2000) synthesized the considerable research using this framework to identify three cumulative stages, namely predisposition, search, and choice, as inputs and enrollment decisions as outputs. Between seventh and ninth grade, students develop a predisposition to attend college based on their family background, academic ability, and school characteristics. Based on this predisposition, students develop career and educational aspirations that steer them toward high school coursework and academic achievement aimed at reaching those aspirations. Beginning in tenth grade, students search for potential colleges and universities based on their

predispositions, aspirations, academic qualifications, and their awareness of viable options. After collecting information and narrowing their list of potential institutions, students choose where to apply and to matriculate during their junior and senior years. Students’ choice processes are influenced by perceived institutional characteristics and affordability in addition to the same factors and outcomes that are associated with predisposition and search. This study concentrates on the early stages of the choice process when students choose where to apply, as opposed to the later stages of the choice process when students decide where to matriculate. Racial Differences in College Preferences Existing research on ethnoracial composition of college is largely based on the experiences of black students. Braddock’s (1980) study that proposed the perpetuation of segregation hypothesis is based on a randomly selected sample of 253 black undergraduates from two predominantly black and two predominantly white colleges in Florida in 1973. He finds a large positive association between attending a desegregated college and graduating from a high school with less than 50 percent black students. Braddock concludes that the “data provide supporting evidence for the self-perpetuating tendencies of early school desegregation” (Braddock 1980: 183), but acknowledges the difficulty in generalizing the results beyond blacks in Florida. Selection bias and retrospective reporting weaken his self-perpetuation argument. Because students were selected for the study from among the undergraduate black population at four colleges, it is difficult to know whether students actively sought colleges based on the ethnoracial composition of their high school or whether majority white colleges were more likely to admit students who attended majority white high

schools. One reason that majority white colleges would admit students from majority white high schools is that greater numbers are likely to qualify for admission compared with graduates from predominantly black schools. Aside from the dichotomous measure of desegregated high school, Braddock’s study lacked any additional school-level measures of academic quality or college-going traditions that might be associated with successful admission. Drawing on intergroup contact theory, Braddock suggested that students from majority-black or all-black schools would avoid predominantly white colleges because they lacked experiences to counterbalance stereotypes regarding interracial differences. Braddock suggested these students would underestimate their academic abilities due to “traditional myths and stereotypes concerning black-white intellectual differences;” “underestimate their skill at coping with strains in future interracial situations;” or “overanticipate the amount of overt hostility to be encountered” (Braddock 1980: 181). 3 In Texas, for example, black and Hispanic students, even those at the top of their class, are less likely than whites to prefer and attend colleges with highly competitive admissions standards (Niu, Tienda and Cortes 2006). Despite their potential for success at predominantly white colleges, high-achieving black and Hispanic students who attend segregated schools might, as Braddock suggested, underestimate their ability to compete at selective colleges that tend to be predominantly white. Furthermore, black and Hispanic students from segregated high schools might also be steered toward colleges by

3 Wells and Crain report that an unpublished paper by Braddock affirmed his previous findings based on a representative national sample from the 1980 cohort of the High School and Beyond survey. In this later study, Braddock reportedly finds that “racial composition of a black student's high school exhibits a much larger effect on the racial makeup of the college he/she will likely attend than any of the other factors measured” (Wells and Crain 1994: 542).

the desire to continue their education amongst their friends. Among a sample of 1,539 incoming freshmen at four colleges, black and Hispanic students reported that the college preferences of peers are a significant factor in their college choice decision (Cho, Hudley, Lee, Barry and Kelly 2008). Relatively few studies of high school students directly address whether campus ethnoracial composition influences college choice, and most are qualitative. Using data from group and individual interviews of 158 black college-bound students, counselors, and parents at twenty Los Angeles high schools, Tobolowsky and associates find that some black students avoided predominantly white institutions based on a fear of racial isolation, while others sought these schools in search of experiences with diversity that mirrored the real world (Tobolowsky, Outcalt and McDonough 2005). Their study does not distinguish between whether high school ethnoracial composition generates the observed differences in these student motivations. Based on group interviews with 70 black high school students in five urban cities, Freeman (1999) finds that knowing a graduate or current student of an HBCU is the most consistent predictor of HBCU matriculation. Freeman also finds that black students who attended predominantly white private high schools are motivated to consider HBCUs by a desire to “search for their roots [or find] a connection to the African American community” (Freeman 1999: 99). Alternatively, black students who attended predominantly black high schools are motivated to consider predominantly white colleges to experience different cultures. These findings provide a counterfactual to the perpetuation of segregation hypothesis.

Nearest College Characteristics Frenette (2004) claims that distance to college deters enrollment because of relocation costs, particularly for economically-disadvantaged students. Black and Hispanic students are more likely to be economically disadvantaged than white or Asian students, and therefore might be more inclined to attend a nearby college. Furthermore, existing research shows that the demographic composition of high schools reflects that of the local community (Cohen and Brawer 2002: 47). It is conceivable, therefore, that claims about self-perpetuated segregation in college choices might largely reflect the characteristics of nearby colleges rather than a preference for segregated settings. Using a nationally representative sample, Turley (2009) finds that living near a college increases the likelihood of applying. In particular, she claims that college proximity exerts a large influence on post-secondary choices for students whose parents wanted their child to live at home while attending college. Among Texas high school students, Desmond and Turley (2009) find that Hispanics are more likely than other demographic groups to live at home while attending college. Using the same sample of Texas high school students, Tienda and Niu (2006a) show that students who consider the school’s distance from home an important factor in their college decision were more likely to enroll at less selective public colleges (e.g., community colleges) than students who did not rank this criterion highly. Hypotheses Braddock’s core argument is that individuals raised in segregated settings will avoid interracial settings and prefer segregated settings, because they lack diversity experiences to counterbalance myths and stereotypes related to interracial differences.

Full document contains 196 pages
Abstract: Existing research suggests that racial segregation in neighborhoods and schools might lead students to develop negative predispositions regarding multiethnic settings and situations. However, very little research exists regarding how the presence or absence of precollege diversity experiences shapes students' future multiethnic college experiences. This dissertation consists of three related empirical analyses based on conceptual frameworks that I develop from research evidence and theories from the fields of sociology, psychology, and education. Using two recent multi-institution student surveys, I evaluate how students' ethnoracial origin and precollege experiences with racial and ethnic diversity influence their college preferences as well as their interracial interactions and perceptions of comfort in college. This dissertation contributes to a growing literature on the enduring consequences of racial and ethnic segregation, particularly how experiences of segregation influence individual attitudes and behavior. I find that both opportunities for and direct experiences with diversity prior to college influence students' future engagement with and interpretation of racial and ethnic experiences. In the concluding chapter, I discuss important areas for research and action related to minimizing the reproduction of social stratification and maximizing the preparation of students for a global community where experience with diversity is becoming a form of capital.