Achievement and Expectations of Immigrant, Second Generation, and Non-immigrant Black Students in U.S. Higher Education

Research on academic achievement contrasting Black immigrant, second generation, and non-immigrant students as distinct groups is surprisingly sparse in the higher education literature.  This study examined Black immigrant and second generation undergraduates from Africa and the Caribbean and non-immigrant Black American undergraduates, using the contrasting lenses of segmented assimilation theory and cultural ecological theory. Results for academic achievement favored second generation students, consistent with cultural ecological theory, while findings concerning expectations were more consistent with segmented assimilation theory.  However, findings were moderated by gender in complex ways.  This research indicates the need for more comprehensive theories of immigrant student achievement and motivation that incorporate consideration of the context surrounding both emigration from the home country and immigration to the host country.

he literature on academic achievement has not always attended to differences in ethnicity and immigration history among Black youth in U.S. schools. Rather, Black students are too often depicted as a monolithic group, and immigration history is ignored. Further, although a great deal of research has been devoted to the academic trajectories of immigrant and second generation adolescents, relatively little of this body of work has addressed students from Africa and the Caribbean; the largest share of this research has been devoted to immigrant children from Asia and Latin America (cf. Kao & Thompson, 2003;Zhou, 1997). Thus we know relatively little about how the academic experiences and outcomes of immigrant and second generation Black youth compare to nonimmigrant Black youth. For the purposes of this investigation, nonimmigrant Black youth are those born in the US to parents who were both born in the US; immigrant Black youth were born abroad; and second generation Black youth are US born with at least one parent born abroad. Understanding possible within-group variability among Black students may provide key insights to address the pervasive underachievement, typically measured by GPA and standardized test scores, experienced by some Black students, and importantly for society, stem the loss of valuable human capital.
Some early research, based on 1980 census data, indicated that nonimmigrant Black men completed more years of education and earned higher incomes than Caribbean Black immigrant men (Farley & Allen, 1987). However, more recent U.S. census data consistently demonstrated that immigrant Black households from the Caribbean and Africa were more economically and academically successful than non-immigrant Black households. Educational attainment for non-immigrant Black adults 25 years of age and older was 30% with some college, 11% with a Bachelor's degree, and 6% with a graduate degree. Poverty rates for non-immigrant Black households approximated 20% (U.S. Census Bureau, 2008). For Black adults 25 years and older who immigrated from the Caribbean, 38% completed at least some college education; 54% of all households earned in excess of $50,000; and 31% earned $75,000 or more (Schmidley, 2001). Finally, for immigrant Black adults from the African continent, 24% of had attained a Bachelor's degree and 19% attained a graduate degree (U.S. T Hudley -Achievement and Expectations 226 Census Bureau, 2003). Average educational attainment for this group was 14 years of formal schooling, greater than the average attainment of all whites (12.9 years) and Asians (13.1 years) (U.S. Census Bureau, 2004). As well, 45% of African immigrants had a household income in excess of $50,000, while only 13% of families had incomes below the poverty level (Newburger & Gryn, 2009). Clearly, these data show that immigrant Black households from Africa and the Caribbean are outpacing their nonimmigrant Black counterparts in both educational attainment and income.

Portraits of Economic and Academic Success
Research treating Black youth as a monolithic group demonstrates that Black children and adolescents underperform academically on traditional indices of academic performance (e.g., standardized tests, grades, college going rates), relative to U.S. averages across all levels of education (NCES, 2007). Lack of successful participation in higher education is one of the especially persistent and troubling findings among Black youth in the aggregate, given that the life chances and income potential of those earning less than a bachelor's degree are greatly diminished relative to their peers who complete a 4 year degree (Hudley, 2009). However, an emergent, more fine-grained body of research demonstrates the substantial variability within this broadly defined racial category.
Using different national databases, Massey, Mooney, Torres, and Charles (2007) and Bennett and Lutz (2009), for example, found that immigrant and second generation Black students are matriculating at elite institutions at higher rates than their non-immigrant Black peers, although these differences sometimes had much to do with SES and achievement differences between the groups (Bennett & Lutz, 2009). However, other data suggest this overrepresentation was only partly explained by high achievement in high school (Keller & Tillman, 2008). Further, the John Harvard Journal (2004) found that over 50% of the undergraduate Black students at Harvard University were immigrant or second generation youths, though these two generations make up less than 10% of the Black population in the U.S. As well, data from an urban, commuter University revealed that second generation Black freshmen persisted longer at the institution than their non-immigrant Black peers (Jenkins, Harburg, Weissberg, & Donnelly, 2004). Postsecondary participation and persistence are clearly greater for immigrant and second generation Black youth than for their non-immigrant Black peers, and differences in participation are especially evident in elite institutions.
However, a variety of personal characteristics and measures of achievement have not been shown to differ by immigration history among Black students who matriculated in elite institutions (Massey, Mooney, & Torres, 2007). Research in an urban American university with a largely minority student body also revealed no significant differences between immigrant, second generation, and non-immigrant Black students in study habits, attitudes, or GPA (Campbell & Cohen, 2004). However, there is a clear advantage in postsecondary enrollment and persistence for immigrant and second generation Black students that may be somehow relevant to immigration history. The variability in postsecondary indicators among Black students will benefit from further research that clarifies within-group differences.

Getting ready for college
Research with younger students has found a mixed picture. An examination of transcripts (Albertini, 2004) revealed that mean cumulative middle school GPA for immigrant Black middle school students from the Caribbean varied from 1.88 to 1.81 on a four-point scale. Research using standardized test scores as outcome variables (Njue & Retish, 2010) revealed that immigrant Black middle school students from Africa outperformed their non-immigrant Black peers in math skills (29% and 25% pass rate, respectively), but in reading the pattern was reversed (39% and 47% pass rate for immigrant and non-immigrant students, respectively). Taken together, findings demonstrate variability in participation and persistence in higher education despite similar constraints in academic preparation for Black students. These findings raise the possibility that success in postsecondary matriculation and persistence but not in academic achievement may be due to influences that are unique to each group (e.g., experiences of racism, individual and family expectations). Deficient K-12 education and academic preparation are too common for Black students regardless of immigration history (Hudley & Duran, 2012). However, immigration history may at least partially explain observed differences in postsecondary indicators through its influence on academic motivation, including future expectations.

Student Expectations
All Black students typically articulate high expectations, defined as beliefs and plans about the future, to persist and thrive academically beyond high school (Cunningham, Corprew, & Becker, 2009), as well as a strong desires for future life success, (Kao & Thompson, 2003;Solorzano, 1991). In higher education, Black undergraduate students who are immigrants or second generation from Africa or the Caribbean have been shown to choose math and science majors more often than their non-immigrant Black peers. These students reported a belief that such majors represented an expectation for greater adult earning power and future economic benefits from the college degree (Tseng, 2006). Conversely, an in-depth examination of students in higher education using qualitative interviews and participant observation (Berg, 2010) demonstrated that low income non-immigrant Black undergraduates generally reported the expectation that a college degree was not sufficient to assure them of future socioeconomic benefits. These students took into consideration the racial hierarchy in America (Hacker, 1993) and anticipated that their adult life chances could be stymied by a variety of social barriers that constitute institutionalized racism. Crosssectional survey data similarly have revealed that non-immigrant Black undergraduate students enrolled in a selective, historically White institution had low expectancies for future economic benefits from earning a college degree, and expectations for future economic benefits were sharply lower for juniors and seniors than for freshman and sophomores. The awareness of possible barriers to success increased with time among college students who saw barriers persist despite their academic progress (van Laar, 2001). Thus, expectations for future benefits to accrue from an undergraduate degree seem to differ between immigrant and second generation Black students and their non-immigrant Black peers, making expectations a useful indicator to pursue in search of within-group differences.
On the other hand, a reanalysis of the High School and Beyond data collected in the 1980's (Beattie, 2002) reported that state level income data did not predict Black students' postsecondary enrollment decisions, suggesting that those decisions may not be driven by the expectation of economic benefits. However, state level aggregate data may be too distal an indicator. As well, the ethnic composition of the racially identified Black student subsample was not reported, clearly demonstrating the need for nuanced examinations of within-group variability among monolithically described Black samples. Research on low-income, non-immigrant Black male high school students' beliefs about school success (Irving & Hudley, 2005) found that students who perceived their academic opportunities to be limited also expressed negative expectations about the future benefits of educational success. Expectations for economic benefits of a college education appear to diverge by immigration history and social class. However, studies that do not report possible ethnic and immigration history in their racially identified Black samples may yield results that obscure differences by immigration history.

Gender
Findings of gender differences in educational attainment for African descent youth have been documented in the literature for several decades. As a monolithic group, Black female students on average achieve higher levels of education than Black male students at almost all levels of education (Grant and Rong, 2002). When one considers within-group variability, immigrant Black women have the highest postsecondary enrollment rates of any race/gender/immigrant group (81%) and non-immigrant Black men the lowest (52%). These gender patterns conform to the current pattern in the overall U.S. population of women attaining more years of school than men (NCES, 2007). Higher achieving girls have also expressed greater aspirations and expectations for future academic success and life attainment than their lower achieving female peers and all male peers (Honora, 2002). Interestingly, as noted above, some of the more debilitating patterns of achievement motivation and future expectations (e.g., perceptions of limited academic opportunities) are more characteristic of non-immigrant Black males but not females. These data clearly point to the strong possibility of an immigration history by gender interaction in the educational achievement and expectations of Black students.

Theoretical Explanations for Differential Academic Success
Existing theories that foreground structural (e.g., Ogbu, 1983) or cultural variables (Portes & Zhou, 1993) have disagreed somewhat on potential trajectories of immigrant and second generation Black students. Ogbu's cultural ecological theory posits that, in a pluralistic society, all ethnic minority groups that have voluntarily joined the society (i.e., immigrated) should have more educational success than their non-immigrant ethnic minority peers. Based on their voluntary incorporation into a pluralistic society, immigrants and second generation minorities are more optimistic about the link between educational preparation and economic success. They have strong expectations that education will pay off once they learn to navigate the language and culture of the host society, and their success is measured relative to their peers in their home countries. In contrast, groups that were incorporated into the society in an involuntary manner (e.g., through enslavement or conquest) are less likely to expect academic achievement to serve as a direct pathway to economic success. Nonimmigrant minorities compare themselves to the dominant society and find persistent inequality.
Involuntary minority youth may expect life opportunities to be suppressed by racial discrimination in social and employment spheres rather than facilitated by educational preparation (Irving & Hudley, 2005), a perspective that somewhat diminishes academic motivation and success in high school (Irving & Hudley, 2008). Even academically successful non-immigrant Black college students attributed negative future outcomes to discrimination, although attributions were significantly more pessimistic for students in the third year of college and beyond (van Laar, 2001). Thus, perceived structural barriers apparently do have an impact on achievement and expectations for non-immigrant Black students, i.e., a minority group with a history of enslavement, even among high achieving students. This typology of immigrant and involuntary minority groups, however, has been criticized as rigid and unable to distinguish effectively between immigrant and involuntary status beyond the initial generation of immigrants.
The theory of segmented assimilation has drawn on somewhat different constructs of incorporation to describe educational adjustment, focusing specifically on second generation youth (Portes & Zhou, 1993). This model theorizes that the segment of American society into which a particular immigrant group assimilates will determine the academic trajectory of their youth, i.e., the second generation. Assimilation into varying segments of society is determined largely according to the race of the immigrant group, the pattern of residence upon first settling in this country, and available economic ladders for upward mobility.
Immigrant Black families with few resources who settle in inner cities in proximity to non-immigrant Black families, according to the theory of segmented assimilation, will find their adolescents under pressure to embrace an oppositional identity similar to non-immigrant youth to cope with structural racism and economic marginalization. This identity choice may block generational upward mobility in ways that are not true for immigrants who enjoy more resources when they arrive, those who are able to settle in suburban or rural areas, or immigrant groups who are not phenotypically similar to minorities victimized by prejudice and discrimination in the U.S. Immigrants who enjoy personal and community resources (education, wealth, previously established strong ethnic communities) may be able to provide access to opportunities in the host culture, protection from potential downward mobility, and opportunities for ethnic solidarity for their youth, irrespective of racial heritage or residential location (Thomas, 2009).
Empirical evidence partially supports segmented assimilation. High school students from the Caribbean have been shown to differ at a rate of 7 to 1 in their identification with their unique ethnic heritage (e.g., Jamaican, Dominican) as a function of SES (57% for middle class vs. 8% for low income youth) (Waters, 1999). More generally, a segmented assimilation perspective draws some support from a sizeable body of research which concludes that in general, youth from lower SES families tend to attain lower levels of education than their more economically advantaged peers (Brooks-Gunn, Duncan, & Aber, 1997;McLoyd, 1998). Thus, it is entirely reasonable to anticipate that immigrant families who arrive with some degree of wealth and join co-ethnic communities with some measure of access to the opportunity structure will enjoy more upward mobility in the second generation than their less advantaged peers. Interestingly, research on academic persistence (Tauriac & Liem, 2012) has demonstrated direct effects of SES and indirect effects of generational status (through high school grades) on Black students' persistence in predominately White institutions. Cultural ecological theory and segmented assimilation theory make distinct predictions about the impact of generational status and family income on academic achievement and future expectations among Black students. The current study will examine these contrasting predictions.

The Current Study
Possible reasons why GPA and future expectations might differ among immigrant Black and non-immigrant Black students, particularly college students, have not been fully clarified in the education literature. The current study examined GPA and future expectations among Black undergraduate students who were either first or second generation immigrants or nonimmigrants. This integration of race and immigrant status in a single study contributes to our understanding of the role these two factors play in Black college students' achievement and future expectations. The study goal was to examine contrasting predictions of cultural ecological theory and segmented assimilation theory in a sample of students who were enrolled in a selective, historically white (now predominantly white and Asian), multi-campus University system. Academic GPA in higher education is an appropriate outcome of interest, as achieving an undergraduate degree has become an important credential for entry into the middle class in the United States (Wilson, 1999).
Further, the diminishing numbers of Black students on historically white college campuses has become a national emergency, so examining this population can yield especially important insights into supporting academic success among this population. As well, student participants have matriculated into a selective institution, indicating some degree of prior academic success among all groups; thus possible group differences will less likely be a function of systematic differences in academic history. Finally, recall that our brief review indicates that very little of the research examining postsecondary success has considered the influence of multiple categories of immigrant history among this population; thus this study integrates immigrant status into an examination of higher education outcomes and expectations for Black students.
Consistent with segmented assimilation theory and with recent research that accounted for within-group diversity in a sample of Black undergraduates (Campbell & Cohen, 2004) but counter to cultural ecological theory, I posed two specific hypotheses. I hypothesized that undergraduate GPA would differ by family income rather than immigrant history. I also hypothesized that students' expectations regarding the future benefits of academic success would differ by parental income, as predicted by segmented assimilation theory, but not by immigrant history (as would be postulated by cultural ecological theory). Based on pervasive findings of gender differences in academic attainment, I also expected these hypothesized relationships to be moderated by gender.

Sample
The sample was drawn from the University of California Undergraduate Experience Survey (UCUES), a census that surveys students from each of eight UC undergraduate campuses. From that database, I selected all of the students who self identified as Black, including either immigrant Black students or non-immigrant Black students. International students were excluded. First-generation participants were those who were foreign-born and immigrated to the United States prior to matriculation at the University. Second-generation youth were born in the United States, but had at least one foreign-born parent. Non-immigrant participants were U.S.-born, and both their parents were U.S.-born. The final sample (N = 820) includes only those cases with data on both parent income and immigration history 1 , and includes all students from freshman to senior class standing. The sample was 11% first generation immigrant (M age = 20.6), 35% second generation immigrant (M age = 20.2), and 54% non-immigrant (M age = 20.3), as well as 36% male, with a mean family income of $55,700.

Procedure
Participants enrolled in 2004/05 responded to an email invitation sent to the entire student population to participate in a web based survey. Data were also obtained from the registrar's office at each campus, including current GPA and self reported family income. Each participant was directed to log in to a secure site maintained by the University. The survey had to be completed in a single session requiring approximately 20 minutes. Once all surveys were completed, all personal identifiers were stripped from the data made available to researchers in order to protect the confidentiality of respondents.

Measures
Participants responded to a core set of items requesting demographic information; participation in and satisfaction with campus activities, programs, and services; self-assessment of academic skills; and educational plans and long-term goals. The analyses reported here concentrate on questions of GPA and academic expectations. Cumulative GPA at the end of spring 2004 and 2003 parent income both were collected from the relevant records at each campus. Parental income, initially reported as 14 categories, was recoded into a 4 level variable. Families were identified as low income if reported family total income was less than $20,000, as moderate income if total family income was between $20,000 and $50,000, as middle income if family income was between $50,001 and $100,000, and families were identified as affluent if total family income was over $100,000. These categories are based roughly on the Federal Poverty Guidelines (U.S. Social Security Administration, 2010) to distinguish low and moderate income, and Federal Census Data to distinguish the 2 lowest categories from the 2 highest categories (U.S. Census, 2008). Three survey items measuring students' goals while in college were used to examine participants' expectations of what they "hope to achieve." One item, "prepare for an advanced degree" measured educational expectations; a second item, "prepare for my chosen career," assessed occupational expectations; and "earn a lot of money in my chosen career" reflected economic expectations. Responses were measured on a scale of 1-6, with higher numbers reflecting stronger endorsement of the statement.

Preliminary Analyses
To identify any systematic group demographic differences, I examined means, standard deviations, and intercorrelations for all of the study variables. Table 1 reports intercorrelations among variables as well as the means for the full sample and separately by immigration history. For parent income, the sample was 19% low income, 28% moderate income, 32% middle income, and 21% affluent. The income distribution did not differ significantly by gender (χ 2 [3, N = 835] = 6.50, p = .09) or generational status (χ 2 [6, N = 830] = 9.17, p = .16).

Hypothesis Tests
I used ANOVA to examine the first hypothesis, which anticipated that GPA would differ by income rather than immigration status. Final 2004 GPA was the dependent variable and sex of student, 3 categories of immigration history, and 4 categories of family income were the grouping factors. A significant 3-way interaction (F[6,169] = 2.58, p < .02, η 2 = .06) emerged (See Figure 1), revealing a much more complex relationship than hypothesized. Partially confirming the hypothesis, family income influenced GPA, and effects were moderated by gender, as expected; however, differences by immigration history were also present. Note that all of the cell sizes for immigrant males and one-half of the cell sizes for immigrant females were very small (n < 10); thus, findings must be interpreted with extreme caution. Post hoc test revealed that among female participants (top panel), immigrant women's GPA was higher for the 2 lowest income groups, non-immigrant women's GPA was highest for the 2 highest income groups, and second generation women had the highest overall GPA among all women. Conversely, immigrant men's GPA was highest for the 2 higher income groups, and second generation men's GPA was highest for the highest income group (bottom panel). Non-immigrant men had the lowest GPA overall, irrespective of family income. Thus, student achievement is influenced by family income and moderated by gender as expected, but the effects of immigration history are also significant.
I used MANOVA to examine the second hypothesis, which stated that student academic expectations would differ by family income, and effects would be moderated by gender. The 3 expectation measures were the jointly dependent variables; sex of student, 3 categories of immigration history, and 4 categories of family income were again the grouping factors. A significant multivariate 3-way interaction (F[6,698] = 2.66, p < .02, η 2 = .06) emerged that was supported by a single significant univariate interaction for occupational expectations (see Figure 2). Again results must be interpreted with extreme caution. For immigrant women, higher family income related to higher occupational expectations. Conversely, second generation low income women expressed the highest occupational expectations. For non-immigrant women, occupational expectations were generally high irrespective of family income. For men, the most affluent and lowest income participants expressed the highest occupational expectations in both immigrant and non-immigrant groups. However for second generation men, the most affluent group expressed the highest expectations, while the low-income group expressed the lowest expectations of any group in the sample. Again findings only partially support the hypothesis, as family income and immigration history both demonstrate significant effects on occupational expectations. However, effects were moderated by gender, as expected. Note that among immigrant participants, only high income women but high and low income men had high expectations. Among second generation participants, low-income women had among the highest expectations but low-income men had the lowest expectations.
Counter to hypotheses two, neither family income (as expected) nor immigration history influenced educational expectations.
Rather, a significant main effect of gender emerged for (F[1,698] = 3.91, p = .05, η 2 = .03); women endorsed this expectation (M = 4.71, sd = 1.2) significantly more strongly than men (M = 4.36, sd = 1.2). Finally, economic expectations revealed a trend for the interaction of immigration history and parent income (F[6,697] = 1.83, p = .09, η 2 = .02), but counter to hypotheses three, the effect was not moderated by gender (see Figure 3). Affluent immigrant students and low and moderate income non-immigrant students endorsed this expectation more strongly than their peers in other income/immigration history groups. In sum, hypotheses were only partially supported, as both family income and immigration history were significantly related to 2 of the 3 measures of expectations. As well, gender effects were present in only 2 of the 3 expectation measures.

Discussion
This study set out to examine the contrasting predictions of cultural ecological theory and segmented assimilation theory in a sample of Black undergraduate students. Research on educational questions has typically not attended to immigration history in this population. The results presented here reveal a more complex picture of the relationship between immigration history and family economic circumstances than either theory would postulate. However, these complex and interesting results must be interpreted very cautiously; recall that cell sizes for some of the gender/immigrant/SES groups were quite small.
In considering the findings for hypothesis 1, results for academic achievement were somewhat consistent with the often reported second generation advantage, or the relative optimism among immigrants postulated by cultural ecological theory. However, these effects often manifested in conjunction with income effects and were moderated by gender. Family income was inversely related to GPA for immigrant women, unrelated to GPA for second generation women, and positively related to GPA for nonimmigrant women. For men's findings, GPA was positively related to family income for first and second generation immigrants, a direct contrast by gender for immigrants, and unrelated to GPA for non-immigrant men, who were the lowest achieving group.
Low income immigrant women's higher achievement may reflect their belief, consistent with cultural ecological theory's prediction of immigrant optimism, that education provides opportunity in the host country to transcend economic barriers. This belief may be particularly encouraging for women immigrating from home societies with traditional gender roles. Further, girls in Caribbean immigrant families, for example, are taught that they must become financially independent in the U.S. (López, 2003), and one might speculate that belief in the economic benefits of an education would be strongest in families without economic means to sponsor their daughters into the opportunity structure. By the third generation, messages about education may be more consistent with the income advantage often found in U.S. populations. The effects of family income for male participants are consistent with a large body of work that demonstrates the particular vulnerability of Black boys and men in our society (Noguera, 2001). Findings suggest that parental economic resources serve as a protective factor for immigrant and second generation males, consistent with segmented assimilation theory but not for non-immigrant males, consistent with cultural ecological theory.
Turning to findings on academic expectations, gender differences in the expectation to prepare for graduate school are unsurprising, given the substantial evidence that women are completing most graduate programs at rates higher than men (NCES, 2010). The findings relating to occupational expectations are more challenging to interpret, as the 3-way interaction yielded very small cell sizes. The two groups (low income immigrant females and low income second generation males) who reported significantly lower expectations than their peers had small cell sizes. The finding would be somewhat consistent with segmented assimilation theory in that low income youth are less certain about their future career prospects, and inconsistent with cultural ecological theory in that these two groups do not display the expected immigrant optimism.
Turning to other groups, family income interacted with immigration history to influence occupational expectations most strongly for low income non-immigrant males, low income second generation women, and affluent immigrants overall. However, cell sizes again were small, particularly for immigrant students. Immigrant males' responses are consistent with segmented assimilation theory, in that more affluent immigrant families may provide more role models of successful professionals and entrepreneurs. The finding of high expectations for second generation, low income women is again somewhat consistent with cultural ecological theory's prediction of immigrant optimism and an expectation that education can be especially valuable for women's self-sufficiency. For second generation women for whom barriers of language have largely been surmounted, expectations of occupational success are unsurprising and consistent with cultural beliefs. It is important to note, however, that the sample overall, irrespective of family income, immigration history, or gender, endorsed this expectation very strongly, in the range of 4-6 on a 6 point scale (scale point definitions ranging from "important" to "essential"). Less than 2% of the entire sample (n = 8) rated this item in the 1-2 range ("not important"; "of little importance", respectively).
Finally, the reported trend for economic expectations revealed differences by income and family history, but differences were not moderated by gender. Immigrant Black students whose families were economically successful rated the expectation that a college education would yield high earnings significantly higher than their peers in the lower income groups. These differences by family income were similar but of a lesser magnitude in the second generation sample. Conversely, non-immigrant Black students demonstrated the reverse pattern, with low and moderate income students rating the expectation for a high income after college significantly higher than their more affluent peers. Immigrant responses are consistent with a segmented assimilation interpretation, in that affluent immigrants and their children expect to continue the family's economic success. The families of affluent first and second generation students might also have been similarly advantaged in their home country, an advantage that could explain their initial ability to emigrate. In particular, immigrant families from African countries bear considerable expense to travel to the United States. Cultural ecological theory would have predicted that all immigrant adolescents, consistent with the principle of immigrant optimism, would have high economic expectations.
The responses of non-immigrant Black students are also inconsistent with cultural ecological theory, which would predict that all students (Ogbu, 2003), and particularly low income students, would have relatively low outcome expectations. Prior research (Irving & Hudley, 2005) found that low income high school adolescents' perceptions of institutional racism relate to lower expectations to gain access to the opportunity structure, regardless of their academic efforts. Similarly, low income college students, who have perhaps attended less advantaged schools or lived in poorly resourced neighborhoods, might also hold lower expectations of material reward for educational efforts. However, I speculate that low income, nonimmigrant Black adolescents who were qualified to matriculate in a selective institution maintain high expectations because they have a strong belief in their own potential (Hudley, 2009).
In sum, cultural ecological theory proposes that the first generation may arrive with dreams of a better life but struggle with language and cultural adjustment, while the second generation has begun to overcome some barriers and can concentrate on attaining opportunities for success that were lacking in their home country (Ogbu, 1983). However, the effects of immigrant optimism appear in conjunction with family income effects in these data, consistent with segmented assimilation theory, and findings are moderated by gender. While more definitive conclusions comparing the relative roles of gender, income, and immigration history must be left to future research with larger sample sizes, these findings suggest that Black students unsurprisingly consider a 4 year institution the place to prepare for a future career. These data are thus consistent with a large body of research that indicates students in 4 year institutions uniformly expect their higher education to prepare them for a chosen future occupation (Metz, Fouad, & Ihle-Helledy, 2009), although ethnic minority students anticipate more barriers to success (Fouad & Byars-Winston, 2005), consistent with cultural ecological theory.

Limitations
These findings are limited by a sample size that makes complex interactions with small cell sizes subject to extremely tentative interpretation. It is quite possible that cell sizes alone can explain the lack of interaction effects detected by these analyses. The declining enrollment of Black students on traditionally White campuses makes this population simultaneously compelling and increasingly elusive for disciplined research. Further, a sample drawn from a single higher education system leaves open the question of generalizability. Given the geographic breadth of the U.S., these findings may not generalize to other regions of the country, although the University of California system increasingly enrolls students from across the country. As well, the database used for this secondary analyses provided only a single measure of each of the 3 dependent variables concerning students' expectations. Given these limitations, the results of this investigation of student expectations, while promising, are exploratory and must await replication with larger national samples and stronger measures that permit more sophisticated and powerful data analytic strategies.
However, this exploratory study is one of only a few to examine withingroup variability as a function of immigration history and family resources and thus makes an important contribution in furthering our understanding of the variability inherent in the population of Black students. In general, findings indicate, consistent with Suarez-Orosco (2001), that theoretical formulations about immigrant motivation and achievement must include consideration of the context surrounding both emigration from the home country and immigration to the host country. I would similarly argue that for all students, not only immigrant students, personal and motivational variables can make visible the within-group variability in academic achievement for Black students in a manner that is not possible using structural explanations (Hudley, 2009). Comprehensive models that can explain the largest amount of variance in academic achievement and motivation for all students will be those that can successfully incorporate historical (e.g. immigration history), contextual (e.g., neighborhood and family resources), and individual (e.g., expectations) variables.

Notes
1 Immigrant and second generation Black students were primarily from Ethiopia, Ghana, and Nigeria in Africa and Jamaica and Belize in the Caribbean.