Parental Characteristics and the Schooling Progress of the Children of Immigrant and U.S.-Born Blacks

Abstract

In this study, I examine disparities in schooling progress among children born to immigrant and U.S.-born blacks. I find that in one- and two-parent families, children born to black immigrants are less likely to fall behind in school than those born to U.S.-born blacks. In two-parent immigrant families, children born to two immigrant parents have a significant schooling advantage over children born to one immigrant parent. While children born to two immigrant parents in the wealthiest black immigrant families do better in the second generation than in the first, the reverse is observed among children in less wealthy families. These findings contribute in two ways to our understanding of the assimilation processes of children born to black immigrant parents. First, they show that there is a positive association between the number of immigrant parents in a family and children’s schooling performance. Second, they suggest that disparities in the assimilation patterns of the children of black immigrants are a likely product of the interaction between their parental characteristics and the socioeconomic circumstances of their families.

Scholarly interest in the educational characteristics of black immigrants in the United States has been renewed in recent years. Although interest in the educational outcomes of children has generally been limited, an increasing number of recent studies have included children born to black immigrants in schooling comparisons of different groups of immigrant children or comparison of immigrant and native-born children in the United States (Hirschman 2001; Kao 2004; Kao and Tienda 1995). More systematic examinations of their schooling indicators are even more recent, some of which have been associated with concerns about the extent to which black immigrant youths are overrepresented in selective higher educational institutions (Massey et al. 2007). This recent interest in black immigrant youths marks a clear departure from their apparent invisibility in the literature on the children of immigrants. Most previous studies have given more focus to immigrant children in other ethnic groups than they have to the children of black immigrants (e.g., Driscoll 1999; Fuligini 1997; Glick 2007; Hernandez and Charney 1998; Portes and MacLeod 1996). Logically, the large number of Hispanic immigrants in the United States justifies this greater focus. However, the growing importance of African and Caribbean immigration to the United States in recent decades (Knight 1994; Konadu-Agyemang and Takyi 2006; Ricketts 1987) suggests that black immigrants will receive increased scholarly attention in the coming years.

Immigration reforms adopted in the 1965 Immigration and Nationality Act and the introduction of the U.S. Diversity Visa Program in 1990 have played a crucial role in increasing the immigration of black populations to the United States (Konadu-Agyemang and Takyi 2006; Lobo 2001). Indeed, a number of new studies now draw attention to the rapid increases in the immigration of blacks to the United States in the second half of the twentieth century and the impacts of these increases on the composition of the U.S. population. Kent (2007), for example, indicated that between 1960 and 2005, the number of foreign-born blacks in the United States increased by a factor of more than 20, from about 125,000 to about 2.8 million. Logan and Deane (2003) estimated that about 25% of the growth of the total U.S. black population between 1990 and 2000 was due to the immigration of Africans and Caribbeans. The relative importance of immigration flows from these two regions is also changing. Before 1990, for example, the majority of black immigrant arrivals were from the Caribbean. However, between 2000 and 2005, more black Africans arrived in the United States than did blacks from Caribbean countries (Kent 2007). Black immigrant sub-populations, however, have differences and similarities that are described in greater detail in other studies (e.g., Butcher 1994; Dodoo 1997; Kent 2007; Logan and Deane 2003).

In this study, I attempt to bridge the gap in literature on the educational outcomes of children born to black immigrants. Given the attention afforded to the outcomes of adolescents in the literature on children born to immigrants (e.g., Bankston and Zhou 1995, 1997; Fuligini 1997; Harker 2001; Warren 1996), this study will focus on the outcomes of teenage children who are also currently enrolled in high school. I mainly concentrate on comparing the schooling progress of first and second children born to all black immigrants (i.e., black immigrants in aggregate) with that of the children of native-born blacks. However, the analysis ends by paying specific attention to the extent to which disparities in schooling progress are conditional on the ethnic identities of immigrant parents from the major black immigrant ethnicities. Particular attention is also given to the effects of parental and family characteristics as important determinants of children’s schooling progress. On the whole, this study will add to the findings of emerging studies on schooling progress indicators among the children of immigrants (e.g., Tillman, Guo, and Harris 2006), by systematically investigating how specific characteristics of the family context mediate disparities in schooling among youths born to black immigrants.

In pursuing the analysis, I attempt to achieve several objectives. First, I consider the findings of previous studies on the role of parents in their children’s schooling performance to be instructive. A number of such studies conducted among immigrants have pointed to the important role played by immigrant parents in promoting the educational success of their children (Aronowitz 1992; Delgado-Gaitan 1992; Goldenberg et al. 2002). As a result, the study examines the extent to which parental immigration characteristics affect disparities in schooling progress among children in black immigrant families. Using data on children in two-parent families, I test the hypothesis that having two immigrant parents will have a more positive impact on schooling progress than having only one. The study’s second objective is to examine the dynamics of schooling progress among the children of immigrants as they assimilate into the United States, and the extent to which parental and family characteristics mediate the effects of assimilation processes. Thus, to examine possible assimilation effects, I investigate how differences in generational status (i.e., differences between first- and second-generation children) affect disparities in schooling progress among children born to immigrants. Furthermore, I attempt to account for the impacts of differential exposure to the U.S. society, measured by duration of residence, on the schooling progress of first-generation children. A third objective of the study involves the investigation of the role of family structure in mediating the disparities in schooling outcomes. Considering the large number of single-parent families among native-born blacks (London 1998; Rendall 1999), I examine whether differences in household structure account for schooling differences in black immigrant and native-born families and examine whether disparities between immigrant and native-born children are different within single-parent and two-parent families.

Much attention has been given to delineating the effects of schooling progress in childhood on subsequent outcomes in later years. Among preschoolers, for example, Dauber, Alexander, and Entwisle (1996) found that repeating a grade was associated with lower math and English scores later on in middle school. Similarly, Hauser (1999) demonstrated that a history of grade retention among students was an important determinant of subsequent dropout rates. Although Jimerson (1999) reported similar findings, he also confirmed the intuitive expectation that such students were also more likely to have less favorable employment outcomes in subsequent years compared with those who stayed in school.

Against this background, previous findings pointing to differences in educational indicators between immigrant and native-born blacks in postsecondary institutions are instructive. Quigley (1996), for example, found that black immigrants are more likely to attend postsecondary schools after high school than are native-born blacks. Massey et al. (2007) reported that, although both immigrant and native-born blacks in selective colleges and universities had lower GPAs than whites, the difference between black immigrants and U.S.-born whites was smaller than the difference between U.S.-born blacks and whites. Two issues arise from the findings of these studies. First, they raise questions concerning whether the educational disparities between immigrant and native-born blacks are also observed in levels of schooling that precede the college level and the extent to which family contexts account for these disparities. Second, while the selective nature of the institutions examined in Massey et al. (2007) is important, it also points to a need to investigate the nature of the educational disparities between immigrant and native-born blacks using less selective samples.

LITERATURE REVIEW

Insights derived from the social learning or role model explanatory framework indicate that parental influence is a major factor that is conceptually associated with the social indictors of their children (Biblarz, Raftery, and Bucur 1997; Chen and Kaplan 2001; Feng et al. 1999). According to this perspective, the desire of children to imitate their parents is crucial to understanding the process by which social indicators are transmitted across generations. Parents with higher levels of education, for example, are hypothesized to have a positive influence on the educational outcomes of their children because of the expectation that such parents will be seen as good role models by their children (Haveman, Wolfe, and Spaulding 1991). Accordingly, the role model hypothesis will predict that children living with single parents who have lower levels of education will have less favorable schooling outcomes because of the absence of the positive model provided by a highly educated parent within the family and the negative influence of their own parents’ lower levels of schooling.1

Related to the effect of parental influence is the effect of parental aspirations. Several studies have found evidence that parental aspirations do matter in terms of the schooling performance of immigrant children. Kao and Tienda (1995), for example, argued that immigrant parents are more likely than their native-born counterparts to promote educational achievement through the imposition of strict rules on expected GPAs and homework assignments. Kao (2004) further reported that while the parents of immigrant youths are generally less likely to talk about schooling, they are more likely to talk about college and have closer relationships with their children than their native-born counterparts. In another study, Fuligni (1997) identified the strong emphasis on education shared by immigrant parents, their children, and their children’s peers, and argued that this emphasis is an important contributor to the educational successes of East Asian, Filippino, and European youths.

Although systematic investigations of the importance of parental influence and aspirations in black immigrant families are limited, Waters’ (1999) study of West Indian immigrants provides useful insights on the mediating effects of socioeconomic status on both processes and the negotiation of this influence by their children. She reported that middle-class immigrant children were more likely to see strong differences between themselves and native-born black youths primarily because of their perception of native youths as having values antithetical to those of their own immigrant parents. These immigrant parents generally stressed the importance of education, discipline, and a strong work ethic to their children and had other negative impressions of poor native-born blacks. Poorer West Indian youths, on the other hand, were more likely to embrace the black American identity and reject their parents’ negative perspectives of the black American culture. Conceptually, such contrasting perspectives have important implications for immigrant schooling outcomes and how they perform relative to their native-born counterparts—implications that have not been investigated in greater detail in empirical demographic studies.

In the economics literature, discussions on the effects of parents on the educational outcomes of their children are usually framed in terms of a human capital investment framework (Becker 1993; Grawe and Mulligan 2002). This framework describes the positive relationship that exists between children’s schooling and the economic resources of their parents (Taubman 1989) and their families (Behrman and Knowles 1999; Ermisch and Francesconi 2001). Not surprisingly, the income effect is intuitive. Children in families with more resources (e.g., incomes and assets) are more likely to live in areas with better schools and have parents who can afford to pay for supplemental tutoring and other auxiliary services (Alderman, Orazem, and Paterno 2001; Gordon, Bridglall, and Meroe 2004). Parents with higher incomes also have a greater ability to mold the behavior of their children using pecuniary incentives compared with their less wealthy counterparts (Weinberg 2001). In the context of black immigrant and native-born comparisons, human capital investment models predict better schooling indicators among the children of some black immigrant groups that have been found to earn higher incomes compared with native-born blacks (Butcher 1994; Kalmijn 1996; Massey et al. 2007).

Family structure is another factor associated with educational attainment in childhood. Children living in single-parent families or in families that include a stepparent have less favorable educational outcomes even after other demographic and background characteristics are controlled (Pong 1997; Thompson, Hanson, and McLanahan 1994). Explanations for this association are normally related to the presence of only one biological parent, less parental attention, higher poverty rates, and more behavioral problems among children living in single-parent and stepparent families (Aston and McLanahan 1991; Entwisle and Alexander 1995; Heiss 1996; Pong 1997). There is also evidence showing that the association between single-parent households and child educational outcomes varies between single mothers and single fathers (Biblarz and Raftery 1999). In general, however, no extensive attention has been given to the examination of how differences in family structure between immigrant and native-born blacks affect the schooling outcomes of their children.

Beyond parental, family, and socioeconomic factors, other immigrant-specific attributes are also conceptually associated with schooling disparities between children in immigrant and native-born families. English-language proficiency, for example, affects the ability of immigrant parents to navigate the vicissitudes of parent-teacher relationships (Fred-Ramirez 2004) and labor market conditions (Chiswick and Miller 2002; Kossoudji 1988) and is also likely to affect their ability to help their children with their homework. Since many black immigrant parents come from French- and Spanish-speaking countries in Africa and the Caribbean, English-language proficiency will likely affect the extent to which parents can affect their children’s schooling outcomes. Furthermore, at the child level, there is a positive relationship between English-language proficiency and schooling outcomes among children born to immigrant parents (Dawe 2004; Feliciano 2001).

Duration of residence is another important immigrant-specific attribute usually associated with immigrant-to-native-born disparities. Comprehensive discussions of these effects revolve around the various perspectives on assimilation theory normally used to understand how U.S. residence affects immigrant outcomes. In fact, most studies on the schooling outcomes of children born to immigrants identify duration of residence in the United States as an important factor affecting schooling performance. Several perspectives on the nature of the association between duration of residence and immigrant children’s outcomes are now well recognized. The conventional approach, which predicts improved immigrant outcomes in sequential generations, has been criticized for being ethnocentric and of limited use in the understanding of most immigrant incorporation processes (e.g., Rumbaut 1997a, 1997b). Other assimilation perspectives are at variance with the conventional approach (Bankston and Zhou 1997; Portes and Zhou 1993; Rumbaut 1994; Zhou 1997a, 1997b). Proponents of alternative perspectives argue that visible minority immigrant groups are likely to experience patterns of segmented assimilation as they become integrated into society. Empirical support for assimilation effects on immigrant schooling outcomes is mixed, however. Kao (2004), for example, found higher GPAs among first-generation children of black immigrants than among other children. On the other hand, Hirschman (2001) did not find significant decreases in nonenrollment with increased residence in the United States among youths from the Dominican Republic and Cuba. Among immigrant children of other ethnicities, Glick and White (2003) also failed to find meaningful changes in academic trajectories by generational status. However, a number of other studies have found declining levels in other social indicators among immigrants as generational status increases (e.g., Livingston and Kahn 2002).

DATA AND METHODS

Most survey data sources used in the analysis of immigrant children’s schooling outcomes have insufficient numbers of black children (Kao 2004; Mendoza and Dixon 1999). Such sample size limitations are not encountered when using census data. As a result, I use data from a 5% public use sample of the 2000 U.S. census, available in the Integrated Public Use Microdata Series (IPUMS) database of the Minnesota Population Center. Information on respondents’ country of birth and racial identity are used to identify black immigrant parents (i.e., household heads and their spouses) as foreign-born parents who are self-identified as black. Information on each household member’s relationship with the household heads is used to identify their children. Thus, the study considers children born to black immigrants as children living in black immigrant households, that is, households in which either the household heads or their spouses were identified as black immigrants. For analytical simplicity, I focus only on biological children and exclude children with other relationships (e.g., grandchildren, adopted, foster, and stepchildren) from the sample. The 2000 U.S. census data also contain information on educational indicators, as well as information on other demographic and social characteristics of the enumerated population. The educational information is used to identify children currently enrolled in high school (i.e., between grades 9 and 12). Among this group, I limit the study’s focus to children between the ages of 13 and 19.

Analytically, the main indicator used to measure schooling progress captures the relationship between a child’s age and his/her highest schooling grade completed (not current grade). As a result, the empirical outcome of interest in this study is delayed schooling progress or the likelihood of attaining a completed grade level that is lower than what the child is expected to have completed at his or her current age. The study’s schooling indicator is therefore similar to those used in other studies to analyze other grade-for-age relationships—for example, scholastic retardation, schooling progress, some grade retention indicators, and estimates of low grade-for-age (Fields and Smith 1998; Oreopolous, Page, and Stevens 2006; Roderick 1994).

To measure delayed schooling progress, we begin by assuming that children enter the first grade at around age 5 years and 6 months, on average. In practice, this estimate is very close to the true mean age because children in the United States usually begin schooling within the 5- to 6-year age range, with a 12-month age spread being observed among children at every grade level (Angrist and Krueger 1991, 1992; Stypek 2003). Because the 2000 U.S. census was conducted in April 2000 (U.S. Census Bureau 2005), almost all first grade students, for example, would have celebrated their sixth birthdays at the time of the census, while still being in the first grade in the 1999–2000 academic year. This is because the timing of the 2000 census adds at least 7 months to the ages of all students at the start of current school year in 1999. For students who had successfully completed at least one grade level at the time of the census (e.g., students enrolled in the second grade in April 2000), the difference between their age in April and their completed grade level will not be more than 6 if they had never experienced delays in their schooling progress. Theoretically, this difference will remain unchanged in each subsequent April as they move through their schooling careers if they do not experience any further interruptions in their schooling. Implicit in this indicator is the fact that due to differences in state rules on age at entry into the first grade (Angrist and Krueger 1991, 1992), some students may have started their schooling careers at exact age 5 or exact age 6. In both instances, the difference between their age in April and their completed grade level would still be less than or equal to 6, and this difference will be also maintained in subsequent years in the absence of any interruptions in their schooling progress.

With regard to the sample population, students entering the 10th grade, for example, at the start of the school year in 1999 and who had never experienced interrupted schooling progress will, on average, be 15 years old at the time of the 2000 census. By April 2000, therefore, the highest grade level successfully completed by these students would have been the 9th grade because they were still in the 10th grade at the time of the census. The difference between their age (15) and their completed grade level (9) is therefore 6. This intuitive process is what informs the choice of the schooling progress indicator. In the empirical analysis, therefore, the indicator of delayed schooling progress is equal to 1 if a child’s age minus his/her highest grade level successfully completed is greater than 6.

The empirical analysis also attempts to estimate the effects of various household and parental characteristics on the probability of delayed schooling progress (1 = yes, 0 = no). However, because children are clustered within households and thus among parents, standard errors of the estimates of the effects of parental and household attributes are likely to be biased. To account for these biases, I use logistic regression models with household-level random effects to examine how the probability of delayed schooling progress is associated with various child-, parental-, and household-level covariates.

To estimate the effects of children’s own immigration status and the possible impacts of assimilation on schooling progress, the analysis makes a distinction between the outcomes of first- and second-generation children born to black immigrants by using information on each child’s country of birth. First-generation children are identified as the foreign-born children of black immigrant parents, and second-generation children are the U.S.-born children of these parents. For first-generation children, the analysis uses information on years of residence in the United States to investigate the extent to which duration of residence is associated with their schooling progress. Consequently, I distinguish between two cohorts of first-generation children: recently arrived first-generation children (those with 0 to 5 years of residence) and long-term first-generation children (those who have been in the United States for more than 5 years). I then compare how these two cohorts of first-generation children perform relative to their second-generation and native-born counterparts in terms of schooling.

RESULTS

The distribution of the characteristics of the children of immigrants and those of native-born blacks is presented in Table 1. About 11% of all children in the sample were born to black immigrants. The table also shows the ethnic composition of black immigrant parents.2 More than half of the children of black immigrants were born to parents from the Caribbean. In addition, children with black African parents accounted for about one-fifth of the sample of immigrants’ children. Many fewer black immigrant parents were from Latin/South America. Further examinations of the data revealed that more than half of the children with parents from these regions—mainly from Guyana and Belize—had parents that self-identified as having a non-Hispanic origin. Similarly, only about 1% of the children with black immigrant parents from the Caribbean had parents who were also Hispanic black immigrants. In general, these patterns appear to be consistent with reports from recent studies (e.g., Kent 2007) that point to much lower levels of fertility among Hispanic black immigrants relative to their non-Hispanic counterparts.3

Table 1.

Summary Characteristics of Children of Immigrant and Native-born Blacks

Variables	Children of Immigrants	Native-born Children
Child Characteristics
Age (mean)	15.8***	15.9
Female (%)	49.2**	50.2
Generational status (%)
Recent first generation	15.6	—
Long-term first generation	21.5	—
Second generation	62.8	—
Proficient English-language skills (%)	91.0***	97.8
Family size (mean)	4.7***	4.2
Single-parent family (%)	40.6***	59.2
Parental origin of immigrants’ children (%)
Africa	19.44
Caribbean	63.71
Latin/South America	11.30
Other blacks	5.55
Household Head Characteristics
Female (%)	47.9***	62.0
Age (mean)	44.6***	42.3
Education (%)
Four-year college graduate	22.3***	11.7
Associate degree	8.2***	6.8
Some college education	19.0***	26.6
High school graduate	32.1***	39.7
Spouse Characteristics
Female (%)	78.5	79.3
Age (mean)	43.6***	42.6
Education (%)
Four-year college graduate	53.3***	65.6
Associate degree	5.9***	3.1
Some college education	11.3	10.7
High school graduate	19.5***	15.9
N	7,481	60,171

^*p < .05,

^**p < .01,

^***p <.001, for t tests of differences between children born to immigrant and U.S.-born parents.

In terms of generational status, the summary distributions indicate that over half of the children born to black immigrants were second-generation children, that is, children born to immigrants after their arrival in the United States. Not surprisingly, a lower proportion of the children of black immigrants had proficient English-language skills compared with their native-born counterparts. For purposes of this study, children’s English-language skills were assumed to be proficient if they spoke only English or spoke English very well. In terms of family size, the children of black immigrants lived in slightly larger families than did the children of native-born blacks.

More striking differences are observed among the parental characteristics of the two groups of children in the sample. For example, the proportion of children who lived in single-parent families was about 19% lower among children born to immigrants than among their native-born counterparts. Immigrant household heads, however, were more likely to have graduated from four-year colleges or have Associate degrees from two-year colleges. This immigrant advantage is not surprising given the documented evidence of black immigrants’ higher educational attainment levels (Butcher 1994; Johnson and Staples 2005). However, the percentage of immigrant household heads that had some college education or were high school graduates was much smaller compared with the respective estimates of native-born household heads.

Given the higher proportion of black immigrant household heads with college or university credentials, the lower percentage of college graduates among the spouses of immigrant household heads compared with spouses in native-born households was unexpected. However, spouses in immigrant households had a slightly higher likelihood of having an Associate degree, having some college education, or being a high school graduate. In terms of age, spouses in immigrant households were slightly older, as were immigrant household heads, compared with their native-born counterparts.

Table 2 gives a preliminary illustration of the association between delayed schooling progress among children and the educational attainment of household heads in immigrant and native-born families. It also differentiates between the schooling progress of children in single-parent and two-parent families in order to examine the extent to which family structure affects schooling disparities. In general, the percentages of children who attained lower than the required completed grade for their ages, presented in Table 2, are not unusual. Several studies have suggested that between 30% and slightly more than 40% of children in the early teenage years are at a grade that is below what would be expected for their current age (e.g., Guo, Brooks-Gunn, and Harris 1996; Morris 1993; Roderick 1994).

Table 2.

Percentage Distribution of Children With a Lower Grade Completed for Their Age by Type of Family and Parental Migration Status

Household Head’s Educational Attainment	Single-Parent Families		Two-Parent Families
	Immigrant	U.S.-born	Immigranta	U.S.-born
Four-Year College Graduate	33.4	37.1	28.3***	35.4
Associate Degree	30.9*	39.0	32.7	37.7
Some College Education	33.1***	40.1	33.1**	38.7
High School Graduate	38.6	41.1	36.0**	39.7
High School and Below	43.9**	49.3	41.5***	47.1

^aThis refers to families with at least one immigrant parent.

^*p < .05,

^**p < .01,

^***p <.001, for t tests of differences within single-parent and two-parent families.

Other conspicuous patterns of schooling disparities are obvious from the results shown in Table 2. First, there is a general and expected positive association between delayed schooling progress and the educational attainment of household heads.4 In short, children born to the least-educated household heads generally do worse than those born to household heads with more education. In terms of family structure, more children living in single-parent families are at a lower completed grade for their age compared with children in two-parent families, except in immigrant families headed by parents with an Associate degree. This advantage of having two parents is generally consistent in both immigrant and native-born families. More important, however, is the fact that children born to black immigrants are less likely to fall behind in school compared with those born to native-born blacks. In addition, the immigrant–native-born disparities are different for children in single-parent and two-parent families. In two-parent families, the advantage of children of black immigrants over those of native-born blacks is generally larger in families headed by parents with the highest and lowest levels of schooling. Nevertheless, Table 2 also suggests that in terms of reductions in delayed schooling progress, children with U.S.-born black parents derive more benefits from living in two-parent families than from living in single-parent families. For those born to immigrant parents, the difference between living in two-parent versus single-parent families is not as great, especially in families headed by parents with either some college education or an Associate degree.

Multiple Regression Results

The association between schooling progress, generational status, and parental arrangements are examined in greater detail in Table 3. In Model 1, the children’s demographic characteristics and levels of language proficiency are controlled. Accounting for these factors, the results show that children born to black immigrants do significantly better than those born to native-born blacks regardless of whether children of immigrants live in two-immigrant-parent or single-immigrant-parent families. Moreover, there is no indication that the schooling progress of children with black immigrant parents converges with that of children born to native-born blacks as immigrant children assimilate into the United States. On the contrary, the largest advantage in terms of schooling progress is observed among second-generation children born to black immigrants. They are the least likely to have a lower completed grade for their age among the children living in both types of immigrant families. Even more important is the fact that the advantage of children born to black immigrants is larger among children in families with two immigrant parents. In other words, the odds of experiencing delayed schooling progress are lowest among second-generation children with two immigrant parents (i.e., 0.60, or exp(−0.515)), and this is about 40% lower compared with those of children born to native-born blacks. Being a long-term first-generation child born to two black immigrant parents is still associated with better progress in school than being a second-generation child with only one immigrant parent, reinforcing the fact that children with two immigrant parents do better than those with one.

Table 3.

The Probability of Delayed Schooling Progress Among the Children of Black Immigrant and Native-born Parents

Variable	Model 1	Model 2
Both Parents Are Immigrants
Recent first generation	0.001	−0.023
Long-term first generation	−0.412***	−0.391**
Second generation	−0.515***	−0.456***
Only One Parent Is an Immigrant
Recent first generation	−0.011	−0.136
Long-term first generation	−0.192	−0.256
Second generation	−0.327***	−0.253*
Only U.S.-born Parents (ref.)	(0.000)	(0.000)
Proficient in English (yes = 1)	0.180*	0.178*
Child’s Age	1.338***	1.334***
Male	0.349***	0.357***
Female (ref.)	(0.000)	(0.000)
Single-Parent Household	—	0.065*
Family Size	—	0.060***
Log of Total Family Income	—	−0.110***
Constant	−22.29	−21.38
N	67,652	66,268
Log-Likelihood	−36,095	−35,359
Rho	0.406	0.402

^*p < .05;

^**p < .01;

^***p < .001.

In Model 2, additional controls are included to account for basic differences in family structure, family size, and family incomes. Consequently, the disparity between long-term first-generation and second-generation children born to two immigrant parents is reduced. This suggests that some of the advantage of second-generation children observed in Model 1 could be explained by differences in factors associated with the family socioeconomic context. Despite the reduced disparity, the advantage of children born to two immigrant parents remains, as does the better performance of second-generation children among children with only one immigrant parent.

In Table 4, the analysis of disparities in schooling progress is stratified by family structure. In addition, apart from controlling for the demographic characteristics of children and family socioeconomic characteristics, additional models are included with controls for the demographic and educational characteristics of parents. In single-parent families, when only the demographic characteristics of children are controlled (Model 1), the results still show a lower likelihood of delayed schooling progress among second-generation children compared with either first-generation immigrant children or children born to native-born blacks. Additional controls are included in Model 2 to account for the characteristics of the household heads in single-parent families and for differences in family sizes and incomes. Including these controls does not affect the robustness of the second-generation advantage previously observed in Model 1. Rather, when these factors are controlled, long-term first-generation children also have a significant schooling advantage over children with only native-born parents. In two-parent families, Model 3 confirms the advantage of children born to two immigrant parents and the additional advantage of second-generation children living in such families. However, the disparity between first- and second-generation children born to two immigrant parents is diminished when other family attributes (e.g., family size and income) and those of household heads are controlled (Model 4). Furthermore, the disparity between long-term first-generation and second-generation immigrant children with two immigrant parents is slightly reversed when even more controls are added for the characteristics of the spouses of household heads (Model 5). This suggests that the advantage of second-generation children over their long-term first-generation counterparts in two-parent families is mostly explained by factors associated with differences in family socioeconomic contexts and parental attributes.

Table 4.

The Probability of Delayed Schooling Progress in Single- and Two-Parent Households

Variable	Single-Parent Families		Two-Parent Families
	Model 1	Model 2	Model 3	Model 4	Model 5
Both Parents Are Immigrants
Recent first generation			0.057	−0.082	−0.150
Long-term first generation			−0.339**	−0.432***	−0.482***
Second generation			−0.448***	−0.451***	−0.481***
Only One Parent Is an Immigrant
Recent first generation	−0.044	−0.214	0.007	−0.058	−0.307
Long-term first generation	−0.215	−0.300*	−0.258	−0.381	−0.419
Second generation	−0.330***	−0.321***	−0.340**	−0.306**	−0.322**
Only U.S.-born Parents (ref.)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Age	1.377***	1.375***	1.291***	1.290***	1.291***
Male	0.358***	0.366***	0.342***	0.349***	0.349***
Female (ref.)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Proficient in English	0.226***	0.259*	0.127	0.167	0.119
Household Head Characteristics
Age		−0.012***		–0.010**	−0.004
Male		0.031		−0.039	−0.058
College graduate		−0.514***		−0.421***	−0.361***
Some college education		−0.401***		−0.325***	−0.260***
Associate degree		−0.413***		−0.352***	−0.289**
High school graduate		−0.356***		−0.312***	−0.255***
High school and below (ref.)		(0.000)		(0.000)	(0.000)
Proficient in English		−0.218		−0.118	−0.085
Spouse Characteristics
Age					−0.009
Sex					—
College graduate					−0.122
Some college education					−0.212*
Associate degree					−0.273***
High school graduate					−0.232*
High school and below (ref.)					(0.000)
Proficient in English					−0.277*
Family Size		0.027*		0.057***	0.050***
Log of Family income		−0.043*		−0.099***	−0.081**
Constant	−22.91	−21.54	−21.54	−19.89	−19.64
N	38,652	37,461	29,000	28,807	28,807
Log-Likelihood	−20,608	−19,960	−12,590	−15,320	−15,306
Rho	0.414	0.409	0.399	0.391	0.390

^*p < .05;

^**p < .01;

^***p < .001.

Three conclusions can be drawn from the results shown in Table 4. First, in both single-parent and two-parent families, children born to immigrants are less likely to experience delays in their schooling progress compared with the children of native-born blacks. Therefore, like most other racial groups in the United States, there is an immigrant advantage in educational attainment among black youths as far as schooling progress is concerned. Second, the results indicate that the observed advantage of children born to black immigrants compared with those with U.S.-born parents is not explained by differences in parental educational attainment. Children born to black immigrants still do better in one- and two-parent families even after the educational attributes of their parents are controlled. Third, the results show no evidence of a likely convergence in the educational progress of black immigrant children with those of native-born blacks as immigrants assimilate into U.S. society. Rather, they suggest that schooling performance has a positive association with the generational status of the children of immigrants: children born to black immigrants do better with a sequential increase in their generational status. But these results also suggest that in two-parent families, this trend is explained by differences in parental and family social and demographic attributes.

Another important dimension in the analysis of immigrant–native-born disparities involves an investigation of how both groups of children compare within families with similar levels of parental education. Such investigations are important because the social learning hypothesis suggests that the social outcomes of children are a reflection of a desire by children to adopt the characteristics and behaviors of their parents. Adding controls for parental educational attainment is one way to control for the greater positive social learning effect in families with more-educated parents; but another, more robust way to test this hypothesis is to examine how children in both immigrant and native-born families perform when their highest levels of parental educational attainment are similar. Comparisons of this nature are presented in Table 5, which stratifies the analysis by the highest level of parental education within families. In addition, all models in Table 5 include controls that account for whether children lived in single-parent families.

Table 5.

The Probability of Delayed Schooling Progress by the Highest Level of Parental Educational Attainment Within Families^a

Variable	College Graduate	Associate Degree	Some College	High School Graduate	High School and Below
Both Parents Are Immigrants
Recent first generation	−0.348	−0.186	0.317	−0.220	0.112
Long-term first generation	−0.182	−0.804	−0.491	−0.509*	−0.649
Second generation	−0.504***	−0.645*	−0.451*	0.312	−0.473
Only One Parent Is an Immigrant
Recent first generation	−0.508	−0.784	−0.274	−0.053	−0.318
Long-term first generation	0.091	−0.165	−0.652*	−0.500*	−0.207
Second generation	−0.109	−0.083	−0.484***	−0.256*	−0.837***
Only U.S.-born Parents (ref.)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Proficient in English	0.087	0.541*	0.259	0.266	0.152
Child’s Age	1.202***	1.367***	0.417***	1.397***	1.515***
Male	0.368***	0.448***	0.417***	0.337***	0.202**
Female (ref.)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Single-Parent Household	−0.139	0.092	−0.029	0.033	0.048
Family Size	0.036	0.100**	0.009	0.049***	0.044*
Log of Family Income	−0.166***	−0.144*	−0.065*	−0.044	−0.021
Other Characteristics of the More-Educated Parent
Age	−0.009	−0.006	−0.018***	−0.013***	0.000
Male	−0.101	0.088	−0.003	−0.051	−0.097
Proficient in English	−0.067	0.175	−0.187	−0.284	−0.361
Constant	−17.97	−22.16	−19.69	−22.19	−24.31
N	11,065	5,654	18,089	23,669	7,791
Log-Likelihood	−5,878	−2,977	−9,724	−12,562	−6,463
Rho	0.375	0.453	0.369	0.416	0.414

^aWhen the levels of education of household heads and their spouses were the same, the sex, age, and English-language proficiency indicators of the household heads were used as controls. In all other cases, the characteristics of the parent with the highest level of education were used.

^*p < .05;

^**p < .01;

^***p < .001.

In general, the findings indicate that even when the highest levels of parental education are similar, the schooling advantage of children in immigrant families remains. As in the preceding analyses, the children of immigrants generally performed better when they had two immigrant parents than when they had only one. However, Table 5 demonstrates a deviation from this pattern in families in which the highest level of parental schooling was some college education. Second-generation children with two immigrant parents also had the largest significant advantage in families with a parent who graduated from a four-year college or had an Associate degree. Long-term first-generation immigrant children with two immigrant parents, on the other hand, were the least likely to experience delayed schooling progress in families in which the highest level of parental education was completed high school education.

Particularly important, however, is the exceptional assimilation effect observed among children born to immigrants in families in which the more-educated parent either had some college education or was a high school graduate. Even though immigrant children are still less likely to experience delayed schooling progress, children with only one immigrant parent in such families are the only group of immigrant children who experience a significant reduction in their schooling advantage between the first and second generation. This reduction is larger in families in which the highest level of parental education was completed high school education. However, although second-generation children living in these families do not perform as well as their first-generation counterparts, they are still less likely to experience delayed schooling progress than the children of native-born blacks. This association between children’s generational status and schooling progress appears to be quite dissimilar from what is observed in Table 3 and suggests that the dynamics of immigrant parental influence among families with parents with lower levels of education may be different. In terms of the social learning hypotheses, these results would suggest that the lack of highly educated immigrant parents in such families contributes to the significant reductions in the immigrant advantage between the first and second generation.

An alternative interpretation of the results presented in Table 5 is that children born to immigrant parents in lower socioeconomic groups, or those with limited economic resources, experience differential effects of assimilation on their schooling progress. At the same time, the income disparities between immigrant and native-born blacks identified in the literature are also likely to mask the extent to which levels of family income account for differences in the impacts of assimilation processes on schooling progress. This background provides a basis for probing the socioeconomic nexus further. As noted earlier, at least with regard to parental influence among West Indian immigrants, family socioeconomic status affects how parental influence is negotiated among first- and second-generation children (Waters 1999). Table 6 therefore tests the hypothesis that the direction of the disparity in educational progress between first- and second-generation children would be different in lower- and upper-income families and thus reflects differences in their patterns of assimilation. I pursue this test by stratifying children in the sample by family income quintiles and estimating the respective schooling progress disparities between children born to immigrant and U.S.-born blacks. The income quintiles range from quintile 1, which contains families with the lowest levels of income, to quintile 5, which contains the wealthiest families. As shown in Table 6, the median family incomes (M.I.) in these quintiles range from $7,000 in the first quintile to $86,200 in the fifth. All models in Table 6 also include controls accounting for the highest level of parental education. The controls are used for reasons that are mainly theoretical: to account for differences in the influence of parental educational role models within families by income quintile. The models in Table 6 also include controls that account for whether children live in single-parent households and for differences in family size.

Table 6.

The Probability of Delayed Schooling Progress by Median Family Income (M.I.)

Variable	Quintile 1: M.I. = $7,000	Quintile 2: M.I. = $19,600	Quintile 3: M.I. = $32,300	Quintile 4: M.I. = $50,800	Quintile 5: M.I. = $86,200
Both Parents Are Immigrants
Recent first generation	0.866	−0.701	−0.045	−0.223	0.002
Long-term first generation	0.059	−1.063*	−0.196	−0.571**	−0.358*
Second generation	−0.827	−0.535	−0.172	−0.528**	−0.454***
Only One Parent Is an Immigrant
Recent first generation	−0.051	−0.480	−0.343	0.239	−0.120
Long-term first generation	0.395	−0.633*	−0.214	−0.208	−0.622*
Second generation	−0.198	−0.596***	−0.432**	−0.021	−0.319*
Only U.S.-born Parents (ref.)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Proficient in English (yes = 1)	0.263	0.117	0.118	0.277	0.331*
Child’s Age	1.514***	1.415***	−1.344***	1.237***	1.204***
Male	0.426***	0.282***	0.381***	0.343***	0.375***
Female (ref.)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Single-Parent Household	0.006	−0.028	−0.038	−0.123	−0.049
Family Size	0.010	0.048*	0.059***	0.061**	0.029
Log of Family Income	0.091**	−0.223	−0.326	−0.368	−0.200**
Highest Level of Parental Schooling
College graduate	−0.639**	−0.366*	−0.394**	−0.322*	−0.518**
Associate degree	−0.181	−0.446**	−0.510***	−0.311*	−0.469*
Some college education	−0.439***	−0.351***	−0.435***	−0.314*	−0.425*
High school graduate	−0.383***	−0.377***	−0.431***	−0.129	−0.460**
High school and below (ref.)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Other Characteristics of the More-Educated Parent
Age	−0.013*	−0.011	−0.008	−0.008	−0.010*
Male	−0.046	0.035	−0.103	−0.045	0.011
Proficient in English	−0.118	−0.329	0.098	−0.325	−0.227
Constant	−24.9	−20.19	−18.45	−23.69	−17.13
N	12,214	13,519	13,521	13,503	13,511
Log-Likelihood	−6,418	−7,156	−7,154	−7,309	−7,189
Rho	0.463	0.429	0.392	0.385	0.324

^*p < .05;

^**p < .01;

^***p < .001.

Although no significant disparity in schooling progress by parental immigration status is observed in families with the lowest incomes (quintile 1), there is a striking difference in the likely assimilation effects (i.e., the implied difference between first- and second-generation children) on schooling between children in wealthier and less-wealthy families. Among families in the second income quintile, the observed disparity in families with only one immigrant parent closely mirrors that found among families with lower levels of parental education (Table 5). In other words, long-term first-generation children with one immigrant parent do better than either second-generation children or children of native-born blacks in the second-poorest quintile. Surprisingly, similar patterns are also observed in two-immigrant parent families in quintile 4 and in families with only one immigrant parent in quintile 5. Consequently, these results also suggest that the effect of family contexts in mediating differential assimilation patterns among black immigrants may extend beyond the influence of socioeconomic factors. In this regard, the decline in the immigrant advantage observed among families in the fourth quintile is instructive. It indicates that the negative impact of assimilation processes on schooling progress may not be limited to families with the lower parental levels of education but may also extend to nonpoor black immigrant families, even after parental levels of education are controlled.

In the wealthiest black families (quintile 5) with two immigrant parents, the observed assimilation effects are quite different. Second-generation children born to the wealthiest immigrants do much better than their first-generation counterparts and children born to the wealthiest native-born blacks. The advantage of second-generation children does not extend to children of immigrants in the wealthiest families who have only one immigrant parent. The empirical evidence, therefore, suggests that the advantage of second-generation children over their first-generation counterparts observed in Table 3 and Table 4 is mostly driven by the schooling performance of children with two immigrant parents who live in the wealthiest black immigrant families.

To explore the extent to which disparities in schooling progress are conditional on the ethnic origins of black immigrant parents, Table 7 focuses on the comparison between children of native-born blacks and the first- and second-generation children born to immigrants of three broad ethnic groups. Controlling only for child demographic characteristics (Model 1), I find that the second-generation children of African immigrants are the least likely to experience delays in schooling progress and that they are closely followed by second-generation children born to Caribbeans. Comparatively, second-generation children born to parents of other black immigrant ethnicities (i.e., black immigrant parents from Latin/South America and other regions) do less well than other second-generation children but better than their counterparts from native-born families. Model 1 also suggests that improvement in schooling outcomes between the first and second generation occurs only among children with African and Caribbean parents; however, this finding is significant only among the latter. In Model 2, results indicate that the advantage of second-generation children relative to their first-generation counterparts among children born to Africans and Caribbeans is robust. Furthermore, although second-generation children born to African parents still have the best schooling performance, the magnitude of their advantage declines after other family characteristics are controlled. More importantly, accounting for other family characteristics leads to dramatic increases in the schooling performance of first-generation children with Caribbean and “other” immigrant parents. For first-generation children born to other parents, including these additional controls results in a significantly lower likelihood of delayed schooling progress compared with their second-generation counterparts. Seemingly, improvements in the performance of first-generation children in Model 2 imply that adverse family socioeconomic circumstances are more likely to affect the schooling progress of first-generation children than of second-generation children. However, even after these factors are controlled, and regardless of parental ethnic origin, children born to immigrants are still less likely to fall behind than those born to native-born blacks.

Table 7.

Predictors of Delayed Schooling Progress Among the Children of Black Immigrants

Variable	Model 1	Model 2
Children Born to African Parents
First generation	−0.036	−0.037
Second generation	−0.584***	−0.465**
Children Born to Caribbean Parents
First generation	−0.274**	−0.364***
Second generation	−0.409***	−0.408***
Children Born to Other Black Immigrant Parents		(0.000)
First generation	−0.329	−0.427*
Second generation	−0.277*	−0.229*
Native-born (ref.)	(0.000)	(0.000)
Child’s Age	1.338***	1.336***
Male	0.350***	0.359**
Proficient in English		0.212***
Children in Single-Parent Household		−0.015
Family Size		0.039***
Log of Family Income		−0.058**
Highest Level of Parental Education
College graduate		−0.488***
Associate degree		−0.433***
Some college education		−0.398***
High school graduate		−0.348***
High school and below (ref.)		(0.00)
Other Characteristics of the More-Educated Parent
Age		−0.011**
Male		−0.040
Proficient in English		−0.204*
N	67,652	66,268
Constant	−22.11	−20.82
Log-Likelihood	−36,098	−35,292
Rho	0.406	0.401

^*p < .05;

^**p < .01;

^***p < .001.

CONCLUSIONS

A considerable amount of knowledge has been generated from previous studies on the outcomes of children born to immigrants in the United States. However, until quite recently, these studies have generally paid little attention to the schooling outcomes of the children born to black immigrants. In this study, the schooling progress of the children of black immigrants has been compared with that of their native-born counterparts, with particular attention given to how parental characteristics, family socioeconomic contexts, and generational status affect differences in delayed schooling progress.

Among currently enrolled high school students, the analytical results reveal a consistently lower likelihood of delayed progress among the children of immigrants compared with children of native-born blacks. In other words, as far as educational progress is concerned, children of black immigrants are less likely to have lower completed grade levels for their age than children with only native-born black parents. Although immigrant household heads, on average, have higher levels of educational attainment, the results demonstrate that the superior educational attainment of immigrant parents does not explain the schooling advantage of their children. The absence of other indicators, such as GPAs and other test scores, limits our ability to determine whether these disparities are solely the product of the better academic performance of immigrants’ children or are a consequence of other psychological and social factors that disproportionately affect the schooling of native-born blacks. However, because other studies suggest that black immigrant adolescents have higher GPAs than native-born blacks (Kao 2004), the role of GPA differences cannot be discounted.

A more plausible explanation for the consistent immigrant advantage is the role of immigrant parental influences and aspirations on their children. Studies of immigrant families of other ethnicities have shown some evidence that immigrant parents and their children share aspirations of high educational attainment (Fuligini 1997). For West Indian immigrants, Waters (1999) indicated that parents stress the importance of education and a strong work ethic to their children. However, selectivity explanations associated with the higher motivation of immigrant parents and their children cannot be discounted. Intuitively, these positive influences are logically multiplied in two-immigrant-parent families. It is this multiplier effect that I consider to be the most likely explanation for the better performance of children with two immigrant parents. By the same token, having only one immigrant parent is associated with a lower likelihood of delayed schooling progress compared with having only native-born parents. However, unlike for children with two immigrant parents, this effect is mostly important for the U.S.-born (i.e., second-generation) children of one black immigrant parent compared with their first-generation counterparts (Tables 3 and 4).

Important assimilation effects have also been observed in the schooling progress of children born to immigrants. Although first-generation immigrant children generally have a lower likelihood of delayed progress than the children of native-born blacks, the study’s findings indicate that the schooling advantage is even greater among the U.S.-born children of immigrants. In other words, according to Table 3 (Models 1 and 2), being born in the United States has positive effects on the schooling of children born to immigrants since it further reduces the likelihood that the children of black immigrants will lag behind in school. Furthermore, although there is no statistically significant association between schooling progress and recent first-generational status (e.g., in Table 3), the magnitudes of their associated coefficients suggest that schooling progress delays are higher among newly arrived first-generation children and that their schooling progress is much closer to that of native-born blacks.

Additional analyses presented in Tables 5 and 6, however, indicate that the assimilation effects on the educational progress of immigrant children are mediated by the educational attainment of parents and levels of family income. Children whose parents have lower levels of schooling and children in nonwealthy families experience a reduction in their schooling advantage between the first and second generation. The implied trajectory of convergence in the schooling outcomes of children of immigrants born to poorly educated parents relative to those of the children of the native-born is mainly found in families with one immigrant parent (Tables 5 and 6). On the other hand, in the wealthiest black immigrant families, the analysis reveals antithetical assimilation effects as generational status increases. Unlike children in other immigrant families, second-generation children born to two immigrant parents and living in the wealthiest families do better than both their first-generation counterparts and the children born to the wealthiest U.S.-born blacks.

These different assimilation effects are likely a consequence of a combination of factors. First, the lower family socioeconomic status may serve as a barrier to the enabling effect of having either one or two immigrant parents in black immigrant families. Second, as suggested in previous studies, children born to black immigrant parents in lower socioeconomic groups, especially those born in the second generation, may be more likely to identify with poorer native-born blacks and be more unyielding to parental influences than their first-generation counterparts. The fact that the study reveals reductions in schooling between the first and second generation among less poor black immigrants suggests that this possibility may not only be limited to children of poor families.

Woldemikael (1989) has also suggested that black immigration to the United States puts educational institutions in a dilemma associated with their possible responses to the children of black immigrants. In short, they are faced with the choice of treating immigrants as either native-born blacks or as a distinct cultural group. Massey et al. (2007) also suggested that the cultivation of “soft skills” by immigrant parents helps to improve the cultural capital of their children. The cultivation of these skills is an important type of parental influence in immigrant families that is likely to have important implications for the academic success of their children. Since these “soft skills” may to contribute to the generation of stereotypes about black immigrants as being less hostile and “easier to get along with,” they may help resolve the dilemma faced by educational institutions in ways that may favor children with black immigrant parents. To the extent that these stereotypes create disparities in teacher and school responses to student needs that may favor black immigrants, the cultural capital of children of immigrants may indirectly affect their schooling progress and give them an advantage over their native-born counterparts.

By and large, this study has demonstrated that the schooling advantage enjoyed by children born to black immigrants is driven by the outcomes of children born to two immigrant parents. In addition, it has pointed to important disparities in the outcomes of children of different generations that are mediated by parental and family socioeconomic contexts. These findings lay the foundation for further research on the schooling outcomes of children in black immigrant families. Questions concerning how their schooling progress compares with that of nonblack immigrant children in other contexts or how they compare with the children of native-born blacks in terms of other schooling indicators need to be examined further. Other studies could, for example, examine whether the origins of the schooling disparities between children with black immigrant and native-born parents can be found in early childhood. In combination, these studies will represent crucial contributions to the research on the children of immigrants and help to increase the visibility of black immigrant families in future research and policy discussions.