ethnic differences in educational achievement essay

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Understanding the Influence of Race/Ethnicity, Gender, and Class on Inequalities in Academic and Non-Academic Outcomes among Eighth-Grade Students: Findings from an Intersectionality Approach

* E-mail: [email protected]

Affiliation Centre on Dynamics of Ethnicity, Department of Social Statistics, University of Manchester, Manchester, United Kingdom

Affiliation Australian National University, Acton, Australia

• Laia Bécares, 
• Naomi Priest

• Published: October 27, 2015
• https://doi.org/10.1371/journal.pone.0141363
• Reader Comments

Socioeconomic, racial/ethnic, and gender inequalities in academic achievement have been widely reported in the US, but how these three axes of inequality intersect to determine academic and non-academic outcomes among school-aged children is not well understood. Using data from the US Early Childhood Longitudinal Study—Kindergarten (ECLS-K; N = 10,115), we apply an intersectionality approach to examine inequalities across eighth-grade outcomes at the intersection of six racial/ethnic and gender groups (Latino girls and boys, Black girls and boys, and White girls and boys) and four classes of socioeconomic advantage/disadvantage. Results of mixture models show large inequalities in socioemotional outcomes (internalizing behavior, locus of control, and self-concept) across classes of advantage/disadvantage. Within classes of advantage/disadvantage, racial/ethnic and gender inequalities are predominantly found in the most advantaged class, where Black boys and girls, and Latina girls, underperform White boys in academic assessments, but not in socioemotional outcomes. In these latter outcomes, Black boys and girls perform better than White boys. Latino boys show small differences as compared to White boys, mainly in science assessments. The contrasting outcomes between racial/ethnic and gender minorities in self-assessment and socioemotional outcomes, as compared to standardized assessments, highlight the detrimental effect that intersecting racial/ethnic and gender discrimination have in patterning academic outcomes that predict success in adult life. Interventions to eliminate achievement gaps cannot fully succeed as long as social stratification caused by gender and racial discrimination is not addressed.

Citation: Bécares L, Priest N (2015) Understanding the Influence of Race/Ethnicity, Gender, and Class on Inequalities in Academic and Non-Academic Outcomes among Eighth-Grade Students: Findings from an Intersectionality Approach. PLoS ONE 10(10): e0141363. https://doi.org/10.1371/journal.pone.0141363

Editor: Emmanuel Manalo, Kyoto University, JAPAN

Received: June 10, 2015; Accepted: October 6, 2015; Published: October 27, 2015

Copyright: © 2015 Bécares, Priest. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Data Availability: All ECLS-K Kindergarten-Eighth Grade Public-use File are available from the National Center for Education Statistics website ( https://nces.ed.gov/ecls/dataproducts.asp#K-8 ).

Funding: This work was funded by an ESRC grant (ES/K001582/1) and a Hallsworth Research Fellowship to LB.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The US racial/ethnic academic achievement gap is a well-documented social inequality [ 1 ]. National assessments for science, mathematics, and reading show that White students score higher on average than all other racial/ethnic groups, particularly when compared to Black and Hispanic students [ 2 , 3 ]. Explanations for these gaps tend to focus on the influence of socioeconomic resources, neighborhood and school characteristics, and family composition in patterning socioeconomic inequalities, and on the racialized nature of socioeconomic inequalities as key drivers of racial/ethnic academic achievement gaps [ 4 – 10 ]. Substantial evidence documents that indicators of socioeconomic status, such as free or reduced-price school lunch, are highly predictive of academic outcomes [ 2 , 3 ]. However, the relative contribution of family, neighborhood and school level socioeconomic inequalities to racial/ethnic academic inequalities continues to be debated, with evidence suggesting none of these factors fully explain racial/ethnic academic achievement gaps, particularly as students move through elementary school [ 11 ]. Attitudinal outcomes have been proposed by some as one explanatory factor for racial/ethnic inequalities in academic achievement [ 12 ], but differences in educational attitudes and aspirations across groups do not fully reflect inequalities in academic assessment. For example, while students of poorer socioeconomic status have lower educational aspirations than more advantaged students [ 13 ], racial/ethnic minority students report higher educational aspirations than White students, particularly after accounting for socioeconomic characteristics [ 14 – 16 ]. Similarly, while socio-emotional development is considered highly predictive of academic achievement in school students, some racial/ethnic minority children report better socio-emotional outcomes than their White peers on some indicators, although findings are inconsistent [ 17 – 22 ].

In addition to inequalities in academic achievement, racial/ethnic and socioeconomic inequalities also exist across measures of socio-emotional development [ 23 – 26 ]. And as with academic achievement, although socioeconomic factors are highly predictive of socio-emotional outcomes, they do not completely explain racial/ethnic inequalities in school-related outcomes not focused on standardized assessments [ 11 ].

Further complexity in understanding how academic and non-academic outcomes are patterned by socioeconomic factors, and how this contributes to racial/ethnic inequalities, is added by the multi-dimensional nature of socioeconomic status. Socioeconomic status is widely recognized as comprising diverse factors that operate across different levels (e.g. individual, household, neighborhood), and influence outcomes through different causal pathways [ 27 ]. The lack of interchangeability between measures of socioeconomic status within and between levels (e.g. income, education, occupation, wealth, neighborhood socioeconomic characteristics, or past socioeconomic circumstances) is also well established, as is the non-equivalence of measures between racial/ethnic groups [ 27 ]. For example, large inequalities have been reported across racial/ethnic groups within the same educational level, and inequalities in wealth have been shown across racial/ethnic that have similar income. It is therefore imperative that studies consider these multiple dimensions of socioeconomic status so that critical social gradients across the entire socioeconomic spectrum are not missed [ 27 ], and racial/ethnic inequalities within levels of socioeconomic status are adequately documented. It is also important that differences in school outcomes are considered across levels of socioeconomic status within and between racial/ethnic groups, so that the influence of specific socioeconomic factors on outcomes within specific racial/ethnic groups can be studied [ 28 ]. However, while these analytic approaches have been identified as research priorities in order to enhance our understanding of the complex ways in which socioeconomic status and race/ethnicity intersect to influence school outcomes, research that operationalizes these recommendations across academic and non-academic outcomes of school children is scant.

In addition to the complexity that arises from race/ethnicity, socioeconomic status, and intersections between them, different patterns in academic and non-academic outcomes by gender have also received longstanding attention. Comparisons across gender show that, on average, boys have higher scores in mathematics and science, whereas girls have higher scores in reading [ 2 , 3 , 29 ]. In contrast to explanations for socioeconomic inequalities, gender differences have been mainly attributed to social conditioning and stereotyping within families, schools, communities, and the wider society [ 30 – 35 ]. These socialization and stereotyping processes are also highly relevant determining factors in explaining racial/ethnic academic and non-academic inequalities [ 35 , 36 ], as are processes of racial discrimination and stigmatization [ 37 , 38 ]. Gender differences in academic outcomes have been documented as differently patterned across racial/ethnic groups and across levels of socioeconomic status. For example, gender inequalities in math and science are largest among White and Latino students, and smallest among Asian American and African American students [ 39 – 43 ], while gender gaps in test scores are more pronounced among socioeconomically disadvantaged children [ 44 , 45 ]. In terms of attitudes towards math and sciences, gender differences in attitudes towards math are largest among Latino students, but gender differences in attitudes towards science are largest among White students [ 39 , 40 ]. Gender differences in socio-developmental outcomes and in non-cognitive academic outcomes, across race/ethnicity and socio-economic status, have received far less attention; studies that consider multiple academic and non-academic outcomes among school aged children across race/ethnicity, socioeconomic status and gender are limited in the US and internationally.

Understanding how different academic and non-academic outcomes are differently patterned by race/ethnicity, socio-economic status, and gender, including within and between group differences, is an important research area that may assist in understanding the potential causal pathways and explanations for observed inequalities, and in identifying key population groups and points at which interventions should be targeted to address inequalities in particular outcomes [ 28 , 46 ]. Not only is such knowledge critical for population level policy and/or local level action within affected communities, but failing to detect potential factors for interventions and potential solutions is argued as reinforcing perceptions of the unmodifiable nature of inequality and injustice [ 46 ].

Notwithstanding the importance of documenting patterns of inequality in relation to a particular social identity (e.g. race/ethnicity, gender, class), there is increasing acknowledgement within both theoretical and empirical research of the need to move beyond analyzing single categories to consider simultaneous interactions between different aspects of social identity, and the impact of systems and processes of oppression and domination (e.g., racism, classism, sexism) that operate at the micro and macro level [ 47 , 48 ]. Such intersectional approaches challenge practices that isolate and prioritize a single social position, and emphasize the potential of varied inter-relationships of social identities and interacting social processes in the production of inequities [ 49 – 51 ]. To date, exploration of how social identities interact in an intersectional way to influence outcomes has largely been theoretical and qualitative in nature. Explanations offered for interactions between privileged and marginalized identities, and associated outcomes, include family and teacher socialization of gender performance (e.g. math and science as male domains, verbal and emotional skills as female), as well as racialized stereotypes and expectations from teachers and wider society regarding racial/ethnic minorities that are also gendered (e.g. Black males as violent prone and aggressive, Asian females as submissive) [ 52 – 57 ]. That is, social processes that socialize and pattern opportunities and outcomes are both racialized and gendered, with racism and sexism operating in intersecting ways to influence the development and achievements of children and youth [ 58 – 60 ]. Socioeconomic status adds a third important dimension to these processes, with individuals of the same race/ethnicity and gender having access to vastly different resources and opportunities across levels of socioeconomic status. Moreover, access to resources as well as socialization experiences and expectations differ considerably by race and gender within the same level of socio-economic status. Thus, neither gender nor race nor socio-economic status alone can fully explain the interacting social processes influencing outcomes for youth [ 27 , 28 ]. Disentangling such interactions is therefore an important research priority in order to inform intervention to address inequalities at a population level and within local communities.

In the realm of quantitative approaches to the study of inequality, studies often examine separate social identities independently to assess which of these axes of stratification is most prominent, and for the most part do not consider claims that the varied dimensions of social stratification are often juxtaposed [ 56 , 61 ]. A pressing need remains for quantitative research to consider how multiple forms of social stratification are interrelated, and how they combine interactively, not just additively, to influence outcomes [ 46 ]. Doing so enables analyses that consider in greater detail the representation of the embodied positions of individuals, particularly issues of multiple marginalization as well as the co-occurrence of some form of privilege with marginalization [ 46 ]. It is important to note that the languages of statistical interaction and of intersectionality need to be carefully distinguished (e.g. intersectional additivity or additive assumptions, versus additive scale and cross-product interaction terms) to avoid misinterpretation of findings, and to ensure appropriate application of statistical interaction to enable the description of outcome measures for groups of individuals at each cross-stratified intersection [ 46 ]. Ultimately this will provide more nuanced and realistic understandings of the determinants of inequality in order to inform intervention strategies.

This study fills these gaps in the literature by examining inequalities across several eighth grade academic and non-academic outcomes at the intersection of race/ethnicity, gender, and socioeconomic status. It aims to do this by: identifying classes of socioeconomic advantage/disadvantage from kindergarten to eighth grade; then ascertaining whether membership into classes of socioeconomic advantage/disadvantage differ for racial/ethnic and gender groups; and finally, by contrasting academic and non-academic outcomes at the intersection of race/ethnicity, gender and socioeconomic advantage/disadvantage. Intersecting identities of race/ethnicity, gender, and socioeconomic characteristics are compared to the reference group of White boys in the most advantaged socioeconomic category, as these are the three identities (male, White, socioeconomically privileged) that experience the least marginalization when compared to racial/ethnic and gender minority groups in disadvantaged socioeconomic positions.

This study used data on singleton children from the Early Childhood Longitudinal Study—Kindergarten (ECLS-K). The ECLS-K employed a multistage probability sample design to select a nationally representative sample of children attending kindergarten in 1998–99. In the base year the primary sampling units (PSUs) were geographic areas consisting of counties or groups of counties. The second-stage units were schools within sampled PSUs. The third- and final-stage units were children within schools [ 62 ]. Analyses were conducted on data collected from direct child assessments, as well as information provided by parents and school administrators.

Ethics Statement

This article is based on the secondary analysis of anonymized and de-identified Public-Use Data Files available to researchers via the Inter-University Consortium for Political and Social Research (ICPSR). Human participants were not directly involved in the research reported in this article; therefore, no institutional review board approval was sought.

Outcome Variables.

Eight outcome variables, all assessed in eighth grade, were selected to examine the study aims: two measures relating to non-cognitive academic skills (perceived interest/competence in reading, and in math); three measures capturing socioemotional development (internalizing behavior, locus of control, self-concept); and three measures of cognitive skills (math, reading and science assessment scores).

For the eighth-grade data collection, children completed the 16-item Self Description Questionnaire (SDQ) II [ 63 ], where they provided self-assessments of their academic skills by rating their perceived competence and interest in English and mathematics. The SDQ also asked children to report on problem behaviors with which they might struggle. Three subscales were produced from the SDQ items: The SDQ Perceived Interest/Competence in Reading, including four items on grades in English and the child’s interest in and enjoyment of reading. The SDQ Perceived Interest/Competence in Math, including four items on mathematics grades and the child’s interest in and enjoyment of mathematics. And the SDQ Internalizing Behavior subscale, which includes eight items on internalizing problem behaviors such as feeling sad, lonely, ashamed of mistakes, frustrated, and worrying about school and friendships [ 62 ].

The Self-Concept and Locus of Control scales ask children about their self-perceptions and the amount of control they have over their own lives. These scales, adopted from the National Education Longitudinal Study of 1988, asked children to indicate the degree to which they agreed with 13 statements (seven items in the Self-Concept scale, and six items in the Locus of Control Scale) about themselves, including “I feel good about myself,” “I don’t have enough control over the direction my life is taking,” and “At times I think I am no good at all.” Responses ranged from “strongly agree” to “strongly disagree.” Some items were reversed coded so that higher scores indicate more positive self-concept and a greater perception of control over one’s own life. The seven items in the Self-Concept scale, and the six items in the Locus of Control were standardized separately to a mean of zero and a standard deviation of 1. The scores of each scale are an average of the standardized scores [ 62 ].

Academic achievement in reading, mathematics and science was measured with the eighth-grade direct cognitive assessment battery [ 62 ].

Children were given separate routing assessment forms to determine the level (high/low) of their reading, mathematics, and science assessments. The two-stage cognitive assessment approach was used to maximize the accuracy of measurement and reduce administration time by using the child’s responses from a brief first-stage routing form to select the appropriate second-stage level form. First, children read items in a booklet and recorded their responses on an answer form. These answer forms were then scored by the test administrator. Based on the score of the respective routing forms, the test administrator then assigned a high or low second-stage level form of the reading and mathematics assessments. For the second-stage level tests, children read items in the assessment booklet and recorded their responses in the same assessment booklet. The routing tests and the second-stage tests were timed for 80 minutes [ 62 ]. The present analyses use the standardized scores (T-scores), allowing relative comparisons of children against their peers.

Individual and Contextual Disadvantage Variables.

Latent Class Analysis, described in greater detail below, was used to classify students into classes of individual and contextual advantage or disadvantage. Nine constructs, measuring characteristics at the individual-, school-, and neighborhood-level, were captured using 42 dichotomous variables measured across the different waves of the ECLS-K.

Individual-level variables captured household composition, material disadvantage, and parental expectations of the children’s success. Measures included whether the child lived in a single-parent household at kindergarten, first, third, fifth and eighth grades; whether the household was below the poverty threshold level at kindergarten, fifth and eighth grades; food insecurity at kindergarten, first, second and third grades; and parental expectations of the child’s academic achievement (categorized as up to high school and more than high school) at kindergarten, first, third, fifth and eighth grades. An indicator of whether parents had moved since the previous interview (measured at kindergarten, first, third, fifth and eighth grades) was included to capture stability in the children’s life. A household-level composite index of socioeconomic status, derived by the National Center for Education Statistics, was also included at kindergarten, first, third, fifth and eighth grades. This measure captured the father/male guardian’s education and occupation, the mother/female guardian’s education and occupation, and the household income. Higher scores reflect higher levels of educational attainment, occupational prestige, and income. In the present analyses, the socioeconomic composite index was categorized into quintiles and further divided into the lowest first and second quintiles, versus the third, fourth and fifth quintiles.

Two variables measured the school-level environment: percentage of students eligible for free school meals, and percentage of students from a racial/ethnic background other than White non-Hispanic. These two variables were dichotomized as more than or equal to 50% of students belonging to each category. Both variables were measured in the kindergarten, first, third, fifth and eighth grade data collections.

To capture the neighborhood environment, a variable was included which measured the level of safety of the neighborhood in kindergarten, first, third, fifth and eighth grades. Parents were asked “How safe is it for children to play outside during the day in your neighborhood?” with responses ranging from 1, not at all safe, to 3, very safe. For the present analyses, response categories were recoded into 1 “not at all and somewhat safe,” and 0 “very safe.”

Predictor Variables.

The race/ethnicity and gender of the children were assessed during the parent interview. In order to empirically measure the intersection between race/ethnicity and gender in the classes of disadvantage, a set of six dummy variables were created that combined racial/ethnic and gender categories into White boys, White girls, Black boys, Black girls, Latino boys, and Latina girls.

Statistical Analyses

This study used the manual 3-step approach in mixture modeling with auxiliary variables [ 64 , 65 ] to independently evaluate the relationship between the predictor auxiliary variables (the combined race/ethnicity and gender groups), the latent class variable of advantage/disadvantage, and the outcome (non-cognitive skills, socioemotional development, cognitive assessments). This is a data-driven, mixture modelling technique which uses indicator variables (in this case the variables described under Individual and Contextual Disadvantage Variables section) to identify a number of latent classes. It also includes auxiliary information in the form of covariates (the race/ethnicity and gender combinations described under Predictor Variables) and distal outcomes (the eight outcome variables), to better explore the relationships between the characteristics that make up the latent classes, the predictors of class membership, and the associated consequences of membership into each class.

The first step in the 3-step procedure is to estimate the measurement part of the joint model (i.e., the latent class model) by creating the latent classes without adding covariates. Latent class analyses first evaluated the fit of a 2-class model, and systematically increased the number of classes in subsequent models until the addition of latent classes did not further improve model fit. For each model, replication of the best log-likelihood was verified to avoid local maxima. To determine the optimal number of classes, models were compared across several model fit criteria. First, the sample-size adjusted Bayesian Information Criterion (BIC) [ 66 ] was evaluated; lower relative BIC values indicate improved model fit. Given that the BIC criterion tends to favor models with fewer latent classes [ 67 ], the Lo, Mendell, and Rubin likelihood ratio test (LMR-LRT) statistic [ 68 ] was also considered. The LMR-LRT can be used in mixture modeling to compare the fit of the specified class solution ( k -class model) to a model with fewer classes ( k -1 class model). A non-significant chi-square value suggests that a model with one fewer class is preferred. Entropy statistics, which measure the separation of the classes based on the posterior class membership probabilities, were also examined; entropy values approaching 1 indicate clear separation between classes [ 69 ].

After determining the latent class model in step 1, the second step of the analyses used the latent class posterior distribution to generate a nominal variable N , which represented the most likely class [ 64 ]. During the third step, the measurement error for N was accounted for while the model was estimated with the outcomes and predictor auxiliary variables [ 64 ]. The last step of the analysis examined whether race/ethnic and gender categories predict class membership, and whether class membership predicts the outcomes of interest.

All analyses were conducted using MPlus v. 7.11 [ 70 ], and used longitudinal weights to account for differential probabilities of selection at each sampling stage and to adjust for the effects of non-response. A robust standard error estimator was used in MPlus to account for the clustering of observations in the ECLS-K.

Four distinct classes of advantage/disadvantage were identified in the latent class analysis (see Table 1 ).

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0141363.t001

Class characteristics are shown in Table A in S1 File . Trajectories of advantage and disadvantage were stable across ECLS-K waves, so that none of the classes identified changed in individual and contextual characteristics across time. The largest proportion of the sample (47%; Class 3: Individually and Contextually Wealthy) lived in individual and contextual privilege, with very low proportions of children in socioeconomic deprived contexts. A class representing the opposite characteristics (children living in individually- and contextually-deprived circumstances) was also identified in the analyses (19%; Class 1: Individually and Contextually Disadvantaged). Class 1 had the highest proportion of children living in socioeconomic deprivation, attending schools with more than 50% racial/ethnic minority students, and living in unsafe neighborhoods, but did not have a high proportion of children with the lowest parental expectations. Class 4 (19%; Individually Disadvantaged, Contextually Wealthy) had the highest proportion of children with the lowest parental expectations (parents reporting across waves that they expected children to achieve up to a high school education). Class 4 (Individually Disadvantaged, Contextually Wealthy) also had high proportions of children living in individual-level socioeconomic deprivation, but had low proportions of children attending a school with over 50% of children eligible for free school meals. It also had relatively low proportions of children living in unsafe neighborhoods and low proportions of children attending diverse schools, forming a class with a mixture of individual-level deprivation, and contextual-level advantage. The last class was composed of children who lived in individually-wealthy environments, but who also lived in unsafe neighborhoods and attended diverse schools where more than 50% of pupils were eligible for free school meals (13%; Class 2: Individually Wealthy, Contextually Disadvantaged; see Table A in S1 File ).

The combined intersecting racial/ethnic and gender characteristics yielded six groups consisting of White boys (n = 2998), White girls (n = 2899), Black boys (n = 553), Black girls (n = 560), Latino boys (n = 961), and Latina girls (n = 949). All pairs containing at least one minority status of either race/ethnicity or gender (e.g., Black boys, Black girls, Latino boys, Latina girls) were more likely than White boys to be assigned to the more disadvantaged classes, as compared to being assigned to Class 3, the least disadvantaged (see Table B in S1 File ).

Racial/Ethnic and Gender Differences in Eighth-Grade Academic Outcomes

Table 2 shows broad patterns of intersecting racial/ethnic and gender inequalities in academic outcomes, although interesting differences emerge across racial/ethnic and gender groups. Whereas Black boys achieved lower scores than White boys across all classes on the math, reading and science assessments, this was not the case for Latino boys, who only underperformed White boys on the science assessment within the most privileged class (Class 3: Individually and Contextually Wealthy). Latina girls, in contrast, outperformed White boys on reading scores within Class 4 (Individually Disadvantaged, Contextually Wealthy), but scored lower than White boys on science and math assessments, although only when in the two most privileged classes (Class 3 and 4). For Black girls the effect of class membership was not as pronounced, and they had lower science and math scores than White boys across all but one instance.

https://doi.org/10.1371/journal.pone.0141363.t002

In general, the largest inequalities in academic outcomes across racial/ethnic and gender groups appeared in the most privileged classes. For example, results show no differences in math scores across racial/ethnic and gender categories within Class 4, the most disadvantaged class, but in all other classes that contain an element of advantage, and particularly in Class 3 (Individually and Contextually Wealthy), there are large gaps in math scores across racial/ethnic and gender groups, when compared to White boys. These patterns of heightened inequality in the most advantaged classes are similar for reading and science scores (see Table 2 ).

Racial/Ethnic and Gender Differences in Eighth-Grade Non-Academic Outcomes

Interestingly, racialized and gendered patterns of inequality observed in academic outcomes were not as stark in non-cognitive academic outcomes (see Table 3 ).

https://doi.org/10.1371/journal.pone.0141363.t003

Racial/ethnic and gender differences were small across socioemotional outcomes, and in fact, White boys were outperformed on several outcomes. Black boys scored lower than White boys on internalizing behavior and higher on self-concept within Classes 2 (Individually Wealthy, Contextually Disadvantaged) and 4 (Individually Disadvantaged, Contextually Wealthy), and Black girls scored higher than White boys on self-concept within Classes 2 and 3 (Individually Wealthy, Contextually Disadvantaged, and Individually and Contextually Wealthy, respectively). White and Latina girls, but not Black girls, scored higher than White boys on internalizing behavior (within Classes 3 and 4 for White girls, and within Classes 1 and 3 for Latina girls; see Table 3 ).

As with academic outcomes, most racial/ethnic and gender differences also emerged within the most privileged classes, and particularly in Class 3 (Individually and Contextually Wealthy), although in the case of perceived interest/competence in reading, White and Latina girls performed better than White boys. White girls also reported higher perceived interest/competence in reading than White boys in Class 4: Individually Disadvantaged, Contextually Wealthy.

This study set out to examine inequalities across several eighth grade academic and non-academic outcomes at the intersection of race/ethnicity, gender, and socioeconomic status. It first identified four classes of longstanding individual- and contextual-level disadvantage; then determined membership to these classes depending on racial/ethnic and gender groups; and finally compared non-cognitive skills, academic assessment scores, and socioemotional outcomes across intersecting gender, racial/ethnic and socioeconomic social positions.

Results show the clear influence of race/ethnicity in determining membership to the most disadvantaged classes. Across gender dichotomies, Black students were more likely than White boys to be assigned to all classes of disadvantage as compared to the most advantaged class, and this was particularly strong for the most disadvantaged class, which included elements of both individual- and contextual-level disadvantage. Latino boys and girls were also more likely than White boys to be assigned to all the disadvantaged classes, but the strength of the association was much smaller than for Black students. Whereas membership into classes of disadvantage appears to be more a result of structural inequalities strongly driven by race/ethnicity, the salience of gender is apparent in the distribution of academic assessment outcomes within classes of disadvantage. Results show a gendered pattern of math, reading and science assessments, particularly in the most privileged class, where girls from all ethnic/racial groups (although mostly from Black and Latino racial/ethnic groups) underperform White boys in math and science, and where Black boys score lower, and White girls higher, than White boys in reading.

With the exception of educational assessments, gender and racial/ethnic inequalities within classes are either not very pronounced or in the opposite direction (e.g. racial/ethnic and gender minorities outperform White males), but differences in outcomes across classes are stark. The strength of the association between race/ethnicity and class membership, and the reduced racial/ethnic and gender inequalities within classes of advantage and disadvantage, attest to the importance of socioeconomic status and wealth in explaining racial/ethnic inequalities; should individual and contextual disadvantage be comparable across racial/ethnic groups, racial/ethnic inequalities would be substantially reduced. This being said, most within-class differences were observed in the most privileged classes, showing that benefits brought about by affluence and advantage are not equal across racial/ethnic and gender groups. The measures of advantage and disadvantage captured in this study relate to characteristics afforded by parental resources, implying an intergenerational transmission of disadvantage, regardless of the presence of absolute adversity in childhood. This pattern of differential returns of affluence has been shown in other studies, which report that White teenagers benefit more from the presence of affluent neighbors than do Black teenagers [ 71 ]. Among adult populations, studies show that across several health outcomes, highly educated Black adults fare worse than White adults with the lowest education [ 72 ]. Intersectional approaches such as the one applied in this study reveal how power within gendered and racialized institutional settings operates to undermine access to and use of resources that would otherwise be available to individuals of advantaged classes [ 72 ]. The present study further contributes to this literature by documenting how, in a key stage of the life course, similar levels of advantage, but not disadvantage, lead to different academic outcomes across racial/ethnic and gender groups. These findings suggest that, should socioeconomic inequalities be addressed, and levels of advantage were similar across racial/ethnic and gender groups, systems of oppression that pattern the racialization and socialization of children into racial/ethnic and gender roles in society would still ensure that inequalities in academic outcomes existed across racial/ethnic and gender categories. In other words, racism and sexism have a direct effect on academic and non-academic outcomes among 8 th graders, independent of the effect of socioeconomic disadvantage on these outcomes. An important limitation of the current study is that although it uses a comprehensive measure of advantage/disadvantage, including elements of deprivation and affluence at the family, school and neighborhood levels through time, it failed to capture these two key causal determinants of racial/ethnic and gender inequality: experiences of racial and gender discrimination.

Despite this limitation, it is important to note that socioeconomic inequalities in the US are driven by racial and gender bias and discrimination at structural and individual levels, with race and gender discrimination exerting a strong influence on academic and non-academic inequalities. Racial discrimination, prevalent in the US and in other industrialized nations [ 38 , 73 ] determines differential life opportunities and resources across racial/ethnic groups, and is a crucial determinant of racial/ethnic inequalities in health and development throughout life and across generations [ 37 , 38 ]. In the context of this study’s primary outcomes within school settings, racism and racial discrimination experienced by both the parents and the children are likely to contribute towards explaining observed racial/ethnic inequalities in outcomes within classes of disadvantage. Gender discrimination—another system of oppression—is apparent in this study in relation to academic subjects socially considered as typically male or female orientated. For example, results show no difference between Black girls and White boys from the most advantaged class in terms of perceived interest and competence in math but, in this same class, Black girls score much lower than White boys in the math assessment. This difference, not explained by intrinsic or socioeconomic differences, can be contextualized as a consequence of experienced intersecting racial and gender discrimination. The consequences of the intersection between two marginalized identities are found throughout the results of this study when comparing across broad categorizations of race/ethnicity and gender, and in more detailed conceptualizations of minority status. Growing up Black, Latino or White in the US is not the same for boys and girls, and growing up as a boy or a girl in America does not lead to the same outcomes and opportunities for Black, Latino and White children as they become adults. With this study’s approach of intersectionality one can observe the complexity of how gender and race/ethnicity intersect to create unique academic and non-academic outcomes. This includes the contrasting results found for Black and Latino boys, when compared to White boys, which show very few examples of poorer outcomes among Latino boys, but several instances among Black boys. Results also show different racialization for Black and Latina girls. Latina girls, but not Black girls, report higher internalizing behavior than White boys, whereas Black girls, but not Latina girls, report higher self-concept than White boys. Black boys also report higher self-concept and lower internalizing behavior than White boys, findings that mirror research on self-esteem among Black adolescents [ 74 , 75 ]. In cognitive assessments, intersecting racial/ethnic and gender differences emerge across classes of disadvantage. For example, Black girls in all four classes score lower on science scores than White boys, but only Latina girls in the most advantaged class score lower than White boys. Although one can observe differences in the racialization of Black and Latino boys and girls across classes of disadvantage, findings about broad differences across Latino children compared to Black and White children should be interpreted with caution. The Latino ethnic group is a large, heterogeneous group, representing 16.7% of the total US population [ 76 ]. The Latino population is composed of a variety of different sub-groups with diverse national origins and migration histories [ 77 ], which has led to differences in sociodemographic characteristics and lived experiences of ethnicity and minority status among the various groups. Differences across Latino sub-groups are widely documented, and pooled analyses such as those reported here are masking differences across Latino sub-groups, and providing biased comparisons between Latino children, and Black and White children.

Poorer performance of girls and racial/ethnic minority students in science and math assessments (but not in self-perceived competence and interest) might result from stereotype threat, whereby negative stereotypes of a group influence their member’s performance [ 78 ]. Stereotype threat posits that awareness of a social stereotype that reflects negatively on one's social group can negatively affect the performance of group members [ 35 ]. Reduced performance only occurs in a threatening situation (e.g., a test) where individuals are aware of the stereotype. Studies show that early adolescence is a time when youth become aware of and begin to endorse traditional gender and racial/ethnic stereotypes [ 79 ]. Findings among youth parallel findings among adult populations, which show that adult men are generally perceived to be more competent than women, but that these perceptions do not necessarily hold for Black men [ 80 ]. These stereotypes have strong implications for interpersonal interactions and for the wider structuring of systemic racial/ethnic and gender inequalities. An example of the consequences of negative racial/ethnic and gender stereotypes as children grow up is the well-documented racial/ethnic and gender pay gap: women earn less than men [ 81 ], and racial/ethnic minority women and men earn less than White men [ 82 ].

In addition to the focus on intersectionality, a strength of this study is its person-centered methodological approach, which incorporates measures of advantage and disadvantage across individual and contextual levels through nine years of children’s socialization. Children live within multiple contexts, with risk factors at the family, school, and neighborhood level contributing to their development and wellbeing. Individual risk factors seldom operate in isolation [ 83 ], and they are often strongly associated both within and across levels [ 84 ]. All risk factors captured in the latent class analyses have been independently associated with increased risk for academic problems [ 10 , 71 , 85 , 86 ], and given that combinations of risk factors that cut across multiple domains explain the association between early risk and later outcomes better than any isolated risk factor [ 83 , 84 ], the incorporation of person-centered and intersectionality approaches to the study of racial/ethnic, gender, and socioeconomic inequalities across school outcomes provides new insight into how children in marginalized social groups are socialized in the early life course.

Conclusions

The contrasting outcomes between racial/ethnic and gender minorities in self-assessment and socioemotional outcomes, as compared to standardized assessments, provide support for the detrimental effect that intersecting racial/ethnic and gender discrimination have in patterning academic outcomes that predict success in adult life. Interventions to eliminate achievement gaps cannot fully succeed as long as social stratification caused by gender and racial discrimination is not addressed [ 87 , 88 ].

Supporting Information

S1 file. supporting tables..

Table A: Class characteristics. Table B: Associations between race/ethnicity and gender groups and assigned class membership (membership to Classes 1, 2 or 4 as compared to Class 3: Individually and Contextually Wealthy).

https://doi.org/10.1371/journal.pone.0141363.s001

Acknowledgments

This work was funded by an ESRC grant (ES/K001582/1) and a Hallsworth Research Fellowship to LB. Most of this work was conducted while LB was a visiting scholar at the Institute for Social Research, University of Michigan. She would like to thank them for hosting her visit and for the support provided.

Author Contributions

Conceived and designed the experiments: LB. Performed the experiments: LB. Analyzed the data: LB. Wrote the paper: LB NP.

• 1. Jencks C, Phillips M. The Black-White Test Score Gap. Washington, D.C.: Brookings Institution Press; 1998.
• 2. (NCES) NCfES. The Nation’s Report Card: Science 2011(NCES 2012–465). Washington, D.C.: Institute of Education Sciences, U.S. Department of Education, 2012.
• 3. (NCES) NCfES. The Nation’s Report Card: A First Look: 2013 Mathematics and Reading (NCES 2014–451). Washington, D.C.: Institute of Education Sciences, U.S. Department of Education, 2013.
• View Article
• Google Scholar
• PubMed/NCBI
• 12. Ogbu J. Low performance as an adaptation: The case of blacks in Stockton, California. In: Gibson M, Ogbu J, editors. Minority Status and Schooling. New York: Grand Publishing; 1991.
• 41. Coley R. Differences in the gender gap: Comparisons across racial/ethnic groups in education and work. Princeton, NJ: Educational Testing Service, 2001.
• 47. Hankivsky O, Cormier R. Intersectionality: Moving women’s health research and policy forward. Vancouver: Women’s Health Research Network; 2009.
• 48. Dhamoon K, Hankivsky O. Why the theory and practice of intersectionality matter to health research and policy. In: Hankivsky O, editor. Health Inequities in Canada: Intersectional Frameworks and Practices. Vancouver: UBC Press; 2011. p. 16–50.
• 50. Crenshaw K. Demarginalizing the intersection of race and sex: a black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum. 1989:139–67.
• 62. Tourangeau K, Nord C, Lê T, Sorongon A, Najarian M. Early Childhood Longitudinal Study, Kindergarten Class of 1998–99 (ECLS-K), Combined User’s Manual for the ECLS-K Eighth-Grade and K–8 Full Sample Data Files and Electronic Codebooks (NCES 2009–004). Washington, DC.: National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education., 2009.
• 63. Marsh HW. Self Description Questionnaire (SDQ) II: A theoretical and empirical basis for the measurement of multiple dimensions of adolescent self-concept. An interim test manual and a research monograph. Sydney: University of Western Sydney: 1992.
• 64. Asparouhov T, Muthén B. Auxiliary Variables in Mixture Modeling: 3-Step Approaches Using Mplus. Mplus Web Notes [Internet]. 2013 April 2014. Available from: www.statmodel.com/examples/webnotes/webnote15.pdf ‎.
• 67. Dayton C. Latent class scaling analysis. Thousand Oaks, CA: Sage, 1998.
• 70. Muthén L, Muthén B. Mplus User's Guide. Seventh Edition. Los Angeles, CA.: 2012.
• 72. Jackson P, Williams D. The intersection of race, gender, and SES: Health paradoxes. In: Schulz A, Mullings L, editors. Gender, Race, Class, & Health: Intersectional Approaches. San Francisco, CA: Jossey-Bass; 2006. p. 131–62.
• 76. Ennis S, Rios-Vargas M, Albert N. The Hispanic Population: 2010. United Census Bureau, 2011.
• 88. Rouse C, Brooks-Gunn J, McLanahan S. School Readiness: Closing racial and ethnic gaps. Washington, DC: Brookings Press, 2005.

Change Password

Your password must have 8 characters or more and contain 3 of the following:.

• a lower case character, 
• an upper case character, 
• a special character 

Password Changed Successfully

Your password has been changed

• Sign in / Register

Request Username

Can't sign in? Forgot your username?

Enter your email address below and we will send you your username

If the address matches an existing account you will receive an email with instructions to retrieve your username

Reframing Educational Outcomes: Moving beyond Achievement Gaps

• Sarita Y. Shukla
• Elli J. Theobald
• Joel K. Abraham
• Rebecca M. Price

School of Educational Studies, University of Washington, Bothell, Bothell, WA 98011-8246

Search for more papers by this author

Department of Biology, University of Washington, Seattle, Seattle, WA 98195

Department of Biological Science, California State University–Fullerton, Fullerton, CA 92831

*Address correspondence to: Rebecca M. Price ( E-mail Address: [email protected] )

School of Interdisciplinary Arts & Sciences, University of Washington, Bothell, Bothell, WA 98011-8246

The term “achievement gap” has a negative and racialized history, and using the term reinforces a deficit mindset that is ingrained in U.S. educational systems. In this essay, we review the literature that demonstrates why “achievement gap” reflects deficit thinking. We explain why biology education researchers should avoid using the phrase and also caution that changing vocabulary alone will not suffice. Instead, we suggest that researchers explicitly apply frameworks that are supportive, name racially systemic inequities and embrace student identity. We review four such frameworks—opportunity gaps, educational debt, community cultural wealth, and ethics of care—and reinterpret salient examples from biology education research as an example of each framework. Although not exhaustive, these descriptions form a starting place for biology education researchers to explicitly name systems-level and asset-based frameworks as they work to end educational inequities.

INTRODUCTION

Inequities plague educational systems in the United States, from pre-K through graduate school. Many of these inequities exist along racial, gender, and socioeconomic lines ( Kozol, 2005 ; Sadker et al. , 2009 ), and they impact the educational outcomes of students. For decades, education research has focused on comparisons of these educational outcomes, particularly with respect to test scores of students across racial and ethnic identities. The persistent differences in these test scores or other outcomes are often referred to as “achievement gaps,” which in turn serve as the basis for numerous educational policy and structural changes ( Carey, 2014 ).

A recent essay in CBE—Life Sciences Education ( LSE ) questioned narrowly defining “success” in educational settings ( Weatherton and Schussler, 2021 ). The authors posit that success must be defined and contextualized, and they asked the community to recognize the racial undercurrents associated with defining success as limited to high test scores and grade point averages (GPAs; Weatherton and Schussler, 2021 ). In this essay, we make a complementary point. We contend that the term “achievement gap” is misaligned with the intent and focus of recent biology education research. We base this realization on the fact that the term “achievement gap” can have a deeper meaning than documenting a difference among otherwise equal groups ( Kendi, 2019 ; Gouvea, 2021 ). It triggers deficit thinking ( Quinn, 2020 ); unnecessarily centers middle and upper class, White, male students as the norm ( Milner, 2012 ); and downplays the impact of structural inequities ( Ladson-Billings, 2006 ; Carter and Welner, 2013 ).

This essay unpacks the negative consequences of using the term “achievement gap” when comparing student learning across different racial groups. We advocate for abandoning the term. Similarly, we suggest that, in addition to changing our terminology, biology education researchers can explicitly apply theoretical frameworks that are more appropriate for interrogating inequities among educational outcomes across students from different demographics. We emphasize that the idea that a simple “find and replace,” swapping out the term “achievement gap” for other phrases, is not sufficient.

In the heart of this essay, we review some of these systems-level and asset-based frameworks for research that explores differences in academic performance ( Figure 1 ): opportunity gaps ( Carter and Welner, 2013 ), educational debt ( Ladson-Billings, 2006 ), community cultural wealth ( Yosso, 2005 ), and ethics of care ( Noddings, 1988 ). Within each of these frameworks, we review examples of biology education literature that we believe rely on them, explicitly or implicitly. We conclude by reiterating the need for education researchers to name explicitly the systems-level and asset-based frameworks used in future research.

FIGURE 1. Research frameworks highlighted in the essay. The column in gray summarizes deficit-based frameworks that focus on achievement gaps. The middle column (in gold) includes examples of systems-based frameworks that acknowledge that student learning is associated with society-wide habits. The rightmost columns (in peach) include examples of asset-based models that associate student learning with students’ strengths. The columns are not mutually exclusive, in that studies can draw from multiple frameworks simultaneously or sequentially.

We will use the phrase “students from historically or currently marginalized groups” to describe the students who have been and still are furthest from the center of educational justice. However, when discussing work of other researchers, we will use the terminology they use in their papers. Our conceptualization of this phrase matches, as near as we can tell, Asai’s phrase “PEERs—persons excluded for their ethnicity or race” ( Asai, 2020 , p. 754). We also choose to capitalize “White” to acknowledge that people in this category have a visible racial identity ( Painter, 2020 ).

Positionality

Our positionalities—our unique life experiences and identities—mediate our understanding of the world ( Takacs, 2003 ). What we see as salient in our research situation arises from our own life experiences. Choices in our research, including the types of data we collect and how we clean the data and prepare it for analysis, adopt analytical tools, and make sense of these analyses are important decision points that affect study results and our findings ( Huntington-Klein et al. , 2021 ). We recognize that it is impossible to be free of bias ( Noble, 2018 ; Obermeyer et al. , 2019 ). Therefore, we put forth our positionality to acknowledge the lenses through which we make decisions as researchers and to forefront the impact of our identities on our research. Still, the breadth of our experiences cannot be described fully in a few sentences.

The four authors of this essay have unique and complementary life experiences that contribute to the sense-making presented in this essay. S.Y.S. has been teaching since 2003 and teaching in higher education since 2012. She is a South Asian immigrant to the United States, and a cisgender woman. E.J.T. has taught middle school, high school, and college science since 2006. She is a cisgender White woman. J.K.A. is a cisgender Black mixed-race man who comes from a family of relatively recent immigrants with different educational paths. He has worked in formal and informal education since 2000. R.M.P. is a cisgender Jewish, White woman, and she has been teaching college since 2006. We represent a team of people who explicitly acknowledge that our experiences influence the lenses through which we work. Our guiding principles are 1) progress over perfection, 2) continual reflection and self-improvement, and 3) deep care for students. These principles guide our research and teaching, impacting our interactions with colleagues (faculty and staff) as well as students. Ultimately, these principles motivate us to make ourselves aware of, reflect on, and learn from our mistakes.

Simply Changing Vocabulary Does Not Suffice

The term “achievement gap” is used in research that examines differences in achievement—commonly defined as differences in test scores—across students from different demographic groups ( Coleman et al. , 1966 ). Some studies replace “achievement gap” with “score gap” (e.g., Jencks and Phillips, 2006 ), because it defines the type of achievement under consideration; others use “opportunity gap,” because it emphasizes differences in opportunities students have had throughout their educational history (e.g., Carter and Welner, 2013 ; more on opportunity gaps later). The shift for which we advocate, however, does not reside only with terminology. Instead, we call for a deeper shift of using research frameworks that acknowledge and respect students’ histories and empower them now.

The underlying framework in research that uses “achievement gap” or even “score gap” may not be immediately apparent. Take for example two studies that both use the seemingly benign term, “score gap.” A close read indicates that one study attributed the difference in test scores between Black and White students to deficient “culture and child-rearing practices” ( Farkas, 2004 , p. 18). Thus, even though the researcher uses what can be considered to be more neutral terminology, the phrase in this context represents deficit thinking and blame. On the other hand, another study uses the term “score gap” to explore differences that have been historically studied through cultures of poverty, genetic, and familial backgrounds ( Jencks and Phillips, 2006 ). While these researchers discuss the Black–White score gap, they present evidence that examines this phenomenon with nuanced constructs, such as stereotype threat ( Steele, 2011 ) and resources available. These authors also mention ways to reduce score gaps, such as smaller class sizes and high teacher expectations ( Jencks and Phillips, 2006 ).

Some researchers who use the phrase “achievement gap” explicitly avoid deficit thinking and instead embrace an asset-based framework. Jordt et al. (2017) address systemic racism, just as Jencks and Phillips (2006) do. Specifically, Jordt et al. (2017) identified an intervention that affirmed student values that might also be a potential tool for increasing underrepresented minority (URM) student exam scores in college-level introductory science courses. The researchers found that this intervention produced a 4.2% increase in exam performance for male URM students and a 2.2% increase for female URM students. Thus, while they use “achievement gap” throughout the paper to refer to racial and gender differences in exam scores, the study focused on ways to support URM student success.

In pursuit of improved language and clarity of intent, the term “achievement gap” should be replaced to reflect the research framework used to interrogate educational outcomes within and across demographic groups.

DEFICIT THINKING

Deficit thinking describes a mindset, or research framework, in which differences in outcomes between members of different groups, generally a politically and economically dominant group and an oppressed group, are attributed to a quality that is lacking in the biology, culture, or mindset of the oppressed group ( Valencia, 1997 ). Deficit thinking has pervaded public and academic discourse about the education of students from different races and ethnicities in the United States for centuries ( Menchaca, 1997 ).

Tenacious deficit-based explanations blame students from historically or currently marginalized groups for lower educational attainment. These falsities include biological inferiority due to brain size or structure ( Menchaca, 1997 ), negative cultural attributes such as inferior language acquisition ( Dudley-Marling, 2007 ), and accumulated deficits due to a “culture of poverty” ( Pearl, 1997 ; Gorski, 2016 ). More recently, lower achievement has been attributed to a lack of “grit” ( Ris, 2015 ) or the propensity for a “fixed” mindset ( Gorski, 2016 ; Tewell, 2020 ). While ideas around grit and mindset have demonstrable value in certain circumstances (e.g., Hacisalihoglu et al. , 2020 ), they fall short as primary explanations for differences in educational outcomes, because they focus attention on perceived deficits of students while providing little information about structural influences on failure and success, including how we define those constructs ( Harper, 2010 ; Gorski, 2016 ). In other words, deficit models often posit students as the people responsible for improving their own educational outcomes ( Figure 1 ).

Deficit thinking, regardless of intent, blames individuals, their families, their schools, or their greater communities for the consequences of societal inequities ( Yosso, 2006 ; Figure 1 ). This blame ignores the historic and structural drivers of inequity in our society, placing demands on members of underserved groups to adapt to unfair systems ( Valencia, 1997 ). A well-documented example of structural inequity is the consistent underresourcing of public schools that serve primarily students of color and children from lower socioeconomic backgrounds ( Darling-Hammond, 2013 ; Rothstein, 2013 ). Because learning is heavily influenced by factors outside the school environment, such as food security, trauma, and health ( Rothstein, 2013 ), schools themselves reflect gross disparities in resourcing based on historic discrimination ( Darling-Hammond, 2013 ). Deficit thinking focuses on student or cultural characteristics to explain performance differences and tends to overlook or minimize the impacts of systemic disparities. Deficit thinking also strengthens the narrative around student groups in terms of shortcomings, reinforces negative stereotypes, and ignores successes or drivers of success in those same groups ( Harper, 2015 ).

Achievement Gaps

The term “achievement gap” has historically described the difference in scores attained by students from racial and ethnic minority groups compared with White students on standardized tests or course exams ( Coleman et al. , 1966 ). As students from other historically or currently marginalized groups, such as female or first-generation students, are increasingly centered in research, the term is now used more broadly to compare any student population to White, middle and upper class, men ( Harper, 2010 ; Milner, 2012 ). Using White men as the basis for comparison comes at the expense of students from other groups ( Harper, 2010 ; Milner, 2012 ). Basing comparisons on the cultural perspectives of a single dominant group leads to “differences” being interpreted as “deficits,” which risks dehumanizing people in the marginalized groups ( Dinishak, 2016 ). Furthermore, centering White, wealthy, male performance means that even students from groups that tend to have higher test scores, like Asian-American students, risk dehumanization as “model minorities” or “just good at math” ( Shah, 2019 ).

Many researchers have highlighted the fact that the term “achievement gap” is a part of broader deficit-thinking models and rooted in racial hierarchy ( Ladson-Billings, 2006 ; Gutiérrez, 2008 ; Martin, 2009 ; Milner, 2012 ; Kendi, 2019 ). Focusing on achievement gaps emphasizes between-group differences over within-group differences ( Young et al. , 2017 ), reifies sociopolitical and historical groupings of people ( Martin, 2009 ), and minimizes attention to structural inequalities in education ( Ladson-Billings, 2006 ; Alliance to Reclaim Our Schools, 2018 ). Gutiérrez (2008) names this obsession with achievement gaps as a “gap-gazing fetish” that draws attention away from finding solutions that promote equitable learning ( Gutiérrez, 2008 ). Under a deficit-thinking model, achievement gaps are viewed as the primary problem, rather than a symptom of the problem ( Gutiérrez, 2008 ), and for decades they have been attributed to different characteristics of the demographics being compared ( Valencia, 1997 ). As such, proposed solutions tend to be couched in terms of remediation for students ( Figure 1 ).

Ignoring the social context of students’ education necessarily limits inferences that can be drawn about their success. Limiting measures of educational success, also conceptualized as achievement, to performance on exams or overall college GPA, often leaves out consideration of other potential data sources ( Weatherton and Schussler, 2021 ; Figure 2 ). This narrow perspective tends to perpetuate the systems of power and privilege that are already in place ( Gutiérrez, 2008 ). The biology education research community can instead broaden its sense of success to recognize the underlying historical and current contexts and the intersections of identities (e.g., racial, gender, socioeconomic) that contribute to those differences ( Weatherton and Schussler, 2021 ).

FIGURE 2. A selection of potential data sources that could inform researchers about within- and between-group differences in educational outcomes. This list does not encompass the full range of possible data sources, nor does it imply a hierarchy to the data. Instead, it reflects some of the diversity of quantitative and qualitative data that are directly linked to student outcomes and that are used under multiple research frameworks.

In biology education research, many papers still use the language of “achievement gap,” even in instances when researchers explicitly or implicitly use other nondeficit frameworks. While some may argue that this language merely describes a pattern, its origin and history is explicitly and inextricably linked to deficit-thinking models ( Gutiérrez, 2008 ; Milner, 2012 ). Thus, we join others in the choice to abandon the term “achievement gap” in favor of language—and frameworks—that align better to the goals of our research and to avoid the limitations and harm that can arise through its use.

Example: Focusing on Achievement Gaps Can Reinforce Racial Stereotypes

Messages of perpetual underachievement can inadvertently reinforce negative stereotypes. For example, Quinn (2020) demonstrated that, when participants watched a 2-minute video of a newscast using the term “achievement gap,” they disproportionately underpredicted the graduation rate of Black students relative to White students, even more so than participants in a control group who watched a counter-stereotypical video. They also scored significantly higher on an instrument measuring bias. Because bias is dynamic and affected by the environment, Quinn concludes that the video discussing the achievement gap likely heightened the bias of the participants ( Quinn, 2020 ).

Education researchers, just like the participants in Quinn’s (2020) study, inadvertently carry implicit bias against students from the different groups they study, and those biases can shift depending on context. Quinn (2020) demonstrates that just using the term “achievement gap” can reinforce the pervasive racial hierarchy that places Black students at the bottom. Researchers, without intending to, can be complicit in a system of White privilege and power if the language and frameworks underlying their study design, data collection, and/or data interpretation are aligned with bias and stereotype. If the goal is to dismantle inequities in our educational systems and research on those systems, the biology education research community must consider the historical and social weight of its literature to address racism head on, as progressive articles have been doing (e.g., Eddy and Hogan, 2014 ; Canning et al. , 2019 ; Theobald et al. , 2020 ).

SYSTEMS-LEVEL FRAMEWORKS

To move away from the achievement gap discourse—because of the history of the term, the perceived blame toward individual students, as well as the deficit thinking the term may imbue and provoke—we highlight some of the other frameworks for understanding student outcomes. We conclude discussion of each framework with an example from education research that can be reinterpreted within it, keeping in mind that multiple frameworks can be applied to different studies. We acknowledge two caveats about these reinterpretations: first, we are adding another layer of interpretation to the original studies, and we cannot claim that the original authors agree with these interpretations; second, each example could be interpreted through multiple frameworks, especially because these frameworks overlap ( Figure 1 ).

In this section, we begin at the systems level by examining opportunity gaps and educational debt. Rather than blaming students or their cultures for deficits in performance, these systems-level perspectives name white supremacy and the concomitant policies that maintain power imbalances as the cause of disparate student experiences.

Opportunity Gaps

The framework of opportunity gaps shifts the onus of differential student performance away from individual deficiencies and assigns solutions to actions that address systemic racism ( Milner, 2012 ; Figure 1 ). Specifically, opportunity gaps embody the difference in performance between students from historically and currently marginalized groups and middle and upper class, White, male students, with primary emphasis on opportunities that students have or have not had, rather than on their current performance (i.e., achievement) in a class ( Milner, 2012 ). Compared with deficit models, the focus shifts from assigning responsibility for the gap from the individual to society ( Figure 1 ).

Some researchers explore opportunity gaps by discussing the structural challenges that students from historically and currently marginalized groups have been facing (e.g., Rothstein, 2013 ). For example, poor funding in K–12 schools leads to inconsistent, poorly qualified, and poorly compensated teachers; few and outdated textbooks ( Darling-Hammond, 2013 ); limited field trips; a lack of extracurricular resources ( Rothstein, 2013 ); and inadequately supplied and cleaned bathrooms ( Darling-Hammond, 2013 ). Additional structural challenges that occur outside school buildings, but impact learning, include poor health and lack of medical care, food and housing insecurity, lead poisoning and iron deficiency, asthma, and depression ( Rothstein, 2013 ).

While the literature about opportunity gaps focuses more on K–12 than higher education ( Carter and Welner, 2013 ), college instructors can exacerbate opportunity gaps by biasing who has privilege (i.e., opportunities) in their classrooms. For example, some biology education literature focuses on how instructors’ implicit biases impact our students, such as by unconsciously elevating the status of males in the classroom ( Eddy et al. , 2014 ; Grunspan et al. , 2016 ).

Example: CUREs Can Prevent Opportunity Gaps.

Course-based undergraduate research experiences (CUREs) are one way to prevent opportunity gaps (e.g., Bangera and Brownell, 2014 ; CUREnet, n.d. ). Specifically, we interpret the suggestions that Bangera and Brownell (2014) make about building CUREs as a way to recognize that some students have the opportunity to participate in undergraduate research experiences while others do not. For example, students who access extracurricular research opportunities are likely relatively comfortable talking to faculty and, in many cases, have the financial resources to pursue unpaid laboratory positions ( Bangera and Brownell, 2014 ). More broadly, when research experiences occur outside the curriculum, they privilege students who know how to pursue and gain access to them. However, CUREs institutionalize the opportunity to conduct research, so that every student benefits from conducting research while pursuing an undergraduate degree.

Educational Debt

Ladson-Billings (2006) submits that American society has an educational debt, rather than an educational deficit. This framework shifts the work of finding solutions to educational inequities away from individuals and onto systems ( Figure 1 ). The metaphor is economic: A deficit refers to current mismanagement of funds, but a debt is the systematic accumulation of mismanagement over time. Therefore, differences in student performances are framed by a history that reflects amoral, systemic, sociopolitical, and economic inequities. Ladson-Billing ( 2006 ) suggests that focusing on debts highlights injustices that Black, Latina/o, and recent immigrant students have incurred: Focusing on student achievement in the absence of a discussion of past injustices does not redress the ways in which students and their parents have been denied access to educational opportunities, nor does it redress the ways in which structural and institutional racism dictate differences in performance. This approach begins by acknowledging the structural and institutional barriers to achievement in order to dismantle existing inequities. This reframing helps set the scope of the problem and identify a more accurate and just lens through which we make sense of the problem ( Cho et al. , 2013 ).

Example: NSF Supports Historically Black Colleges and Universities.

From my own (yet to be published) research, a participant described the HBCU where he studied physics as providing a “dome of security and safety.” In contrast, he recounted that when he attended a predominantly White institution, he constantly needed to be guarded and employ “his body sense,” an act that made him tense, defensive, and unable to listen. ( Rankins, 2019 , p. 50)

Example: Institutions Can Repay Educational Debt.

Institutions can repay educational debt by ensuring that their students have the resources and support structures necessary to succeed. The Biology Scholars Program at the University of California, Berkeley, is a prime example ( Matsui et al. , 2003 ; Estrada et al. , 2019 ). This program, begun in 1992 ( Matsui et al. , 2003 ) and still going strong ( Berkeley Biology Scholars Program, n.d. ), creates physical and psychological spaces that support learning: a study space and study groups, paid research experiences, and thoughtful mentoring. The students recruited to the program are from first-generation, low-socioeconomic status backgrounds and from groups that are historically underrepresented. When the students enter college, they have lower GPAs and Scholastic Aptitude Test scores than their counterparts with the same demographic profile who are not in the program. And yet, when they graduate, students in the Biology Scholars Program have higher GPAs and higher retention in biology majors than their counterparts ( Matsui et al. , 2003 ), perhaps because of the extended social support they receive from peers ( Estrada et al. , 2021 ). Moreover, students in this program report lower levels of stress and a greater sense of well-being ( Estrada et al. , 2019 ).

ASSET-BASED FRAMEWORKS

In this section, we continue to explore frameworks that move away from the achievement gap discourse, now focusing on models that build from students’ strengths. We have chosen two frameworks whose implications seem particularly relevant to and coincident with anti-racist research in biology education: community cultural wealth ( Yosso, 2005 ) and ethics of care ( Noddings, 1988 ). As before, we reinterpret articles from the education literature to illustrate these frameworks, and we once again include the caveats that we extend beyond the authors’ original interpretations and that other frameworks could also be used to reinterpret the examples.

Community Cultural Wealth

One asset-based way to frame student outcomes is to begin with the strengths that people from different demographic groups hold ( Yosso, 2005 ). Rather than focusing on racism, this approach focuses on community cultural wealth. The premise is that everyone can contribute a wealth of knowledge and approaches from their own cultures ( Yosso, 2005 ).

Community cultural wealth begins with critical race theory (CRT; Yosso, 2005 ). CRT illuminates the impact of race and racism embedded in all aspects of life within U.S. society ( Omi and Winant, 2014 ). CRT acknowledges that racism is interconnected with the founding of the United States. Race is viewed in tandem with intersecting identities that oppose dominant ones, and the constructs of CRT emerge by attending to the experiences of people from communities of color ( Yosso, 2005 ). Therefore, the experiences of students of color are central to transformative education that addresses the overrepresentation of White philosophies. CRT calls on research to validate and center these perspectives to develop a critical understanding about racism.

Community cultural wealth builds on these ideas by viewing communities of color as a source of students’ strength ( Yosso, 2005 ). The purpose of schooling is to build on the strengths that students have when they arrive, rather than to treat students as voids that need to be filled: students’ cultural wealth must be acknowledged, affirmed, and amplified through their education. This approach is consistent with those working to decolonize scientific knowledge (e.g., Howard and Kern, 2019 ).

Example: Community Cultural Wealth Can Improve Mentoring.

Thompson and Jensen-Ryan (2018) offer advice to mentors about how to use cultural wealth to mentor undergraduate students in research. They identify the forms of scientific cultural capital that research mentors typically value, finding that these aspects of a scientific identity are closely associated with majority culture. They challenge mentors to broaden the forms of recognizable capital. For example, members of the faculty can actively recruit students into their labs from programs aimed to promote the diversity of scientists, rather than insisting that students approach them with their interest to work in the lab ( Thompson and Jensen-Ryan, 2018 ). They can recognize that undergraduate students may not express an interest in a research career–especially initially—but that research experience is still formative. They can recognize that students who are strong mentors to their peers are valuable members of a research team and that this skill is a form of scientific capital. They can value the diverse backgrounds of students in their labs, rather than insisting that they come from families that have prioritized scientific thinking and research. In sum, the gaps that Thompson and Jensen-Ryan (2018) identify are in research mentors’ attitudes, rather than in student performance.

Assets can also be developed in the classroom. We interpret Parnes et al. ’s (2020) analysis of the Connected Scholars program as stemming from community cultural wealth. The Connected Scholars program normalized help-seeking and increased the help network available to first-generation college students, 90% of whom were racial or ethnic minorities, in a 6-week summer program that bridged students from high school to college. First-generation college students were provided explicit instruction on how to sustain these two types of support. The Connected Scholars intervention promoted help-seeking behaviors and seemed to mediate higher GPAs. Additionally, students in the intervention reported through a survey that they had better relationships with their instructors than students in the control group ( Parnes et al. , 2020 ). In other words, cultural wealth can be amplified in college for first-generation students (see also the Biology Scholars Program, discussed in the Opportunity Gaps section; Matsui et al. , 2003 ; Estrada et al. , 2019 ).

Ethics of Care

As a framework, ethics of care complements community cultural wealth, in that both are asset-based. A key difference is that community cultural wealth focuses on the assets that students bring, and ethics of care focuses on the assets that an instructor brings to create a classroom of respect and confidence in students.

A foundation of biology education research is that instructors want their students to learn, and it is buttressed by literature concerning students’ emotional well-being. For example, the field considers how students with disabilities experience active learning ( Gin et al. , 2020 ) and how group work promotes collaboration and learning ( Wilson et al. , 2018 ). Studies like these echo the philosophy of ethics of care developed by Noddings (1988) .

The premises of teaching through the ethics of care are that everyone—including students and instructors—has both an innate desire to learn and the capacity to nurture ( Pang et al. , 2000 ). In teaching, these premises form the basis for student–instructor relationships. Nieto and Bode (2012) caution against the oversimplification that caring means being nice: the ethics of care encompasses niceness, in addition to articulating high standards of performance. Instructors must also support and respect students as they meet those standards, especially when students did not recognize that they could meet those goals at the outset. This framework is about nurturing students to accomplish more than they thought possible.

Combining an inclusive culture, for example, through positive instructor talk ( Seidel et al. , 2015 ; Harrison et al. , 2019 ; Seah et al. , 2021 ), growth mindset ( Canning et al. , 2019 ), or increased course structure ( Eddy and Hogan, 2014 ), with evidence-based practices for teaching content ( Freeman et al. , 2014 ; Theobald et al. , 2020 ) has garnered recent attention as a way to create a powerful ethic of care in classrooms. For example, instructor talk, that is, what instructors say in class other than the content they are teaching, addresses student affect. Seidel et al. (2015) and Harrison et al. (2019) analyzed classroom transcripts to identify different categories of instructor talk. While further research can probe the impacts of instructor talk on student outcomes, the idea is consistent with the principles of ethics of care: for example, one category of talk describes the instructor–student relationship as one of respect, fostered through statements such as “People are bringing different pieces of experience and knowledge into this question and I want to kind of value the different kinds of experience and knowledge that you bring in” ( Seidel et al. , 2015 , p. 6). Instructor talk also generates a classroom culture of support and validation for marginalized students and overall builds classroom community ( Ladson-Billings, 2013 ).

Example: Departments Can Implement Care.

Gutiérrez (2000) presents an example of an entire department applying ethics of care to support how African-American students learn math. This study is an ethnography of a particularly successful STEM magnet program in a public high school with a population that is majority African American. In her analysis of the math department, Gutiérrez avoids the phrase “achievement gap,” while also recognizing that people outside the school assume a deficit model when considering the students . Instead, she illustrates how researchers can use an asset-based lens to build from knowledge about differences in performance ( Gutiérrez, 2000 ).

Gutierrez ( 2000 ) examines pedagogy that supports African-American students. She documents how a culture of excellence is developed within a school setting that promotes student achievement. This culture is complex, in that there are multiple layers of support that provide students with repertoires for advancement ( Gutiérrez, 2000 )—the emphasis is on how teachers create an environment where students are both challenged through the curriculum and supported along the way. The teachers in this study have a dynamic conception of their students, and they demonstrate a unified commitment to support the broadest array of students at their school. The institution itself, represented in part through the departmental chair, has values that empower teachers to support students, proactive commitment from teachers to find innovative practices to serve students, and a supportive chairperson.

The math department exhibited a student-centered approach that epitomizes ethics of care. The teachers in the math department rotated through all of the courses and were therefore familiar with the entire curriculum. This knowledge helped them support one another, sharing successful strategies and working to improve the courses. It set up an environment in which they prioritized making decisions collectively. This collaboration led to a sense of togetherness among teachers and a sense of investment in individual students’ successes. As a result, the teachers decided to remove less-challenging courses from the curriculum and replaced them with more advanced courses—against the recommendations of the school district. The chair of the department worked with the faculty to support student learning, consider course assignments, and choose topics for and frequency of faculty meetings. The chair also attended to teachers’ emotional needs, for example, by talking to teachers every day, working with teachers to determine the best strategies for evaluating teaching practices, and enacting a teaching philosophy that valued problem solving over achieving correct answers.

The support that the teachers provided each other coincided with strong support for students. For example, students attended the magnet program because they were interested in science; they notably did not have to take entrance exams or maintain a certain GPA. If students struggled with a subject, they received tutoring. The teachers also invited graduates of the program to come back and visit, keeping the students motivated by showing them success.

Example: Biology Instructors Can Adopt an Ethics of Care.

In much of the research on differential performance in our field, researchers focus on identifying strategies that help students, regardless of their histories, in their learning success. This asset-based approach acknowledges that students start at different places, but also that instructors can implement strategies that support all students in a trajectory toward common learning goals. This argument is often posited in terms of inclusive teaching (e.g., Dewsbury and Brame, 2019 ).

Some papers that measure the effect of inclusive teaching practices may use “gap” language, perhaps as a historical artifact of our discipline. These papers emphasize the just mission to “close the gap”—or, in anti-deficit language, for all students to learn the material and perform well on assessments. For example, Theobald et al. (2020) conducted a meta-analysis of undergraduate STEM classes, drawing on 26 studies of courses reporting failure rates (44,606 students) and 15 studies (9238 students) that reported exam scores. Within these samples, they compared instruction in lecture format with instruction using active-learning strategies. The analysis compared the success of students from minoritized groups using these two teaching strategies and found conclusive evidence of the efficacy of active teaching for underrepresented student success in STEM courses. The powerful implication of this study is that college STEM instructors can mitigate some of the effects of oppression that students have experienced in their lifetime.

In another study demonstrating the philosophy of ethics of care, Canning et al. (2019) found narrower racial disparities in performance in courses taught by instructors who had a growth mindset about their students’ ability to learn, compared with instructors who viewed level of achievement as fixed. In fact, they found that the instructor mindset had a bigger impact on student performance than other faculty characteristics ( Canning et al. , 2019 ). While they focused on the negative consequences of instructors’ fixed mindset, the corollary is that a growth mindset can reflect an ethics of care that both motivates students and generates a positive classroom environment.

The successful instructors will also work to recognize their implicit biases and to ensure that they support a growth mindset for all students, regardless of demographic. This is particularly relevant, because implicit biases have “more to do with associations we’ve absorbed through history and culture than with explicit racial animus” ( Eberhardt, 2019 , p. 160). Realizing how our own socialization may have conditioned us to automatically produce harmful but hidden narratives warrants our attention ( Eberhardt, 2019 ).

MOVING FORWARD

Ladson-Billings (2006) reframed the performance of students from historically and currently marginalized groups from achievement gap to educational debt; this reframing has contributed to a movement to critically examine the term. At the same time, however, the term “achievement gap” has become a catchall used by researchers untethered from its deeper historical context.

Researchers choose words to describe their research that reflect their personal worldviews and research frameworks; in turn, these worldviews and frameworks influence future researchers. Every discipline grapples with terminology, and phrases that were common historically may fall out of use. In some instances, the terms themselves no longer suffice, so a simple “search and replace” may be all that is required to address the issue. The term “achievement gap,” however, is tied to specific frameworks that need to be acknowledged and redressed; it affects how research is designed, how results are interpreted, and what conclusions are drawn. Simply replacing “achievement gap” would not address the undermining nature of deficit-based research frameworks.

Researchers who used the term “achievement gap” may not have intended to use a deficit-thinking framework in their study. In fact, as we have demonstrated with our examples, some powerful articles exist in biology education research that used the term and also implicitly used one of the systems-level or asset-based frameworks we identified.

In these examples, we have reinterpreted the results of primary research with the frameworks we identified. This leads to two points of caution. The first is that we are adding another layer of interpretation, one that the original authors may not have intended. The second is that each example could be interpreted through multiple frameworks, especially because these frameworks overlap ( Figure 1 ). For example, Bangera and Brownell (2014) identify barriers to participating in independent undergraduate research experiences. Course-based undergraduate research experiences (CUREs) offer research opportunities to students who previously could not access them. As discussed earlier, we posited CUREs as an example of a way to reduce opportunity gaps. However, we could also have interpreted the act of implementing a CURE as repaying an educational debt by repairing a form of bias typical within the academy ( Figure 1 ).

Addressing educational inequities requires that biology education researchers quantify differences in performance across demographic groups ( Figure 2 ) and must be done with the utmost care. Disaggregating data is necessary, as is analyzing those data with a just framework that dismantles racial hierarchies and carefully considers the sources of data used to understand those inequities. The frameworks we choose affect our analysis; we must avoid the common trap of assuming that quantitative data and data analysis are free from bias. To illustrate the degree of subjectivity that enters data analysis, Huntington-Klein et al. (2021) found that when seven different researchers received copies of the same data set, each reported different levels of statistical significance, including one researcher who found an effect that was opposite to what the others found. Moving away from analyses based on the phrase “achievement gap” will avoid unintentionally reinforcing the racial bias and better reflect the intention of disaggregating data to quantify differences in performance across demographic groups to actively dismantle persistent educational inequities.

In addition to disaggregating and diversifying data on outcomes ( Figure 2 ), the biology education research community must consider how definitions of success may center White, middle-class ways of knowing and performing ( Weatherton and Schussler, 2021 ). In their recent essay, Weatherton and Schussler (2021) reported that, in articles published in LSE between the years 2015 and 2020, the word “success,” when defined, largely meant high GPAs and exam scores. This narrow definition of success prioritizes scientific content, whereas there are additional admirable goals by which success could be measured ( Figure 2 ; see also Weatherton and Schussler, 2021 and references therein). Moreover, the scientific skills that are valued are Eurocentric, rather embodying a diversity of scientific approaches ( Howard and Kern, 2019 ). In addition to the limitations of narrowly defining success as exam performance, it should be noted that tests themselves are not always fair or equitable across all student populations ( Martinková et al. , 2017 ); success measured in this way should be interpreted with caution, particularly when comparing students across different courses, institutions, or identities.

As we discussed earlier, instructors’ and researchers’ deep beliefs about educational success and achievement necessarily impact their actions. For this reason, we propose that interrogating the frameworks we use is necessary and that such interrogation should acknowledge harm that may have been inflicted. While writing this essay, for example, our understandings of the frameworks underlying our own research, teaching, and other engagements have grown. Much like the research studies we discuss, our intentions, actions, and frameworks can be and have been out of alignment. For example, our own actions with respect to departmental policies, course designs, and program structures have not always reflected the principles to which we subscribe. Although this essay focuses on frameworks in research, we provide a list of some questions that we have asked of ourselves and that could catalyze reflection in all areas of our professional work ( Table 1 ).

In conclusion, we have presented four ways to frame differences in academic performance across students from different demographic groups that firmly reject deficit-based thinking ( Figure 1 ). The notions of opportunity gaps and educational debt demonstrate how systems thinking can recognize socio-environmental barriers to student learning. Asset-based frameworks that include community cultural wealth and ethics of care can help identify actions that institutions, instructors, and students can take to meet learning goals. We hope that researchers in the field move forward by 1) avoiding, or at least minimizing, deficit thinking; 2) explicitly stating asset-based and systems-level frameworks that celebrate students’ accomplishments and move toward justice; and 3) using language consistent with their frameworks.

ACKNOWLEDGMENTS

We thank Starlette Sharp and our external reviewers for helpful feedback on this article. We live and work on the lands of the Kizh/Tongva/Gabrieleño, Duwamish, and Willow (Sammamish) People past, present, and future. We also acknowledge the people whose uncompensated labor built this country, including many of its academic institutions.

• Alliance to Reclaim Our Schools . ( 2018 ). Confronting the education debt: We owe billions to Black, Brown and low-income students and their schools (p. 25). Retrieved February 23, 2022, from http://educationdebt.reclaimourschools.org/wp-content/uploads/2018/08/Confronting-the-Education-Debt_FullReport.pdf Google Scholar
• Asai, D. J. ( 2020 ). Race matters . Cell , 181 (4), 754–757. https://doi.org/10.1016/j.cell.2020.03.044 Medline ,  Google Scholar
• Bangera, G., & Brownell, S. E. ( 2014 ). Course-based undergraduate research experiences can make scientific research more inclusive . CBE—Life Sciences Education , 13 , 602–606. https://doi.org/10.1187/cbe.14-06-0099 Link ,  Google Scholar
• Berkeley Biology Scholars Program . ( n.d. ). Home . Retrieved March 25, 2021, from https://bsp.berkeley.edu/home Google Scholar
• Canning, E. A., Muenks, K., Green, D. J., & Murphy, M. C. ( 2019 ). STEM faculty who believe ability is fixed have larger racial achievement gaps and inspire less student motivation in their classes . Science Advances , 5 (2), eaau4734. https://doi.org/10.1126/sciadv.aau4734 Medline ,  Google Scholar
• Carey, R. L. ( 2014 ). A cultural analysis of the achievement gap discourse: Challenging the language and labels used in the work of school reform . Urban Education , 49 (4), 440–468. https://doi.org/10.1177/0042085913507459 Google Scholar
• Carter, P. L., & Welner, K. G. ( 2013 ). Closing the opportunity gap: What America must do to give every child an even chance . New York, NY: Oxford University Press. Google Scholar
• Cho, S., Crenshaw, K. W., & McCall, L. ( 2013 ). Toward a field of intersectionality studies: Theory, applications, and praxis . Signs: Journal of Women in Culture and Society , 38 , 785–810. Google Scholar
• Coleman, J. S., Campbell, E. Q., Hobson, C. J., McPartland, J., Weinfeld, F. D., & York, R. L. ( 1966 ). Equality of educational opportunity . Washington, DC: U.S. Department of Health, Education, and Welfare. Google Scholar
• CUREnet . ( n.d. ). Home page . Retrieved May 28, 2021, from https://serc.carleton.edu/curenet/index.html Google Scholar
• Darling-Hammond, L. ( 2013 ). Inequality and school resources: What it will take to close the opportunity gap? In Carter, P. L.Welner, K. G. (Eds.), Closing the opportunity gap: What America must do to give every child an even chance (pp. 77–97). : Oxford University Press. Google Scholar
• Dewsbury, B. M., & Brame, C. J. ( 2019 ). Inclusive Teaching . CBE—Life Sciences Education , 18 (2), fe2. https://doi.org/10.1187/cbe.19-01-0021 Link ,  Google Scholar
• Dinishak, J. ( 2016 ). The deficit view and its critics . Disability Studies Quarterly , 36 (4). http://dsq-sds.org/article/view/5236/4475 Google Scholar
• Dudley-Marling, C. ( 2007 ). Return of the deficit . Journal of Educational Controversy , 2 (1), 14. Google Scholar
• Eberhardt, J. ( 2019 ). Biased: Uncovering the hidden prejudice that shapes what we see, think, and do . New York, NY: Viking. Google Scholar
• Eddy, S. L., Brownell, S. E., & Wenderoth, M. P. ( 2014 ). Gender gaps in achievement and participation in multiple introductory biology classrooms . CBE—Life Sciences Education , 13 (3), 478–492. https://doi.org/10.1187/cbe.13-10-0204 Link ,  Google Scholar
• Eddy, S. L., & Hogan, K. A. ( 2014 ). Getting under the hood: How and for whom does increasing course structure work? CBE—Life Sciences Education , 13 (3), 453–468. https://doi.org/10.1187/cbe.14-03-0050 Link ,  Google Scholar
• Estrada, M., Eppig, A., Flores, L., & Matsui, J. T. ( 2019 ). A longitudinal study of the Biology Scholars Program: Maintaining student integration and intention to persist in science career pathways . Understanding Interventions , 10 , 26. Google Scholar
• Estrada, M., Young, G. R., Flores, L., Yu, B., & Matsui, J. ( 2021 ). Content and quality of science training programs matter: Longitudinal study of the Biology Scholars Program . CBE—Life Sciences Education , 20 (3), ar44. https://doi.org/10.1187/cbe.21-01-0011 Link ,  Google Scholar
• Farkas, G. ( 2004 ). The Black-White test score gap . Context , 3 , 12–19. https://doi.org/10.1525/ctx.2004.3.2.12 Google Scholar
• Freeman, S., Eddy, S. L., McDonough, M., Smith, M. K., Okoroafor, N., Jordt, H., & Wenderoth, M. P. ( 2014 ). Active learning increases student performance in science, engineering, and mathematics . Proceedings of the National Academy of Sciences USA , 111 (23), 8410–8415. https://doi.org/10.1073/pnas.1319030111 Medline ,  Google Scholar
• Gin, L. E., Guerrero, F. A., Cooper, K. M., & Brownell, S. E. ( 2020 ). Is active learning accessible? Exploring the process of providing accommodations to students with disabilities . CBE—Life Sciences Education , 19 (4), es12. https://doi.org/10.1187/cbe.20-03-0049 Link ,  Google Scholar
• Gorski, P. C. ( 2016 ). Poverty and the ideological imperative: A call to unhook from deficit and grit ideology and to strive for structural ideology in teacher education ,. Journal of Education for Teaching , 42 (4), 378–386. https://doi.org/10.1080/02607476.2016.1215546 Google Scholar
• Gouvea, J. S. ( 2021 ). Antiracism and the problems with “achievement gaps” in STEM education . CBE—Life Sciences Education , 20 (1), fe2. https://doi.org/10.1187/cbe.20-12-0291 Link ,  Google Scholar
• Grunspan, D. Z., Eddy, S. L., Brownell, S. E., Wiggins, B. L., Crowe, A. J., & Goodreau, S. M. ( 2016 ). Males under-estimate academic performance of their female peers in undergraduate biology classrooms . PLoS ONE , 16. Google Scholar
• Gutiérrez, R. ( 2000 ). Advancing African-American, urban youth in mathematics: Unpacking the success of one math department . American Journal of Education , 109 (1), 63–111. https://doi.org/10.1086/444259 Google Scholar
• Gutiérrez, R. ( 2008 ). A “gap-gazing” fetish in mathematics education? Problematizing research on the achievement gap . Journal for Research in Mathematics Education , 39 (4), 357–364. Google Scholar
• Hacisalihoglu, G., Stephens, D., Stephens, S., Johnson, L., & Edington, M. ( 2020 ). Enhancing undergraduate student success in stem fields through growth-mindset and grit . Education Sciences , 10 (10), 279. https://doi.org/10.3390/educsci10100279 Google Scholar
• Harper, S. R. ( 2010 ). An anti-deficit achievement framework for research on students of color in STEM . New Directions for Institutional Research 148 , 63–74. Google Scholar
• Harper, S. R. ( 2015 ). Success in these schools? Visual counternarratives of young men of color and urban high schools they attend . Urban Education , 50 , 139–169. https://doi.org/10.1177/0042085915569738 Google Scholar
• Harrison, C. D., Nguyen, T. A., Seidel, S. B., Escobedo, A. M., Hartman, C., Lam, K. , ... & Tanner, K. D. ( 2019 ). Investigating instructor talk in novel contexts: Widespread use, unexpected categories, and an emergent sampling strategy . CBE—Life Sciences Education , 18 (3), ar47. https://doi.org/10.1187/cbe.18-10-0215 Link ,  Google Scholar
• Howard, M. A., & Kern, A. L. ( 2019 ). Conceptions of wayfinding: Decolonizing science education in pursuit of Native American success . Cultural Studies of Science Education , 14 , 1135–1148. https://doi.org/10.1007/s11422-018-9889-6 Google Scholar
• Huntington-Klein, N., Arenas, A., Beam, E., Bertoni, M., Bloem, J. R., Burli, P. , ... & Stopnitzky, Y. ( 2021 ). The influence of hidden researcher decisions in applied microeconomics . Economic Inquiry , 59 (3), 944–960. https://doi.org/10.1111/ecin.12992 Google Scholar
• Jencks, C., & Phillips, M. ( 2006 ). The Black-White test score gap: An introduction . In The Black-White test score gap (pp. 1–51). Washington, DC: Brookings Institution Press. Google Scholar
• Jordt, H., Eddy, S. L., Brazil, R., Lau, I., Mann, C., Brownell, S. E. , ... & Freeman, S. ( 2017 ). Values affirmation intervention reduces achievement gap between underrepresented minority and white students in introductory biology classes . CBE—Life Sciences Education , 16 (3), ar41. https://doi.org/10.1187/cbe.16-12-0351 Link ,  Google Scholar
• Kendi, I. X. ( 2019 ). How to be an antiracist . New York, NY: One World. Google Scholar
• Kozol, J. ( 2005 ). The shame of the nation . New York, NY: Crown Publishing. Google Scholar
• Ladson-Billings, G. ( 2006 ). From the achievement gap to the education debt: Understanding achievement in U.S. schools . Educational Researcher , 35 (7), 3–12. https://doi.org/10.3102/0013189X035007003 Google Scholar
• Ladson-Billings, G. ( 2013 ). The dreamkeepers (2nd ed.). San Francisco, CA: Jossey-Bass. Google Scholar
• Martin, D. B. ( 2009 ). Researching race in mathematics education . Teachers College Record , 111 , 295–338. Google Scholar
• Martinková, P., Drabinová, A., Liaw, Y.-L., Sanders, E. A., McFarland, J. L., & Price, R. M. ( 2017 ). Checking equity: Why differential item functioning analysis should be a routine part of developing conceptual assessments . CBE—Life Sciences Education , 16 (2), rm2. https://doi.org/10.1187/cbe.16-10-0307 Link ,  Google Scholar
• Matsui, J. T., Liu, R., & Kane, C. M. ( 2003 ). Evaluating a science diversity program at UC Berkeley: More questions than answers . Cell Biology Education , 2 (2), 117–121. https://doi.org/10.1187/cbe.02-10-0050 Link ,  Google Scholar
• Menchaca, M. ( 1997 ). Early racist discourses: The roots of deficit thinking . In Valencia, R. R. (Ed.), The evolution of deficit thinking: Educational thought and practice ( The Stanford series on education and public policy ) (pp. 13–40). Washington, DC: Falmer Press/Taylor & Francis. Google Scholar
• Milner, H. R. ( 2012 ). Beyond a test score: Explaining opportunity gaps in educational practice . Journal of Black Studies , 43 (6), 693–718. https://doi.org/10.1177/0021934712442539 Google Scholar
• National science Foundation . ( 2020 ). Historically Black Colleges and Universities - Undergraduate Program (HBCU-UP) . Retrieved February 23, 2022, from https://beta.nsf.gov/funding/opportunities/historically-black-colleges-and-universities-undergraduate-program-hbcu Google Scholar
• Nieto, S., & Bode, P. ( 2012 ). Affirming diversity: The sociopolitical context of multicultural education (6th ed.). Boston, MA: Pearson Education. Google Scholar
• Noble, S. ( 2018 ). Algorithms of oppression (Illustrated ed.). New York: NYU Press. Google Scholar
• Noddings, N. ( 1988 ). An ethic of caring and its implications for instructional arrangements . American Journal of Education , 96 (2), 215–230. https://doi.org/10.1086/443894 Google Scholar
• Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. ( 2019 ). Dissecting racial bias in an algorithm used to manage the health of populations . Science , 366 , 447–453. https://doi.org/10.1126/science.aax2342 Medline ,  Google Scholar
• Omi, M., & Winant, H. ( 2014 ). Racial formation in the United States (3rd ed.). New York, NY: Routledge. Google Scholar
• Painter, N. I. ( 2020 , July 22). Why “White” should be capitalized, too . Washington Post . Retrieved February 23, 2022, from www.washingtonpost.com/opinions/2020/07/22/why-white-should-be-capitalized Google Scholar
• Pang, V. O., Rivera, J., & Mora, J. K. ( 2000 ). The ethic of caring: Clarifying the foundation of multicultural education . Educational Forum , 64 , 25–32. Google Scholar
• Parnes, M. F., Kanchewa, S. S., Marks, A. K., & Schwartz, S. E. O. ( 2020 ). Closing the college achievement gap: Impacts and processes of a help-seeking intervention . Journal of Applied Developmental Psychology , 67 , 101121. https://doi.org/10.1016/j.appdev.2020.101121 Google Scholar
• Pearl, A. ( 1997 ). Cultural and accumulated environmental models . In Valencia, R. R. (Ed.), The evolution of deficit thinking: Educational thought and practice ( The Stanford series on education and public policy ) (pp. 132–159). Washington, DC: Falmer Press/Taylor & Francis. Google Scholar
• Quinn, D. M. ( 2020 ). Experimental effects of “achievement gap” news reporting on viewers’ racial stereotypes, inequality explanations, and inequality prioritization . Educational Researcher , 49 (7), 482–492. https://doi.org/10.3102/0013189X20932469 Google Scholar
• Rankins, C. ( 2019 ). HBCUs and Black STEM student success . Peer Review , 21 , 50–51. Google Scholar
• Ris, E. W. ( 2015 ). Grit: A short history of a useful concept . Journal of Educational Controversy , 10 (1). Retrieved February 23, 2022, from https://cedar.wwu.edu/jec/vol10/iss1/3 Google Scholar
• Rothstein, R. ( 2013 ). Why children from lower socioeconomic classes, on average, have lower academic achievement than middle-class children . In Carter, P. L.Welner, K. G. (Eds.), Closing the opportunity gap: What America must do to give every child an even chance (pp. 61–74). New York, NY: Oxford University Press. Google Scholar
• Sadker, D., Sadker, M. P., & Zittleman, K. R. ( 2009 ). Still failing at fairness: How gender bias cheats girls and boys in school and what we can do about it . New York, NY: Scribner. Google Scholar
• Seah, Y. M., Chang, A. M., Dabee, S., Davidge, B., Erickson, J. R., Olanrewaju, A. O., & Price, R. M. ( 2021 ). Pandemic-related instructor talk: How new instructors supported students at the onset of the COVID-19 pandemic . Journal of Microbiology & Biology Education , 22 . https://doi.org/10.1128/jmbe.v22i1.2401 Medline ,  Google Scholar
• Seidel, S. B., Reggi, A. L., Schinske, J. N., Burrus, L. W., & Tanner, K. D. ( 2015 ). Beyond the biology: A systematic investigation of noncontent instructor talk in an introductory biology course . CBE—Life Sciences Education , 14 (4), ar43. https://doi.org/10.1187/cbe.15-03-0049 Link ,  Google Scholar
• Shah, N. ( 2019 ). “Asians are good at math” is not a compliment: STEM success as a threat to personhood . Harvard Educational Review , 89 , 661–686. https://doi.org/10.17763/1943-5045-89.4.661 Google Scholar
• Steele, C. M. ( 2011 ). Whistling Vivaldi: How stereotypes affect us and what we can do (Reprint ed.). New York: Norton. Google Scholar
• Takacs, D. ( 2003 ). How does your positionality bias your epistemology? Thought & Action , 19 , 27–38. Google Scholar
• Tewell, E. ( 2020 ). The problem with grit: Dismantling deficit thinking in library instruction . Portal: Libraries and the Academy , 20 , 137–159. Google Scholar
• Theobald, E. J., Hill, M. J., Tran, E., Agrawal, S., Arroyo, E. N., Behling, S. , ... & Freeman, S. ( 2020 ). Active learning narrows achievement gaps for underrepresented students in undergraduate science, technology, engineering, and math . Proceedings of the National Academy of Sciences USA , 117 (12), 6476–6483. https://doi.org/10.1073/pnas.1916903117 Medline ,  Google Scholar
• Thompson, J. J., & Jensen-Ryan, D. ( 2018 ). Becoming a “science person”: Faculty recognition and the development of cultural capital in the context of undergraduate biology research . CBE—Life Sciences Education , 17 (4), ar62. https://doi.org/10.1187/cbe.17-11-0229 Link ,  Google Scholar
• Valencia, R. R. ( 1997 ). The evolution of deficit thinking: Educational thought and practice ( The Stanford series on education and public policy ). Washington, DC: Falmer Press/Taylor & Francis. Google Scholar
• Weatherton, M., & Schussler, E. E. ( 2021 ). Success for all? A call to re-examine how student success is defined in higher education . CBE—Life Sciences Education , 20 (1), es3. https://doi.org/10.1187/cbe.20-09-0223 Link ,  Google Scholar
• Wilson, K. J., Brickman, P., & Brame, C. J. ( 2018 ). Group Work . CBE—Life Sciences Education , 17 (1), fe1. https://doi.org/10.1187/cbe.17-12-0258 Link ,  Google Scholar
• Yosso, T. J. ( 2005 ). Whose culture has capital? A critical race theory discussion of community cultural wealth . Race Ethnicity and Education , 8 (1), 69–91. https://doi.org/10.1080/1361332052000341006 Google Scholar
• Yosso, T. J. ( 2006 ). Critical race counterstories along the Chicana/Chicano educational pipeline . New York, NY: Routledge. Google Scholar
• Young, J. L., Young, J. R., & Ford, D. Y. ( 2017 ). Standing in the gaps: Examining the effects of early gifted education on black girl achievement in STEM . Journal of Advanced Academics , 28 (4), 290–312. https://doi.org/10.1177/1932202X17730549 Google Scholar

Submitted: 3 June 2021 Revised: 20 December 2021 Accepted: 2 February 2022

© 2022 S. Y. Shukla et al. CBE—Life Sciences Education © 2022 The American Society for Cell Biology. This article is distributed by The American Society for Cell Biology under license from the author(s). It is available to the public under an Attribution–Noncommercial–Share Alike 3.0 Unported Creative Commons License (http://creativecommons.org/licenses/by-nc-sa/3.0).

Educational Inequalities in the UK: Gender, Ethnicity and Social Class

• First Online: 28 February 2022

Cite this chapter

• Aqsa Saeed 2  

1025 Accesses

1 Altmetric

Chapter 2 looks at the trends in changing intergenerational gender, ethnic and class differences in overall educational achievement. Schools have been significant in widening the expectations of appropriate careers for women and supporting women to obtain the qualifications necessary for a greater choice of occupational recruitment. I also consider why gender inequalities in career development persist in the workplace. I present the broad trends in educational achievement of British Pakistanis, considering whether this is translated into more professional jobs. Finally I look at the ways in which class can affect educational outcomes and employment prospects. People who come from more deprived communities have lower educational attainment levels. Pakistanis have lower GCSE attainment rates in comparison with most other ethnic minority groups, and a significant proportion of British Pakistanis leave school with no qualifications. Nevertheless increasing numbers now stay on to pursue further study, including to degree level.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

• Available as EPUB and PDF
• Read on any device
• Instant download
• Own it forever
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Data shows that in 2017/18, 46.7% of girls and 40% of boys got a strong pass (Grade 5 or above) in GCSE English and Maths. Source Data: GOV.UK Ethnicity facts and figures statistics (education-skills-and training/11-to-16-year-old/a-to-c-in-english-and-maths-gcse-attainment-for-children-aged-14-to-16-key-stage-4).

Source: ONS SFR 37/2004 National Statistics http://www.education.gov.uk/rsgateway/DB/SFR/s000528/sfr37-2004.pdf ). I am using older data because it is closer to the cohort that my informants belong to.

The data do not show whether they are unemployed and ‘seeking employment’ or just ‘not employed’.

Abbas, T. (2002). Teacher Perceptions of South Asians in Birmingham Schools and Colleges. Oxford Review of Education, 28 (4), 447–471.

Article   Google Scholar  

Anwar, M. (1979). Myth of Return: Pakistanis in Britain . Heinemann Educational Publishers.

Google Scholar  

Arnot, M. (1993). A Crisis in Patriarchy? British Feminist Educational Politics and State Regulation of Gender. In M. Arnot & K. Weiler (Eds.), Feminism and Social Justice in Education: International Perspectives . The Falmer Press.

Arnot, M., David, M., & Weiner, G. (1999). Closing the Gender Gap Postwar Education and Social Change . Polity.

Basit, T. (1997). Eastern Values, Western Milieu . Ashgate.

Broecke, S., & Nicholls, T. (2007). Ethnicity and Degree Attainment . Research Report RW92. Department for Education and Skills.

Brown, J. (2006). Global South Asians Introducing the Modern Diaspora . Cambridge University Press.

Book   Google Scholar  

Dale, A., Shaheen, N., Kalra, V., & Fieldhouse, E. (2002). Routes into Education and Employment for Young Pakistani and Bangladeshi Women in the UK. Ethnic and Racial Studies, 25 (6).

DFES. (2007). Gender and Education: The Evidence on Pupils in England . DfES Publications.

EOC. (2006). Moving on Up? Bangladeshi, Pakistani and Black Caribbean Women and Work . Equal Opportunities Commission.

Fisher, K., Egerton, M., Gershuny, J. I., & Robinson, A. (2007). Gender Convergence in the American Heritage Time Use Study (AHTUS). Social Indicators Research, 82 (1), 1–33.

Garner, S., & Bhattacharyya, G. (2011). Poverty, Ethnicity and Place . Joseph Rowntree Foundation.

Gershuny, J. (2000). Changing Times: Work and Leisure in Post-Industrial Society . Oxford University Press.

Ghuman, P. (2002). South Asian Adolescents in British Schools: A Review. Educational Studies, 28 (1), 47–59.

Gillborn, D., & Gipps, C. (1996). Recent Achievements of Ethnic Minority Pupils (HMSO, Ed.). London.

Heath, A., & Cheung, S. Y. (2006). Ethnic Penalties in the Labour Market: Employers and Discrimination . Department for Work and Pensions.

Heath, A., & McMahon, D. (1997). Education and Occupational Attainments: The Impact of Ethnic Origins . The Stationery Office.

Hussain, Y., & Bagguley, P. (2007). Moving on Up. South Asian Women and Higher Education . Trentham Books Limited.

Kalra, V. (2000). From Textile Mills to Taxi Ranks: Experiences of Migration, Labour and Social Change . Ashgate.

Kay, T. (2007). Daughters of Islam, Sisters in Sport. In C. Aitchison, P. Hopkins, & M. Kwan (Eds.), Geographies of Muslim Identities . Ashgate.

Khan, D., & Shaheen, F. (2017). Introduction: Analysing and Responding to Racial and Class Inequalities. In D. Khan & F. Shaheen (Eds.), Minority Report: Race and Class in Post-Brexit Britain . Runnymede.

Leese, M. (2010). Bridging the Gap: Supporting Student Transitions into Higher Education. Journal of Further and Higher Education, 34 (2), 239–251.

Lindley, J. K., Dale, A., & Dex, S. (2006). Ethnic Differences in Women’s Employment. Oxford Economic Papers, 58 (2), 351–378.

Modood, T. (2005). The Educational Attainments of Ethnic Minorities in Britain. In G. Loury, T. Modood, & S. Teles (Eds.), Ethnicity, Social Mobility and Public Policy: Comparing the US and the UK . Cambridge University Press.

Modood, T., & Shiner, M. (1994). Ethnic Minorities and Higher Education – Why Are There Different Rates of Entry? PSI.

Oakley, A. (1974). Housewife . Penguin.

ODPM. (2006). The State of the English Cities Report . ODPM Publications.

Platt, L. (2011). Understanding Inequalities Stratification and Difference . Polity Press.

Purvis, J. (1989). Hard Lessons. The Lives and Education of Working-Class Women in Nineteenth-Century England . Polity Press.

Richardson, J. (2008). The Attainment of Ethnic Minority Students in UK Higher Education. Studies in Higher Education, 33 (1), 33–48.

Robinson, J. (2010). Bluestockings the Remarkable Story of the First Women to Fight for an Education . Penguin.

Rupp, L. (2001). Feminist Movements. In N. Smelser & P. Baltes (Eds.), International Encyclopedia of the Social and Behavioral Sciences (Vol. 8). Elsevier Science.

Shain, F. (2020). Navigating the Unequal Education Space in Post-9/11 England: British Muslim Girls Talk About Their Educational Aspirations and Future Expectations. Educational Philosophy and Theory . https://doi.org/10.1080/00131857.2020.1769604

The Change Institute. (2009). The Pakistani Muslim Community in England Understanding Muslim Ethnic Communities . Queen’s Printer and Controller of Her Majesty’s Stationery Office.

Weiner, G., & Arnot, M. (1987). Gender Under Scrutiny: New Inquiries in Education . Hutchinson in association with The Open University.

Whittier, N. (2007). Social Movements and Gender. In G. Ritzer (Ed.), The Blackwell Encyclopedia of Sociology Volume IV F-HE . Blackwell Publishing.

Download references

Author information

Authors and affiliations.

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Aqsa Saeed .

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this chapter

Saeed, A. (2022). Educational Inequalities in the UK: Gender, Ethnicity and Social Class. In: Education, Aspiration and Upward Social Mobility. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-030-82261-3_2

Download citation

DOI : https://doi.org/10.1007/978-3-030-82261-3_2

Published : 28 February 2022

Publisher Name : Palgrave Macmillan, Cham

Print ISBN : 978-3-030-82260-6

Online ISBN : 978-3-030-82261-3

eBook Packages : Education Education (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Live revision! Join us for our free exam revision livestreams Watch now →

Reference Library

Collections

• See what's new
• All Resources
• Student Resources
• Assessment Resources
• Teaching Resources
• CPD Courses
• Livestreams

Study notes, videos, interactive activities and more!

Sociology news, insights and enrichment

Currated collections of free resources

Browse resources by topic

• All Sociology Resources

Resource Selections

Currated lists of resources

• Study Notes

Differential Education Achievement by Ethnicity - Statistics

Last updated 26 Nov 2019

• Share on Facebook
• Share on Twitter
• Share by Email

Differential educational achievement by ethnicity refers to the fact that pupils from some ethnic backgrounds perform better in school than others.

While some sociologists point to cultural and material deprivation – outside school factors – to explain these differences, others look to processes inside the school such as labelling and institutional racism.

Black pupils statistically underperform in school while pupils of Indian or Chinese heritage often “over-perform”. However, the picture is not straightforward. Female black pupils are more likely to go into higher education than girls from several other ethnicities (including white British) and Bangladeshi pupils achieve above the national average at GCSE but are among the groups least likely to go to university. As such, these statistics are unlikely to be able to be explained by one factor – like teacher racism – and a combination of factors are likely to be at play.

In terms of achieving 5 A*-C grades at GCSE, pupils from a Chinese heritage tend to perform best (74% in 2014), Indian next (73% in 2014) with white British trailing behind (56% in 2014, a little below the national average) and children from Pakistani backgrounds and African Caribbean behinds further behind still (51.4% and 47% respectively in 2014). The lowest performing ethnic groups are Irish traveller and Roma/Gypsy groups (at 14% and 8% respectively).

• Differential Educational Achievement

You might also like

Example answer for question 2 paper 1: a level sociology, june 2017 (aqa).

Exam Support

Educational Divide

10th January 2018

Learning in Syria

27th April 2018

School Segregation

21st January 2019

Bowles and Gintis on Education

Topic Videos

Differential Educational Achievement by Ethnicity - Parental Support

The international students struggling to feed themselves in lockdown.

11th February 2021

Quick Revise - Education and Attainment | AQA GCSE Sociology Revision Blast Video (Paper 1)

Our subjects.

• › Criminology
• › Economics
• › Geography
• › Health & Social Care
• › Psychology
• › Sociology
• › Teaching & learning resources
• › Student revision workshops
• › Online student courses
• › CPD for teachers
• › Livestreams
• › Teaching jobs

Boston House, 214 High Street, Boston Spa, West Yorkshire, LS23 6AD Tel: 01937 848885

• › Contact us
• › Terms of use
• › Privacy & cookies

© 2002-2024 Tutor2u Limited. Company Reg no: 04489574. VAT reg no 816865400.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• HHS Author Manuscripts

Racial/Ethnic Differences in the Educational Expectations of Adolescents: Does Pursuing Higher Education Mean Something Different to Latino Students Compared to White and Black Students?

Viana y. turcios-cotto.

1 Department of Psychology, University of Connecticut, Storrs, CT

Stephanie Milan

There are striking disparities in the academic achievement of American youth, with Latino students being a particularly vulnerable population. Adolescents’ academic expectations have been shown to predict educational outcomes, and thus are an important factor in understanding educational disparities. This article examines racial/ethnic differences in the future expectations of adolescents, with a particular focus on how expectations about higher education may differ in frequency and meaning for Latino youth. Participants included 375 urban ninth-grade students (49% Latino, 23% White, 22% Black, and 6% other; 51% female) who gave written descriptions of how they pictured their lives in five years. Responses were subsequently coded for content and themes. Results demonstrate that Latino youth were less likely to picture themselves attending college when compared to Black and White youth, and more likely to hold social goals, such as starting their own family. Ethnic/racial differences also were found in the themes present in responses, with Latino and Black students more likely than White students to describe individuation and materialistic goals, and to give more unrealistic responses. For Latino youth only, higher education goals were associated significantly with individuation themes. In addition, for Latino youth, adolescents who wished to pursue higher education reported more depressive symptoms and emotional distress than those who did not picture going to college, whereas the opposite pattern was evident for Black and White youth. These differences may reflect cultural values, such as familismo . Practice implications include the importance of culturally tailoring programs aimed at promoting higher education.

Introduction

There are striking disparities in the academic achievement of American youth. In 2008, approximately 81% of White students graduated from high school within four years, or on time, compared to only 62% of Black students, and 64% of Latino students. Of students who graduated, 72% of White, 56% of Black, and 64% of Latino students immediately enrolled in some form of higher education ( NCES, 2010a ; NCES, 2010b ). Among students who did go on to post-secondary education, students of color were more likely to attend junior or community college and less likely to complete their degrees ( NCES, 2010b ). Such disparities have significant consequences in the U.S., where academic achievement is strongly linked to lifelong income and socioeconomic status ( NCES, 2010a ). Lower SES is associated, in turn, with worse physical and mental health and more difficult life circumstances (e.g., increased exposure to violence) throughout the adult years. Given these serious consequences, identifying and addressing the factors that contribute racial and ethnic disparities in academic achievement is critical.

Historically, most research on educational disparities focused on Black-White comparisons. In the past decade, however, national studies on educational outcomes have provided statistics on White, Black, and Latino youth to better reflect the American demographic composition. During this time, an increasing number of studies have documented the structural reasons for lower educational achievement among Latino youth ( Flores-Gonzalez, 1999 ; Nieto, 2005 ). In contrast, less empirical attention has been given to cultural and psychological factors that may contribute to lower educational achievement among Latinos. Given that Latinos are the fastest growing minority group in the U.S. and that there are important cultural differences between Latinos and Blacks, more research in this area is needed. The goal of this article is to understand better one potential psychological factor that may contribute to higher education achievement among Latino youth, namely adolescents’ future educational expectations.

Structural, cultural, and individual factors associated with educational disparities

Educational disparities are the result of structural, cultural, and individual factors. Structural factors influencing Latino youth are largely the result of poverty, and include attendance of schools that have higher student-to teacher ratios. Overcrowding reduces the amount and quality of resources available to these already underprivileged students ( Flores-Gonzalez, 1999 ; Ready, Lee, & Welner, 2004 ; Rivera-Batiz & Marti, 1995 ). Additionally, inequalities in educational funding perpetuate the lack of resources (e.g., textbooks, extracurricular activities) for the neediest children, further contributing to their academic underachievement ( Beese & Liang, 2010 ; Carey, 2004 ). Latinos also face challenges associated with bilingual education. In a time when Latinos are the fastest growing minority group in the United States, bilingual education is being eradicated from public schools at a feverish pace ( Nieto, 2005 ).

The classroom curriculum and atmosphere also may contribute to less engagement in youth of color. For example, the curriculum in most school systems may be less personally relevant to many students of color, as it often acknowledges the histories and prominence of White individuals throughout the school year ( Nieto, 2000 ). American teachers in American schools and families from Latin American backgrounds who attend these schools also may hold differing beliefs and interpretations of events in the world ( Greenfield, Keller, Fuligni, & Maynard, 2003 ). Similarly, cultural differences in pedagogy preferences may lead to the disengagement of Latino youth. For example, there is evidence of racial/ethnic differences in intergroup instruction, with Latino children more likely to participate in dyadic and triadic processes in attempts to learn about different activities and roles at home than White children ( Rogoff et al., 2007 ). This is in stark contrast to a typical public school, in which students are expected to work independently and quietly for large portions of the day.

Cultural values such as familismo and ser buen educado also may impact educational attainment among Latinos. In many Latin American communities, familismo is a firm belief in strong family ties, with the family as the primary source of support and loyalty to the family taking precedence over one’s personal desires ( Halgunseth, Ispa, & Rudy, 2006 ; Negy & Woods, 1992 ). This belief has implications for children leaving the home as adults, especially when it is time to attend college. Though the majority of Latino youth and their parents believe a college education is important for a successful future, many youth report the need to support their family as the reason for not continuing with their education ( Pew Hispanic Center, 2009 ). Latino students also may have difficulty adjusting to life away from family, on whom they have depended as their primary source of support ( Kenny & Stryker, 1996 ).

Although ser buen educado can be literally translated to “be well educated,” the Spanish phrase carries much greater meaning than that. Ser buen educado not only means to have a good formal education, but also means to be respectful, well mannered, and have high morals ( Valdes, 1996 ; Halgunseth et al., 2006 ). This understanding of education implicates a much more social definition than thought of in English. This may lead to Latino children and adolescents formulating much more social goals rather than academic goals, especially goals around family life, whether it be assisting in their current family or starting a family of their own.

Finally, there are individual reasons for differences seen in academic achievement independent of differences in actual ability. Indeed, many of the structural and cultural factors previously discussed likely contribute to achievement disparities through their impact on individual psychological factors. Several researchers have examined psychological factors that may influence academic motivation and achievement in students of color, including stereotype threat ( Steele & Aronson, 1995 ; Sackett et al., 2004), fear of “acting white” ( Fordham & Ogbu, 1986 ), and future expectations ( Cunningham, Corprew, & Becker, 2009 ; Oyerserman, Bybee, & Terry, 2006). Stereotype threat refers to the theory that one’s underperformance may stem from the worry of being judged or treated negatively due to the stereotypes of one’s group or fulfilling those stereotypes ( Steele & Aronson, 1995 ). When “acting White” one is perceived as denying their minority identity by behaving in ways that are congruent with the majority culture, which often includes studying in the library and getting good grades ( Fordham & Ogbu, 1986 ). Unlike stereotype threat and “acting white,” which are theories directly related to race, future expectations refers to one’s belief of who one will become and what might occur in the future. In general, stereotype threat and fear of “acting white” have been viewed as potential barriers to achievement among Black youth. In contrast, the concept of future expectations has been applied more widely, although research on future expectations as related to educational achievement among Latinos is still in its early stages.

The concept of future expectations has been studied as an aspect of adolescent’s possible selves , the future oriented aspect of one’s self-concept that is believed to serve as a motivational source and guide to behavior ( Markus & Nurius, 1986 ). Possible selves measures typically involve having adolescents describe what they hope for and fear they will be like in the future using open-ended questions that are later coded for content (see Oyserman & Fryberg, 2006 for a review). Through one’s possible selves, people are able to project who they would like to be (and not be) in the future, which is presumed to create a goal to work towards. Consistent with this assumption, adolescents’ possible selves do predict increases in grade point average ( Anderman, Anderman, & Griesinger, 1999 ). Similarly, students’ educational aspirations in 9 th and 12 th grade are correlated positively to their academic outcomes during early adulthood (Crockett & Beal, 2012). Although these studies suggest that having educational future goals is linked to actual achievement, many students are not likely to reach their desired goals without proper planning and interventions that provide them the appropriate strategies ( Oyserman, Bybee, Terry, & Hart-Johnson, 2004 ).

Future expectations and the educational expectations of Latinos

Possible selves research has shown group differences in the number of adolescents reporting educational or occupational goals, with White students reporting higher education expectations more often than students of color ( Anderman et al., 1999 ; see Oyserman & Fryberg, 2006 for a review). Studies including only Latino adolescents have found that youth report academic future expectations more than any other type ( Cansler, Updegraff, & Simpkins, 2012 ; Yowell, 2000 , 2002 ); however, these studies only examined Latinos and thus do not speak to group differences. In a review of the possible selves literature, Oyserman and Fryberg (2006) concluded that Latino youth appear less likely to hold educational future expectations than youth from other backgrounds, although comparative studies are still limited.

Although these studies are informative, they do not provide insight into the context or potential meaning of higher education future goals, and how these may differ by race/ethnicity because of cultural influences. For example, the cultural values of familismo or ser buen educado may lead Latino adolescents to hold more social future goals when compared to youth from other backgrounds, such as starting a family or having a romantic partner. These goals may indirectly impact educational achievement (e.g., starting a family may inhibit educational attainment). Thus, it is important to examine not only group differences in future educational goals, but also differences in the types of goals given in addition to or instead of higher education.

Relatedly, youth may have the same goals for the future, but have different reasons for that goal. For example, one student may want to go to college in order to help out his family whereas another student may want to go to college to be out on her own and have independence. Research on future expectations has not examined themes or reasoning that also may be part of responses (e.g., individuation, family assistance, materialism). These are important themes during the adolescent years, and they may lead youth of different backgrounds to navigate this developmental period in specific ways that reflect cultural values or norms ( Greenfield et al., 2003 ). For example, Latino adolescents report more family obligation than White youth ( Fuligni, Tseng, & Lam, 1999 ), which may in turn influence decisions they make about pursuing higher education ( Suarez-Orozco & Suarez-Orozco, 1995 ) and employment ( Fuligni & Pederson, 2002 ). Although generally not studied, the reasons why a student wants to be in higher education or in a certain job in the future likely impacts whether he or she reaches this goal. Moreover, programs aimed at promoting higher education among diverse youth are likely to be more effective if underlying values associated with different future goals are incorporated.

Another important consideration of the content of adolescent’s future expectations is how realistic the envisioned self is in the projected timeline. A high school student may picture himself in medical school working to be a doctor in a year’s time. If this student is only in high school, however, this expectation is not realistic. Students who live in families and communities where few adults have gone to college may have a less clear understanding of what college involves, even if they envision themselves pursuing higher education ( Abraham, Lujan, López, & Walker, 2002 ; Destin & Oyserman, 2009 ; Oyserman, Johnson, & James, 2011 ). These types of inaccuracies may impact the likelihood of achieving a stated future goal. Most studies, however, have not examined whether the type of educational or occupational future expectation described was—in fact—attainable or realistic within the allotted period.

Finally, the underlying assumption in much of the future expectations literature as applied to educational achievement is that having an academic future expectation is desirable and positive. However, this may not always be the case for students of color. In explaining their possible selves theory, Markus and Nurius (1986) proposed that one’s future expectations may contribute to psychological adjustment as these expectations have positive (e.g., graduating, having money, etc.) or negative (e.g., being a loser, doing drugs) valences. More importantly, these goals may be viewed as unattainable or unavoidable, which changes the emotional implications that the goals may carry. For example, if an adolescent wishes to attend college but is unsure of the application process, is unable to afford college, or does not feel supported by family members in this endeavor, postsecondary education may then become a stressful, saddening, or anxiety-provoking goal. Similarly, given different cultural values, higher education goals may have different implications for psychological well-being for adolescents of different racial/ethnic groups ( Fuligni & Pederson, 2002 ).

Much also has been written about the potential negative consequences of academic achievement among Black youth (e.g., “acting white”; Fordham & Ogbu, 1986 ; Fryer & Torelli, 2006 ). The basis for this assumption is that Black youth who are high achievers in school face negative peer consequences, including being accused of trying to be better than others or “acting white.” There has been debate about whether this process actually occurs in schools, and it may be only in integrated schools in which there are negative consequences of achievement for students of color ( Fryer & Torelli, 2006 ). Nonetheless, this line of inquiry highlights how there may be negative consequences for youth of color when they act in ways that seem inconsistent among family or peers. For example, Black students accused of “acting white” report greater symptoms of anxiety than peers who have not faced the same accusations ( Murray, Neal-Barnett, Demmings, & Stadulis, 2012 ).

Among Latino families, there is evidence that having more individualistic goals than parents may be a source of acculturative stress and family conflict for adolescents ( Hwang & Wood, 2009 ; Lau, McCabe, & Yeh, 2005 ). Familismo may compel Latino youth to feel torn between helping their family financially versus exercising their autonomy by attending college, as many Latino youth report desires to attend college, as well as the need to support their family as the reason for not doing so ( Pew Hispanic Center, 2009 ). Additionally, the need to be respectful to elders, an aspect of ser buen educado , may lead adolescents to do what their family thinks is best, rather than pursue their own goals. If this is the case, then having academic future goals may be a source of stress for Latino youth if it creates conflict with parents or family members, who may have different goals for their children. Thus, we also examine whether having academic future goals is differentially associated with emotional distress or mental health among Latino adolescents compared to their Black and White peers.

Current Study

The goal of this study is to provide a richer contextual picture of future expectations as they may relate to educational achievement, particularly among Latinos. Specifically, this study builds on existing findings by addressing three research questions.

First, do future expectations differ by race/ethnicity? In particular, do Latino students differ from White and Black youth in holding higher education future expectations or in types of future goals (e.g. social, starting family) that may influence academic achievement? Although there is evidence that youth of color hold fewer academic expectations than White youth, fewer studies have compared Latinos with both White and Black youth from similar economic backgrounds. There is also reason to believe that Latino youth may have more social future goals because of cultural values such as familismo and ser buen educado . Thus, we examine not only educational goals but also alternative goals that may be particularly important among Latino adolescents.

Second, do adolescents’ future expectation responses differ in theme (e.g., family assistance, individuation, materialism, negativistic) and how realistic they are by race/ethnicity, particularly in regards to higher education? Latino adolescents report more family obligation than youth from other backgrounds (e.g., Fuligni et al., 1999 ); however, less is known about how developmental themes such as family obligation and individuation may be tied to future educational goals. Similarly, students from different backgrounds may hold the same broad future expectation (e.g., higher education), but differ in what they believe pursuing this goal involves.

Third, how do higher education future expectations relate to emotional adjustment, and does this differ by race/ethnicity? Higher educational goals are generally viewed as positive pursuits for adolescents. The meaning of this goal is likely shaped by cultural, social, and economic contextual factors. Consequently, the relationship between higher education goals and psychological well-being may be moderated by race/ethnicity.

Participants and Procedure

Participants for this study were drawn from a larger study (n=537) of ninth grade students attending a large, diverse urban high school in a low income, central Connecticut city. The larger sample reflected about 84% of all eligible 9 th grade students. In the school population, the majority of the student body comes from low-income households and 79% qualify for free/reduced-price lunch. About 40% of the students have a non-English speaking home and 10% participate in bilingual education and ESL services. A large group of students are from immigrant families, with the majority of these newcomers being from Latin America and Poland. The current sample was comprised of the 375 ninth graders who answered the future expectation question, the item of interest for this study. Because of an unscheduled early dismissal, a subset of students were not able complete both days of the survey. Students with and without the future expectation survey item did not differ on any demographic factors.

Participants were 14 to 17 years old with a mean age of 14.92 (SD = 0.65) and an equal proportion of boys (n=185) and girls (n=190). The sample was 49% Latino (88% Puerto Rican, 5% Mexican, 4% Central American, 2% Dominican), 23% White (30% 1 st or 2 nd generation Polish), 22% Black (20% 1 st or 2 nd generation Jamaican, 15% 1 st or 2 nd generation African) and 6% other. The “other” category included adolescents who identified as Arabic, Asian, and multiracial. Genders were represented equally within each racial/ethnic group. Students included in the current sample broadly matched their high school population which, according to school demographic data, includes 45% Latino, 30% White, and 20% Black students. About 85% of students reported living with their biological mother and 48% reported living with their biological father. Based on student reports, 18% of the students had mothers with less than a high school degree, 53% with a high school degree, and 16% with a bachelor’s degree; 13% of the students reported not knowing. For fathers, 21% of the students had fathers without a high school degree, 45% with a high school degree, and 10% with a bachelor’s degree; 24% reported not knowing.

The grade-wide survey was conducted in conjunction with the school-based health center (SBHC) at the high school, which serves as primary care provider to 70% of the student population. The purpose of the survey was to obtain information about health knowledge and behaviors in order to refine the health class curriculum and to develop more targeted health promotion services within the school. Data were collected through anonymous in-class surveys administered during all of the regularly scheduled health classes, a required course for ninth grade students and one in which students in ESL and special education participate. After receiving a verbal and written description of the purpose of the study, as well as the voluntary and anonymous nature of the study, students were given time to ask questions. Students then signed a written consent form, which was collected separately from de-identified surveys. Students were given the option of completing alternative Health curriculum materials during the class period, although no students did so. The survey included various questions on health knowledge and behaviors, emotional stress, future goals, and demographic information. Because the survey was anonymous and collected for the purpose of curriculum development by school personnel, parental consent was not required. However, parents were informed that students would be participating in an anonymous survey in health class and given a choice to refuse their child’s participation. No parents did so. The school administration and University of Connecticut’s Institutional Review Board approved the study design and procedures.

Future Expectations Question

Students responded to the open-ended future expectation question, “Picture what you would like to be doing 5 YEARS FROM NOW. Please write a few sentences about what you think your life will be like at that time.” Responses were later typed verbatim into a separate file, and then coded by two trained, independent research assistants blind to all other information. Content codes previously used in possible selves studies include school achievement, interpersonal relationships, jobs, material goods and negative selves such as “homeless,” “junkie,” etc. ( Oyserman et al., 2004 ; Oyserman & Markus, 1990 ; Oyserman & Saltz, 1993 ). In the current study, codes were based on these previous content codes and were also made more explicit. For example, in this study the broad category of interpersonal relationships was separated into 3 categories (romantic, own family, and social) to gather more detailed information about this social domain. Moreover, additional codes (individuation, family assistance) were added a priori to reflect themes or tones in the response (see Table 1 ). These categories were deemed relevant to adolescent development given the centrality of seeking autonomy during this period, coupled with the idea that families of color may navigate this developmental period differently due to cultural differences ( Greenfield, 2003 ; Fuligni et al., 1999 ). Lastly, an unrealistic category was used to examine students’ responses and understandings of what might be possible to accomplish in five years, considering research that shows adolescents from socioeconomically disadvantaged neighborhoods might not have a good understanding of paths to achieving future goals ( Oyserman et al., 2011 ).

Operational Definitions of Coded Content and Themes

Responses were coded into area categories, which were not mutually exclusive, including: Academic, Higher Education, Work, Specific Career, Romantic, Own Family, and Social. Students’ responses could receive more than one content code (e.g., “I will be going to college and working part time”). They were also coded for themes including: Individuation, Family Assistance, Materialistic, and Negativistic. Theme codes were given only when the participant explicitly responded in a way that expressed that theme (e.g., “I will be making lots of money and driving a Mercedes”; “I plan to be in a good job so I can help out my family”). In addition, all responses were also coded on a 0–2 scale for how realistic they were for the given amount of time of “5 years from now”, with 0 being realistic (e.g. freshman in college), 1 being unlikely but not impossible (e.g. starting point guard for Duke University), and 2 being unrealistic/impossible given the time frame (e.g., working as a pediatrician). Given the low number of 2’s and to increase reliability across coders, responses of 1 and 2 were collapsed into one unrealistic category. As shown in Table 1 , coding yielded good to high Kappas values for most categories, with moderate values for family assistance ( Landis & Koch, 1977 ). Disagreements for all codes were resolved by a consensus meeting between two doctoral students and a Ph.D. faculty not involved in the first coding.

Depressive Symptoms

Using criteria from the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV; APA, 1994 ), the Adolescent Psychopathology Scale (APS; Reynolds, 1998 ) is a self-report measure that evaluates a broad array of psychological disorders and distress in adolescents 12–19 years old. Questions measuring suicidality were excluded at the request of the school. Twelve items (= .916) were used, with a response format that included three options for evaluating their frequency of occurrence over the past two weeks. Answer choices were 1 = almost never, 2 = sometimes, and 3 = nearly every day. Mean scores were calculated for each participant with higher scores indicating a greater amount of depressive symptoms. Extensive evidence supporting the measure’s reliability and validity has been reported ( Reynolds, 1998 ).

Emotional Distress

A widely used self-report measure, the Brief Symptom Inventory -S (BSI-S; Derogatis, 1993 ) screens psychological symptom patterns of mental health difficulties in three domains: Anxiety, Depression, and Hostility. Eighteen items (= .934) from these scales were aggregated to reflect emotional distress (and reflect the BSI-S scale). The items were rated on a scale from 1 = not at all to 5 = extremely, indicating how much the adolescent felt bothered by the problem in the past 7 days including that day. Mean scores were calculated for each participant with higher scores indicating a greater amount of emotional distress. The BSI has also demonstrated sound psychometric properties ( Derogatis, 1993 ).

Demographic Characteristics

Students self reported on race/ethnicity and country-of-origin for themselves and their parents. They also reported on languages spoken in the household, family structure, and parental education level.

Data Analysis

To address RQ1 and RQ2 about racial/ethnic differences in the content and themes of future expectations, chi-square tests were conducted to examine overall racial/ethnic differences in the 7 categories of future goals and the 4 categories of themes/tones. In categories that differed in chi-square analysis, logistic regression was used to explore specific racial/ethnic group differences controlling for demographic factors. To examine whether gender moderated effects, logistic regression was used to test for race/ethnicity and gender interactions in the likelihood of each type of future goal following Jaccard (2001) . To address RQ3, how higher education future expectations relate to emotional adjustment by race/ethnicity, a Multivariate ANCOVA was used with the three categories of race/ethnicity (Latino, Black, White) and two levels of higher education future expectation (yes, no) as independent variables. Dependent variables included the APS Major Depression Subscale and the BSI emotional distress scores, again with demographic factors as covariates. Follow-up univariate F tests were conducted to determine the nature of significant MANCOVA results.

Preliminary analyses examined the frequency of responses for each category of coded future goals. Across the sample, the most commonly cited future expectations were academic, further education, career, and work. Gender differences were found for romantic and own family future expectations and mentioning individuation as a theme. Specifically, girls were more likely than boys to endorse having their own children and family (13% vs. 5%, χ 2 (1, n = 360) = 6.96, p<.05) and being in a romantic relationship (16% vs. 9%, χ 2 (1, n = 360) = 3.73, p<.05). Girls also were much more likely to specifically express a desire to move away from their current family as part of their response, regardless of whether they indicated this happening through education, occupation, or starting their own family (22% vs. 13%, χ 2 (1, n = 360) = 5.28, p<.05).

Besides gender, only maternal education level was associated with any future expectation. Since no other demographic factors (e.g., parental education, parent marital status, language, family size) were related to outcomes, only gender and maternal education were used as covariates in subsequent analysis.

Racial/Ethnic Differences in Future Expectations Content

The next set of analyses examined racial/ethnic differences in future expectations. Table 2 presents the frequencies of endorsement of each category by race/ethnicity and results from chi-square analyses of overall group difference. Logistic regression was then used to explore specific group differences controlling for parent educational level and gender. Within logistic regression analyses, race/ethnicity was coded so that the White and Black subsamples were compared to the Latino group. In terms of specific types of goals, racial/ethnic differences were found for higher education and own family goals. Specifically, Latinos were less likely to hold higher education expectations than Black students (57.2% vs. 72.8%, AOR=0.55 [0.31–0.99]) and White students (57.2% and 80.4%, AOR=0.34 [0.19–0.63]). For own family future expectations, a greater percentage of Latino students mentioned having children and their own family in five years than did White students at 13.3% and 4.1% (AOR=3.42 [1.14–10.28]), respectively. However, Latinos did not differ significantly than Blacks in this category (13.3% vs. 4.9%, AOR=2.54 [0.84–7.72]), although their rate of endorsement was over twice as high.

Frequency of Responses for Each Category of Coded Possible Selves by Race/Ethnicity

Note: Percentages do not add up to 100% as participants could receive a code for more than one category.

Because Latinos were least likely to report expecting to be in school in five years and most likely to report having their own family, follow-up analyses were conducted to examine if Latino students gave a goal of having their own family as an alternative to higher education, or in addition to higher education. Among Latino youth who did not endorse higher education, 21% stated having an own family future expectation; in contrast, only 8% of Latino students who endorsed expectations for higher education also expressed a desire to have their own family in five years χ 2 (1, n = 180) = 6.32, p=.01. This finding suggests that having your own family is more often viewed as an alternative rather than co-occurring expectation among Latinos.

As a follow-up analysis, we also tested whether contents differed within the Latino subsample by proxies of acculturation status, including nativity and language spoken at home. No within group differences were found.

Racial/Ethnic Differences in Future Expectation Themes

In addition to differences in content of future expectations, we also examined differences in themes (i.e., individuation, family assistance, materialism, negativistic) and how realistic goals were. Thirty-four percent of participants had a response that was coded for some theme. Latino (39%) and Black (34%) students had themes in their responses more often than White students (23%), χ 2 (2, n = 360) = 7.16, p=.01. As shown in Table 2 , racial/ethnic differences were seen in individuation and materialistic themes. Latino adolescents endorsed individuation goals more often than did White adolescents (20.4% vs. 9.3%, AOR=2.55 [1.17–5.57]), but not more than did Black adolescents (20.4% vs. 21.0%, AOR=0.94 [0.49–1.82]). This same pattern was exhibited with the materialistic theme. More Latino students stated materialistic goals than White students at 10.5% and 1.1% (AOR=10.99 [1.44–84.09]), respectively; however, Latino and Black adolescents did not differ significantly (10.5% vs. 7.8%, AOR=1.43 [0.54–3.84]). These results suggest similarities between Latino and Black students in individual and materialistic future goals, with both of these two groups giving responses that included these themes more than White students. No differences were found in family assistance or negativistic tones in responses. In general, these themes were uncommon.

As a post-hoc exploratory analysis, the association between endorsement of higher education future expectation and each of the four themes within each racial/ethnic group was examined. Specifically, we examined whether higher education was more closely associated with individuation, family assistance, materialism, or negativistic themes for a particular racial/ethnic group to explore the possibility that higher education goals mean something different to Latino youth. The differing sample sizes limit power to detect differences within the three racial/ethnic subsamples; however, effect sizes were considered in interpretation. Results are presented in Table 3 .

Percent students with specific themes by higher education goals and race/ethnicity

Higher education was not associated with family assistance or negativistic themes for any racial/ethnic group, although the low rate of these themes limits interpretation. Higher education goals were associated with materialism and individualism. Across all groups, youth who endorsed higher education goals were less likely to express materialistic themes than youth who did not endorse higher education goals. This difference was significant for Latino and Black youth. Among those who did not report higher education goals, 20.8% of Latino and 20% of Black youth had materialistic themes in their responses. In contrast, among those who did report higher education goals, 3.1% of Latino and 3.0% of Black youth had materialism themes. In other words, among students of color, youth who did not have higher education goals were more likely to explicitly state materialistic future goals in their response.

Higher education expectations also were associated significantly with individuation themes, but only for Latino students. Latino students who held a higher education expectation were more likely than those who did not have higher education goals to explicitly express individuation goals, χ 2 (1, n = 180) = 4.72, p <.05. It is worth noting that, although Black and Latino youth were equally likely to express individuation desires, individuation was related differentially to higher education for the two groups. Black youth with and without higher education expectations were equally likely to express individuation themes in their response (20% vs. 23%). In contrast, among Latinos, youth with higher education goals were twice as likely to express individuation desires as youth without higher education goals (26% vs. 13%). Across all youth, Latinos who imagined higher education in their future were the most likely to explicitly describe individuation within their response, suggesting that the desire to attend college may mean something different to Latino youth compared to other adolescents.

In addition to developmental themes, responses were coded for how realistic or attainable they were given the stated timeline. Unrealistic future goals were fairly common, with 28% of students giving a response that was deemed at least somewhat unrealistic. Latino students did not differ significantly in giving unrealistic responses than Black students (31.3% vs. 32.5%, AOR=0.93 [0.53–1.64]), but were more likely to respond with an unrealistic goal than White youth (31.3% vs. 16.5%, AOR=2.26 [1.21–4.21]). Among youth who gave higher education goals, 15% had responses that were coded as unrealistic. This rate did not differ by race/ethnicity. Among all the content codes, unrealistic codes were most often given to responses focusing on career goals, with 45% of these responses judged as being unrealistic. Among youth who stated a specific career in their future expectation, 52% of Latino, 50% of Black, and 22% of White students gave goals judged as unrealistic, χ 2 (2, n = 161) = 9.53, p <.01.

Racial/Ethnic Differences in Academic Future Expectations and Emotional Adjustment

Although having higher education goals is generally thought to be associated with better adjustment, this may not be the case for different racial/ethnic groups if educational goals have different meanings (e.g., separation from families). The final set of analyses examined whether the relationship between higher education future expectations and emotional adjustment differed by race/ethnicity. MANCOVA was used to test this possibility, with race/ethnicity and academic future expectations used as independent variables and emotional adjustment variables as dependent variables, controlling for parents’ educational level and gender. There were no significant main effects of race [ F =.91 (4, 654) p=.46] or higher education future expectations [ F =1.42 (2, 327) p=.24]; however, there was a significant interaction [F=3.67 (4, 654) p=.006]. As shown in Table 4 , the interaction was evident for both the BSI emotional distress [F=7.32 (2, 328) p=.001] and APS depression [F=4.09 (2, 328) p=.018]. Specifically, having a higher education goal was related to having a more positive emotional adjustment for Black and White youth when compared to youth who did not have higher education goals. In contrast, Latino youth who endorsed a higher education goal demonstrated worse emotional adjustment than Latino youth without stated higher education goals.

Emotional Adjustment Differences by Race With and Without Higher Education Possible Selves

This study examined racial/ethnic differences in the types of future expectations held by high school freshmen projecting five years into the future. Previous studies have shown that students of color are less likely than White students to endorse higher education goals, and this may be one factor that contributes to lower educational achievement among students of color ( Anderman et al., 1999 ). This study extends existing research on future expectations in youth of color by focusing specifically on Latinos and providing a more complete picture of aspects of academic future expectations that may interfere with or promote educational attainment. More specifically, this study extends existing findings in four ways.

First, instead of looking only at education and employment related expectations, we also examined another domain of future expectations: socially oriented goals (e.g., getting married or living with a romantic partner, having children, being with friends, etc.). Given evidence of an increased interdependent focus among Latinos, this additional type of future expectation may be especially relevant among Latino youth. Second, we examined various themes evident within future expectation responses. In particular, we examined whether adolescents’ responses included themes like autonomy strivings, family assistance, uncertainty, materialism, or negativity based on the assumption that adolescents of different racial and ethnic groups may endorse the same type of goal (e.g., “I will be in college”) but for different reasons (e.g., “I will be in college so I can be on my own” vs. “I will be in college so I can get a job to help my family out”). To date, however, the notion of different themes associated with future expectations has not been examined. Finally, we examined whether having an education related goal is differentially linked to emotional adjustment for Latino, Black, and White students. While focusing on higher education is generally assumed to be a positive indicator of adjustment, this may not be true for youth of all racial and ethnic groups; however, this possibility has not been explored in the future expectations literature.

Consistent with previous research ( Oyserman et al., 2004 ; Shepard & Marshall, 1999), the content of the adolescents’ future expectations centered on educational and work-oriented aspirations, regardless of race. The majority of students in each racial/ethnic category did identify an academic future goal and a higher education future goal. However, significant differences by race/ethnicity were found in that Latinos held such expectations at lower rates than did Black and White students. Given previous literature ( Oyserman & Fryberg, 2006 ), this difference between Latino and White students was expected. However, it is notable that, even among students of color, Latino students were less likely than Black students to endorse higher education expectations. In this sample, both groups faced similar socioeconomic challenges, yet fewer Latino students envisioned education in their futures. Indeed, in Connecticut, Latino students fare worse than Black students in regards to high school graduation rates with 54% versus 64%, respectively, graduating within 4 years in 2008 ( EPE Research Center, 2011 ). This data demonstrates that Latino students, and perhaps Puerto Rican students in particular, may be a vulnerable population in need of more tailored interventions to address educational disparities.

In contrast, no racial/ethnic differences were found in the work or specific career goals. This may mean that students equally picture a future that involves work or some kind of career, but that Latinos may be less likely to see themselves in an educational setting in pursuit of these work goals. Alternatively, these findings suggest that Latino students may hold other types of future expectations. Consistent with this latter idea, Latinos were more likely to endorse goals that were socially oriented. These socially-oriented goals may be more significant or more desired than educational goals or may compete with the ability to continue their education.

Latino youth endorsed having their own family in five years significantly more often than White and Black youth. Although only a small portion of Latino youth in general described having their own children and family as a future expectation, it was at far greater rates than did other youth (13.3% Latino vs. 4.9% Black and 4.1% White). Further, this pattern was found regardless of gender, showing that Latino boys also held their own family expectations at greater rates than did Black and White boys. This finding is congruent with national statistics demonstrating that, while overall teen birth rates decreased from 1981 to 2006, birth rates for Latina teens increased ( Wingo, Smith, Tevendale, & Ferré, 2011 ). Similarly, results from a Pew Hispanic Center study (2009) show that Latino youth reported younger ideal ages for having a first child compared to youth from other backgrounds. This difference could be due to specific cultural factors of Latinos, such as familismo , which emphasizes strong family ties and the importance of family. In this context, having a family may be seen as a highly regarded goal for the future, and a viable alternative to higher education as a path to adulthood. While the desire to have a family at a younger age may be rooted in positive cultural values, the link between early parenthood and chronic poverty is well-established.

Racial/ethnic Differences in Future Expectations Themes

In this study, different “themes” relevant to the adolescent period also were coded with the idea that students could give future goals with similar content (e.g., working) but associate different meaning to it (e.g., feeling negative or positive, doing it to be away from family or for materialistic goals, etc.). Four developmentally relevant themes were coded: individuation (i.e., autonomy strivings), family assistance, materialism (i.e., wanting nice things), and negativistic. Racial and ethnic differences emerged in individuation and materialism.

Latino and Black adolescents endorsed the individuation goal more often than did White adolescents. This finding was counterintuitive given that Latino and Black families are typically thought to have more collectivistic attitudes ( Triandis, 1989 ). However, although Latinos do seem to be higher on collectivism than Blacks and Whites, Black Americans actually have been found to be higher on individualism than Latinos and White Americans ( Oyserman, Coon, & Kemmelmeier, 2002 ). Further, thinking about the transition to adulthood might cue-up a more individualistic mindset for Black youth, leading to a greater endorsement of individuation ( Oyserman, 2011 ). Yet, values such as familismo , which may be reflective of collectivism in Latino culture, may be why Latino students stated an expectation or a hope of living on their own and being independent of their family. Seeking autonomy and individuality from parents is the hallmark of American adolescence and the transition to adulthood. Individuation may be a more salient construct for Latino youth if they regularly have received direct and indirect messages of the importance of staying with the family. In contrast, in White families, an adolescent’s developmentally typical desire for independence may not come into “conflict” with the desires of other family members, and thus be less salient an issue.

Although Latino and Black youth were equally likely to express individuation goals, there was one important difference. For Black youth, the desire for higher education was not associated with individuation goals, with 21% of youth who stated higher education goals and 20% of youth who did not state higher education goals including individuation themes in their response. In contrast, the desire for higher education and individuation were linked for Latino youth: Latino adolescents who expressed higher education goals were twice as likely as youth without higher education goals to include individuation in their response (26% vs. 13%). This finding implies that pursuing an education after high school may mean something different to Latino adolescents: going to college might also mean separating from family. The belief that higher education inevitably causes one to leave home may create conflict and distress for Latino families, especially when considering cultural values like familismo .

There were also racial/ethnic differences in materialistic themes, with Latino and Black youth again more likely to have responses with this type of theme present. This code was given when students specifically said something like “I’ll be rich and have a nice car” or “I’ll be living in a mansion”, or another response that specifically described a desire for material wealth and possessions. Although we controlled for some markers of socioeconomic differences (e.g., parental education) we did not have comprehensive information on family SES. In the larger community, families of color do have lower incomes than White families. Consequently, the desire for material wealth may be the result of a lack of material resources now. Interestingly, for both Latino and Black youth, materialistic responses were given more often by youth who did not also state expecting to go on to higher education. Thus, higher education may not be viewed as a path to acquire material goods for many adolescents of color. This may be reflective of youth of color’s understanding of how one earns large sums of money for these material goods. Perhaps they believe that other venues that may not require many years of education, such as playing professional sports, will allow for greater material goods ( Simons, 1997 ). Moreover, students with educational expectations may not be pursuing these goals for material gain, but instead for genuine interest in a specific academic field or career.

Finally, Latino and Black students were more likely than White students to give responses judged as potentially unrealistic, although these two groups did not differ. Most often, this code was given when students responded that in five years they would be in an established career in a field that requires years of study beyond college, such as a lawyer or pediatrician. These outcomes would be impossible in five years given that participants were freshmen in high school. This finding is consistent with other research demonstrating that low-income youth of color often lack the knowledge of what is involved in certain career goals, especially those that require advanced educations ( Abraham et al., 2002 ; Destin & Oyserman, 2009 ). Particularly among male students, unrealistic responses also often included athletic accomplishments unlikely for any 19 or 20 year old, such as being a “starting quarterback in the NFL”. While such goals may enhance school involvement through participation in school sports, they may be detrimental to youth of color in the long run ( Simons, 1997 ).

The final analyses demonstrated that there were racial/ethnic differences in the relationship between higher education future expectations and emotional adjustment, with Latino students displaying a different pattern than Black and White youth. Specifically, Black and White students who did not endorse a higher education future expectation reported more depressive symptoms and emotional distress than peers of the same race who did endorse higher education expectations. In contrast, Latino youth who did endorse a higher education possible self had greater levels of depressive symptoms and distress than those who did not. This finding implies that having educational aspirations like attending college might be a source of stress and emotional difficulty for some Latino adolescents in contrast to students of other racial/ethnic backgrounds.

Why might educational future goals be associated with more distress among Latino students? As described above, Latino youth with higher education goals were also significantly more likely to have individuation themes as part of their response when compared to Black and White students. This may mean that Latino students, moreso than other students, associate pursuing higher education as leaving home and family or being on your own. This interpretation of what it means to go to college may conflict with desires explicitly or implicitly expressed by parents for the adolescent to stay close to family. This may be one form of intergenerational acculturative stress that leads to emotional distress among Latino adolescents ( Hwang & Wood, 2009 ).

There also may be socio-economic reasons for the association between emotional distress and higher education goals for Latino youth. Latino youth are more likely to live in poverty than any other group ( Lopez & Velasco, 2011 ). Although analyses controlled for broad markers of SES, it is possible that Latino youth actually had fewer economic means or felt more impoverished than youth from other backgrounds in this sample. If so, then Latino students who want to go to college may be more likely to feel that it is not economically feasible, and thus become emotionally distressed at not being able to pursue their goals. Alternatively, some Latino youth may have limits on the ability to go to college because of other external factors, such as their immigration status (i.e., being from an undocumented family). Although the majority of Latino youth in this study were Puerto Rican, the influence of immigration policies on the educational aspirations and achievement of Latino youth is an important area for study.

Practice Implications

Results from this study have implications for educators and clinicians. When working with Latino students, educators need to involve families in discussions about future educational goals, particularly in considering cultural values such as familismo . Families and their children should be allowed to express their fears, concerns, and overall beliefs about what it would mean to them if the adolescent pursues higher education. Such conversations should include the topic of family planning, being supportive of students who desire to have children but encouraging them to do so after furthering their own education. Information on the benefits of higher education also should be discussed with a focus on all the possibilities for higher education, including local colleges/universities that allow students to commute while still living at home or to live on campus and frequently visit family if desired. Similarly, youth development programs can help low-income adolescents gain insight into a broader array of possible career paths and the intermediate steps necessary to reach these goals through mentoring, job shadowing, and apprentice programs.

In-school interventions in which students are directed to picture themselves in a successful academic context in the future have been shown to result in improvement in grades, fewer absences, greater concern about doing well in school, and fewer depressive symptoms ( Oyserman et al., 2002 ; Oyserman et al., 2006 ). Similarly, when used as a self-regulation tool with a specific action plan, having students picture an academic possible self helped low-income eighth graders improve their grades, increase time spent on homework and class participation, and decrease referrals to summer school ( Oyserman et al., 2004 ). Thus, programs in which youth are actively directed to envision an academic future appears to impact aspirations and behaviors. This type of intervention also may be beneficial with Latino youth; however, such efforts need to address simultaneously emotions and meanings that may be associated with higher education in this population. In particular, the pursuit of higher education may come at an emotional cost if this goal is in conflict with other personal, familial, or cultural goals. Mental health workers, especially those serving high school and college students, should consider these factors as potential reasons for distress exhibited by Latino high school students. Finally, it is important for professionals who work with Latinos to conceptualize familismo as a strength of families rather than a sign of enmeshment or family problems.

Limitations and Future Directions

Although this study has many strengths, several limitations should be noted. Like most previous research examining future expectations, this study was cross-sectional. Generalizeability may be limited because of the convenience sampling since all participants came from one high school in one city of Connecticut; however, it provides insight into racial/ethnic differences among youth living in similar economic situations. Also, it is important to note that Latinos in this sample were overwhelmingly Puerto Rican. Examining future expectations within groups of Latinos from differing countries of origin might produce different results. Moreover, a sizable minority of Black and White youth in this sample were in 1 st or 2 nd generation immigrant families, which may have impacted racial/ethnic group comparisons. Relatedly, there has been some evidence that educational aspirations may be associated with more distress in Asian-American youth, although this has been conceptualized as being the result of achievement pressure from family rather than individuation stress ( Qin, 2008 ). A study that looked at interactions between race/ethnicity and educational goals on emotional distress in an even more diverse sample of youth may provide a different picture. Additionally, although a strength of this study is the open-ended nature in which adolescents were allowed to express their goals, the percentage of students who responded with some theme was relatively low. Future studies with a more in-depth qualitative perspective on the meaning of higher education for different youth, in which students are asked to explicitly provide their reasons for certain goals, would offer additional insight into potential cultural differences. Also, though the Kappa value of family assistance was in the moderate range, the lack of greater interrater reliability may limit interpretation in this study. Finally, more longitudinal research should be conducted to see if and how adolescent future expectations are linked to actual adult outcomes and if that differs by race/ethnicity.

This study extends research on future expectations and academic disparities among Latinos in several ways. First, we examined goals beyond academic and career oriented, including more social goals that may be particularly important to diverse youth. Similarly, we coded for themes in addition to content, which can help determine if types of future expectations (e.g., education, work) have different meaning (e.g., as a means for individuation, family assistance, or material gain) for youth of different backgrounds. Finally, we also found that race/ethnicity moderated the link between educational future goals and emotional adjustment, highlighting the importance of considering the different meaning of higher education for diverse youth. In these three ways, this study provides a more contextualized understanding of future expectations and how these may impact subsequent educational achievement for Latinos. Educational disparities have a considerable cost to individuals, communities, and families. Understanding psychological factors that contribute to these disparities is an important step in delivering interventions that are developmentally and culturally tailored to adolescents of diverse backgrounds.

Acknowledgments

This research was conducted while the lead author was supported by NIH T32 (5T32MH074387-07).

Biographies

Viana Y. Turcios-Cotto is a Doctoral Candidate at the University of Connecticut in the Clinical Psychology division. She received her M.A. in Clinical Psychology from the University of Connecticut and her Ed.M. in Human Development and Psychology from the Harvard Graduate School of Education. Her major research interests include educational and health disparities, the mental health of urban youth and families of color, and positive development of urban youth and families of color.

Stephanie Milan is an Associate Professor at the University of Connecticut. She received a joint Ph.D. in Clinical and Quantitative Psychology from Vanderbilt University. Her major research interests include developmental psychopathology in the context of poverty and mental and physical health disparities among adolescent girls.

VTC participated in the interpretation of the data and drafted the manuscript; SM conceived of the study, and participated in its design and coordination, performed the statistical analysis, and helped to draft the manuscript. All authors read and approved the final manuscript.

• Abraham JP, Lujan SS, López EE, Walker MT. Graduating students’ perceptions of outcomes of college experiences at a predominantly Hispanic university. Journal of Hispanic Higher Education. 2002; 1 (3):267–276. [ Google Scholar ]
• American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: Author; 1994. [ Google Scholar ]
• Anderman EM, Anderman LH, Griesinger T. The relation of present and possible academic selves during early adolescence to grade point average and achievement goals. Elementary School Journal. 1999; 100 :3–17. [ Google Scholar ]
• Beese J, Liang X. Do resources matter? PISA science achievement comparison between students in the United States, Canada and Finland. Improving Schools. 2010; 13 (3):266–279. [ Google Scholar ]
• Cansler E, Updegraff KA, Simpkins SS. Mexican American seventh graders' future work and family plans: Associations with cultural experiences and adjustment. The Journal of Early Adolescence. 2012; 33 :313–339. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Carey K. The funding gap: Many states still shortchange low-income and minority students. The Education Trust; 2004. [Retrieved May 27, 2011]. from http://www.edtrust.org/dc/publication/the-funding-gap-report-2004 . [ Google Scholar ]
• Cohen AB. Many forms of culture. American Psychologist. 2009; 64 :194–204. [ PubMed ] [ Google Scholar ]
• Cunningham M, Corprew CS, III, Becker JE. Associations of future expectations, negative friends, and academic achievement in high-achieving African American adolescents. Urban Education. 2009; 44 :280–296. [ Google Scholar ]
• Derogatis LR. Brief Symptom Inventory: Administration, scoring, and procedures manual. 3rd ed. Minneapolis: National Computer Systems; 1993. [ Google Scholar ]
• Destin M, Oyserman D. From assets to school outcomes: How finances shape children’s perceived possibilities and intentions. Psychological Science. 2009; 20 (4):414–418. [ PubMed ] [ Google Scholar ]
• Editorial Projects in Education Research Center. [Retrieved on July 19, 2012]; National graduation rate rebounds; 1.2 million students still fail to earn diplomas. 2011 from http://www.edweek.org/media/diplomascount2011_pressrelease.pdf . [ Google Scholar ]
• Flores-Gonzalez N. Puerto Rican high achievers: An example of ethnic and academic identity compatibility. Anthropology and Education Quarterly. 1999; 30 :343–362. [ Google Scholar ]
• Fordham S, Ogbu JU. Black students’ school success: Coping with the “burden of ‘acting white’” Urban Review. 1986; 18 (3):176–206. [ Google Scholar ]
• Fryer RG, Torelli P. Cambridge, MA: National Bureau of Economic Research; 2006. An empirical analysis of ‘acting white’. NBER Working Paper 11334. [ Google Scholar ]
• Fuligni AJ, Tseng V, Lam M. Attitudes toward family obligations among American adolescents from Asian, Latin American, and European backgrounds. Child Development. 1999; 70 :1030–1044. [ Google Scholar ]
• Fuligni AJ, Pederson S. Family obligation and the transition to young adulthood. Developmental Psychology. 2002; 38 :856–868. [ PubMed ] [ Google Scholar ]
• Greenfield PM, Keller H, Fuligni A, Maynard A. Cultural pathways through universal development. Annual Review of Psychology. 2003; 54 :461–490. [ PubMed ] [ Google Scholar ]
• Halgunseth LC, Ispa JM, Rudy D. Parental control in Latino families: An integrated review of the literature. Child Development. 2006; 77 :1282–1297. [ PubMed ] [ Google Scholar ]
• Hwang WC, Wood J. Acculturative family distancing: Links with self-reported symptomatology among Asian Americans and Latinos. Child Psychiatry and Human Development. 2009; 40 (1):123–138. [ PubMed ] [ Google Scholar ]
• Jaccard J. Quantitative Applications in the Social Science. Vol 135. CA: Sage Publishers; 2001. Interaction effects in logistic regression. [ Google Scholar ]
• Kenny M, Stryker S. Social network characteristics and college adjustment among racially and ethnically diverse first-year students. Journal of College Student Development. 1996; 37 :649–658. [ Google Scholar ]
• Landis J, Koch G. The measurement of observer agreement for categorical data. Biometrics. 1977; 33 :159–174. [ PubMed ] [ Google Scholar ]
• Lau AS, McCabe KM, Yeh M. The acculturation gap distress hypothesis among high-risk Mexican American families. Journal of Family Psychology. 2005; 19 :367–375. [ PubMed ] [ Google Scholar ]
• Lopez MH, Velasco G. The toll of the great recession: Childhood poverty among Hispanics sets record, leads nation. [Retrieved on November 28, 2011]; 2011 from: http://pewhispanic.org/reports/report.php?ReportID=147 . [ Google Scholar ]
• Markus H, Nurius P. Possible selves. American Psychologist. 1986; 41 :954–969. [ Google Scholar ]
• Murray MS, Neal-Barnett A, Demmings JL, Stadulis RE. The acting White accusation, racial identity, and anxiety, in African American adolescents. Journal of Anxiety Disorders. 2012; 26 :526–531. [ PubMed ] [ Google Scholar ]
• National Center for Education Statistics. [Retrieved on October 29, 2010]; The Condition of Education 2009. 2009 from http://nces.ed.gov/pubs2009/2009081.pdf .
• National Center for Education Statistics. [Retrieved on October 29, 2010]; Public school graduates and dropouts from the common core of data: School year 2007–08. 2010a from http://nces.ed.gov/pubs2010/2010341.pdf .
• National Center for Education Statistics. [Retrieved on October 29, 2010]; Student effort and educational progress: Transition to college. 2010b from http://nces.ed.gov/programs/coe/2010/section3/indicator20.asp .
• Negy C, Woods D. The importance of acculturation in understanding research with Hispanic-Americans. Hispanic Journal of Behavioral Sciences. 1992; 14 :224–247. [ Google Scholar ]
• Nieto S. Affirming Diversity. 3rd ed. New York, NY: Longman; 2000. [ Google Scholar ]
• Nieto S. Public education in the twentieth century and beyond: High hopes, broken promises, and an uncertain future. Harvard Educational Review. 2005; 75 :43–64. [ Google Scholar ]
• Oyserman D, Bybee D, Terry K. Possible selves and academic outcomes: How and when possible selves impel action. Journal of Personality and Social Psychology. 2006; 91 (1):188–204. [ PubMed ] [ Google Scholar ]
• Oyserman D, Bybee D, Terry K, Hart-Johnson T. Possible selves as roadmaps. Journal of Research in Personality. 2004; 38 :130–149. [ Google Scholar ]
• Oyserman D, Fryberg S. The possible selves of diverse adolescents: Content and function across gender, race and national origin. In: Dunkel C, Kerpelman J, editors. Possible Selves: Theory, Research and Application. New York: Nova Science Publishers, Inc.; 2006. pp. 17–39. [ Google Scholar ]
• Oyserman D, Johnson E, James L. Seeing the destination but not the path: Effects of socioeconomic disadvantage on school-focused possible self content and linked behavioral strategies. Self and Identity. 2011; 10 :474–492. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Oyserman D, Markus HR. Possible selves and delinquency. Journal of Personality and Social Psychology. 1990; 59 :112–125. [ PubMed ] [ Google Scholar ]
• Oyserman D, Saltz E. Competence, delinquency, and attempts to attain possible selves. Journal of Personality and Social Psychology. 1993; 65 :360–374. [ PubMed ] [ Google Scholar ]
• Oyserman D, Terry K, Bybee D. A possible selves intervention to enhance school involvement. Journal of Adolescence. 2002; 25 :313–326. [ PubMed ] [ Google Scholar ]
• Pew Hispanic Center. [Retrieved May 27, 2011]; Between two worlds: How young Latinos come of age in America. 2009 from http://pewhispanic.org/reports/report.php?ReportID=117 . [ Google Scholar ]
• Ready DD, Lee VE, Welner KG. Educational equity and school structure: School size, overcrowding, and schools-within-schools. Teachers College Record. 2004; 106 (10):1989–2014. [ Google Scholar ]
• Reynolds WM. Adolescent Psychopathology Scale: Administration and interpretation manual. Odessa, FL: Psychological Assessment Resources; 1998. [ Google Scholar ]
• Rivera-Batiz FL, Marti L. A school system at risk: A study of the consequences of overcrowding in New York City Public Schools. Institute for Urban and Minority Education Research Report No. 95-1. 1995:1–19. [ Google Scholar ]
• Rogoff B, Moore L, Najafi B, Dexter A, Correa-Chavez M, Solis J. Children’s development of cultural repertoires through participation in everyday routines and practices. In: Grusec JE, Hastings PD, editors. Handbook of Socialization: Theory and Research. New York, NY: The Guilford Press; 2007. pp. 490–515. [ Google Scholar ]
• Qin DB. Doing well vs. feeling well: Understanding family dynamics and the psychological adjustment of Chinese immigrant adolescents. Journal of Youth and Adolescence. 2008; 37 (1):22–35. [ Google Scholar ]
• Simons J. Improbable Dreams: African-Americans are a dominant presence in professional sports. Do blacks suffer as a result? U.S. News and World Report. 1997 Mar;:46–52. Retrieved from http://www.usnews.com/usnews/culture/articles/970324/archive_006530.htm .
• Steele CM, Aronson J. Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology. 1995; 69 (5):797–811. [ PubMed ] [ Google Scholar ]
• Suarez-Orozco C, Suarez-Orozco MM. Transformations:Immigration, family life, and achievement motivation among Latino adolescents. Stanford, CA: Stanford University Press; 1995. [ Google Scholar ]
• Triandis HC. The self and social behavior in differing cultural contexts. Psychological Review. 1989; 96 :269–289. [ Google Scholar ]
• Valdes G. Con respeto: Bridging the distances between culturally diverse families and schools. An ethnographic portrait. New York: Teachers College Press; 1996. [ Google Scholar ]
• Wingo PA, Smith RA, Tevendale HD, Ferré C. Recent changes in the trends of teen birth rates, 1981–2006. Journal of Adolescent Health. 2011; 48 :281–288. [ PubMed ] [ Google Scholar ]
• Yowell CM. Possible selves and future orientation: Exploring hopes and fears of Latino boys and girls. Journal of Early Adolescence. 2000; 20 :245–280. [ Google Scholar ]
• Yowell CM. Dreams of the future: The pursuit of education and career possible selves among ninth grade Latino youth. Applied Developmental Science. 2002; 6 :62–72. [ Google Scholar ]

Numbers, Facts and Trends Shaping Your World

Read our research on:

Full Topic List

Regions & Countries

• Publications
• Our Methods
• Short Reads
• Tools & Resources

Read Our Research On:

• Half of Latinas Say Hispanic Women’s Situation Has Improved in the Past Decade and Expect More Gains

3. Educational and economic differences among Latinas today

Table of contents.

• Assessing the progress of Hispanic women in the last 10 years
• Views of Hispanic women’s situation in the next 10 years
• Views on the gender pay gap
• Latinas’ educational attainment
• Latinas’ labor force participation
• Latinas’ earnings
• Latinas as breadwinners in their relationships
• Bachelor’s degrees among Latinas
• Labor force participation rates among Latinas
• Occupations among working Latinas
• Earnings among Latinas
• Latinas as breadwinners in 2022
• Appendix: Supplemental charts and tables
• Acknowledgments
• The American Trends Panel survey methodology
• Methodology for the analysis of the Current Population Survey

Though Latinas have collectively seen socioeconomic gains, their educational and economic circumstances are varied. Younger Latinas and U.S.-born Latinas, for instance, are more likely to report having a bachelor’s degree than older and immigrant Latinas, respectively. This chapter explores how other characteristics such as spouse or partner ethnicity and presence of their children at home are associated with differences in educational and economic outcomes.

Some Latinas are more likely than others to have a bachelor’s degree.

• Age: Younger Latinas (ages 25 to 29) are about twice as likely as older Latinas (ages 65 or older) to hold a bachelor’s degree (30% vs. 14%).
• Nativity: U.S.-born Latinas are more likely than those born outside the U.S. to hold a bachelor’s degree (30% vs. 19%).

Though labor force participation rates have increased in the last two decades for Latinas overall, some are more likely to be employed or seeking work. Among civilians ages 25 to 64:

• Education: Latinas with a bachelor’s degree or higher are more likely than those with a high school education or less to participate in the labor force (82% vs. 60%).
• Nativity: U.S.-born Latinas are more likely than Latinas born outside the U.S. to participate in the labor force (75% vs. 64%).
• Spouse or partner: Latinas who are living with a Hispanic spouse or partner are less likely to work or seek work than those living with a non-Hispanic partner (63% vs. 77%).
• Children at home: Latinas with children in the home are less likely to work or seek work than Latinas without (67% vs. 73%).

Among civilian Latinas ages 25 to 64 who were employed or looking for work in 2023, 15% work in office and administrative support occupations. Similar shares work in management, business and financial occupations (13%) and education, legal, community service, arts and media occupations (12%).

The kinds of occupations Latinas most recently worked in are also associated with whether they have a bachelor’s degree. Among civilian Latinas ages 25 to 64 who were employed or looking for work in 2023:

• Those with a bachelor’s are most likely to have education, legal, community service, arts and media jobs (27%) or management, business and financial jobs (24%).
• For those without a bachelor’s, the most common occupational groups are office and administrative support (17%) and health care support, protective service, and personal care and service (15%).
• Those with a bachelor’s degree are less likely than those without one to work in health care support, protective service, and personal care and service occupations (6% vs. 15%, respectively) and building and grounds cleaning and maintenance occupations (3% vs. 12%).

Though wages have increased for Latinas overall in the last two decades, some earn more than others. Among Latinas ages 25 to 64 who are not self-employed:

• Education: Latinas with a bachelor’s degree make $28.85 per hour (at the median) while those with a high school education or less earn $16.67 per hour.
• Nativity: U.S.-born Latinas make more per hour than immigrant Latinas ($21.25 vs. $17.90).
• Spouse or partner: Hispanic women who live with a spouse or partner earn roughly the same as those without a spouse or partner. However, Hispanic women living with a non- Hispanic spouse or partner make significantly more at the median than those living with a Hispanic spouse or partner ($25.00 vs. $19.00).
• Children at home: Latinas living with their children earn about the same as Latinas not living with their children ($18.50 vs. $20.00).

Overall, 13% of Hispanic women living with their spouse or partner are the breadwinners of their couples. Another 28% of Latinas are in financially egalitarian relationships, while the remaining 59% are living with a breadwinner spouse or partner.

Some Latinas are more likely than others to be either their relationships’ breadwinners or in financially egalitarian relationships with their spouse or partner.

• Education: Latinas with a bachelor’s degree or higher were more likely than those with a high school education or less to be breadwinners (16% vs. 9%, respectively) or in financially egalitarian relationships (35% vs. 24%).
• Spouse or partner: Hispanic women living with a partner or spouse who is not Hispanic were more likely than those with a Hispanic spouse or partner to be the breadwinner of their relationship (16% vs. 12%, respectively). They were also less likely than their Hispanic-partnered counterparts to say their spouse or partner was the breadwinner (54% vs. 61%).

Sign up for our weekly newsletter

Fresh data delivery Saturday mornings

Sign up for The Briefing

Weekly updates on the world of news & information

• Economics, Work & Gender
• Education & Gender
• Educational Attainment
• Gender & Work
• Gender Equality & Discrimination
• Gender Pay Gap
• Higher Education
• Hispanics/Latinos
• Hispanics/Latinos & Education

Key facts about U.S. Latinos with graduate degrees

Hispanic enrollment reaches new high at four-year colleges in the u.s., but affordability remains an obstacle, u.s. public school students often go to schools where at least half of their peers are the same race or ethnicity, what’s behind the growing gap between men and women in college completion, for u.s. latinos, covid-19 has taken a personal and financial toll, most popular, report materials.

1615 L St. NW, Suite 800 Washington, DC 20036 USA (+1) 202-419-4300 | Main (+1) 202-857-8562 | Fax (+1) 202-419-4372 |  Media Inquiries

Research Topics

• Age & Generations
• Coronavirus (COVID-19)
• Economy & Work
• Family & Relationships
• Gender & LGBTQ
• Immigration & Migration
• International Affairs
• Internet & Technology
• Methodological Research
• News Habits & Media
• Non-U.S. Governments
• Other Topics
• Politics & Policy
• Race & Ethnicity
• Email Newsletters

ABOUT PEW RESEARCH CENTER  Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of  The Pew Charitable Trusts .

Copyright 2024 Pew Research Center

ReviseSociology

A level sociology revision – education, families, research methods, crime and deviance and more!

Ethnicity and Education – The Role of Cultural Factors

Cultural factors include parental attitudes, peer-group pressure, language barriers and student aspirations.

Table of Contents

Last Updated on May 17, 2023 by Karl Thompson

Cultural Factors are mostly part of a students’ home background and cultural differences between ethnic groups go some way to explaining different levels of educational achievement by ethnicity .

Cultural factors which may explain why Chinese and Indian children do well in school and why Black Caribbean Children and White children do not do so well include:

• Parental control and expectation, and the value parents place on education.
• Single parent families, and the absence of a male role model (for boys)
• Peer group pressure and an anti-school ‘street’ culture

Language barriers

• Student aspirations to go on to higher education.

The post below explores the above cultural factors and then goes on to evaluate the importance of such factors in relation to in-school factors and structural racism in society.

Parental Control and Expectation

Indian and Chinese families have higher levels of Parental control and expectation .

Strand’s (2007)’s analysis of data from the 2004 Longitudinal Study of Young People found that Indian students are the ethnic group most likely to complete homework five evenings a week and the group where parents are most likely to say they always know where their child is when they are out.

Francis and Archer (2007) found that a high value is placed on education by Chinese parents, coupled with a strong cultural tradition of respect for one’s elders. High educational aspiration transmits from parents to children, and students derive positive self-esteem from constructing themselves as good students.

(Although in a later 2010 paper (1), Francis warned against the stereotype that all Chinese parents are pushy, most middle class white parents are also pushy!))

Basit (2013) researched British Asian families focussing on British Indians and British Pakistanis (both Hindus and Muslims). She studied three generations within the families, using focus groups to collect data from the children and in-depth interviews with the parents and grandparents.

She found that all generations placed a high value on education and the grandparents especially saw free state education as a ‘blessing’ because they did not have such opportunities in their countries of origin. Grandparents and parents thus put special effort into ensuring the children had the resources to study at school. Even the relatively poor children had access to computers at home and their own quiet, independent study spaces.

Grandparents and parents alike viewed education as a form of capital that would transform the lives of the younger generation, opening up opportunities for them, so they were happy to provide them the resources to make the most of these educational opportunities.

There were actually two generations of aspiration being passed down to the children: from the grandparents who had helped their children succeed in education and then from the parents themselves!

Single Parent Households

The New Right argues that the high proportion of lone parents fail to ‘provide a home environment conducive to learning’. There have also been concerns about the development of ‘gangsta’ culture with the absence of positive Black male role models at home as well as in schools (Abbott, 2002).

Historically Caribbean households did have the highest proportion of lone parent households, but according to recent g overnment data on ethnicity and family-structure this is no longer the case.

20.7% of Black African households are lone parent with dependent children, compared to only 16.6% of Black Caribbean households. However Black African children do better at GCSEs than Black Caribbean children. (48% compared to 30% get 5 GCSES grades A*-C including English and Maths, so the difference is massive).

The only thing that might explain the difference in relation to family structure is if Black Caribbean Single Parent Households have more children, which might skew the results if this is a causal factor (but I doubt it!).

The culture of anti-school black masculinity

Tony Sewell (1997) observes that Black Caribbean boys may experience considerable pressure by their peers to adopt the norms of an ‘urban’ or ‘street’ subculture. More importance is given to unruly behaviour with teachers and antagonistic behaviour with other students than to high achievement or effort to succeed.

However Sewell as been criticised for blaming Black Caribbean children for their own failure, rather than taking into account possible racism within the education system itself, more on that when we look at the role of in-school factors .

Acting white and acting black

Fordham and Ogbu (1986) further argue that part of an anti-school black masculinity was what they called ‘acting black’ and ‘acting white’. Notions of ‘acting White’ or ‘acting Black’ become identified in opposition to one another. Hence because acting White includes doing well at school, acting Black necessarily implies not doing well in school.

Crozier (2004) found that Pakistani and Bangladeshi parents ‘kept their distance’ from their children’s schools because they trusted the professionals to do their jobs; they lacked confidence in use of English and there were no translators.

Educational Aspirations

White children have lower educational aspirations than most ethnic minorities.

Research by Connor et al (2004) found that year 13 students from all ethnic minority groups had stronger aspirations to go onto higher education than white children, with the aspiration being strongest for Black African children.

Professor Simon Burgess and Dr Deborah Wilson (2008) found that among Indian, Pakistani, Bangladeshi, Black Caribbean and Black African families, over 90 per cent of parents want their child to stay on at school at age 16, compared with 77 per cent of white families – which correlates with lower numbers at university.

The Immigrant Paradigm

Ogbu (1978), summarised in Strand (2021) (see ‘2’ below) developed the theory that first generation immigrants are enthusiastic about education, seeing it as a real opportunity to help their children progress in a new country, whereas this enthusiasm wheres off for second and especially third generations.

This can go some some way to explaining why Black-Africans overachieve compared to whites while Black-Caribbeans underachieve.

Data from the 2011 census shows that 66.7% of Black Africans are ‘optimistic’ first wave immigrants, while only 39.8% of Black-Caribbeans are first wave immigrants.

Part of the theory is that Black Caribbean families have become assimilated into mainly poor working class neighbourhoods and so their children have adopted the same lack of enthusiasm that White working class children have for education, thus a combination of social class and ethnic background is at work here to explain the low educational achievement of Black Caribbean students.

South Asian women go to university despite cultural pressures

Bagguley and Hussain (2007) found that aspirations to higher education for Pakistani and Bangladeshi women were often complicated by cultural pressures. Many had to negotiate decisions around marriage and the expectations of their parents.

Many Muslim students consequently studied at a local university in order to placate their parents’ concerns about morality, being in the company of men and their family honour or ‘izzat’. In contrast, Indian students currently at university appeared to have had the option of leaving home. Indian women often spoke of a natural progression into higher education that was assumed by both their parents and their schools

How important are cultural factors in educational achievement?

While there are statistical correlations between factors such as parental control and pupil aspirations and educational achievement by ethnicity, it is important to remember that these are just overall averages and that there are variations within each ethnic group.

In other words, be careful not to fall into the stereotype trap of thinking that all Chinese parents or all white children are the same. There are some Chinese parents who don’t value education and some white children (even working class ones) who have high educational aspirations.

There are variations in educational achievement by gender within ethnic groups, for example the cultural barriers to achievement SE Asian women are greater than for boys, and the cultural barriers for AC boys are greater than for AC girls.

Cultural barriers can’t explain all of the variation in educational achievement by ethnicity. Social class and material deprivation also play a role.

In-school factors generally play less of a role in explaining educational differences but where black boys are concerned there is evidence that racist banding and streaming policies may play a role in explaining their relative underachievement, which happens in school and is not to do with cultural background.

Signposting

This post has primarily been written for students studying the education module as part of A-level sociology.

Related posts on the topic of ethnicity and education include:

Material Deprivation and Ethnicity  

In school factors and institutional racism

Please click here to return to the homepage – ReviseSociology.com

(1) Francis et Al (2010) The Construction of British-Chinese Educational Success .

(2) GOV.UK (2021) Ethnic, socio-economic and sex inequalities in educational achievement at age 16, by Professor Steve Strand .

Share this:

• Share on Tumblr

Leave a Reply Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed .

Discover more from ReviseSociology

Subscribe now to keep reading and get access to the full archive.

Continue reading

Cookies on GOV.UK

We use some essential cookies to make this website work.

We’d like to set additional cookies to understand how you use GOV.UK, remember your settings and improve government services.

We also use cookies set by other sites to help us deliver content from their services.

You have accepted additional cookies. You can change your cookie settings at any time.

You have rejected additional cookies. You can change your cookie settings at any time.

• Society and culture
• Equality, rights and citizenship
• The report of the Commission on Race and Ethnic Disparities: supporting research
• Commission on Race and Ethnic Disparities

Ethnic, socio-economic and sex inequalities in educational achievement at age 16, by Professor Steve Strand

Updated 28 April 2021

© Crown copyright 2021

This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3 or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU, or email: [email protected] .

Where we have identified any third party copyright information you will need to obtain permission from the copyright holders concerned.

This publication is available at https://www.gov.uk/government/publications/the-report-of-the-commission-on-race-and-ethnic-disparities-supporting-research/ethnic-socio-economic-and-sex-inequalities-in-educational-achievement-at-age-16-by-professor-steve-strand

Ethnic, socio-economic and sex inequalities in educational achievement at age 16: An analysis of the Second Longitudinal Study of Young People in England (LSYPE2) Report for the Commission on Ethnic and Racial Disparities (CRED) by Professor Steve Strand, Department of Education, University of Oxford

You can download a PDF version of this report .

This report analyses ethnic, socio-economic and sex differences in educational achievement at age 16. It uses the Second Longitudinal Study of Young People in England (LSYPE2), a nationally representative sample of 9,704 students who completed GCSE examinations at the end of year 11 in summer 2015.

The LSYPE2 includes ethnic minority boosts so that sample sizes are sufficient to make robust estimates, and is the most recent dataset from which a comprehensive measure of students socio-economic status (SES) can be derived.

The analysis uses regression modelling to explore the achievement of the 9 biggest ethnic groups, at 3 levels of SES and separately for boys and girls, thus considering a total of 54 estimates for all combinations of ethnic group, SES and sex.

The key results are shown in table 1 and figure 1. The key findings are as follows:

The groups with the lowest achievement at 16 years old are White British, and Black Caribbean and Mixed White and Black Caribbean (MWBC) students from low SES backgrounds, who have mean scores well below the average for all students. This is most pronounced for boys (-0.77 SD and -0.68 SD respectively), but low SES girls of Black Caribbean and Mixed White and Black Caribbean, and White British ethnicity are also the lowest scoring groups of girls (-0.54 SD and -0.39 SD respectively).

Low SES boys of Pakistani, White Other and Any Other ethnic group also have a mean score well below the grand mean, but still score substantially higher than comparable White British and Black Caribbean and Mixed White and Black Caribbean boys.

Among students from average SES backgrounds, only Black Caribbean and Mixed White and Black Caribbean boys and White British boys have mean scores below the average for all students.

The overwhelming picture is therefore of ethnic minority advantage in relation to educational achievement at age 16. At low and average SES, no ethnic minority has a mean score substantially (less than 0.20 SD) lower than White British students, and in 23 of the 32 contrasts the ethnic minority mean is substantially (greater than 0.20 SD) above White British students of the same SES and sex.

There are only 2 instances of ethnic under-achievement compared to White British students of the same SES and sex. First, Black Caribbean and Black African boys from high SES families score lower than comparable White British high SES boys. Second, Pakistani girls from high SES backgrounds do not achieve as well as White British high SES girls, and also substantially below high SES Pakistani boys, who have the highest mean score of all groupings.

The results are discussed in relation to theories of “immigrant optimism” (Kao and Thompson, 2003), “segmented assimilation” (Portes and Zhou, 1993), and teacher expectations and cultural norms.

Table 1: Mean best 8 score by ethnic group, SES and sex, and ethnic achievement gaps relative to White British

Notes: Mean Best 8 scores show the difference between the mean score for the group and the grand mean score across all pupils (which is set to 0). Gap vs White British shows the difference in the mean score between the ethnic minority and White British students of the same sex and SES. Ethnic groups are sorted in order of the mean Best8 score for pupils of average SES.

Figure 1: Mean best 8 score by ethnic group, socio-economic status (SES) and sex

Education is the key to future life outcomes. Success in education at 16 years old is strongly predictive of later occupational, economic, health and well-being outcomes and to future social mobility: this is why 13 of the 17 social mobility indicators drawn up by the government in England are measures of educational attainment (Cabinet Office, 2011).

In the 2019 GCSE examinations, the average Attainment 8 score for Black Caribbean (39.4) and Mixed White and Black Caribbean (41.0) pupils was over 5 points lower than the average for White British pupils (46.2), or over half a grade lower in each of the 8 subjects included. At the same time, the average scores for Indian, Pakistani, Bangladeshi and Black African ethnic groups were above the White British average. What factors underpin such variation?

It is widely documented that socio-economic status (SES) is strongly implicated in low educational achievement. SES may have a direct influence, for example through poorer nutrition and an increased risk of a range of health and developmental problems, and an indirect influence through limited financial resources in the home, low parental education, reduced ability to help with homework, unemployment, maladjustment or neglect, housing instability or homelessness, greater family stress and poorer neighborhood quality in terms of services and crime (for example, Bradley and Corwyn, 2002; McLloyd, 1998; Reis, 2013; Spencer, 1996).

The greater socio-economic deprivation experienced by ethnic minority groups compared to the White ethnic group has also been well documented. For example, in England in 2016, 14% of White British pupils were eligible for a free school meal (FSM) but this doubled to 25% of Black African, 28% of Black Caribbean and 29% of Mixed White and Black Caribbean pupils (Strand and Lindorff, 2021).

This unevenness extends across many socio-economic dimensions in employment, income, housing and health (Kenway and Palmer, 2007; Strand, 2011). Ethnic minority pupils may therefore be more at risk of low achievement because of the greater socio-economic disadvantage they experience relative to White pupils.

The purpose in taking the socio-economic factors into account is not to ‘explain away’ any ethnic achievement gaps, but to better understand the root causes and therefore identify relevant policy interventions and action. For example, if ethnic achievement gaps reflect the socio-economic disparities between ethnic groups, then a focus on in-service training to address racism by secondary school teachers would be unlikely to deliver substantial change, whereas a focus on increased resourcing for disadvantaged pupils (such as the pupil premium grant) may have a greater likelihood of success.

It is therefore important that any analysis looks not just at ethnicity in isolation, but looks simultaneously at ethnicity and socio-economic status as well as gender. Previous analyses of the first Longitudinal Study of Young People in England (LSYPE) have looked at these 3 factors simultaneously in relation to educational achievement at 11, 14 and 16 year old (See Strand 2011; 2012; 2014). A summary of the results at age 16 is reported in Appendix B. The results indicated average scores for ethnic minority groups were higher than for White British pupils of the same SES and sex, such that ethnic minority status was a facilitator, not a barrier, to achievement. However, the LSYPE cohort took their GCSEs in summer 2006, some time ago. This report analyses recent data from LSYPE2 which provides the most up-to-date data to analyse the combined effect of ethnicity, sex and socio-economic status in relation to students’ educational achievement at 16 year olds.

Methodology

We place the vast amount of information on the methodology in the detailed methodology, so that we can focus immediately on the key findings and discussion. We summarise here only those features that are essential to interpretation of the results and key findings. Detailed description of the dataset and analysis is also given in the detailed methodology.

The dataset

The LSYPE2 recruited a nationally representative sample of 13,000 young people aged 14 in year 9 in the 2012 to 2013 school year, and conducted detailed 45-minute interviews with them and with their parents in their own homes, as well as drawing from linked administrative sources such as the National Pupil Database (NPD).

Importantly, the LSYPE2 includes ethnic minority boosts with a target of 1,000 respondents from each of the main ethnic minority groups, so that the sample size is large enough to support robust national estimates for ethnic minority groups. The students and their families were interviewed again in wave 2 in year 10 and in wave 3 in year 11. Of the 10,396 students who completed wave 3, a total of 9,704 gave their permission for linkage to the NPD so we can analyse their GCSE results from the end of year 11 in summer 2015.

The measures

Ethnic group.

In 2019, one-third (32.9%) of the school population in England were from ethnic minority groups. We present a summary at the highest level of aggregation (White, Mixed, Asian, Black, Other) but believe there is value in a more differentiated analysis in relation to the 9 main ethnic groups in England (White British, White Other, Indian, Pakistani, Bangladeshi, Asian Other, Black Caribbean, Black African and Any Other group). We include the Mixed ethnic groups in the ethnic minority part of their heritage – for example, we combine Black Caribbean and Mixed White and Black Caribbean students. The rationale is explained in the detailed methodology.

Socio-economic status

For descriptive purposes we focus on parental occupation as the single most frequently cited measure of social class (Raffe et al, 2006). We use the Office for National Statistics (ONS) Socio-Economic Classification (ONS-SEC), and indicative examples of the classification are given in Appendix B. We employ the dominance method (Erikson, 1984) taking either the father’s or the mother’s occupation, whichever is the highest. We do the same for parents’ educational qualifications and family income. For subsequent statistical modelling, we create a comprehensive measure of socio-economic status incorporating all 3 measures: parental occupational status, parental educational qualifications, and average family income. To do this we take the loading on the first factor of a principal component analysis of the 3 measures.

Educational outcomes

We calculate each pupil’s Best 8 point score, which is the total score across the best 8 examination results achieved by the pupil. The points are calculated on the QCA scale which is not a very familiar metric, and the score distribution is slightly negatively skewed, so for ease of interpretation we have applied a normal score transformation so the outcome is expressed in standard deviation (SD) units.

Therefore, the average score across all students is indicated by zero, and two-thirds of students score in the range between -1 and +1. For a threshold measure we report the percentage of pupils achieving a GCSE grade A* to C in both English and maths. This measure is still reported in secondary school performance tables (based on the percentage achieving a Grade 5 or above using the new 1 to 9 scale first examined from summer 2017) so is more useful than the headline measure in use in 2015, which was 5 or more GCSEs at A* to C including English and maths.

Key findings

Descriptive statistics for achievement by ethnicity, sex and ses.

Table 1 and Figure 1 presents the mean Best 8 points score and the percentage achieving GCSE A* to C in both English and mathematics by ethnicity, sex and 3 measures of SES (parental occupation, parental education and family income).

The key points are as follows.

At the highest level of ethnic aggregation, the mean Best 8 score was 0.05 for White students and -0.06 for Black students, giving a Black-White difference of -0.11 SD. This Black-White gap is statistically significant but small. By way of comparison, Cohen’s (1988) effect size thresholds suggest 0.20 SD is small, 0.50 SD is medium and 0.80 is large.

The results contrast strongly with those from the US, where in the 2017 National Assessment of Educational Progress (NAEP), Black students scored -0.81, -0.83 and -0.89 SD below the mean for White students in mathematics at age 10, 14 and 18 respectively. They also scored approximately -0.72 SD below the mean for White students for reading at the same ages (US Department of Education, 2019).

When the ‘Black’, ’Asian’ and ‘White’ groups are disaggregated, some slightly larger gaps are found. However, the only ethnic group with an average score significantly below the White British mean is Black Caribbean and Mixed White and Black Caribbean students, with a gap of -0.29 SD, while Black African and Mixed White and Black African students have a mean score near identical to White British. All other ethnic groups score as well as, or in the case of Indian and Asian Other ethnic groups significantly better than, the White British average.

This Black Caribbean achievement gap is the same magnitude as the gender gap which is also 0.29 SD, with girls scoring higher than boys. However, both gaps are dwarfed by the parental occupation gap, which is over 3 times larger at 0.97 SD. The family income gap is 0.93 SD and the parental education gap is 1.14 SD.

If we take a conservative analysis, comparing the results for the 22% of students in the lowest 3 parental occupational groups (LTU, routine and semi-routine occupations) against the average for the 45% of students with a parent in the highest groupings (higher technical, higher managerial and professional occupations), the gap is 0.81 SD, still 3 times larger than either the Black Caribbean or gender gaps.

Table 2: KS4 results by ethnicity, sex and parental SEC

Notes: SEC is the ONS Socio-economic classification (SEC) of the occupation of the highest classified parent. Parent Educ. is the highest educational qualification held by the most qualified parent. Family income is average family income expressed in quintiles.

Figure 2: Mean Best 8 points score by ethnic group, sex and parental SEC

Figure 3: mean best 8 points score by ethnic group, sex and parental sec, ethnicity and socio-economic status (ses).

Considering the 3 factors of ethnic group, sex and SES separately is limited, because there is significant confounding between these variables. Most particularly, levels of socio-economic disadvantage are substantially higher among ethnic minority groups than among the White British majority. Table xxx presents averages for a wide range of socio-economic measures separately for each ethnic group.

The key findings are:

Parental occupation (r= 0.38), parental education (r= 0.38) and family income (r= 0.38) were all positively correlated with KS4 Best 8 score, but the overall SES measure gave the highest correlation (r= 0.45). Therefore, SES is the best single measure in relation to exam success.

In terms of SES, White British (0.22 SD), Indian (0.21 SD) and Asian Other (0.11 SD) had mean SES scores above average, Black Caribbean (-0.15 SD), Black African (-0.12 SD) and White Other (-0.14 SD) were closely grouped, while Pakistani (-0.53 SD) and Bangladeshi (-0.83 SD) had substantially the lowest SES.

The gaps in the underlying measures are often stark:

• for 20% of White British students the highest parent occupation is ‘LTU, routine or routine occupation’, but this more than doubles to over 40% for each of the Black African, Pakistani and Bangladeshi ethnic groups
• for 29% of White British students at least one parent has a degree, compared to Black African (40%), Indian (43%) and Asian Other (52%) students, but just 13% of Bangladeshi students
• White British students had the highest annualised family income (£40,785), followed by Indian (£36,246) and Asian Other (£33,862), but is more than one-third lower for Black Caribbean (£29,485) and Black African (£28,405) students, and half as high for Pakistani (£22,693) and Bangladeshi (£19,828) students
• 24% of White British students have been entitled to a free school meal at some time in the last 6 years, but this is more than doubled for Black Caribbean (47%), Any Other (49%), Black African (53%) and Bangladeshi (61%) students

While Black Caribbean and Black African students had similar overall SES (-0.15 and -0.12 respectively), they differed in their profile across the 3 underlying components: Black African students have a higher proportion of parents in ‘LTU, routine or semi-routine’ occupations (41% vs. 31%) and slightly lower family annualised income (£28,405 vs. £29,475), but had a higher proportion of parents educated to degree level (40% vs. 23% respectively).

Table 3: socio-economic variation between ethnic groups

You may need to scroll horizontally to see all columns.

Notes. ‘SES’ is a standardised score of the loading on the first factor from a principal components analysis of parental occupation, parental education and average family income. ‘Parental occupation’ as coded by the ONS Socio-Economic Classification (ONS-SEC) 3 category version. ‘LTU’ means long term unemployed, defined as 6 months or more. ‘Parental education’ is the highest qualification assessed on a 7 point scale ranging from no educational qualifications through to university degree. ‘Family income’ is average equivalised income per annum. ‘FSM’ indicates eligibility for free school meals in January of year 11. ‘EVER6’ indicates entitlement to free school meals at any point during the last 6 years (Y6-Y11). ‘IDACI’ is the Income Deprivation Affecting Children Index quartile, based on the proportion of children in the neighbourhood from families entitled to state benefits. ‘X’ indicates fewer than 10 cases in the cell so the value is suppressed following ONS rules.

Interactive effects of ethnicity, sex and SES with achievement

Given these results, we complete a regression analysis to look at the combined associations of achievement with ethnicity, sex and SES. There were several highly significant ethnic and SES interactions, one ethnic and sex interaction and a 3-way ethnic, SES and sex interaction. Therefore, a full-factorial model was specified and effects were assessed using estimated marginal means. The parameters from the model are given in Appendix C.

Table 3 and Figure 4 present the mean Best 8 score for each ethnic, SES and sex combination, along with the ethnic achievement gap showing the difference between the average score for the ethnic minority compared to White British pupils of the same sex and SES.

The key findings are as follows.

Mean Best 8 score

The groups with the lowest achievement at age 16 are White British, and Black Caribbean and Mixed White and Black Caribbean students from low SES backgrounds, who are scoring substantially below the average for all students (which is set at zero). This is most pronounced for boys (-0.77 SD and -0.68 SD respectively), but low SES girls in the Black Caribbean and Mixed White and Black Caribbean, and White British ethnic groups are also the lowest scoring groups of girls (-0.54 SD and -0.39 SD respectively).

Low SES boys in the Pakistani, White Other and Any Other ethnic groups also score well below the mean, but still score substantially higher than comparable White British, and Black Caribbean and Mixed White and Black Caribbean peers.

Among students from average SES backgrounds, only Black Caribbean and Mixed White and Black Caribbean boys and White British boys score below the grand mean.

Beyond the above, no ethnic, SES and sex combination scores substantially below the grand mean, with the majority scoring well above the average.

Ethnic gaps relative to White British

The overwhelming picture is that ethnic minority groups have higher educational achievement at age 16 than White British students of the same sex and SES. This is particularly notable at low and average SES, where no ethnic minority groups have a significantly lower score than White British students, and indeed in 23 of the 32 comparisons the mean score for ethnic minority students is substantially higher than for comparable White British students.

There are only 2 instances of ethnic under-achievement compared to White British students of the same SES and sex. First, Black Caribbean and Black African boys from high SES families score more than 0.20 SD lower than comparable White British boys. Second, Pakistani girls from high SES backgrounds do not achieve as well as White British high SES girls, and substantially below high SES Pakistani boys, who have the highest mean score of all groupings.

Table 4: Mean best 8 score by ethnic group, SES and sex, and ethnic achievement gaps relative to White British

Figure 4: mean best 8 score by ethnic group, level of ses and sex, ethnicity and low educational achievement.

The key finding is that White British and Black Caribbean students, both boys and girls, from low SES backgrounds are the lowest achieving groups of all students. While low SES boys from Pakistani, White Other and Any Other ethnic groups also score below the overall average, they are still scoring significantly higher than White British and Black Caribbean low SES boys. It is also notable that at mean SES, it is again only White British and Black Caribbean boys who score substantially below the average. A key question therefore is why most ethnic minority groups are so much more resilient compared to White British and Black Caribbean students.

The ‘immigrant paradigm’ (Kao and Thompson, 2003) suggests that recent immigrants devote themselves more to education than the native population because they lack financial capital and see education as a way out of poverty. In a similar vein, Ogbu (1978) makes a distinction between ‘voluntary minorities’ (such as immigrant groups who may be recent arrivals to the country and have very high educational aspirations) and ‘involuntary’ or ‘caste like’ minorities (such as African Americans or Black Caribbean and White working class pupils in England) who hold less optimistic views around social mobility and the transformative possibilities of education.

This theory can, for example, account for the substantial contrast between Black Caribbean and Mixed White and Black Caribbean pupils on the one hand and Black African and Mixed White and Black African pupils on the other, whose achievement is substantially higher despite the same or higher levels of risk in terms of low SES, neighbourhood deprivation, and poverty. Most Black Caribbean and Mixed White and Black Caribbean pupils are third generation UK born, while many Black African pupils are more recent immigrants, some of whom have arrived directly from abroad. For example, the 2011 national population Census indicates that one-third (66.7%) of the Black African population were born outside of the UK, compared to 39.8% of the Black Caribbean population (ONS, 2013).

But if ‘immigrant optimism’ is the explanation, why does the achievement of Black Caribbean and Mixed White and Black Caribbean students more closely match that of White British students, particularly at low SES, rather than matching other ethnic minority groups? Partly this may be because they are one of the longer-standing migrant groups, with the largest waves of migration in the 1950s and early 1960s.

Ogbu (1978) suggests that those minorities who have been longest established in a country, particularly in a disadvantaged context, may be the least likely to be optimistic about the possibilities of education to transform their lives, and several studies have noted this ‘second generation’ gap (for example, Rothon et al, 2009). But Indian and Pakistani migration was also high during the 1950s and 1960s, why is the achievement profile for these ethnic groups not also closer to White British students?

Perhaps relevant here is “selective assimilation theory”. Black Caribbean migrants in the 1960’s predominantly moved into poor urban and inner city areas populated by the White British working class. The intersecting of the communities is reflected in the high level of inter-ethnic partnerships and births, with there now more being students in school from the Mixed White and Black Caribbean ethnic group than there are from the Black Caribbean ethnic group (1.6% vs. 1.1% of the school population) (DfE, 2019). Thus, Black Caribbean and Mixed White and Black Caribbean students may have cultural attitudes that parallel their (predominantly) White British working class neighbours.

In contrast, other long standing ethnic minority groups have different patterns of migration. Indian migrants were more likely to be of high SES in their host countries, many were professionals and managers, and migrated to a more varied and diverse selection of geographical areas. Other groups such as Pakistani migrants, while also tending to move predominantly to poor areas of inner cities where housing was cheap, tended to have higher levels of ethnic segregation, retaining greater cultural homogeneity.

The most direct support for the ‘immigrant optimism’ thesis comes from Strand (2011; 2014), in his analysis of the original LYSPE, which identified 4 key factors underlying the greater resilience of low SES ethnic minority pupils:

• high educational aspirations on the part of students to continue in education post-16 and to attend university, placing education in central role for achieving their future goals
• high educational aspirations by parents and strong ‘academic press’ at home
• high levels of motivation and homework completion
• strong academic self-concept

There is insufficient time to undertake further analysis at present before the deadline for this report, but further analysis will be completed later in the year to see if these results from LSYPE are replicated for LSYPE2.

Ethnic minority underachievement

The overwhelming picture is that ethnic minority groups have higher average levels of achievement than White British peers of the same SES and sex. While they were very much exceptions to the rule, there were 2 specific instances of ethnic under-achievement.

First, Black Caribbean and Black African boys from high SES homes underachieved relative to White British high SES boys. What underlies this particular finding is not known, and worthy of further investigation. Previous research has indicated that Black Caribbean pupils are under-represented by their teachers in entry to higher tier examinations, after a wide range of controls for prior attainment, SES, attitudes and behaviour (Strand, 2012), and that Black Caribbean and Mixed White and Black Caribbean pupils are more often subject to disciplinary sanctions like exclusion than other ethnic groups, again after control for covariates (Strand and Fletcher, 2014).

It may be that in school settings, negative expectations about Black boys lead to greater surveillance and pre-emptive disciplining by teachers, which may be particularly disproportionately felt by Black middle class boys (Gillborn et al, 2012). Alternatively, it may be that White British middle class families use their financial resources to purchase advantages, like private schooling, to a greater extent than Black middle class families. In the LSYPE2 we found 6.7% of White compared to 2.2% of Black pupils attended independent schools, although analysis of the British Social Attitudes survey suggests no significant difference (Evans and Tilley, 2012).

Out of school factors may also be influential. For example, Foster et al. (1996) and Sewell (2009) argue that Black boys experience considerable pressure by their peers to adopt the norms of an ‘urban’ or ‘street’ subculture where more prestige is given to unruly behaviour with teachers than to high achievement or effort to succeed (for example, Foster et al., 1996; Sewell, 2009). Gangster culture and hyper-masculinity may be shared to greater extent by White and Black boys within working class contexts, more so than in middle class spaces. Issues of identity could also be felt particularly by Black middle class boys, with some researchers suggesting Black middle-class families often express “an unease about middleclassness which was viewed by some as a White social category” (Ball et al, 2013, p270, see also Archer, 2010; 2011). Of course, these arguments are not mutually exclusive, both in-school and out-of-school factors may well play a role.

Second, Pakistani high SES girls underachieved compared both to White British high SES girls, and indeed achieved less well than high SES Pakistani boys. It may be that traditional attitudes to gender roles, lower perceived benefits of daughters’ relative to sons’ education, and threats to respectability and modesty expressed by parents in Pakistan (Purewal and Hashmi, 2015) also apply in England. However, Fleischmann and Kristen (2014) looking at second generation immigrants in 9 European countries (including England and Wales) indicate that gender gaps favouring males in countries of origin are largely reversed in the second generation, transforming to the patterns of female achievement advantage seen in the host countries. This is a small group within the LSYPE2 dataset, because of the very skewed SES distribution for Bangladeshi and Pakistani students. For example, the number of Pakistani pupils in the top top 20% of SES is just 17 and fewer than 10 Bangladeshi pupils (the comparable figure for White British pupils is 1667 cases). The finding should therefore be treated with caution, but is worthy of further investigation.

These results indicate that ethnic minority groups on average achieve higher levels of success in education at age 16 than White British pupils. To the extent that there is a small gap for Black Caribbean students, this seems to reflect structural inequality in SES, with fewer parents in managerial and professional roles and lower average family income. Gaps in achievement at age 16 related to SES are large and persistent, and represent by far the greatest challenge to equity and social mobility agendas.

Educational achievement at age 16 is crucial, in that it acts as a gatekeeper to higher education and employment opportunities later in life. Nevertheless, ethnic variation in outcomes at later ages still remains. For example, in access to high-tariff universities (Boliver, 2016), in entry to work (Heath and Di Stasio, 2019) and to the highest occupational groups (UK Government, 2020).

Detailed methodology

Ethnic minority groups.

Table xxx indicates the unweighted number of pupils within each ethnic group as recorded in the LSYPE2 wave 3 dataset and with valid linkage to the NPD. The third column of the table shows the percentage that each ethnic group represents in the whole school population, sourced from the 2019 school census. This shows that one-third of the school population in England (32.9%) are from an ethnic minority group (DfE, 2019).

Table 5: Ethnic coding for purposes of analysis of LSYPE2

Table 5(a): full set of ethnic codes, table 5(b): ethnic groups used in the analysis.

Notes: (a) less than 10 pupils so number suppressed.

In analysing the LSYPE2 data, a balance needed to be struck between the number of ethnic groups, the size of these groups in the school population and the number of cases in the specific LSYPE2 sample.

The largest ethnic minority groups (Indian, Pakistani, Bangladeshi, Black African, Black Caribbean, White Other and Asian Other) were retained.

The Mixed ethnic groups have been shown to be extremely heterogenous with little in common in terms of the achievement profile among the sub-groups (see Strand, 2015). In term of their achievement profile, there is greater similarity with the ethnic minority side of their Mixed ethnicity. For example, the achievement of Mixed White and Black Caribbean pupils is similar to that of Black Caribbean pupils, the achievement of Mixed White and Black African pupils is similar to the Black African, and the achievement of Mixed White and Asian (MWAS) pupils is similar to that of Asian Other pupils. This is shown in Figure xxx, which is drawn from Strand (2015), p32.

Source: Strand (2015). ‘Ethnicity, deprivation and educational achievement at age 16 in England: trends over time.’ DfE Research Report 439B, p32.

Therefore, to more accurately reflect the patterns of achievement, and to maximise the analytic samples, the Mixed ethnic groups were included with the relevant ethnic minority group.

Smaller ethnic groups were merged. Thus, White Irish and Gypsy Roma Travellers (GRT) were included in White Other; Chinese were included in Asian Other and MWAS; and Black Other and Mixed Other groups were included in Any Other ethnic group.

Table 5(b) shows the 9 ethnic groups used for this analysis, the unweighted number of cases in each group and the percentage the groups represent in the whole school population (school census 2019).

Family socio-economic classification (SEC)

We utilised the ONS Socio-Economic Classification (SEC). A family SEC variable is included in LSYPE2 based upon the household reference person (HRP), but in a large number of cases the HRP was not interviewed (n=487) or the individual was not classifiable (n=121). We therefore created our own family SEC measure. First, we took the SEC for the main parent, which had fewer missing or unclassifiable instances (n=116). Second, to create a family measure, we substituted the SEC of the second parent (if present) if it was higher than for the main parent. As a robustness check we completed the same process taking the highest of the mother’s or father’s SEC. This measure was very highly correlated (r=0.996) with the MP/SP version, but the MP/SP version had fewer missing cases (n=116 as opposed to n=502) so was preferred.

Table 6: ONS Socio-economic classification (SEC) categories: LSYPE2 Sample

We also looked in wave 2 and wave 3 for SOC2010 values if there was no SEC record in the wave 1 file. These employ 9 major groups and 25 sub-major groups (see SOC2010 volume 1: structure and descriptions of unit groups ). We converted codes between SOC2000 and SOC2010 where needed (see https://www.bls.gov/soc/soc_2000_to_2010_crosswalk.xls ). We were able to find valid values for all but 33 cases.

Parental educational qualifications

We took the highest educational qualification of the main parent, substituting the highest qualification of the second parent (where present) if it was higher, termed the dominance method (Erikson, 1984). If we could not find a value in the wave 1 file we again sourced the variable from the wave 2 or wave 3 file. We were able to find valid values for all but 27 cases. A small number of cases (n=37) which were coded as ‘entry level qualifications’ were combined with ‘Other qualifications’. This created a 7 point scale ranging from ‘No educational qualifications’ through to ‘Degree or equivalent’. Descriptive statistics showing the relationship with student achievement are given in Table 1.

Family income

Household income is based on a survey response, with respondents picking a band from a list to represent the annual household income from all sources. The results have been edited to take account of implausible responses, primarily through the use of self-reported earnings data.

Earnings data was generally more credible, not least because parents reported their own earnings, over the time period of their choice, rather than having to combine sources and annualise the results. This data has also been edited where implausible, such as where what looked like an annual salary for the stated occupation was reported as being paid weekly.

Where the plausible earnings of a household were greater than the annual income selected, the earnings have been used instead. This is likely to underestimate the true income, as it excludes other sources such as benefits, but should still represent an improvement on the self-reported estimate.

The data was collected in 15 bands allowing a high degree of differentiation. For descriptive purposes we used the midpoint of the ranges as the data value rather than the band number to give a mean income in pounds per annum. It should be noted that income data is notoriously difficult to collect accurately via household surveys, and LSYPE2 is no exception, with a high level of non-response. To deal with this, we took the average income over all 3 waves of the LSYPE2, this reduced the missing cases to n=437 (or 4.5%) of our sample. To avoid losing these cases, we imputed the value predicted from a regression of income on other variables closely related to income (entitlement to a FSM, IDACI score and parental SEC), so only had one missing value in the final analysis.

Income deprivation affecting children index (IDACI)

IDACI is produced by the Ministry of Housing, Communities and Local Government. The index is based on 32,482 super output areas (SOAs) in England, which are geographical regions of around 1,500 residents, designed to include those of similar social backgrounds.

The IDACI score is the percentage of under-16s in the SOA living in income deprived households (primarily defined by being in receipt of certain benefits). This variable is highly skewed and so for the purpose of the current analysis the measure was normal score transformed to give a variable with a mean of 0 and SD=1. A score above 0 indicate greater than average deprivation, and score below 0 indicate less than average deprivation, relative to the average for the LSYPE2 sample. Both 2001 and 2007 IDACI measures were included in the LSYPE2 file. The means of the 2 were nearly identical (24.7% and 25.7%) and they correlated r=0.97, so the more recent 2007 values were used. You can see more information about IDACI .

FSM and EVER6

We took from the January census of year 11 whether the pupil was entitled to a free school meal (FSM) or had ever been entitled over the last 6 years (EVER6).

The LSYPE2 sample

The primary sample frame for LSYPE2 was the England school census, which was used to identify sample members in state-funded education. This provides access to pupil-level characteristics information about these young people, which was used to stratify the sample.

The stratification has been designed to maintain minimum numbers in certain subgroups of interest right through to the planned end of the survey, to ensure robust analyses of these groups can continue. These subgroups include those with free school meals, those with special educational needs (SEN), and certain ethnic groups. The sample also included pupils from independent schools and pupil referral units (PRUs), these schools and settings were sampled first and then asked to supply contact details for pupils.

Interviews took place with both the young person and at least one parent in the first 3 waves (until the young person was 15 or 16 years old). In wave 1 the interviews took place over a 5-month period, starting in early April 2013 and finishing in early September 2013. In wave 1 LSYPE2 achieved a response rate of 71%, representing an achieved sample of 13,100.

The analytic sample

As stated, there were 13,100 responding young people in wave 1 of LSYPE2. Of these, 12,152 responded in wave 2 and 10,396 in wave 3. Of those responding in wave 3, a total of 9,307 gave permission for linkage and were matched to results in the NPD. Some of those giving permission were in independent schools (n=410) who were missed by the DfE in the initial data match, and so are not yet included in our analysis (at 18/12/20). 9.307 was the total sample available, and we had complete observations for ethnic group and sex, but a small number of cases that were missing parental SEC (n=49), parental education (n=22), family income (n=26), SES (n=69), entitlement to a FSM/EVER6 (n=17) or IDACI (n=7), had to be excluded on a pairwise basis. The ONS-SRS does not have the SPSS Missing Values module, so we cannot impute missing values for these cases, but we will explore whether this might be possible through other means at a later date.

Approach to analysis

We were primarily interested in the relationship between variables, not in simply recapturing descriptive statistics for the relevant population. In these cases, the use of weights is sometimes argued to be problematic (Solon, Haider and Woodridge, 2015). However, given the extent of attrition from wave 1 to wave 3 of LSYPE2, we considered it important to use weights that are meant to limit the effect of differential attrition, and used the combined design and non-response scaled sampling weights from wave 3 in all analyses (LSYPE2_W3_Weight_scaled).

The ONS-SRS has not purchased the SPSS Complex Samples module, and so, despite the software being available to university staff and students throughout the country, we were not able to use it to simultaneously account for weight and for clustering at the school level.

However, we also ran all our models using a complex survey design using the svydesign() and svyglm() functions contained within version 3.35-1 of the Survey package (Lumley, 2019) in version 3.6.1 of R (R Core Team, 2019). These models used the students’ KS4 school URN as the cluster ID and the LPYSE2_W3_Weight_scaled as the sampling weight. In all cases there were no substantive differences in results, means were near identical. Although SEs tended to be marginally higher, all results that were statistically significant in our SPSS regressions were also significant in the R versions. Therefore, we do not consider this a problem for the analysis.

Archer, L. (2010). ‘We raised it with the Head’: the educational practices of minority ethnic, middle-class families. British Journal of Sociology of Education, 31(4), 449 to 469.

Archer, L. (2011). Constructing Minority Ethnic Middle-class Identity: An Exploratory Study with Parents, Pupils and Young Professionals. Sociology, 45(1), 134 to 151. doi:10.1177/0038038510387187

Ball, S. J., Rollock, N., Vincent, C., and Gillborn, D. (2013). Social mix, schooling and intersectionality: identity and risk for Black middle class families. Research Papers in Education, 28(3), 265 to 288. doi:10.1080/02671522.2011.641998

Boliver, V. (2016). Exploring Ethnic Inequalities in Admission to Russell Group Universities. Sociology, 50(2), 247 to 266. doi:10.1177/0038038515575859

Bradley, R. H., and Corwyn, R. F. (2002). Socioeconomic status and child development. Annual Review of Psychology, 53(1), 371 to 399. https://doi.org/10.1146/annurev.psych.53.100901.135233

Cabinet Office (2009). Unleashing aspirations: The final report of the panel on fair access to the professions . London: Cabinet Office.

Cabinet Office (2011). Opening Doors, Breaking Barriers: A Strategy for Social Mobility. London: Cabinet Office.

Cabinet Office (2017). Race Disparity Audit: Summary findings from the Ethnicity facts and figures website. London: Cabinet Office.

Cohen, J. (1988). Statistical power analysis for the behavioural sciences (2nd Edition). Hillsdale, NJ: Erlbaum.

DfE (2019). Schools, pupils and their characteristics: January 2019 . London: Department for Education.

DfE (2020). Key stage 4 performance, 2019 (Revised) . London: Department for Education.

DfE (2020). Secondary accountability measures: Guide for maintained secondary schools, academies and free schools . London: Department for Education.

Erikson, R. (1984). Social class of men, women and families. Sociology, 18, 500 to 514.

Evans, G., and J., T. (2012). Private schools and public divisions: the influence of fee-paying education on social attitudes . British Social Attitudes, 28th Report.

Foster, P., Gomm, R., and Hammersley, M. (1996). Constructing educational inequality. London: Falmer Press.

Gillborn, D., Rollock, N., Vincent, C., and Ball, S. J. (2012). ‘You got a pass, so what more do you want?’: race, class and gender intersections in the educational experiences of the Black middle class. Race Ethnicity and Education, 15(1), 121 to 139. doi:10.1080/13613324.2012.638869.

Heath, A. F., and Di Stasio, V. (2019). Racial discrimination in Britain, 1969–2017: a meta-analysis of field experiments on racial discrimination in the British labour market. The British Journal of Sociology, 70(5), 1774 to 1798. doi:10.1111/1468-4446.12676

Kao, G., and Thompson, J. S. (2003). Racial and ethnic stratification in educational achievement and attainment. Annual Review of Sociology, 29(1), 417 to 442. doi:https://doi.org/10.1146/annurev.soc.29.010202.100019

Lumley, T. (2019) ‘Survey: analysis of complex survey samples’. R package version 3.35-1.

McLoyd, V. C. (1998). Socioeconomic disadvantage and child development. American Psychologist 53(2), 185 to 203. https://psycnet.apa.org/doi/10.1037/0003-066X.53.2.185

Office for National Statistics (ONS) (2013). DC2205EW – Country of birth by ethnic group by sex . Office for National Statistics.

Portes, A., and Zhou, M. (1993). The new second generation: Segmented assimilation and its variants. The Annals of the American Academy of Political and Social Science, 530(1), 74 to 96. doi:10.1177/0002716293530001006

Purewal, N., and Hashmi, N. (2015). Between returns and respectability: parental attitudes towards girls’ education in rural Punjab, Pakistan. British Journal of Sociology of Education, 36(7), 977 to 995. doi:10.1080/01425692.2014.883274

R Core Team (2019) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, URL https://www.R-project.org

Raffe, D., Croxford, L., Iannelli, C., Shapira, M., and Howieson, C. (2006). Social-Class Inequalities in Education in England and Scotland, CES Briefing No. 40 . Centre for Educational Sociology, University of Edinburgh.

Reiss, F. (2013). Socioeconomic inequalities and mental health problems in children and adolescents: A systematic review. Social Science and Medicine, 90, 24 to 31. https://doi.org/10.1016/j.socscimed.2013.04.026

Rothon, C., Heath, A., and Lessard-Phillips, L. (2009). The Educational Attainments of the “Second Generation”: A Comparative Study of Britain, Canada, and the United States. Teachers College Record, 111(6), 1404 to 1443.

Sewell, T. (2009). Generating Genius: Black Boys in Search of Love, Ritual and Schooling. Stoke on Trent: Trentham books.

Spencer, N. (1996). Poverty and child health. Medical Press.

Strand, S. (2011). The limits of social class in explaining ethnic gaps in educational attainment. British Educational Research Journal, 37(2), 197 to 229. 10.1080/01411920903540664.

Strand, S. (2014). Ethnicity, gender, social class and achievement gaps at age 16: intersectionality and ‘getting it’ for the white working class. Research Papers in Education, 29(2), 131 to 171. 10.1080/02671522.2013.767370.

Strand, S. (2015). Ethnicity, deprivation and educational achievement at age 16 in England: Trends over time (DfE Research Report 439B) . London: Department for Education.

UK Government (2020). Ethnicity fact and figures service, work, pay and benefits: Employment by occupation - GOV.UK Ethnicity facts and figures.

US Department of Education (2019). National Centre for Education Statistics, National Assessment of Educational Progress (NAEP). Retrieved from the Main NAEP Data Explorer . (Table prepared October 2019).

Appendix A: Age 16 points score by ethnic group, gender and SES from LSYPE (Strand, 2014)

The KS4 exam results for all pupils in England are available as part of the National Pupil Database (NPD), but there is only very limited data on socio-economic status (SES). The NPD contains only a single measure of SES sourced directly from the pupil, which is whether the pupil is, or is not, entitled to a free school meal (FSM), or whether they have ever been entitled to a FSM at some time in the last 6 years (EVER6). There are often criticisms that some pupils do not claim a FSM even if entitled because of the stigma, but perhaps more problematic is that a simple binary measure tells us nothing about the huge differences in home circumstances among the 85% of pupils who are not entitled to a FSM, which can range from families only just over the income threshold for FSM to those from extremely well-off circumstances.

Fortunately, there is good data on both ethnicity and SES is some of the England longitudinal datasets. For example, Strand (2014) used the Longitudinal Study of Young People in England (LSYPE) to draw on data on parents’ occupational classification, their educational qualifications, whether they owned their own home, the deprivation of the neighbourhood in which they lived as well as whether the student was entitled to a FSM, in order to create a robust and differentiated measure of the family socio-economic status (SES). The LSYPE also includes ethnic minority boosts with a target of 1,000 respondents from each of the main ethnic minority groups, so that the sample size is large enough to support robust national estimates for ethnic minority groups.

The results of the analysis are presented below.

Notes: (1). The outcome (total points score) was drawn from examinations completed in 2006, and is a measure of achievement based on all examinations completed by the young person at age 16, expressed on a scale where 0 is the mean (average) score for all Young People at age 16 and two-thirds of YP score between -1 and 1. (2). The SES measure also has a mean (average) of zero and the effects for low SES are estimated at -1SD and of high SES at +1SD. Source: See Strand (2014) for full details.

Appendix B: Indicative examples of professions in the ONS statistics socio-economic classification (ONS-SEC)

Table 7: indicative examples of professions in each reduced ns-sec class.

Table source: Office for National Statistics

Long Term Unemployed (LTU) are defined as those who have been out of work for 6 months or longer and are included as an eighth category.

Most recently this has been highlighted in the Government’s Racial Disparity Audit (RDA), as reported on the government’s Ethnicity fact and figures website. Black African pupils are 3 times more likely than White British pupils to be entitled to a free school meal, Black Caribbean pupils are 3 times more likely to live in persistent poverty than White British pupils, pupils in the Pakistani and Bangladeshi ethnic groups are more likely than other groups to live in the most disadvantaged neighbourhoods, and so on (for example, Strand, 2011).

Appendix C: Full factorial regression of Best8 score: regression coefficients and parameters

Notes: Estimated with adjustments for LSYPE3 weights and clustering at the school level.

Is this page useful?

• Yes this page is useful
• No this page is not useful

Help us improve GOV.UK

Don’t include personal or financial information like your National Insurance number or credit card details.

To help us improve GOV.UK, we’d like to know more about your visit today. Please fill in this survey .

U.S. Department of the Treasury

Post 5: racial differences in educational experiences and attainment.

This is the fifth installment in a series of blog posts on racial inequality produced by the Office of Economic Policy.  The other posts can be found at these links:  1. Racial Inequality in the United States ,  2. Racial Differences in Economic Security: The Racial Wealth Gap ,  3. Racial Differences in Economic Security: Housing , 4. Racial Differences in Economic Security: Non-housing Assets

Introduction

Free public primary and secondary education in the United States was established to ensure that all Americans have access to educational opportunity and are equipped to fully participate in our democracy. However, laws banning enslaved people from being taught to read, exclusionary Jim Crow laws, and the ruling in Plessy v. Ferguson entrenched racial segregation of public schools in the South, and, while not mandated by law, a de facto system of segregation became commonplace in Northern states at the same time. These systems were used to deprive people of color of the educational resources required to prosper in society throughout the 19th and 20th centuries.

Nearly 70 years after the landmark ruling in Brown v. Board of Education that ended legal school segregation, substantial racial disparities in educational opportunity and attainment still exist. Recognizing these disparities and understanding their determinants is important because they have stark implications for labor market outcomes, including employment, wages and earnings, and occupations and job quality, all key factors in individuals’ economic wellbeing.

In addition, since education is the bedrock of labor productivity, policies that raise the quality and quantity of education for underserved groups have and would boost productivity for the country as a whole.  One well-cited study finds that 40 percent of per capita GDP growth from the period of 1960 to 2010 can be accounted for by women and Black men entering into highly skilled occupations. [1]   Although disparity in access to high quality education is only one of many barriers faced by workers of color, these findings demonstrate the potential magnitude of the gains that would come from better fostering the talent innate in our future workforce. Indeed, researchers from the Federal Reserve Bank of San Francisco estimate that removing racial gaps in educational attainment alone—separate from any effect on employment or hours—would have increased GDP in 2019 by $190 billion. [2]

In this blog post, we discuss racial differences in K-12 and postsecondary educational experiences, highlighting current disparities and changes in these disparities during the recent past. We then discuss their key determinants and show these determinants are factors outside students’ control, including the socioeconomic status of their parents, the schools they attend, the neighborhoods in which they live, discrimination in disciplinary actions, the race of the teacher to which they are assigned, and implicit bias in teacher expectations.

Racial Disparities in Elementary and Secondary Education

Childhood educational experiences have been shown to shape academic outcomes in adolescence and later in life, impacting indicators of well-being far into adulthood. Racial differences in childhood educational experiences thus have the potential to place children of different races on different trajectories at an early age, making it more difficult for some to achieve economic security in adulthood than others.  In future blog posts in this series, we will discuss the disparities in labor market outcomes and how they are intertwined with the educational disparities outlined in today’s post.

Although school enrollment rates are similar across race and ethnicity for three- to five-year-old children, [3] research indicates that substantial gaps in reading and math achievement exist at the beginning of kindergarten as shown in Table 1. Black-white gaps in reading and math are about one-half and three-quarters of a standard deviation, respectively, and Hispanic-white gaps in both subjects are similarly large. [4] Gaps between students of any other race, a diverse category which includes Asian, Native Hawaiian/Other Pacific Islander, and American Indian/Alaska Native children in addition to children of multiple races, and white students are about 0.4 standard deviations for both reading and math. [5]

Table 1. Racial Achievement Gaps in Elementary and Middle School

Notes : Table shows estimated achievement gaps in standard deviations for each racial and ethnic group relative to white students. Standard deviation units are based on the standard deviation across all students for a given test-grade-year combination. The Asian category is comprised of Asian-origin students who are proficient in oral English at the beginning of kindergarten. Data are from the Early Childhood Longitudinal Study – Kindergarten Class of 1998-99 public-use K-9 longitudinal data set. See Reardon, Robinson-Cimpian, and Weathers (2015) for details.

Source : Table 3 of Reardon, Sean F., Joseph P. Robinson-Cimpian, and Ericka S. Weathers. 2015. “Patterns and Trends in Racial/Ethnic and Socioeconomic Academic Achievement Gaps.” In Handbook of Research in Education Finance and Policy , edited by Helen F. Ladd and Margaret E. Goertz, 499-518. New York: Routledge.

Although the Black-white and Hispanic-white gaps for reading and math are similar at the beginning of elementary school, the Black-white gaps widen as students progress through secondary school while the Hispanic-white gaps shrink. The Black-white gaps in reading and math expand from 0.53 and 0.73 standard deviations, respectively, in the fall of kindergarten to 0.95 and 1 standard deviations, respectively, in the spring of 8th grade. In contrast, the Hispanic-white reading and math gaps fall from 0.48 and 0.76 to 0.36 and 0.44, respectively, over the same period. [6]

Most of the widening of the Black-white gaps occurs in elementary school, with evidence of little change after eighth grade. [7] Similarly, most of the narrowing of the Hispanic-white gaps occurs early in elementary school, and research shows the gaps continue to narrow (less rapidly) as students progress through middle and high school. [8] These patterns, coupled with evidence that gains early in elementary school are concentrated among recent immigrants and students with low levels of English proficiency, have led some scholars to conclude that the improvements among Hispanic students are likely driven in part by the development of English language skills among these students. [9]

In addition to highlighting trends in achievement gaps as students progress through school, it is important to understand how achievement gaps have changed over time for students in a given grade. To that end, Figures 1 and 2 display differences in scores from the National Assessment of Educational Progress (NAEP) in reading and mathematics, respectively, for students of different races and ethnicities relative to white students. In 2003, the gaps between white and Black 4th grade reading and math scores were 31 and 27 points, respectively. By 2009, both gaps had closed to 26 and stayed relatively constant throughout the 2010s before widening to 28 (reading) and 29 (math) in 2022, likely due to differential impacts of pandemic-related learning disruptions. In contrast, the Hispanic-white gaps in reading and math have been steadily closing over the last 20 years, improving from 28 and 22, respectively, in 2000 to 22 and 21 in 2022.

Prior to 2011, the scores of Asian and Native Hawaiian/Other Pacific Islander students were reported together, and students in this combined group performed similarly to if not slightly better than white students in both reading and math. However, as evident from the trends for students in these two groups reported separately after 2011, this aggregation masks important heterogeneity, with Asian students consistently outperforming white students in both reading and math and Native Hawaiian/Other Pacific Islander students consistently lagging behind.

Figure 1. Racial Differences in 4th Grade Reading Achievement

Notes : Figure displays differences between the National Assessment of Educational Progress (NAEP) Reading composite scale score for 4th graders in each racial/ethnic group relative to the score for white 4th graders.

Source : Data from the U.S. Department of Education, National Center for Education Statistics, National Assessment of Educational Progress (NAEP) Reading Assessments, retrieved from the Main NAEP Data Explorer .

Figure 2. Racial Differences in 4th Grade Math Achievement

Notes : Figure displays differences between the National Assessment of Educational Progress (NAEP) Mathematics composite scale score for 4th graders in each racial/ethnic group relative to the score for white 4th graders.

Source : Data from the U.S. Department of Education, National Center for Education Statistics, National Assessment of Educational Progress (NAEP) Mathematics Assessments, retrieved from the Main NAEP Data Explorer .

High school completion rates are another important measure of academic achievement in secondary school since high school completion is required for many jobs and to pursue postsecondary education. Figure 3 shows differences in high school completion by race and ethnicity over the last three decades. Since 1992, the percent of adults 25-years-old and older who have graduated from high school has increased for all racial and ethnic groups, with the largest improvements occurring among Black and Hispanic individuals, the groups with the lowest completion rates in 1992. High school completion rates increased from 68 percent in 1992 to 91 percent in 2021 for Black individuals and from 53 percent to 74 percent over that same period for Hispanic individuals. While there is still a relatively large gap in the Hispanic high school completion rate relative to others, it is clear that progress has been and continues to be made on this dimension.

Figure 3. High School Completion Rates by Race and Ethnicity

Notes : Figure displays the percent of people 25-years-old and older who have graduated from high school by race and ethnicity.

Source : Data from the U.S. Department of Commerce, U.S. Census Bureau, Current Population Survey, Annual Social and Economic Supplement, 1990 through 2021, prepared by the National Center for Education Statistics in November 2021 and retrieved from Digest of Education Statistics 2021 , table 104.10 .

Racial Disparities in Postsecondary Education

Racial disparities in education persist beyond high school and into postsecondary education and are evident in the differences in college enrollment rates visible in Figure 4, which shows the percent of 18- to 24-year-olds in each group enrolled in college or graduate school.  College premiums remain high, such that these differences in enrollment and graduation rates will contribute to large disparities in lifetime earnings. [10]

Overall, the share of young adults enrolled in college has increased over the last three decades, with the largest gains occurring for Black and Hispanic individuals. In 1990, 25 percent of Black 18- to 24-year-olds were enrolled in college or graduate school, and by 2020, that number had risen to 36 percent. Over the same period, the share of Hispanic young adults enrolled in college or graduate school more than doubled from 16 percent to over 35 percent. White and Asian young adults, who had the highest rates of college enrollment in 1990, experienced smaller gains of roughly 6 to 7 percentage points, as did American Indian/Alaska Native young adults. [11] In contrast, the enrolled share of Native Hawaiian/Other Pacific Islander young adults and those belonging to two or more races declined between 2010 and 2020 from 36 percent to 34 percent and from 38 percent to 34 percent, respectively. [12]

Figure 4. College Enrollment by Race and Ethnicity

Notes : Figure displays the percent of 18- to 24-year-olds enrolled in college or graduate school by race and ethnicity. Data for Asian and Native Hawaiian/Other Pacific Islander individuals separately and for individuals of two or more races are not available until 2003.

Source : Data from the U.S. Department of Commerce, U.S. Census Bureau, Current Population Survey, October, 1990 through 2020 prepared by the National Center for Education Statistics in August 2021 and retrieved from Digest of Education Statistics 2021 , table 302.60 .

In addition to comparing overall college enrollment rates, it is important to consider the types of schools students enroll in. Students may enroll in institutions of different levels (two-year or four-year) and type (not-for-profit or for-profit). Given evidence of higher returns to an additional year of schooling at four-year institutions than at two-year institutions [13] and evidence that attending a for-profit institution may lead to worse post-college labor market and financial outcomes than not attending college at all, [14] differential enrollment patterns by race and ethnicity may lead to lasting differences in economic security post-college.

Figure 5 shows, for each racial and ethnic group, the distribution of undergraduates across institution levels and control of institution in Fall 2020. Asian and white undergraduates, along with undergraduates of two or more races are most likely to be enrolled at four-year institutions. In contrast, Hispanic, American Indian/Alaska Native, Black, and Native Hawaiian/Other Pacific Islander undergraduates are most likely to be enrolled at two-year institutions, although, notably, the share of Hispanic students enrolled at two-year institutions is nearly 10 percentage points larger than the share of Black students enrolled at these institutions. Black and Native Hawaiian/Other Pacific Islander students are much more likely than any other group to enrolled at for-profit institutions, both two- and four-year.

Figure 5. Racial Differences in Undergraduate Enrollment by Level and Type

Notes : Figure displays the percent of 18- to 24-year-olds enrolled in college or graduate school by race and ethnicity.

Source : Data from the U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary Education Data System (IPEDS, Spring 2021, Fall Enrollment component and retrieved from Digest of Education Statistics 2021 , table 306.50 .

These differences in enrollment patterns contribute to the racial differences in bachelor’s degree attainment that are evident in Figure 6. Since 1990, the percent of people 25 years old and older with a bachelor’s degree has increased for all racial and ethnic groups, but Black, Hispanic, Native Hawaiian/Other Pacific Islander, and American Indian/Alaska Native individuals still lag behind Asian and white individuals and those of two or more races.  

Figure 6. Racial Differences in Bachelor’s Degree Attainment

Notes : Figure displays the percent of people 25 years old and older who have earned a bachelor’s degree by race and ethnicity.

Determinants of Racial Disparities in Educational Experiences and Attainment

The role of socioeconomic status.

In this section, we will discuss the determinants of the racial disparities in education outlined above. Recall that sizeable achievement gaps exist at the beginning of kindergarten despite relatively similar pre-school enrollment rates. Therefore, these early disparities in achievement must be driven by something other than differences in access to formal schooling prior to elementary school. Research suggests that nearly all of the Black-white reading gap and over 80 percent of the Black-white math gap at the beginning kindergarten can be explained by differences in socioeconomic status and observed family background characteristics. [15] Similarly, there is evidence that socioeconomic factors and family background account for 85 percent of the Hispanic-white gap in reading achievement and 75 percent of the Hispanic-white math gap in kindergarten. [16] Socioeconomic status is measured using total family income and parental education and occupations, and family background characteristics include the age of the child’s mother at first birth, indicators for receiving aid from government social safety net programs, and the number of children’s books in the home.

The role that socioeconomic status plays in kindergarten achievement suggests that race-neutral policies designed to provide resources to underserved families may aid in closing racial achievement gaps early in elementary school. Head Start is one such policy, created in 1965 to improve school readiness of low-income three- and four-year-olds. While there is evidence that Head Start is effective at improving school readiness, there is also evidence that its effects on test scores fade out by the end of kindergarten. [17] These findings highlight both the benefits of supporting under-resourced families with young children and the limitations of temporary assistance. [18]

While most of the achievement gaps at the beginning of kindergarten are attributable to differences in socioeconomic status, evidence on the extent to which differences in socioeconomic factors can explain Black-white achievement gaps later in elementary school is somewhat mixed. Older studies find that as students progress through school, socioeconomic factors explain a smaller and smaller share of the Black-white gap, with these factors accounting for just 35 percent (or less) of the Black-white achievement gaps from third grade onward. [19] In contrast, more recent work that uses permanent (long-run) income as opposed to current income as a measure of socioeconomic status finds that these factors explain 40 to 75 percent of the gaps at the end of elementary school. [20] Similarly, recent research that takes into account the fact that test scores are imperfect measures of true skills finds that most, if not all, of the gaps throughout elementary and middle school can be explained by socioeconomic and family background factors. [21]

Research on the importance of socioeconomic factors in explaining Hispanic-white achievement gaps in elementary school is less mixed. Most research finds that these factors can explain an increasing share of Hispanic-white achievement gaps as students move through elementary school, so much so that estimates of the gap that are adjusted for differences in socioeconomic status disappear completely by fifth grade. [22]

The Role of Schools

While research suggests observable family background characteristics contribute substantially to racial achievement gaps in elementary and secondary school, there is an active debate about the extent to which schools reduce or exacerbate those disparities.

First, there is evidence that racial segregation in school districts is strongly associated with the magnitude of racial achievement gaps. [23] This is particularly concerning because although it has been nearly 70 years since the end of legal segregation, many of America’s public schools remain segregated by race and ethnicity today: 31 and 23 percent of Hispanic and Black students, respectively, attended a predominately same-race-or-ethnicity school, where 75 percent or more of the students are of their own race or ethnicity, in the 2020-2021 school year. [24] Additional research suggests that school segregation itself may have no effect on achievement gaps independent of the effects of neighborhood segregation, [25] so understanding the mechanisms through which school and neighborhood segregation impact racial achievement gaps (e.g., school and teacher quality, school funding, peer effects, crime, etc.) together and in isolation and their relative importance is a critical area for future research.

Second, differences in discipline rates by race and ethnicity may contribute to differences in achievement and educational attainment across groups. There is evidence both that being formally disciplined in school negatively impacts educational attainment, [26] and that Black public school students are disproportionately disciplined (via suspensions and expulsions) relative to public school students of other racial and ethnic groups for the same violations.  These disparities persist no matter the type or poverty level of the public school attended. [27]

Finally, schools may impact racial differences in achievement and completion through the (non-disciplinary) interactions teachers have with students. For example, there is evidence that same-race elementary school teachers increase the likelihood of high school graduation and college enrollment among Black students, with role model effects as a potential mechanism, [28] and that teacher expectations – even overly optimistic ones – impact the likelihood of college graduation. [29]

Existing research on the determinants of racial college enrollment and completion gaps indicates that nearly all these gaps are attributable to differences in pre-college characteristics, including family socioeconomic status, other family background variables, and pre-college test scores, the latter of which, as discussed earlier in this blog post, are also largely determined by family characteristics. For example, there is evidence that once pre-college characteristics are taken into account, Black and Hispanic students are more likely to attend and complete colleges of all quality levels than white students. [30] These findings suggest that to improve racial gaps in college attendance and completion, policymakers should focus on improving educational opportunity much earlier in life and economic opportunity for families more broadly.

Racial disparities in educational experiences and attainment begin early in life and persist as individuals progress through school. Recognizing these disparities and understanding their determinants is important because they have stark implications for labor market outcomes, including employment, wages and earnings, and occupations and job quality, all of which are key determinants of economic wellbeing.

In particular, it is important to recognize that the key determinants of many of these disparities are factors outside students’ control, including the socioeconomic status of their parents, the schools they attend, the neighborhoods in which they live, discrimination in disciplinary actions, the race of the teacher to which they are assigned, and implicit bias in teacher expectations. Reducing these economic and educational disparities is an important policy goal, not only because of the benefits for individual students, but also because of the benefits to our national economy.

Overall, the findings in this blog post highlight both the limited progress that has been made over the last three decades in closing racial gaps in educational attainment, as well as the substantial work that remains to ensure equal access to high quality educational opportunities for all Americans.

[1] Hsieh, Chang-Tai, Erik Hurst, Charles I. Jones, and Peter J. Klenow. 2019. “The Allocation of Talent and U.S. Growth.” Econometrica, 87 (5): 1439-1474.

[2] See Table 3 of Buckman, Shelby R., Laura Y. Choi, Mary C. Daly, Lily M. Seitelman. 2021. “The Economic Gains From Equity.” Federal Reserve Bank of San Francisco Working Paper Series, Working Paper 2021-11. Retrieved from https://www.frbsf.org/economic-research/publications/working-papers/2021/11/ .

[3] School enrollment rates for three- to four-year-olds were relatively similar across race and ethnicity in the decade leading up to the COVID-19 pandemic, ranging from 57 percent for Asian children to 46 percent for Hispanic children. Enrollment rates vary even less across race and ethnicity for five-year-olds, who are more likely to be enrolled in elementary school than younger children, ranging from 90 to 92 percent. In 2020, enrollment rates for all racial, ethnic, and age groups dropped due to the pandemic, with Hispanic, Asian, and American Indian/Alaska Native three- and four-year-olds experiencing the largest declines. Enrollment rates for white and Asian three-and four-year-olds rebounded in 2021 to or beyond their 2019 levels, but rates for Black and Hispanic young children and young children of two or more races have yet to fully recover. It is also important to note that young children with no formal schooling prior to elementary school may have received educational instruction from a stay-at-home parent or other caretaker, so differences in school enrollment rates prior to elementary school may not fully capture differences in educational experiences for young children. Source: U.S. Department of Commerce, U.S. Census Bureau, Current Population Survey, October, 2010 through 2021 prepared by the National Center for Education Statistics in October 2022 and retrieved from Digest of Education Statistics 2022, table 202.20 .

[4] Standard deviation units are based on the standard deviation across all students for a given test-grade-year combination. Source: Reardon, Sean F., Joseph P. Robinson-Cimpian, and Ericka S. Weathers. 2015. “Patterns and Trends in Racial/Ethnic and Socioeconomic Academic Achievement Gaps.” In Handbook of Research in Education Finance and Policy, edited by Helen F. Ladd and Margaret E. Goertz, 499-518. New York: Routledge.

[5] See footnote 2.

[6] See footnote 2.

[7] See the following studies:

• Clotfelter, Charles T., Helen F. Ladd, and Jacob L. Vigdor. 2009. “The Academic achievement Gap in Grades 3 to 8.” The Review of Economics and Statistics 91 (2) 398-419.
• Neal, Derek. 2006. “Why Has Black-White Skill Convergence Stopped?” In Handbook of the Economics of Education, edited by Eric A. Hanushek and Finis Welch, 511-576. New York: Elsevier.

It is important to note that differential dropout patterns in high school may bias estimates of achievement gaps beyond eighth grade.

[8] Clotfelter, Charles T., Helen F. Ladd, and Jacob L. Vigdor. 2009. “The Academic achievement Gap in Grades 3 to 8.” The Review of Economics and Statistics 91 (2) 398-419.

[9] See the following studies:

• Reardon, Sean F., and Claudia Galindo. 2009. “The Hispanic-White achievement Gap in Math and Reading in the Elementary Grades.” American Educational Research Journal 46 (3): 853-96.
• Reardon, Sean F., Joseph P. Robinson-Cimpian, and Ericka S. Weathers. 2015. “Patterns and Trends in Racial/Ethnic and Socioeconomic Academic Achievement Gaps.” In Handbook of Research in Education Finance and Policy, edited by Helen F. Ladd and Margaret E. Goertz, 499-518. New York: Routledge.

[10] Hendricks, Lutz and Oksana Leukhina. 2018. “The Return to College: Selection and Dropout Risk.” International Economic Review 59 (3): 1077-1102.

[11] The sample size of the American Indian/Alaska Native group is small, such that enrollment rate estimates have standard errors above 5 in each year.  The large measurement error may contribute to the anomalous reading in 2010. 

[12] Due to small sample sizes, the standard errors for the enrollment rates of both of these groups are quite high—greater than 8 for Native Hawaiian/Other Pacific Islander, and greater than 3 for two or more races.

[13] Kane, Thomas J., and Cecilia Elena Rouse. 1995. “Labor Market Returns to Two-Year and Four-Year Schools. American Economic Review 85 (3): 600–614.

[14] See the following studies:

• Armona, Luis, Rajashri Chakrabarti, and Michael F. Lovenheim. 2018. “How Does For-Profit College Attendance Affect Student Loans, Defaults and Labor Market Outcomes?” NBER Working Paper 25041.
• Cellini, Stephanie Riegg, and Latika Chaudhary. 2014. “The Labor Market Returns to For-Profit College Education.” Economics of Education Review 43: 125-140.
• Cellini, Stephanie Riegg, and Nicholas Turner. 2019. “Gainfully Employed? Assessing the Employment and Earnings of For-Profit College Students Using Administrative Data.” The Journal of Human Resources 52(4): 342-370.
• Cottom, Tressie McMillan. 2017. Lower Ed: The Troubling Rise of For-Profit College in the New Economy. New York: The New Press.
• Deming, David J., Claudia Goldin, and Lawrence F. Katz. 2012. “The For-Profit Postsecondary School Sector: Nimble Critters or Agile Predators?” Journal of Economic Perspectives 26 (1): 139-64.

[15] See the following studies:

• Fryer, Ronald G., and Steven D. Levitt. 2004. “Understanding the Black-White Test Score Gap in the First Two Years of School.” Review of Economics and Statistics 86 (2): 447-64.
• Fryer, Ronald G., and Steven D. Levitt. 2006. “The Black-White Test Score Gap through Third Grade.” American Law and Economics Review 8 (2): 249-281.
• Murnane Richard J., John B. Willet, Kristen L. Bub, and Kathleen McCartney. 2006. “Understanding Trends in the Black-White achievement Gaps during the First Years of School.” Brookings-Wharton Papers on Urban Affairs 97-135.

[16] See footnote 4.

[17] See the following studies:

• Kline, Patrick, and Christopher R. Walters. 2016. “Evaluating Public Programs with Close Substitutes: The Case of Head Start.” The Quarterly Journal of Economics 131 (4): 1795-1848.
• U.S. Department of Health and Human Services, Administration for Children and Families. 2010. Head Start Impact Study Final Report. Washington, DC. Retrieved from https://www.acf.hhs.gov/sites/default/files/documents/opre/executive_summary_final_508.pdf .

It is also important to note that recent work provides evidence that Head Start positively impacts later-in-life outcomes including years of education and wage income with stronger effects for Black and Hispanic individuals. See:

• De Haan, Monique and Edwin Leuven. 2020. “Head Start and the Distribution of Long-Term Education and Labor Market Outcomes.” Journal of Labor Economics 38 (3) 727-765.

[18] Johnson and Jackson (2019) find that the benefits of Head Start were larger when followed by access to better-funded schools, highlighting the potential benefits of early educational investments that are sustained.

• Johnson, Rucker C., and C. Kirabo Jackson. 2019. “Reducing Inequality through Dynamic Complementarity: Evidence from Head Start and Public School Spending.” American Economic Journal: Economic Policy 11 (4): 1-40.

[19] See the following studies:

[20] Rothstein, Jesse, and Nathan Wozny. 2013. “Permanent Income and the Black-White Test Score Gap.” The Journal of Human Resources 48 (3): 510-544.

[21] Bond, Timothy N., and Kevin Lang. 2018. “The Black-White Education Scaled Test-Score Gap in Grades K-7.” The Journal of Human Resources 53 (4): 891-917.

[22] See the following studies:

[23] Reardon, Sean F., Erick S. Weathers, Erin M. Fahle, Heewon Jang, and Demetra Kalogrides. 2022. “Is Separate Still Unequal? New Evidence on School Segregation” Stanford Center for Education Policy Analysis (CEPA) Working Paper No. 19-06. Retrieved from https://cepa.stanford.edu/sites/default/files/wp19-06-v082022.pdf .

[24] U.S. Government Accountability Office. 2022. “K-12 Education: Student Population Has Significantly Diversified, but Many Schools Remain Divided Along Racial, Ethnic, and Economic Lines.” GAO-22-104737. Retrieved from https://www.gao.gov/products/gao-22-104737 .

[25] Card, David, and Jesse Rothstein. 2007. “Racial Segregation and the Black-White Test Score Gap.” Journal of Public Economics 91 (11-12): 2158-2184.

[26] Bacher-Hicks, Andrew, Stephen B. Billings, and David J. Deming. 2019. “The School to Prison Pipeline: Long-Run Impacts of School Suspensions on Adult Crime.” NBER Working Paper 26257.

[27] U.S. Government Accountability Office. 2028. “K-12 Education: Discipline Disparities for Black Students, Boys, and Students with Disabilities.” GA)-18-258. Retrieved from https://www.gao.gov/products/gao-18-258 . 

[28] Gershenson, Seth, Cassandra M. D. Hard, Joshua Hyman, Constance A. Lindsay, and Nicholas W. Papageorge. 2022. “The Long-Run Impacts of Same-Race Teachers.” American Economic Journal: Economic Policy 14 (4): 300-342.

[29] Papageorge, Nicholas W., Seth Gershenson, and Kyung Min Kang. 2020. “Teacher Expectations Matter.” The Review of Economics and Statistics 102 (2): 234-251.

[30] See the following studies on the importance of pre-college characteristics for racial gaps in college attendance and completion:

• Flores, Stella M., Toby J. Park, and Dominique J. Baker. 2017. “The Racial College Completion Gap: Evidence from Texas.” The Journal of Higher Education 88 (6): 894-921.
• Hinton, Ivora, Jessica Howell, Elizabeth Merwin, Steven N. Stern, Sarah Turner, Ishan Williams, and Melvin Wilson. 2010. “The Educational Pipeline for Health Care Professionals.” The Journal of Human Resources 45 (1): 116-156.
• Light, Audrey, and Wayne Strayer. 2002. “From Bakke to Hopwood: Does Race Affect College Attendance and Completion?” The Review of Economics and Statistics 84 (1): 34-44.

Brown v. Board of Education: 70 Years of Progress and Challenges

• Share article

After 70 years, what is left to say about Brown v. Board of Education ?

A lot, it turns out. As the anniversary nears this week for the U.S. Supreme Court’s historic May 17, 1954, decision that outlawed racial segregation in public schools, there are new books, reports, and academic conferences analyzing its impact and legacy.

Just last year, members of the current Supreme Court debated divergent interpretations of Brown as they weighed the use of race in higher education admissions, with numerous references to the landmark ruling in their deeply divided opinions in the case that ended college affirmative action as it had been practiced for half a century.

Meanwhile, some school district desegregation cases remain active after more than 50 years, while the Supreme Court has largely gotten out of the business of taking up the issue. There are fresh reports that the nation’s K-12 schools, which are much more racially and ethnically diverse than they were in the 1950s, are nonetheless experiencing resegregation .

At an April 4 conference at Columbia University, speakers captured the mood about a historic decision that slowly but steadily led to the desegregation of schools in much of the country but faced roadblocks and new conditions that have left its promise unfulfilled.

“I think Brown permeates nearly every aspect of our current modern society,” said Janai Nelson, the president and director-counsel of the NAACP Legal Defense and Educational Fund, the organization led by Thurgood Marshall, who would later become a Supreme Court justice, during the Brown era.

“I hope that we see clearly now that there is an effort to roll back [the] gains” brought by the decision, said Nelson, whose organization was a conference co-sponsor. “There is an effort to recast Brown from what it was originally intended to produce. If we want to keep this multiracial democracy and actually have it fulfill its promise, because the status quo is still not satisfactory, we must look at the original intent of this all-important case and make sure we fulfill its promise.”

Celebrations at the White House, the Justice Department, and a Smithsonian Museum

On May 16, President Joe Biden will welcome to the White House descendants of the original plaintiffs in the cases that were consolidated into Brown , which dealt with cases from Delaware, Kansas, South Carolina, and Virginia. (The companion decision, Bolling v. Sharpe , decided the same day, struck down school segregation in the District of Columbia.) On May 17, the president will deliver remarks on the historic decision at the Smithsonian Institution’s National Museum of African American History and Culture.

Attorney General Merrick B. Garland and U.S. Secretary of Education Miguel Cardona marked the anniversary at an event at the U.S. Department of Justice on Tuesday.

“ Brown vs. Board and its legacy remind us who we want to be as a nation, a place that upholds values of justice and equity as its highest ideals,” Cardona said. “We normalize a culture of low expectations for some students and give them inadequate resources and support. Today, it’s still become all too normal for some to deny racism and segregation or ban books that teach Black history when we all know that Black history is American history.”

On May 17, 1954, then-Chief Justice Earl Warren announced the decision for a unanimous court that held that “in the field of public education, ‘separate but equal’ has no place. Separate educational facilities are inherently unequal.”

That opinion was a compromise meant to bring about unanimity, and the court did not even address a desegregation remedy until a year later in Brown II , when it called for lower courts to address local conditions “with all deliberate speed.”

“In short, the standard the court established for evaluating schools’ desegregation efforts was as vague as the schedule for achieving it was amorphous,” R. Shep Melnick, a professor of American politics at Boston College and the co-chair of the Harvard Program on Constitutional Government, says in an assessment of the Brown anniversary published this month by the American Enterprise Institute.

The paper distills a book by Melnick published last year, The Crucible of Desegregation: The Uncertain Search for Educational Equity , which takes a fresh look at the 70-year history of post-Brown desegregation efforts.

Melnick argues that even after 70 years, Brown and later Supreme Court decisions remain full of ambiguities as to even what it means for a school system to be desegregated. He highlights two competing interpretations of Brown embraced by lawyers, judges, and scholars—a “colorblind” approach prohibiting any categorization of students by race, and a perspective based on racial isolation and equal educational opportunity. “Neither was ever fully endorsed or rejected by the Supreme Court,” Melnick writes in the book. “Both could find some support in the court’s ambiguous 1954 opinion.”

The Supreme Court issued some 35 decisions on desegregation after Brown , but hasn’t taken up a case involving a court-ordered desegregation remedy since 1995 and last spoke on the issue of integration and student diversity in the K-12 context in 2007, when the court struck down two voluntary plans to increase diversity by considering race in assigning students to schools.

Citations to Brown pervade last year’s sharply divided opinions over affirmative action

Chief Justice John G. Roberts Jr., in his plurality opinion in that voluntary integration case, Parents Involved in Community Schools v. Seattle School District , laid the groundwork for last year’s affirmative action decision, which fully embraced Brown’s “race-blind” interpretation.

Last term, the high court ruled that race-conscious admissions plans at Harvard and the University of North Carolina violated the 14th Amendment’s equal protection clause. (The vote was 6-2 in Students for Fair Admissions v. President and Fellows of Harvard College , with Justice Ketanji Brown Jackson not participating because of her recent membership on a Harvard governing board. The vote was 6-3 in SFFA v. University of North Carolina .)

The Brown decision was a running theme in the arguments in the case, and in the some 230 pages of opinions.

Roberts, in the majority opinion, said a fundamental lesson of Brown in 1954 and Brown II in 1955 was that “The time for making distinctions based on race had passed.”

Brown and a generation of high court decisions on race that followed, in education and other areas, “reflect the core purpose of the Equal Protection Clause: doing away with all governmentally imposed discrimination based on race,” the chief justice wrote.

Justice Clarence Thomas, who succeeded Thurgood Marshall, joined the majority opinion and wrote a lengthy concurrence that touched on views he had long expressed about the 1954 decision. He cited the language of legal briefs filed by the challengers of segregated schools in the Brown cases (led by Marshall) that embraced the view that the 14th Amendment barred all government consideration of race.

Thomas said those challenging segregated schools in Brown “embraced the equality principle.”

Justice Brett M. Kavanaugh also joined the majority and acknowledged in his concurrence that in Brown , the court “authorized race-based student assignments for several decades—but not indefinitely into the future.”

(The other justices in the majority were Samuel A. Alito Jr., Neil M. Gorsuch, and Amy Coney Barrett.)

Writing the main dissent, Justice Sonia Sotomayor rejected the view that Brown was race-blind.

“ Brown was a race-conscious decision that emphasized the importance of education in our society,” she wrote, joined by justices Elena Kagan and Jackson. “The desegregation cases that followed Brown confirm that the ultimate goal of that seminal decision was to achieve a system of integrated schools that ensured racial equality of opportunity, not to impose a formalistic rule of race-blindness.”

Jackson, in a separate dissent (joined by Sotomayor and Kagan), said, “The majority and concurring opinions rehearse this court’s idealistic vision of racial equality, from Brown forward, with appropriate lament for past indiscretions. But the race-linked gaps that the law (aided by this court) previously founded and fostered—which indisputably define our present reality— are strangely absent and do not seem to matter.”

Amid reports on resegregation, some legal efforts continue

As the Brown anniversary arrives, there are fresh reports about resegregation of the schools. Research released this month by Sean Reardon of Stanford University and Ann Owens of the University of Southern California found that students in the nation’s large school districts have become much more isolated racially and economically in recent years.

The Civil Rights Project at the University of California, Los Angeles, which has been sounding the alarm about resegregation for years, says in a new report that Black and Latino students were the most highly segregated demographic groups in 2021. Though U.S. schools were 45 percent white, Blacks, on average, attended 76 percent nonwhite schools, and Latino students went to 75 percent nonwhite schools.

The CRP says the Brown anniversary is worth celebrating, but “American schools have been moving away from the goal of Brown and creating more ‘inherently unequal’ schools for a third of a century. We need new thought about how inequality and integration work in institutions and communities with changing multiracial populations with very unequal experiences.”

At the Columbia conference, Samuel Spital, the litigation director and general counsel of the Legal Defense Fund, noted that many jurisdictions are still under desegregation orders, some going back decades.

He highlighted one where LDF lawyers have been in federal district court, involving the 7,200-student St. Martin Parish school district in western Louisiana. Black plaintiffs first sued over segregated schools in 1965. In a 2022 decision, the U.S. Court of Appeals for the 5th Circuit, in New Orleans, noted that the case had been pending for “five decades,” though largely inactive for long stretches. The court nonetheless affirmed the district court’s continued supervision of a desegregation plan that addressed disparities in graduation tracks and student discipline, though it said the court overstepped in ordering the closure of an elementary school in a mostly white community.

As recently as this month, the LDF and the Department of Justice’s civil rights division joined with the St. Martin Parish school board in a proposed consent order for revised attendance zones for the district’s schools. The proposed order suggests that court supervision of student assignments could end sometime after June 2027.

“We try to make sure that with the vast docket of segregation cases we have, that we have not lost sight of what Brown’s ultimate intent was,” said LDF’s Nelson, which was not just “to make sure that Black and white children learn together” but also to foster principles of equity and citizenship.

With a hostile federal court climate, advocates more recently have turned to state constitutions and state courts to pursue desegregation. Last year, a state judge in New Jersey allowed key claims to proceed in a lawsuit that seeks to hold the state responsible for remedying racial segregation in its many “racially isolated” public schools. In December, the Minnesota Supreme Court allowed a suit under the state constitution to move forward, ruling that there was no need for plaintiffs to prove that the state itself had caused segregation in its schools.

“We see a path forward through state courts with the very specific goal of trying to challenge state practices, which really boil down to segregative school district lines,” Saba Bireda, the chief legal counsel of Brown’s Promise , said at the Columbia conference. Bireda, a former civil rights lawyer in the Education Department under President Barack Obama’s administration, co-founded the Washington-based organization last year to help address diversity and underfunding in public schools.

A Supreme Court exhibit offers the idealized take on Brown

At the Supreme Court, there has been no formal recognition of the 70th anniversary of Brown . But the court did open an exhibit on its ground floor late last year that tells the story of some of the first desegregation cases, including Brown .

The exhibit is primarily about the Little Rock integration crisis of 1957, when Arkansas Gov. Orval Faubus defied a federal judge’s order to desegregate Central High School. The exhibit is built around the actual bench used by Judge Ronald N. Davies when he heard a challenge to Faubus’ use of the Arkansas National Guard to prevent the nine Black high school students from entering the all-white high school that year. (Davies withstood threats and intense opposition from desegregation opponents, but he ruled for the Black students. The Supreme Court itself supported desegregation in Little Rock with its 1958 decision in Cooper v. Aaron .)

To tell the Little Rock story, the exhibit starts with Brown (and some of the prior history). A central feature is a 15-minute video featuring all current members of the court.

In the video, the justices set aside their differences over the meaning of Brown and provide a more idealized perspective on the 1954 decision.

“ Brown was a godsend,” Thomas says in the video. “Because it said that what was happening that we thought was wrong, they now know that this court said it was also wrong. It’s wrong not just morally, but under the Constitution of the United States. It was like a ray of hope.”

Kavanaugh says: “ Brown vs. Board of Education is the single greatest moment, single greatest decision in this court’s history. And the reason for that is that it enforced a constitutional principle, equal protection of the laws, equal justice under law. It made that real for all Americans. And it corrected a grave wrong, the separate but equal doctrine that the court had previously allowed.”

Jackson, the court’s third Black justice, who has spoken of her family moving in one generation from “segregation to the Supreme Court,” reflects in the video on Brown ‘s legacy.

“I think I’m most grateful for the fact that my parents have lived to see me in this position, after a history of them and others in our family and people from my background not having the opportunity to live to our fullest potential,” she says.

As the video comes to a close, Roberts speaks with evident pride in his voice.

“The Supreme Court building stands as a symbol of our country’s faith in the rule of law,” the chief justice says. “ Brown v. Board of Education , the great school desegregation case, was decided here.”

Sign Up for EdWeek Update

Edweek top school jobs.

Sign Up & Sign In