INTRODUCTION
The academic performance of Black students has consistently been of high interest among education scholars and many sociologists since school integration in the mid-twentieth century. With American higher education consistently expanding and becoming more racially and ethnically diverse, examining student group performance is a constant need. Recent scholarship on the academic performance of Black students at elite colleges and universities has glossed over possible disparities among these students (Massey et al., Reference Massey, Mooney, Torres and Charles2007). Specifically, Black U.S. citizens and Black international students in college are assumed to be the only two groups within the larger Black student group. This research has generally lumped Black immigrant students from different regions together, not differentiating between groups. The current study uses a Du Boisian approach to examine the diversity among Black students, but also the similar experiences these students have in relation to being “Black on campus.”
The expanding literature on colorism and the differential treatment of people based on skin tone and other physiological features, the influence of stereotype threat situations that influence African American students’ academic performance, and different global understandings of Blackness and race indicates that more disaggregation of data is needed to examine differences among various Black immigrant populations in the United States. The current study clarifies differences in academic performance by examining four Black student groups at elite colleges and universities in the United States: native Blacks, Black immigrants from Africa, Black immigrants from the Caribbean and Latin America, and Black immigrants from other parts of the globe.Footnote 2 The authors found significant differences not only between Black native students and Black immigrant students, but also differences between different Black immigrant groups of students. These differences existed for both the characteristics of these students and the factors that can influence their academic performance during college. Lastly, and unique to this study, the analyses indicate the possible association between colorism and academic performance among Black students at elite colleges and universities. Overall, our study points to the diversity among Black students in college, but yet it demonstrates a shared experience of “the veil” (Du Bois 2011[1903]) prior to college has lasting effects on their academic performance in college.
AFRICAN-DESCENDED GROUPS AND THE U.S. EDUCATIONAL SYSTEM
The framing of different Black, African-descended populations as a relatively homogenous and indistinct group has characterized the views of the general U.S. public (Feagin Reference Feagin2010), as well as influenced the perspectives taken in immigration and assimilation research dating back to the turn of the twentieth century (Steinberg Reference Steinberg2007). However, the African diaspora includes many people from around the world with distinct histories and experiences. Below, the authors present research on the different educational attainment and experiences of Black, African-descended groups in the United States. Additionally, the possible influence of stereotype threat and colorism on the differential educational experiences and outcomes of these groups is discussed.
In the 1990s, the number of Blacks in the United States who were born in sub-Saharan Africa tripled. During the same time period, the number of Blacks with Caribbean ancestry increased by 60%. According to Census 2000, there were 1.5 million Afro-Caribbeans and over 600,000 Africans living in the United States (Logan Reference Logan, Shaw-Taylor and Tuch2007). This is even more significant given that previous research (Halter Reference Halter, Shaw-Taylor and Tuch2007) has suggested that the Census underestimates the number of African immigrants in the United States. Over 66% of Afro-Caribbeans and almost 80% of Africans living in the United States were foreign born (Logan Reference Logan, Shaw-Taylor and Tuch2007). As compared to Afro-Caribbeans and Africans, African Americans have lower educational attainment and median household income but higher unemployment and poverty rates (Logan Reference Logan, Shaw-Taylor and Tuch2007). In contrast, African immigrants are the most educated immigrant group in the United States (Massey et al., Reference Massey, Mooney, Torres and Charles2007). On average, Africans have higher educational attainment (14 years) than both Afro-Caribbeans (12.6 years) and African Americans (12.4 years). In fact, their attainment is even higher than Whites (13.5 years) and Asians (13.9 years) (Logan Reference Logan, Shaw-Taylor and Tuch2007). In relation to educational achievement, there is evidence of a possible “immigrant advantage” which can be traced to elementary school. Grace Kao and Marta Tienda (Reference Kao and Tienda1995) find that first-generation Black students earn higher grades in elementary school and higher math test scores than native-born Black students. Similarly, second-generation Black students had higher reading scores than native peers. Immigrant parents also had significantly higher aspirations for their children than native-born parents (Kao and Tienda, Reference Kao and Tienda1995; Raleigh and Kao, Reference Raleigh and Kao2010). Elizabeth Raleigh and Grace Kao (Reference Raleigh and Kao2010) find that a larger proportion of Black immigrant parents maintained consistently high aspirations for their children.
Scholars have also documented variation among and between Black immigrant and Black native students in high school. Previous research (Hirschman Reference Hirschman2001) has highlighted country of origin differences in adolescents’ high school enrollment among Black immigrants. Immigrant students are also more likely than native Black students to attend private high schools (Bennett and Lutz, Reference Bennett and Lutz2009; Massey et al., Reference Massey, Mooney, Torres and Charles2007). Extant research (Charles et al., Reference Charles, Torres, Brunn and Gallagher2008) has demonstrated that African immigrant students earn higher GPAs in high school than Caribbean immigrants and native-born Black students. They also earn higher scores on the SAT (Charles et al., Reference Charles, Torres, Brunn and Gallagher2008).
The positive experiences in elementary and high school for Black immigrant students can lead to increased likelihood of college attendance. Black immigrants have higher rates of overall college enrollment than native Blacks (Bennett and Lutz, Reference Bennett and Lutz2009; Hagy and Staniec, Reference Hagy and Staniec2002). However, immigrant and native Blacks differ with regard to the kind of postsecondary institutions that they attend. A larger proportion of native Black students attend Historically Black Colleges and Universities (HBCUs) than their immigrant peers, while almost four times as many immigrant Blacks attend selective colleges than their native counterparts (Bennett and Lutz, Reference Bennett and Lutz2009). African immigrants are more likely to be enrolled in the top ten selective institutions in the National Longitudinal Survey of Freshmen (NLSF) (Charles et al., Reference Charles, Torres, Brunn and Gallagher2008). Overall and among those who enroll in college, immigrant Blacks have higher enrollment rates than Whites at two-year colleges, HBCUs, and selective colleges (Bennett and Lutz, Reference Bennett and Lutz2009).
The noted differences in enrollment rates at selective colleges and universities warrants an examination of the experiences and academic performance of native Black students as well as Black immigrant students from different global regions attending these institutions. Students that attend elite institutions differ from most college students because of their high levels of academic success before entering college. For example, the median SAT score for the students who participated in the NLSF—the dataset used in the current study—was 1243, and 71% of students graduated in the top 10% of their high school class (Massey et al., Reference Massey, Charles, Lundy and Fischer2003). The NLSF does include a variety of different types of elite institutions that vary in their degree of selectivity (as determined by median SAT score), and type (e.g., small liberal arts colleges, flagship state institutions, Ivy League institutions, and private universities). These elite institutions are home to arguably the most heterogeneous Black student populations in American higher education. Douglas Massey and colleagues (2007) suggest that postsecondary institutions’ emphasis on diversifying their student bodies benefited first- and second-generation immigrant Black students, and led to their increased attendance at elite institutions in the United States. According to Massey and colleagues (2003, 2007), nearly a quarter of Black students entering selective colleges at the time the NLSF was collected were immigrants or children of immigrants compared to approximately 5% of Blacks in the general U.S. population, and 13% of all Black eighteen- to nineteen-year-old adults. In fact, the overrepresentation of Black immigrant students (as compared to native-born African American students) at elite institutions became national news following an alumni gathering at Harvard (Rimer and Arenson, Reference Rimer and Arenson2004). Additionally, this overrepresentation at elite institutions does not hold true for all students who report immigrant origins. The percentage of Hispanic and Latino, and Asian Pacific Islander students of immigrant origin are comparable to their share in the U.S. population (Charles et al., Reference Charles, Torres, Brunn and Gallagher2008; Massey et al., Reference Massey, Mooney, Torres and Charles2007). Thus, examining the students attending these selective institutions can provide more information about the different experiences and academic performance of Black student groups, and it could also provide a baseline for future examinations of different Black student groups at other institutions.
Despite differences within the Black immigrant population, survey researchers often ignore information on Black respondents’ place of birth thereby masking divergent ethnic identities (O’Connor et al., Reference O’Connor, Lewis and Mueller2007). When scholars do acknowledge divergent experiences within the Black population, West Indian immigrants have been portrayed as the “model minority” that native-born African Americans should imitate (Model Reference Model2008). This is done even though other scholars have pointed out that generational differences may account for educational attainment disparities (Waters Reference Waters1999). The experiences of African immigrants receive considerably less attention than Caribbean immigrants. Nevertheless, a growing body of research has attempted to explain why Black immigrants tend to have more positive educational outcomes than their native Black counterparts. For example, Massey and colleagues (2007) find that immigrant Black students reported less exposure to violence in their neighborhoods and schools than their African American counterparts.
Some scholars have argued that native-born African Americans may be more susceptible to stereotype threat, which has been shown to depress academic performance (Steele Reference Steele1997, Reference Steele2010; Steele and Aronson, Reference Steele and Aronson1995). In stereotype threat situations, African American students perform worse on academic tests following prompts that hint at possible racial differences in academic performance on tests, or trigger race-related stereotypes about the intellectual ability of African American students. In these situations, African American students are self-conscious of these links between stereotypes of group members’ abilities and the possibility that their performance on academic tests may conform to these stereotypes. With this “performance burden” in mind, African American students do not perform well on these tests as they attempt to avoid conforming to group stereotypes. Kay Deaux and colleagues (2007) completed an experimental study of stereotype threat among students attending college in the City University of New York system. Second-generation West Indian students tested at levels similar to African American students when stereotype threat conditions existed, and both groups were outperformed by first-generation West Indian students. When stereotype threat conditions were absent, there was no difference in the performance of first- and second-generation West Indian students. All of the students in the study believed that West Indians were regarded more favorably than native-born African Americans (Deaux et al., Reference Deaux, Bikmen, Gilkes, Ventuneac, Joseph, Payne and Steele2007).
Suzanne Model (Reference Model2008) tests several theoretical explanations for West Indian immigrants’ socioeconomic success. The only theory that yielded significant empirical results was positive selection. Other scholars have also examined how immigrants to the United States might be more positively selected than those who chose not to immigrate. Cynthia Feliciano (Reference Feliciano2005) compares the educational selectivity of adult immigrants to the United States from thirty-one countries and Puerto Rico. She finds that immigrants from all countries (with the exception of Puerto Rico) are more educated than the general population in their countries of origin. Latin American and Caribbean countries of origin included Jamaica, Haiti, Peru, Columbia, Guatemala, Ecuador, Dominican Republic, Honduras, Cuba, and Mexico (Feliciano Reference Feliciano2005). The countries with the highest representations of Black immigrants in the NLSF include Nigeria, Jamaica, Ghana, Trinidad, and Haiti (Massey et al., Reference Massey, Mooney, Torres and Charles2007). Unfortunately, Feliciano does not examine the educational selectivity of African immigrants. However, she finds modest gender differences in educational selectivity. Differences tend to be greater among immigrants from countries of origin that are farther away from the United States. Female immigrants from distant countries tend to be less selective than males (Feliciano Reference Feliciano2005).
The possible existence of colorism among Black students at elite colleges and universities may be related to differential educational attainment and performance among Black student groups. Colorism is generally a hierarchy in society built on differential perceptions and discrimination of people based mainly on differences in skin color and physiological characteristics associated with European-, African-, and Indigenous-descended peoples. In this hierarchy, or pigmentocracy (Bonilla-Silva and Dietrich, Reference Bonilla-Silva, Dietrich and Glenn2009), people with more European-associated physiological features and lighter skin complexion are associated with higher socioeconomic status (SES) positions and treated more favorably, while people with darker skin complexions and more African- or Indigenous-associated physiological features have lower SES positions and are discriminated against more readily in society. Research has found that colorism exists in a variety of national contexts, and influences the health and SES positions of African-descended populations, Hispanics and Latinos, and Asian and Pacific Islanders (Bonilla-Silva Reference Bonilla-Silva2010; Glenn Reference Glenn2009; Herring et al., Reference Herring, Keith and Horton2004; Hughes and Hertel, Reference Hughes and Hertel1990; Telles Reference Telles2004). Societal norms of “who looks Black” and “who looks Latino” and discrimination have also been found to influence self-identification among multiracial people (Lee and Bean, Reference Lee and Bean2010), and self-identification among Hispanics and Latinos in the United States (Golash-Boza and Darity, Reference Golash-Boza and Darity2008). For the purposes of this study, colorism among elite Black college students may exist in the form of differential academic performance, whereby the skin color measure is significant despite the inclusion of academic preparation, socioeconomic, and other important measures in the analyses.
The current study aims to clarify possible differences in the academic performance and influences on such academic performance in college among African-descended populations enrolled in elite colleges in the United States. As a review of past research indicated, the perception of all Black immigrant groups (and native-born Black students) having similar educational experiences and outcomes is misguided and stereotypically homogenizes a diverse group of people. Our study poses the following questions: What differences exist among Black student populations at elite colleges? What differences exist between Black student populations at elite colleges in relation to their academic performance? The methodological approach taken by the authors to examine the academic performance of Black native and three Black immigrant groups at elite colleges is described below.
DATA AND METHODS
The current study analyzed data from the National Longitudinal Survey of Freshman (NLSF), a project designed by Douglas S. Massey and Camille Z. Charles and funded by the Mellon Foundation and the Atlantic Philanthropies. The NLSF had five waves of data that followed 3098 students from freshman year (1999–2000) until their graduation from college in the spring of 2003 from twenty-eight of the most selective colleges and universities in the United States (based on student SAT scores and class rank and the U.S. News and World Report College Rankings).Footnote 3 Twelve of these institutions are located in the Northeast and eight in the Midwest region of the United States. Five institutions are located in the South and three in the West. Massey and colleagues (2003) chose the sample of institutions to be comparable to William Bowen and Derek Bok’s (Reference Bowen and Bok1998) College and Beyond Survey. The principal modification was the addition of the University of California at Berkeley (Massey et al., Reference Massey, Charles, Lundy and Fischer2003). The demographics of the students who completed the study are as follows: 382 White men, 432 White women, 261 Black men, 537 Black women, 328 Asian American men, 437 Asian American women, 293 Latino men, and 428 Latina women.Footnote 4 Wave 1 corresponded with college entrance of students. Waves 2, 3, 4, and 5 corresponded with the end of freshman, sophomore, junior, and senior years, respectively.
The current study focuses specifically on Black students at these institutions, which totaled 1051 students in the dataset. However, not every student completed college on time, within six years, or at all. Preliminary analyses found significant differences between the student groups regarding transferring, overall college completion, and dropping out. Specifically, Black native students were more likely to transfer from their original college compared to African students and students from Europe, Asia, and the Pacific. Additionally, Afro-Caribbean studentsFootnote 5 were more likely to transfer compared to students from Europe, Asia, and the Pacific. Black native students were the least likely to graduate college within four years, and less likely to graduate college within six years compared to the three Black immigrant student groups. Students from Europe, Asia, and the Pacific were more likely than the other three Black student groups to graduate from college within six years of college entrance. Black native students were more likely to drop out of college, regardless of transferring or not, than Afro-Caribbean students and students from Europe, Asia, and the Pacific. Students from Europe, Asia, and the Pacific were less likely than all three groups to drop out of college.
The preliminary analyses regarding college completion, transferring, and dropping out indicate somewhat of a complex picture. Given that the study examines college academic performance and uses cumulative grade point average as the outcome measure (and students who transfer or drop out will not have complete information on cumulative GPA), the decision was made to only include students who had graduated college within six years of entrance. This resulted in 834 total students in the sample, or 79.4% of the original sample.Footnote 6 Of the students that were included in this final sample, 600 students were Black natives, seventy-four students were African, 113 students were Afro-Caribbean, and forty-seven students were from either Europe, Asia, or the Pacific. This decision resulted in 76% of Black native students, 85.1% of the African students, 88.3% of Afro-Caribbean students, and 100% of students from Europe, Asia, and the Pacific from the original NLSF sample to be used in the current study. Appendix A contains means comparisons between the finishers contained in this study, and the non-finishers who are not analyzed below.
Measures
Many of the measures used in Massey and colleagues’ (2007) research are also used in this study. However, we build on that work, which examined sophomore year GPA by examining students’ cumulative GPA at the end of their senior year of college. Importantly, whereas Massey and colleagues (2007) examined native Black students compared to Black immigrant students, we disaggregate the immigrant student groups to compare each group’s academic and social characteristics, and how these factors may differently influence their academic performance throughout college. Most of these measures examine the precollege social background and environment of Black students from the United States and abroad. The remaining measures examine the friends and academic characteristics of Black students at these elite colleges and universities.
Cumulative Grade Point Average.
Students’ cumulative grade point average (GPA) serves as the dependent variable in this study. The cumulative GPA represents the grades a student received throughout their undergraduate studies. Each student’s GPA is a non-weighted number ranging from 0.00 to 4.00, and represents their final GPA upon graduation.
Native and Immigrant Group Status.
Four dichotomous variables (yes = 1) were used to identify Black students and their group status in this study. These groups included the following: 1) U.S. native students; 2) students from Africa; 3) students from Latin America and the Caribbean; and 4) students from other regions of the world.
Generational Status.
Three dichotomous variables (yes = 1) were used to identify Black students and their generational status in this study. Third-generation students are U.S.-born children of U.S.-born parents. If a student reported that their mother and/or father was born outside of the United States and that the student was born in the United States, they were classified as a second-generation immigrant. When a respondent reported being born outside of the United States, they were categorized as a first-generation immigrant. All NLSF respondents are U.S. citizens or legal residents (Massey et al., Reference Massey, Charles, Lundy and Fischer2003).
Social Characteristics.
Gender was coded as a dichotomous variable (1 = female, 0 = male). Data on students’ skin color were collected in Wave 1 by interviewers. Each interviewer selected the darkness of each student’s skin on a continuum from very light (0) to very dark (10). In line with the colorism literature, we would predict that the differential treatment of students with darker skin complexions would result in lower academic performance compared to peers with lighter skin complexions. The remaining family composition, religion and religiosity, family SES, interracial friendships, closeness, high school performance and characteristics, college major, campus racial climate, whether a student transferred from their original institution, and college selectivity measures are discussed in Appendix B and Appendix C. However, a word of explanation regarding the neighborhood and school segregation measure and the performance burden scale is in order.
Neighborhood and School Segregation.
The segregation levels for the neighborhoods and high schools of the students are based on precollege information collected in Wave 1. Students were asked for the approximate percentage of Whites, Blacks, Asians, and Latinos in their neighborhood when they were finishing high school and among the students in their high school. Following studies by Charles and Massey (Charles et al., Reference Charles, Dinwiddie and Massey2004, Reference Charles, Torres, Brunn and Gallagher2008; Massey et al., Reference Massey, Charles, Lundy and Fischer2003), the percent of Black and Hispanic and Latino students were added together to create a “percent minority.” Following the education and segregation literature (Kozol Reference Kozol1992, Reference Kozol2005; Orfield and Lee, Reference Orfield and Lee2007), students who come from more segregated neighborhood and school environments may have lower levels of academic performance during college.
Performance Burden and Stereotype Externalization.
Generally speaking, stereotype threat is social psychological situation whereby the performance of a stereotyped group, such as Black students, is lowered by triggering the stereotypical beliefs of intellectual and academic inferiority prior to academic tasks (Steele Reference Steele1997, Reference Steele2010; Steele and Aronson, Reference Steele and Aronson1995). A scale that measures performance burden, which is part of stereotype threat situations, was included in this study. Originally constructed by Charles and colleagues (2009), this scale examines the influence of others’ perceptions of a student’s academic performance. A scale of stereotype externalization was also constructed and used in this study. Similar to the scale constructed by Charles and colleagues (2009), this scale examines students’ perceptions that the negative stereotypes held by instructors and other students will affect their evaluation of individual students from that group. Given the high-stress academic environment of the institutions included in this study, the influence of stereotype threat and feelings of a performance burden among Black student populations may significantly hinder students’ academic performance.
Missing Data
Like most survey research projects, the NLSF has missing data that should be addressed prior to analyses. The large amount of missing data in this study warranted a method of imputation to be conducted prior to the analysis phase of the project. These data were used only to run the regression analyses in this study. All descriptive statistics reported below used non-imputed data.
To resolve the issue of missing data, the EM algorithm was used to complete the dataset in LISREL 8.8 (Jöreskog and Sörbom, Reference Jöreskog and Sörbom2007). The EM algorithm is a two-step iterative process to impute or fill in missing observations in a dataset. The first step (the E step) consists of replacing the missing value with a predicted score that results from a series of regressions where all other variables serve as predictors of the missing value for the variable that contains missing data in each case (Duncan et al., Reference Duncan, Duncan and Lu1998; Enders Reference Enders2001). The sums, sums squares, and cross products are then calculated. In the second step (the M step) maximum likelihood estimation produces a covariance matrix and regression coefficients using the raw and imputed data that are used to calculate new estimates for the missing data points for the next E step, when the process begins again (Enders Reference Enders2001). The EM algorithm cycles through these steps until the changes in the covariance matrices resulting from the M step falls below a preset criterion, indicating that the changes are small and trivial.
Analytic Strategies
A three-pronged approach was used to examine Black students at elite colleges and universities. First, using non-imputed data from the NLSF, the descriptive statistics for each of the four Black student groups were calculated. Then means tests (t-tests) were conducted to identify any significant differences between each of the groups.
Second, using the imputed data, regression models tested the factors contributing to Black students’ academic performance in college. These models include variables for students’ perceived skin color, ratings of closeness to Whites and Blacks, student perceptions of the campus racial climate, interracial friendships before and during college, a scale measuring perceived performance burden among students, college major, and the selectivity level of the college students attended. The first set of models estimated the influence of four blocks of variables on Black students’ cumulative college GPA. The first block included only the dichotomous variables identifying whether students were an immigrant from Africa, Latin America or the Caribbean, or another region of the world to gain a baseline understanding of the association between different immigrant statuses and academic performance in college. The dichotomous variable of whether a student is an U.S. native or not served as the reference group. The second block included the student’s gender, their perceived skin color, and whether they were a first- or second-generation immigrant to consider other important student characteristics that are known to influence students’ academic performance in college. The dichotomous variable measuring whether a student is male or not served as reference group.
The third block of variables included dichotomous variables for students’ religion (Catholic, other religion) with Protestant serving as a reference category, religiosity, family income, whether students’ parents had one or two college degrees, whether students’ parents had one or two graduate degrees, the segregation levels of students’ neighborhoods and high schools, and the proportions of White, Black, Asian, and Latino friends students had prior to starting their freshman year of college. Previous research using the NLSF (Charles et al., Reference Charles, Torres, Brunn and Gallagher2008; Massey et al., Reference Massey, Charles, Lundy and Fischer2003, Reference Massey, Mooney, Torres and Charles2007) have indicated the possible influence of these characteristics on students’ academic performance in college. The fourth block of variables included measures of students’ high school academic preparation, which are among the most influential predictors of students’ college academic performance. A dichotomous variable for whether students attended a private high school (public high school attendance served as the reference category), students’ self-rated preparation for college work, students’ high school GPA, and students’ SAT composite score were included. The fifth block of variables measured various college-related aspects that could influence students’ academic performance. These variables included a series of dichotomous variables for students’ college majors (business majors served as the reference category), scales of interracial friendships with Whites, Blacks, Asians, and Latinos during college, feelings of closeness to Whites and Blacks at the beginning and end of college, students’ perception of the campus racial climate, a scale of performance burden, a scale of stereotype externalization, a dichotomous variable indicating whether a student transferred from their original NLSF institution, and a dichotomous variable measuring whether students attended one of the ten most selective schools in the NLSF.
The last aspect of the analyses consisted of group regression models. These models were similar to the regression models described above with a few differences. The regressions for Black U.S. native students did not contain generational status variables. Additionally, group regression models were not conducted for Black immigrant students from other regions of the world. These students could have been from Europe, Asia, Australia, or the Pacific islands regions. These are regions distinct from each other, and have different racial and colonial histories, economic conditions, and educational norms, among many other features. These differences among students and their homes would not justify regression analysis of a “cohesive” group of other immigrant students for this study.
RESULTS
Descriptive Statistics
Table 1 presents descriptive statistics about the social characteristics of Black students from the United States and abroad. Women composed between 64 and 68% of all four Black student groups in the NLSF. In relation to student’s skin color, African students had the darkest skin color as perceived by the interviewers, followed by Afro-Caribbean students, U.S native students, and students from other world regions. All of the means comparisons for skin color were significant. Regarding generational status, 35% of African students were first generation, while 28% of students from other world regions and 27% of Afro-Caribbean students were first generation. 65% of African students, 73% of Afro-Caribbean students, and 72% of students from other world regions were second generation.
Social Characteristics of Black Students at NLSF Colleges and Universities
Note: Analyses conducted with non-imputed NLSF dataset.
a – significant difference (p < .05) between U.S. natives and African immigrants
b – significant difference (p < .05) between U.S. natives and Latin American/Caribbean immigrants
c – significant difference (p < .05) between U.S. natives and immigrants from other global regions
d – significant difference (p < .05) between African and Latin American/Caribbean immigrants
e – significant difference (p < .05) between African immigrants and immigrants from other global regions
f – significant difference (p < .05) between Latin American/Caribbean immigrants and immigrants from other global regions
Turning to Black students’ family composition, African students were significantly more likely to have both parents present in the home (73%) compared to U.S native and Afro-Caribbean students (52%). African students were significantly less likely to come from mother-only homes (20%) compared to U.S. native (41%) and Afro-Caribbean students (44%). All groups had low percentages of students coming from father-only homes with a range between 2–6%.
The families of Black students had 2002 incomes that ranged from $65,416 to $80,560. No significant difference was found between the four student groups’ average family income. U.S. native students (30%) were significantly more likely than African students (19%) to have fathers with college degrees. African students had the highest percentage of fathers with an advanced graduate degree (63%), followed by students from other world regions (36%), Afro-Caribbean students (32%), and U.S. native students (27%). U.S. native students (28%) were significantly more likely than students from other world regions (15%) to have mothers with a college degree. For mothers with an advanced graduate degree, African students had the highest percentage (41%), followed by U.S. native students (27%), and then Afro-Caribbean students (24%), and students from other world regions (23%).
Table 1 also reports the level of integration and segregation of Black students’ neighborhoods before they entered college. African students (52%) were significantly more likely than Afro-Caribbean students (35%) to live in predominately White neighborhoods. African students (23%) were significantly less likely than U.S native (36%) and Afro-Caribbean students (40%) to live in predominately minority neighborhoods.
Table 2 presents the social interactions and perceptions of groups among Black students at elite colleges and universities. Black students from the United States and abroad typically entered college with 26% to 33% of all their friends being White. However, Black native students had a larger range of their friends being Black. U.S. native students had the highest percentage of Black friends at college entrance (59%), followed by Afro-Caribbean students (53%), African students (51%), and students from other world regions (40%). Black U.S native students had a significantly higher percentage of precollege Black friends than students from other world regions, and out of the students from other world regions, Afro-Caribbean students had a significantly higher percentage of precollege Black friends. Students from other world regions had the highest percentage of Asian friends at college entrance (15%), followed by African students (11%), Afro-Caribbean students (8%), and U.S native students (6%). U.S. native students had a significantly lower percentage of precollege Asian friends than African students and students from other world regions, and Afro-Caribbean students also had a significantly lower percentage of precollege Asian friends than students from other world regions. U.S. native students (5%) had a significantly lower percentage of precollege Latino friends than Afro-Caribbean students (8%).
Social Interactions and Perceptions of Black Students at NLSF Colleges and Universities
Note: Analyses conducted with non-imputed NLSF dataset.
a – significant difference (p < .05) between U.S. natives and African immigrants
b – significant difference (p < .05) between U.S. natives and Latin American/Caribbean immigrants
c – significant difference (p < .05) between U.S. natives and immigrants from other global regions
d – significant difference (p < .05) between African and Latin American/Caribbean immigrants
e – significant difference (p < .05) between African immigrants and immigrants from other global regions
f – significant difference (p < .05) between Latin American/Caribbean immigrants and immigrants from other global regions
U.S native students (22%) had a significantly lower percentage of White friends during college than Afro-Caribbean students (28%) and students from other world regions (33%), while African students (22%) had a significantly lower percentage of White friends compared to students from other world regions. Black students from the United States and abroad had similar percentages of Black friends during college, ranging from 30 to 35%. U.S. native and Afro-Caribbean students (5%) had a significantly lower percentage of Asian friends during college compared to students from other world regions (12%). Black native students (3%) also had a significantly lower percentage of Latino friends than students from other world regions (6%).
At college entrance, Black students from the United States and abroad generally felt mildly close to Whites, while they were much closer to other Blacks. U.S. native and African students had significantly higher levels of closeness to other Blacks than students from other world regions. Additionally, Black native students had significantly higher levels of closeness to other Blacks than Afro-Caribbean students. At the end of college, U.S. native students, African students, and students from other world regions had lower levels of closeness to Whites, but Afro-Caribbean students had a slightly higher level of closeness to Whites compared to college entrance. All four Black student groups had lower levels of closeness to other Blacks at the end of college compared to college entrance, although students still felt much closer to Blacks than to Whites. Students from other world regions had a significantly lower level of closeness to other Blacks than the three other Black student groups.
In relation to perceptions of the campus racial climate, all four groups of Black students perceived little discrimination on elite college campuses. However, Black students did have moderately high feelings of performance burden, associated with stereotype threat. African students had the highest level of performance burden, followed by Afro-Caribbean students, students from other world regions, and U.S. native students. U.S. native students had a significantly lower level on the performance burden scale than African students. African students also had the highest level of stereotype externalization, followed by students from other world regions, Afro-Caribbean students, and then U.S. native students. African students had a significantly higher level on the stereotype externalization scale than students from other world regions.
Table 3 presents data on Black students’ academic characteristics in high school and college. In regards to the demographic composition of Black students’ high schools, U.S. native students (43%) were significantly less likely than African students (51%) to have attended predominately White high schools. Students from other world regions were more likely to have attended racially mixed high schools (36%), followed by U.S. native students (32%), Afro-Caribbean students (26%), and African students (27%). U.S. native students (25%) were significantly more likely than students from other world regions (14%) to have attended predominately minority high schools.
Academic Characteristics of Black Students at NLSF Colleges and Universities
Note: Analyses conducted with non-imputed NLSF dataset.
a – significant difference (p < .05) between U.S. natives and African immigrants
b – significant difference (p < .05) between U.S. natives and Latin American/Caribbean immigrants
c – significant difference (p < .05) between U.S. natives and immigrants from other global regions
d – significant difference (p < .05) between African and Latin American/Caribbean immigrants
e – significant difference (p < .05) between African immigrants and immigrants from other global regions
f – significant difference (p < .05) between Latin American/Caribbean immigrants and immigrants from other global regions
Turning to high school academic performance, African students had the highest high school GPA (3.71), followed closely by students from other world regions (3.66), U.S. native students (3.59), and Afro-Caribbean students (3.57). Black students of all groups had average SAT composite scores from 1211 to 1236 points. In general, all Black students felt relatively prepared by their high school work for college-level work.
Regarding where students attended college, U.S. native students were more likely to attend public universities (35%), followed by students from other world regions (30%), Afro-Caribbean students (26%), and African students (23%). African students were more likely to attend private college or universities (77%), followed by Afro-Caribbean students (74%), students from other world regions (70%), and U.S native students (65%). U.S. native students (32%) were also significantly less likely than African (55%) and Afro-Caribbean students (46%) to attend the ten most selective colleges and universities in the NLSF. Conversely, U.S. native students (36%) were significantly more likely than African students (23%) to attend the ten least selective colleges and universities in the NLSF. African and Afro-Caribbean students were more likely to attend Ivy League institutions (24%), followed by students from other world regions (13%), followed by U.S. native students (11%). Black students had low transfer rates, ranging from 2 to 5% of each group.
In relation to college majors, between 4 and 11% of Black students majored in business, and approximately 2 to 4% of Black students majored in communication. African students had the highest percentage of engineering majors (18%), followed by the other three Black student groups (6%). However, African students significantly differed only from U.S. native and Afro-Caribbean students. Humanities majors composed 12 to 19% of Black students at elite colleges. U.S. native and African students had the highest percentage of math and science majors (8%), followed by Afro-Caribbean students (4%), and students from other world regions (less than 1%). Students from other world regions were significantly less likely to major in mathematics or the sciences compared to the other three Black student groups. Social science majors composed between 16 and 20% of Black students. Black students from the United States and abroad had similar percentages of students majoring in other or multidisciplinary majors, ranging from 4 to 7%. Lastly, Afro-Caribbean students had the highest cumulative college GPA (3.17), followed by African students (3.15), and then students from other world regions and U.S. native students (3.03).
Black Student College Academic Performance
Table 4 reports the regression models examining Black student academic performance at elite colleges and universities. The first model contains only the world region immigrant students originate from in comparison to native Black students. This model only explained 0.3% of variance in Black students’ college academic performance, and was not significant. The only group that significantly differed from native Black students in their academic performance was Afro-Caribbean students who had a higher end-of-college GPA. In the second model, students’ gender, perceived skin color, group status, and generational status were entered into the model with men serving as a reference category. This model significantly explained 3.8% of the variance in Black students’ college academic performance. African students and Afro-Caribbean students had higher college GPAs than Black U.S. native students. Black women had significantly higher college GPAs than Black men. Black students with darker skin color had lower college GPAs than Black students with lighter skin color.
Black Student Academic Performance at Elite Colleges and Universities
† p < .1 (one tail test); *p < .05 **p < .01 ***p < .001 (two-tailed tests).
The third model added students’ religious background, family SES, community-level segregation, and precollege friendships to the model. Students who identified as Protestant and families with no college degrees or advanced graduate degrees served as reference categories. The third model significantly explained 10.2% of the variance in Black students’ college academic performance. African immigrants were no longer significantly different than Black natives in the third model, while Afro-Caribbean students had higher college GPAs than Black native students. Black women had significantly higher college GPAs than Black men. Black students with darker skin color had lower college GPAs than Black students with lighter skin color. Black students who came from families with higher family incomes and Black students whose parents had advanced graduate degrees had higher college GPAs. Black students who attended more segregated high schools had lower college GPAs. Black students who had more White, Black, Asian, and Latino friends prior to college had higher college GPAs.
The fourth model added students’ high school academic performance. Students who attended public high schools served as a reference category. This model significantly explained 23% of the variance in Black students’ college academic performance. Similarly to the third model, gender, Latin American/Caribbean immigrant status, family income, high school segregation level, and precollege friendships were significant. However, students’ perceived skin color, and coming from homes with both parents attaining advanced graduate degree were not significant. Although the variable indicating whether students were from Latin America or the Caribbean was not significant at the 95% confidence interval, Latin American/Caribbean students had higher college GPAs than Black native students. Black women had significantly higher college GPAs than Black men. Black students who attended more segregated high schools had lower college GPAs. Black students who had more White, Black, Asian, and Latino friends prior to college had higher college GPAs. Black students who felt that their high school work prepared them for college-level work had significantly higher college GPAs. The higher a student’s high school GPA the higher their college GPA. Additionally, the higher a student’s composite SAT score the higher their college GPA.
The fifth model added college friendships, fields of study, closeness measures, and college-related characteristics. Business majors and non-top ten selective colleges and universities in the NLSF served as reference categories. This model significantly explained 25.5% of the variance in Black students’ college academic performance. Similarly to the fourth model, gender, high school segregation level, precollege friendships, self-rated academic preparation, high school GPA, and SAT score were significant in the final model. Black women had significantly higher college GPAs than Black men. Students’ skin color returned to significance, as Black students with darker skin color had lower college GPAs than Black students with lighter skin color. Additionally, Black students who came from homes where both parents had advanced college degrees had higher college GPAs. Black students who came from more segregated high schools had lower college GPAs. Black students who had more White, Black, Asian, and Latino friends prior to college had higher college GPAs. Black students who felt that their high school work prepared them for college-level work had significantly higher college GPAs. The higher a student’s high school GPA the higher their college GPA. Additionally, the higher a student’s composite SAT score the higher their college GPA. Black students who majored in engineering had significantly lower college GPAs than students who majored in business. Students who majored in the humanities, the social sciences, and other or interdisciplinary majors had higher college GPAs than students who majored in business. Black students who were closer to Whites at the beginning of college had lower college GPAs. None of the remaining closeness measures, college friendship measures, perceptions of the campus racial climate, performance burden and stereotype externalization scales, transferring to another college, or attending one of the ten most selective institutions in the NLSF were significant.
Differences in College Academic Performance among Black Students
Similar regression models to those discussed above were tested for U.S. native, African, and Afro-Caribbean students. The small number of students from other world regions hindered a group analysis from being conducted. Table 5 presents these final group models. The final model of Black native students significantly explained 27.7% of the variance in their academic performance in college. Black women had higher GPAs than Black men. U.S. native students who attended predominately minority high schools had lower college GPAs. U.S. native students with more precollege friendships with Whites, Asians, and Latinos had higher college GPAs. Additionally, U.S. native students who felt their high school work prepared them for college-level work had higher college GPAs. U.S. native students who had higher high school GPAs and SAT composite scores had higher college GPAs. U.S. native students who had more friendships with other Blacks during college had higher college GPAs. U.S. native students who majored in the humanities, the social sciences, and other or interdisciplinary majors had higher college GPAs than students of the same group who majored in business. Lastly, U.S. native students with higher levels of closeness to Whites at the beginning of college had lower college GPAs.
Black Student Group Regression Models
† p < .1 (one tail test); *p < .05 **p < .01 ***p < .001 (two-tailed tests).
Turning to African students, the final model significantly explained 40.5% of the student group’s academic performance in college. African students with more friendships with Asians during college had higher college GPAs, while African students with more Latino friends during college had lower college GPAs. African students who majored in engineering and math or the sciences had lower college GPAs than students of the same group who majored in business. Lastly, African students who had higher levels of closeness to Blacks prior to entering college had lower college GPAs.
In relation to Afro-Caribbean students, the final model significantly explained 21.4% of the variance in the student group’s academic performance in college. Afro-Caribbean students who were more likely to feel their high school work prepared them for college-level work had higher college GPAs. Lastly, Latin American/Caribbean students who had more Latino friends during college had lower college GPAs.
DISCUSSION
This study points to the different social and academic backgrounds of Black students from around the African diaspora that have not been fully explored for these college-going students. Overall, this study found a wide variety of differences in relation to the academic and social preparation for college, and the influences on the academic performance during college among Black student populations at elite postsecondary institutions. This section discusses some of the findings that standout from these analyses and compares findings with that of Massey and colleagues’ (2007) study of Black student groups at elite colleges and universities. Additionally, the authors note several limitations to the current study. Lastly, a few concluding thoughts are presented in relation to future studies of Black undergraduate populations.
The current study expanded on research conducted by Massey and his colleagues (2007) that examined differences among Black native and immigrant students’ social and academic characteristics as well as their academic performance relative to White students at elite colleges and universities. Several differences were found between the current study and their study. These differences were found in relation to skin color, educational attainment of Black students’ mothers and fathers, religion and religiosity, the level of neighborhood and school segregation experienced by students prior to college, high school GPA, and friendship networks entering college. This paper identified significant differences between the Black student groups with regard to parents’/guardians’ educational attainment. For example, African immigrant students’ mothers and fathers are both significantly more likely to hold advanced graduate degrees than native Black students, Afro-Caribbeans, or students from other parts of the world. African immigrant students were significantly less likely than both Black native students and Afro-Caribbean students to reside in predominantly minority neighborhoods. Conversely, African immigrant students are more likely than their native Black peers to have attended predominantly White high schools. All three of these factors (parents’ educational attainment, lower likelihood of residence in minority neighborhoods, and increased likelihood of attending high school with White peers) are associated with better odds of finishing college within six years. On average, Afro-Caribbean students and Black native students have lower high school grade point averages than African immigrants and immigrants from other parts of the world. Generally, African immigrants and native Black students reported having fewer close friendships with White students during college than the other Black student groups. Immigrants from other global regions tended to have more close Asian friends than other groups. With these findings in mind, the current study indicates the importance of disaggregating data to examine different groups; that these students, while having many similar experiences in college illustrated by many of the “non-findings” from our analyses, also have a diversity of experiences that influence their academic performance. Lumping the student groups together, as shown in the analyses above, can often conceal these different experiences and influences on groups in American society, including their experiences in college. One possible reason our results differ from Massey and colleagues’ (2007) research is that our study examined Black students who graduated within six years of entering college, whereas their study examined students through their sophomore year, which is a slightly larger sample size than the current study used in analyses given students who dropped out or transferred since their sophomore year.
According to the results of this study, the persistence of school segregation has lasting effects on Black students’ academic performance in college, particularly U.S. native students. That is, the more segregated a student’s high school, the worse their college academic performance. (In fact, as shown in Appendix A.2, high school segregation also impacts which students finish college within six years of entrance). The significant finding in the regression models that examine all Black student groups at elite colleges may reflect the experience of Black native students given the significant finding in their group-specific regression models. Higher levels of segregation in schools have been associated with resource deficits and lower amounts of academic opportunities for students, particularly for Black and Latino students (Kozol Reference Kozol1992, Reference Kozol2005). Although students may highly rate their academic preparation for college as found in the final model among all Black student groups, these perceptions may not accurately reflect the opportunities and resources (associated with more segregated schools) they had in high school that influence their academic performance once they enter college.
A second group of findings that stands out from the regression analyses relates to the level of closeness Black students had to Whites when they entered college. In the final model among all Black students, and the final models for U.S. native students, students with higher levels of closeness to Whites at college entrance had significantly lower college GPAs. Although this variable was not significant in the final model for African and Afro-Caribbean students, it was negative as well. These findings may indicate the power of the socialization of people around race with many negative stereotypes and views of people from the African diaspora that can develop early in a person’s life (Glenn Reference Glenn2009; Van Ausdale and Feagin, Reference Van Ausdale and Feagin2001). That this finding exists despite the use of many other variables in the models, particularly those associated with academic preparation for college, is surprising and concerning at once. Although the campus racial climate and performance burden scales were not significant in the regression models, these students may perceive a debilitating situation at these institutions prior to stepping onto campus, such as stereotype threat (Steele Reference Steele1997, Reference Steele2010; Steele and Aronson, Reference Steele and Aronson1995). The stereotype threat situation, which is associated with the influence of racial views of Black student academic and intellectual ability by Whites on academic performance, has been found especially among high-achieving students, which happen to be the category of the students in this study. It is also noteworthy that it is precisely those Black student groups that have the highest grade point averages in high school (African immigrants) and college (both African and Afro-Caribbean immigrants) that report the highest levels of performance burden. Therefore, although students may not often experience the overt forms of racial discrimination as measured by the campus racial climate scale or readily connect their academic classroom performance with their group’s outlook, these closeness measures to Whites may indicate the constant awareness of stereotypes about Black intellect and serve as the foundation for experiencing or perceiving the “threat in the air” (Steele Reference Steele1997, p. 614) of stereotype threat that they carry with them as they go through college. Thus, stereotype threat for these students may not inherently be related to Black students’ academic performance and ability in the classroom, but rather their understanding of Whites’ perception of the “worthiness” of Black students to be on elite college campuses in general, somewhat like a metastereotype (Shelton et al., Reference Shelton, Richeson and Salvatore2005; Torres and Charles, Reference Torres and Charles2004). That is, Black students from around the diaspora are aware of the “veil” (Du Bois 2011[1903]) and Whiteness in society and on college campuses that influences the perceptions and treatment (overt or subtle) of them while pursuing their college degree.
Related to stereotype threat as a lived experience in college for Black students of all groups is the possibility that there may be different comparison groups for Black immigrant students and Black native students. That is, Black immigrants may compare their academic achievements to other Blacks in their home countries, whereas native Black students compare their educational experiences to Whites in the United States. If NLSF students were making comparisons in this way, we would expect native Black students to be more susceptible to stereotype threat than Black immigrant student groups. This relates to the oppositional culture perspective which suggests that Black immigrant students are “voluntary minorities” and native Black students are “involuntary minorities” with differing relations toward education and Whites such that native Black students somewhat reject the importance of education and perform worse than Black immigrant students (Fordham and Ogbu, Reference Fordham and Ogbu1986; Ogbu Reference Ogbu2004). It should be noted that all of these students are committed to education as is seen by their degree completion at highly selective colleges. This in-depth connection with education and learning counters the oppositional culture perspective. Moreover, as mentioned earlier, it is the voluntary immigrant groups that are most committed to education (as evidenced by their higher GPAs) that are most impacted by performance burden. This finding both supports existing research on stereotype threat and raises doubts about the applicability of oppositional culture as a cause of academic underachievement. Additionally, a wealth of research has noted the inadequacies of this cultural perspective for explaining the racial disparities in academic achievement, and its dismissal of external factors that contribute to differing levels of educational achievement such as teacher perceptions and tracking (Ainsworth-Darnell and Downey, Reference Ainsworth-Darnell and Downey1998; Downey Reference Downey2008; Harris Reference Harris2011; Tyson Reference Tyson2011). As Angel Harris (Reference Harris2011) aptly titled his volume on this topic, Kids Don’t Want to Fail, yet some Black students still perform poorly despite their positive attitudes and efforts. As mentioned above, Whiteness on college campuses places White students as the proverbial “bar” for a comparison because these institutions were literally built for them (Karabel Reference Karabel2005; Soares Reference Soares2007; Thelin Reference Thelin2004); and their “worthiness” to pursue degrees at such selective institutions goes unquestioned, but the opposite situation faces Black students when they arrive on campus. What can be garnered from these measures is that whether Black students are immigrant or native to the United States, they are navigating the structural depth of Whiteness within education, and this structure can suppress their academic achievement regardless of their educational orientation and efforts in college.
Previous research has stressed the importance of generational status on educational outcomes, particularly the lowering of academic performance with later generational and native status. This study found that generational status did not significantly influence Black students’ grades during college. The findings hint at the possibility that Black students who are immigrants, regardless of generational status, perform better than native Black students in elite colleges. This finding does not counter some of the research discussed above in relation to the influence of generational status, but points to differences in academic performance between Black immigrant and Black native students in college. Thus, these findings still support previous studies (Bennett and Lutz, Reference Bennett and Lutz2009; Charles et al., Reference Charles, Torres, Brunn and Gallagher2008; Everett et al., Reference Everett, Rogers, Hummer and Krueger2011; Kao and Tienda, Reference Kao and Tienda1995; Massey et al., Reference Massey, Mooney, Torres and Charles2007) that have found Black immigrant students to outperform Black native students in education, particularly African and Afro-Caribbean students.
The results of this study indicate the possibility of colorism among Black students at elite college campuses in the United States. The interviewer ratings of students’ skin color did result in significant differences among all four Black student groups in this study with students from other world regions such as Europe and Asia having the lightest skin color, followed by U.S. native students, Afro-Caribbean students, and then African students. In the final model that examined all Black student groups at these colleges, students with darker skin complexions had significantly lower college GPAs than students with lighter skin complexions. This finding was present even with the inclusion of academic preparation and socioeconomic measures in the models. The findings of skin color significance while controlling for students’ SES position is consistent with the colorism findings of the studies mentioned above whereby skin color and SES position are closely connected, and reflect the societal hierarchy of a pigmentocracy and treatment of people based on physiological features.
Limitations
Several limitations to the current study should be mentioned. First, the variables used to construct the campus racial climate scale did not relate to more subtle forms of discrimination. Most of the variables used to construct the scale were associated with more overt discriminatory behaviors, such as those associated with traditional racism and the Jim Crow era. However, as noted above, modern racism is more covert and subtle. Thus, the campus racial climate scale may not capture all of the components of the climate relating to “everyday racism” experienced by racial-ethnic minority students on college campuses (Chou and Feagin, Reference Chou and Feagin2008; Essed Reference Essed1991; Feagin Reference Feagin1991, Reference Feagin2010; Mueller et al., Reference Mueller, Dirks and Picca2007; O’Brien Reference O’Brien2008; Picca and Feagin, Reference Picca and Feagin2007; Wise Reference Wise and Wise2008a, Reference Wise and Wise2008b). An indication that this component of racial-ethnic minority students’ experiences may be missing from the scale is that average scores were low. Nevertheless, they do not indicate perfect disagreement on the items, which would suggest that these students experienced a “discrimination free” campus.
A second limitation the authors would like to mention is the inability to examine separately the college experiences and performances of all Black student groups, specifically Black students from Europe, Asia, and the Pacific. Low numbers of students from these world regions were represented in the NLSF dataset used in this study. Given that the development of race and the treatment of people based on their perceived race has developed differently throughout the world,Footnote 7 it would be inappropriate to analyze a group that combines these students from other world regions together in regression analyses. Although the authors did examine descriptively this combined student group, it was decided that presenting a picture of what other Black immigrant students “look like” in relation to their characteristics would be helpful for future research. These students may have similar experiences at elite colleges and universities compared with other Black student groups, but given the different racial histories of these groups the authors can only speculate how similar these students are to other students on the same campuses. Future research could provide an enlightening perspective on these small student groups among Black students in American higher education.
A third limitation relates to the selectivity of the colleges and universities used in the current study. As most college students are not enrolled in highly selective institutions such as those included in this study, similar examinations of Black student groups at less selective four-year institutions may find different results than those found by the authors. These highly selective institutions often have different support programs and resources for students to use while pursuing their college degrees that may not exist at other, less selective institutions, and can influence their academic performance and college experiences. However, as mentioned above, a large proportion of Black immigrant students in U.S. colleges and universities are enrolled at highly selective institutions. Thus, how much similarity or difference would be found in the examination of Black student groups at non-highly selective colleges and universities is left to be determined. Future research could shed light on this issue and identify how Black student groups at other institutions pursue their college degrees, and the similarities and differences in their academic and social characteristics, particularly among Black immigrant students from different regions of the world.
A fourth, and related, limitation that should be mentioned is the limited disaggregation of African and Afro-Caribbean students in this study. The authors were unable to examine differences among students from specific nations, which would likely result in different findings from those reported above. For example, the academic and social experiences of Haitians and members of the Haitian diaspora are unique compared to other Caribbean populations,Footnote 8 and we would be remiss not to point out that the findings in this study may not generalize to groups at the nation-state level. More research could identify what findings from the current study more readily apply to different immigrant groups and their college experiences.
A final limitation relates to the decision to only examine students who completed their degree within six years of entering college at an NLSF institution. As seen in Appendix A, there are distinct differences between finishers and non-finishers in this data that require future study beyond considering whether a student was African American or from outside of the United States. For example, Black women were more likely to complete their college degrees in six years compared to Black men. Additionally, students who finished their degrees within six years at an NLSF institution felt more prepared for academic work in college, had higher high school GPAs, and attend the most selective institutions in this study. Although the authors did not examine these two groups of students further in the current study due to our interest in cumulative college grade point average, these differences, along with the others noted in the appendix tables, are worth exploring in future studies concerning such areas as college student persistence and retention, support programs, and precollege academic preparation, among other areas of inquiry.
Future Research
The current study found significant differences among four Black student groups at elite colleges and universities in relation to their characteristics, and the factors that can influence their academic performance at those institutions. The approach used by the authors in which Black immigrant groups were not treated as a singular, homogenous group (as past research on academic performance in college has [Massey et al., Reference Massey, Mooney, Torres and Charles2007]), but composed of different groups that have different experiences at the same colleges indicate how much researchers are missing from such approaches. Although this study could not thoroughly examine Black European, Asian or Pacific Islander students or students at the nation-state level, this study found many significant differences among Black immigrant students at elite colleges. Not only were these differences found among immigrant groups, but also significant differences were found between U.S. native students and immigrant students in differing ways. Thus, there is not a singular Black student or Black immigrant student experience at elite colleges, but multiple experiences and types of students that need further examination. Such studies could indicate how to create and sustain a more supportive structure for various student groups of different races and ethnicities and immigrant origins.
More research is needed to understand the influence of the closeness to Whites at college entrance findings in this study. These findings may indicate whether the “threat in the air” of stereotype threat is relevant across campus, not just in test-taking and classroom settings. As the authors suggest above, these findings may relate to the perceived “worthiness” of Black students on elite college campuses, which were established for and continue to be bastions of White privilege and exclusion (Bowen et al., Reference Bowen, Kurzweil and Tobin2005; Feagin et al., Reference Feagin, Vera and Imani1996; Karabel Reference Karabel2005; Soares Reference Soares2007; Thelin Reference Thelin2004). Additional research may clarify the importance of these measures to the study of the campus racial climate and their influence on high-achieving Black students’ academic performance. Moreover, how students perceive these colleges prior to entering in relation to racial discrimination and expectations of race-related experiences would assist with understanding why colleges act as life-changing experiences in various ways relating to race, ethnicity, and cross-race interactions for some students, and not for others.
Future research would benefit from examining the possibility of colorism (differential treatment based on skin complexion and related phenotypical features) in American higher education. The analyses in this study indicated that colorism may influence high-achieving Black students’ collegiate achievement. These findings existed despite the inclusion of an array of other variables such as previous academic performance. More research on the possibility of colorism in American higher education, and how it functions on college campuses could lead to more support for racial and ethnic minority students and reduce the racial disparities found in higher education. Colorism studies may also indicate how racial discrimination persists in a more subtle, covert, and colorblind era of race relations (Bonilla-Silva Reference Bonilla-Silva2010; Feagin Reference Feagin2010), and assist researchers and administrators with identifying ways to combat such aspects of their own campuses.
The current study provides a unique look at the academic and social differences among Black student populations that are present during high school, at the time of college entrance, and while attending highly selective colleges and universities in the United States. The Black student populations examined in this study indicate the diversity among an often homogenized population. The lasting influence of racial segregation, perceptions of Black students’ academic abilities and intellect, and discriminatory treatment based on skin color found in this study also indicate the continuing inequality faced by Black students in education. As can be seen from our study, understanding how students experience college and are influenced while pursing their degrees is an ever-evolving pursuit with a large degree of complexity that still needs further examination in the future.
Appendix A.1. Social Characteristics of Finishers and Non-Finishers among Black Students at NLSF Institutions
Appendix A.2. Social Interactions, Closeness, and Precollege Academic Performance of Finishers and Non-Finishers among Black Students at NLSF Colleges and Universities
Appendix B: Definitions of Variables
Appendix C.1: Construction of Interracial Friendship Scales
Appendix C.2: Construction of Ingroup and Outgroup Closeness Scales
Appendix C.3: Construction of Campus Racial Climate Scale
Appendix C.4: Construction of Performance Burden Scale
Appendix C.5: Construction of Stereotype Externalization Scale
