research paper on class differences

Class differences

Social status isn't just about the cars we drive, the money we make or the schools we attend — it's also about how we feel, think and act, psychology researchers say.

By Tori DeAngelis

February 2015, Vol 46, No. 2

Print version: page 62

11 min read

University of California, Irvine, professor Paul Piff, PhD, starts his courses on class differences by asking students about their consumer habits: Do they shop at J.C. Penney or Neiman Marcus? What kind of car do they drive, if they drive at all? What is their preferred breakfast, a fruit smoothie from Starbucks or a Dunkin' Donut?

"As people reconstruct their days, it's clear that in every single decision they make, class is an essential feature," says Piff.

The implications are larger than breakfast choice, he adds. "Class affects whether someone is going to be accepted into a particular kind of school, their likelihood of succeeding in that school, the kinds of jobs they have access to, the kinds of friends they make" — in essence, the degree of status, power and perks people enjoy or lack in their daily lives. 

But until the last decade or so, the concept of class has generally eluded psychological inquiry. While sociologists and epidemiologists have examined its effects in broad domains such as health outcomes and mortality, few researchers have explored how we process class internally and psychologically.

Yet several factors make the psychology of class an increasingly important topic to study, some researchers say. One is the widening gulf between rich and poor, and the potentially negative effects this gap has in areas including health, well-being, self-image, relationships, stereotyping and prejudice.

Studying the psychology of class is also important because it puts a contextual spin on what has largely been an individually oriented view of psychological processes, says Michael Kraus, PhD, who studies class at the University of Illinois at Urbana-Champaign. "It suggests that the contexts we grow up in and are socialized in are an important part of what shapes the self," he says.  

That said, these researchers see class on a continuum, rather than as a fixed distinction among upper, middle and lower class. In their view, the higher in socioeconomic status you are, the more independently oriented you are likely to be, while the lower in status you are, the more group-minded you are likely to be, for example.

"At least in the studies we've run so far, we've found that middle-class folks are more independent than lower-class folks, but less so than their upper-class counterparts," Kraus explains.

A theory of classism

In a 2012 paper in Psychological Review , Kraus, Piff, University of California, Berkeley, psychology professor Dacher Keltner, PhD, and colleagues posit that social class — which they define as "a social context that individuals inhabit in enduring and pervasive ways over time" — is a fundamental lens through which we see ourselves and others. Because lower ranking people have fewer resources and opportunities than those of relatively high rank, they tend to believe that external, uncontrollable social forces and others' power have correspondingly greater influence over their lives. Success for them, therefore, depends on how well they can "read," rely on and help out others, the psychologists' theory holds.

By contrast, those who enjoy more resources and greater class status live in contexts that enhance their personal power and freedom — larger and safer living spaces, the means to buy high-priced goods and experiences, and education that provides access to influential people, ideas and venues. These conditions give rise to a more self-focused approach to life, the theory states.

"With wealth and privilege comes this island of sorts, this increased insularity from others," as Piff puts it. 

Another important aspect of the theory is that rank is, in part, subjective and relative. All relationships are marked by class scrutiny: Am I higher or lower than this person? Research also shows that people tend to be quite accurate in their assessment of their own and others' class rank, and that this self-assessment likewise predicts outcomes. For example, people who perceive themselves as lower in rank have worse health outcomes overall than those who see themselves as higher ranking, research finds.

Class effects

Given the advantages that come with higher class, it's not surprising that those of higher rank tend to deploy actions and attitudes that maintain or justify their position. A 2013 paper by Kraus and Keltner in the Journal of Personality and Social Psychology , for example, found that people who see themselves as relatively high class are more likely than those who see themselves as lower in rank to view class as inherent, innate and fixed. Higher-class people also are more likely to endorse punishment over rehabilitation for criminal offenses, and to see the world as a just and fair place.

"We're finding that the super wealthy tell a story about why they have what they have," says Keltner. "In essence, they believe they're a different kind of person, with genes more suitable to success."

People of higher self-reported socioeconomic status and more educated parents also score higher on measures of entitlement and narcissistic personality tendencies than people not in their class, finds a 2014 article by Piff in Personality and Social Psychology Bulletin . Research suggests that perhaps because of this sense of entitlement, higher-class people can behave more selfishly and less ethically than lower-income peers. In a 2012 article in the Proceedings of the National Academy of Sciences , Piff and colleagues describe two studies in which observers watching from the street recorded driving behavior over two days. In the first study, drivers of high-status vehicles were far more likely than others to cut off other drivers at a busy four-way intersection. In the second study, approximately half of the drivers in the highest status vehicles drove illegally through a crosswalk as a pedestrian was waiting to cross, versus none in the lowest status vehicles.

These findings suggest that cultural context and its resulting mental habits allow people of higher classes to disconnect from others' concerns, says Keltner. "To be compassionate, you have to carefully attend to other people — to what they're thinking, feeling and saying," Keltner says. "The wealthy don't do that as well as poorer people — not because they don't have those capabilities, but because the context of their lives allows them to disengage." In other words, having more space, material goods, money and free time makes it easier for wealthy people to buy their way out of problems, take a vacation when things get stressful, or otherwise avoid or mitigate everyday stresses. Consider the person who can afford to have a contractor re-do her kitchen, versus someone who must borrow money for the job, try to do the job herself, or simply live with old equipment that doesn't function properly or is even dangerous.

One outcome of these differences is that people of lower rank tend to be more emotionally attuned to others, these researchers contend. A 2010 paper in Psychological Science by Kraus and colleagues, for example, reports that less educated people are better than more educated peers at identifying emotions on faces. They also are more accurate at reading a stranger's emotions during a group job interview. Another study finds that people with less income and education are more generous, trusting and helpful than their wealthier, more educated counterparts ( Journal of Personality and Social Psychology , 2010).

Not all psychological factors associated with being on the higher end of the social hierarchy are negative, however. One psychological plus may be that people with power and influence have more freedom to be themselves without worrying about adjusting to others' expectations or wishes. A 2011 study by Kraus and colleagues in the Journal of Experimental Social Psychology , for instance, finds that people with more self-defined power — an ingredient often associated with higher classes — were more likely than low-power people to report having a coherent self-view.

"The good news about having influence and control is it's really freeing," Kraus says. "High-power people stay authentically the same person no matter the context. But people who are relatively low-power change little aspects of themselves because having low power means having to adapt and fit in to different contexts."

Fortunately, it's also relatively easy for wealthier people to relearn connection and compassion. In Piff's study of entitlement, he asked high-status people either to list three benefits of regarding others as equals or three things they did during an average day, testing them with measures of narcissism and sense of entitlement before and after the exercise. The scores of wealthier participants in the "equal" group dropped significantly, he found, while those of wealthier people in the second group remained at the same high levels.

"If you bridge the island that separates the wealthy from the rest of the world, then all of a sudden empathy gets restored," Piff says.

Class culture

In another line of research, Stanford University psychologist Hazel Markus, PhD — well known for her work on how sociocultural factors such as race, gender and ethnicity influence our thoughts, feelings and self-perceptions — has been applying this framework to class.

In a 2005 paper in the Journal of Personality and Social Psychology reporting on three studies, Markus and her then-graduate student Alana Conner Snibbe, PhD, showed that college-educated participants were much more upset when they didn't receive an item they had hand-picked than were working-class peers. Like the theory proposed later by Kraus, Piff and Keltner, they also found that working-class people tend to believe that maintaining relationships and fitting in are more important than expressing preferences and standing out in a crowd ( Journal of Personality and Social Psychology , 2007).

Markus and an international team of researchers are now examining how class intersects with other forms of culture, such as nationality. In a 2013 article in Emotion , Jiyoung Park, PhD, Markus and colleagues found that class influences the way people in different countries view and express emotions. Comparing American and Japanese respondents, for instance, they corroborated an earlier finding that lower-status Americans are more likely than higher-status Americans to express anger, especially when it involves frustration. In Japan, however, those of higher social standing were more likely to express anger than lower-status participants, especially when it involved making important decisions.

As this work suggests, class is not set in stone, an observation some researchers are using to help people from working-class backgrounds better understand their culture and hence perform better in school.

In the June 2012 issue of the Journal of Personality and Social Psychology , Northwestern University psychologist Nicole M. Stephens, PhD, Markus and colleagues demonstrated that simply giving a certain kind of welcoming message to incoming first-generation college students from working-class families can profoundly influence their performance. One study found that when these students received independent messages highlighting college as a place for personal exploration and individual achievement, they performed worse than middle-class peers on verbal and spatial tasks. But when they received interdependent messages — greetings that included references to their families and to college as a place to collaborate with their peers — they performed as well as students from college-educated families, a finding replicated at several universities using a number of different tasks.

Similarly, Stephens and colleagues report in an April study in Psychological Science on what happened when they randomly referred incoming freshmen from working-class backgrounds to one of two student-led discussion panels. In one panel, juniors and seniors talked informally about how their class backgrounds raised obstacles to college success and how they overcame those obstacles. In the other, they talked about obstacles and overcoming them, but without reference to class.

At the end of their first year, working-class students who attended the "class" condition had much higher grade-point averages than those in the control condition, and about the same GPAs as students from higher class backgrounds, the team found.

The studies suggest that "if we can raise people's awareness about how people's social class backgrounds matter in college," says Stephens, "we can give them insights that can help them to better navigate their college experience."

Meanwhile, Indiana University sociologist Jessica McCrory Calarco, PhD, has been looking into what might cause cultural differences in academic attitudes and performance in the first place. For two years, she observed a cadre of working- and middle-class kids from the third to the fifth grade, and interviewed the kids, their parents and teachers. In a study in the October American Sociological Review , she reports that middle-class youngsters who were struggling received more attention from teachers because they more actively sought it out, while working-class kids tended to keep quiet because they didn't want to bother the teachers.

Interviews with parents shed further light on these behaviors: Middle-class parents perceived it as their right and duty to take part in the system, while working-class parents felt it rude to insert themselves too much in their children's schooling. As a result, working-class parents "tended to be less aware of what teachers expect today, and hence less apt to encourage their children to seek help with their school challenges," Calarco says.

As the world continues to shrink, it's more important than ever that we understand the subjective nature of such cultural dichotomies, Markus adds.

"Social class differences come about because of the ideas and values you are surrounded by, the types of social interactions you have at home, school and work, and the sorts of institutional practices and policies that are common in your community," she says. "That means that these differences are not immutable."

Tori DeAngelis is a writer in Syracuse, New York.

Further reading

• Fiske, S. & Markus, H. R. (2012).  Facing social class: How societal rank influences interaction . New York: Russell Sage Foundation.
• Jensen, B. (2012). Reading Classes: On Culture and Classism in America . Ithaca, N.Y.: ILR Press. 
• Markus, H. R., & Conner, A. C. (2014).  Clash! How to Thrive in a Multicultural World . New York: Plume. 

Letters to the Editor

• Send us a letter

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.

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
• Springer Nature - PMC COVID-19 Collection
• PMC10155160

The Effects of Different Types of Classism on Psychological Outcomes: Preliminary Findings

Klaus e. cavalhieri.

1 Educational and Counseling Psychology, University at Albany, SUNY, Catskill 260, 1400 Washington Ave., Albany, NY 26012222 USA

2 Counseling Psychology and Community Services, University of North Dakota, 231 Centennial Dr Stop 7189, Grand Forks, ND USA

Amanda Willyard

Justin c. phillippi.

In interpersonal relationships, people make assumptions about others’ social class standing and interact with them based on these assumptions, which constitutes classism. Classism has an adverse impact on people’s overall functioning, although scholarly attention on the unique impact of different types of classism, as proposed the Social Class Worldview Model-Revised (SCMW-R; Liu, 2011), has lagged behind. To address this gap in the literature, we explored how different types of classism (i.e., downward, upward, and lateral) can account for unique variance as predictors of psychological outcomes. Overall, our findings indicate that there is a unique impact of different types of classism on psychological outcomes (i.e., stress, anxiety, well-being, attitudes toward mental health care), beyond social status and overall discrimination alone.

Introduction

Social class has a significant impact on people’s overall development, as it relates to one’s access to wealth, privilege, and status (Liu, 2011a , b ; Noonan & Liu, 2022 ). Historically, research in social class has focused on sociological approaches, mostly using distal variables (e.g., income, occupation) to characterize social class groups (Lau et al., 2013 ). Albeit useful, these objective measures rely on the assumption that people with the same income experience the world similarly. However, earning US$ 40,000.00 a year in metropolitan Chicago is significantly different from earning the same amount in rural North Dakota. Furthermore, using occupation as a proxy for social class may be problematic, as it does not account for professions that have high income and low perceived social status (e.g., plumbers, welders). An alternative to the distributional model of social class (i.e., based on income, education, possessions) is the relational model, through which social class is understood in the dimensions of authority, oppression, exploitation, acculturation, and stress, based on which groups control resources (Liu, 2011a , b ).

In line with the relational model, Liu ( 2011a , b ) proposed a subjective model of social class (Social Class Worldview Model-Revised; SCWM-R), in which social class is understood as a worldview through which people perceive the world around them, filtering the information they receive. As people become aware of their own social position and status in the economic hierarchy, they interact with their environment accordingly, based on the norms and values (i.e., economic culture) from their own perceived social class group. From this approach, people make assumptions regarding the social class standing of other people based on contextual variables (e.g., fashion choices, race, language), and interact with them accordingly. The process of categorizing others and interacting with them according to their perceived social class standing characterizes classism.

Social Class

Although social class is an important part of one’s identity, the construct has been defined in a myriad of ways in research. Liu and colleagues ( 2004 ) found that over 400 different words had been used to describe social class in a 19-year span. More recently, researchers have found an upward trend in research on social class (Cook et al., 2019 ), culminating in the Guidelines for Psychological Practice for People with Low-Income and Economic Marginalization (APA, 2019). The Social Class Worldview Model (SCWM-R; Liu, 2011a , b ) is a phenomenological approach to the study of social class, focusing on subjective worldviews that serves as a lens on how people perceive information and interact with others.

The factors people take in consideration in deciding which social class group someone belongs to are contextual. As people who are wealthy are more likely to create and maintain wealth due to intergenerational inheritance, real estate, and stock market fluctuation, structural barriers (e.g., institutional racism) can serve as obstacles to wealth accumulation (Keister & Moller, 2000 ). As wealth is transmitted intergenerationally through inheritance, racial inequalities are maintained through racial segregation, systemic discrimination, and the stripping of the health and wealth of African Americans as a result of slavery (Gaskin et al., 2005 ; Strand, 2010 ). These intergenerational processes and the inherent social stratification changes how people categorize others in social class groups, as it maintains stereotypical assumptions of how wealthy or poor people look like, permeated by racialized assumptions.

According to the SCWM-R, people become aware of their own social position in the economic hierarchy and create in- and out-groups within the hierarchy. As people classify others in a social class group, they interact with them accordingly (Liu, 2011a , b ). Given this subjective nature, different types of classism exist (i.e., downward, upward, lateral, internalized). Downward classism refers to classism toward those perceived to be from a lower social class, such as believing they are unintelligent, or dirty. Upward classism refers to classism toward those perceived to be from a higher social class, such as believing they are out of touch, or elitist. Lateral classism refers to classism toward those perceived to be in the same social class group, such as receiving and communicating that one does not fit in their social class group, reminding people of one’s shortcomings that are not congruent with the expectations of their social class group. Lastly, internalized classism refers to internalized assumptions and myths about social class (e.g., upward mobility bias, meritocracy) that can lead to experiences of frustration and anxiety, particularly as one feels unable to maintain their social class status (Liu, 2011a , b ). Whereas downward, upward, and lateral classism are experienced in interpersonal relationships, internalized classism is associated with self-devaluation and overall anxiety regarding one’s social class standing (Liu & Cavalhieri, 2022 ; Liu, 2011a , b ).

Although the impact of economic expectations is pervasive in capitalist societies, the implications of downward and upward classism cannot and should not be compared (Cavalhieri & Chwalisz, 2020 ; Liu & Cavalhieri, 2022 ; Liu, 2011a , b ). People from lower social classes do not have the same societal power, and even though they might hold prejudice toward people in higher social classes, they do not have political power to support institutions that marginalize or discriminate against the wealthy (Smith, 2005 ). Nevertheless, recognizing how living in a capitalist society impacts people’s economic assumptions is important, leading to inherent biases that affect interpersonal relationships in every social class group (APA, 2019 ).

Furthermore, although internalized classism has a significant impact on one’s behavior, the experience is significantly different from other types of classism. Whereas internalized classism is an intrapsychic experience, in which distress stems from internalized classist myths and feeling unable to maintain one’s social class standing (Liu & Cavalhieri, 2022 ), downward, upward, and lateral classism are all experienced interpersonally, based on how people perceive one’s social class standing, and the distress associated with these classist assumptions (Liu, 2011a , b ). As such, internalized classism is a qualitatively different experience, and cannot be measured the same way as downward, upward, and lateral classism (Cavalhieri & Chwalisz, 2020 ). Garrison and colleagues ( 2022 ) have found that Chinese international students build their social class worldview partially based on how they construe and cope with their experiences of classism, highlighting how one’s subjective social class is not related solely to one’s income, but also how they perceive others’ behaviors. For that reason, in the current study, we focused on classism experienced interpersonally, opposed to the intrapsychic experience of internalized classism.

Effects of Classism

To our knowledge, the differential impact of classism types (e.g., downward, upward, lateral) has not been empirically addressed, partially due to the lack of measurement tools designed to measure the different types of classism proposed by the SCWM-R (Lau et al, 2013 ; Liu, 2011a , b ). However, classism has been found to be a significant overall stressor, impacting people’s overall mental health.

Downward Classism

Choi and Miller ( 2018 ) found in a large sample of undergraduate students ( N  = 2230) that subjective social status was associated with experiences of classism, which was a significant predictor of mental health stigma and more negative attitudes toward seeking mental health care. On the other hand, Duffy and colleagues ( 2021 ) recently found that undergraduate students who were marginalized and economically constrained were more likely to identify as a student of color, and more likely to seek counseling while in college. These mixed findings underscore the importance of investigating how classism and social class experiences impact attitudes toward mental health care, as economically marginalized groups are more likely to experience psychological distress, at the same time that existing group norms might prevent them from seeking counseling. Furthermore, both of these studies were conducted in college settings, with samples who are likely high academic achievers, which might not accurately represent the experiences of community dwelling adults.

People who experience downward classism were found to have more restrict work volition and career adaptability (Kim & Allan, 2021 ), and to be less connected and valued by their undergraduate institution (Alan et al., 2016 ; Garriott et al., 2021 ). Experiencing institutional classism has been found to be a predictor of interpersonal classism and lower work volition, which in turn was associated with greater life and academic satisfaction (Allan et al., 2021 ). Furthermore, people who are economically marginalized were less likely to comply with shelter-in-place recommendations during the COVID-19 pandemic, increasing their risk of contagium (Cavalhieri, 2021 ). Chronic pain patients’ perceived social status has also been associated with dehumanizing assumptions made by nurses, which adversely impacted pain care and management, assuming low-income patients were “passive” and had poor prospects of improvement (Diniz et al., 2020 ). Furthermore, classism was found to be a significant stressor for African Americans, with greater experiences of classism been significantly associated with higher stress, higher depression, and lower well-being (Cavalhieri & Wilcox, 2022 ).

Upward Classism

In a large cross-cultural study, Zitelmann ( 2020 ) found that the wealthy tend to be perceived as morally corrupt, profit-hungry, and cold-hearted. Brockmann and colleagues ( 2021 ) conducted a text analysis on the twitter feed of tech elites (defined as the 100 wealthiest people in the tech industry) and found that they tend to hold more meritocratic view in comparison to general population who use twitter. These findings highlight the uniquely different social class experiences of the wealthy, and how assumption made about them (i.e., upward classism) could also have a psychological impact. Nevertheless, the impact of downward and upward classism cannot be equated—as the poor do not have power to discriminate against the wealthy or support institutions that harm them. But as the wealthy are part of the economic hierarchy, assumptions made about them continue to maintain a subservient position of the poor and a system that prevents upward mobility (Liu & Cavalhieri, 2022 ).

Affluent teens have been found to be susceptible to parental criticism, which contributed to internalizing and externalizing psychopathology (Stiles et al., 2020 ). Parental control strategies and high achievement expectations have also been found to be a predictor of risk behaviors for adolescents from higher socioeconomic status (Romm et al., 2020 ), and affluent adolescent girls have been found to experience stress and psychosomatic symptoms in response to increased parental criticism in relation to expected achievement (Williams et al., 2018 ). Furthermore, Black youth with self-reported high subjective social status have also been found to be more vulnerable to developing depressive symptoms, partially due to an increase in perceived experiences of discrimination (Assari et al., 2018 ). Assumptions placed on affluent adolescents appear to be associated with expectations of high achievement and of being spoiled and selfish, highlighting the likely adverse impact of upward classism on overall psychological symptoms (e.g., depression, stress, anxiety).

Lateral Classism

Lateral classism has been scarcely researched, so hypotheses about its impact on psychological functioning are tentative. As a unique type of social class discrimination, lateral classism is expressed toward others perceived to be in the same social class group (Liu, 2011a , b ). Bellet ( 2019 ) found that satisfaction with one’s house size was closely associated with upward comparisons with the size of other houses in the neighborhood. The author found that once large houses (i.e., “McMansions”; larger than 90% of local house size distribution) were built, neighbors’ satisfaction with their own house significantly dropped. In a similar study, Kuhlmann ( 2020 ) investigated the impact of house relative size in its neighborhood on one’s overall satisfaction with their residence. The author found that the relative size and positionality of house in the community was significantly related to satisfaction, even after controlling for tenure in the house, number of rooms, income, and absolute unit size. Their results imply that one’s social position in relation to a particular reference group (e.g., neighborhood) significantly impact their decisions where to live and how satisfied they are with their residence. Although studies on the impact of relative house size (Bellet, 2019 ; Kuhlmann, 2020 ) may provide some insight into how comparison to neighbors impacts one’s well-being, the construct of lateral classism is not directly measured, which poses a significant limitation.

Present Study

Classism is a multifaceted construct, as people across the economic hierarchy appear to be affected by social class assumptions. Although there is significant variability on how social class and classism have been operationalized in the literature (Allan et al., 2021 ; Diniz et al., 2020 ; Kuhlmann, 2020 ; Romm et al., 2020 ; Zitelmann, 2020 ), the different types of classism have been widely associated with attitudes toward seeking mental health care (Choi & Miller, 2018 ; Duffy et al., 2021 ), stress (Assari et al., 2018 ; Cavalhieri & Wilcox, 2022 ; Garriott et al., 2021 ), anxiety (Stiles et al., 2020 ; Williams et al., 2018 ), and worse well-being (Allan et al., 2021 ; Kuhlmann, 2020 ). Overall, classism has been investigated as a unidimensional construct in the literature, neglecting how context and social class assumptions impact one’s functioning. As such, it is paramount to investigate the multidimensionality of classism, and partial out how different types of classism may have a unique effect on psychological outcomes, particularly anxiety, stress, well-being, and attitudes toward mental health care.

The present study was designed to investigate whether different types of classism (e.g., downward, upward, and lateral) would be unique predictors of mental health outcomes and attitudes toward mental health services. The differential impact of classisms has not been empirically addressed, due to the lack of psychometrically sound measures. To address this gap, Cavalhieri and Chwalisz ( 2020 ) developed a scale of perceived experiences of classism, based on the Social Class Worldview Model. Their scale (i.e., the Perceived Classism Experiences Scale; PCES) has three subscales: downward, upward, and lateral classism. Overall, classism has been found to be associated with more negative attitudes toward mental health care (Choi & Miller, 2018 ), lower well-being and life satisfaction (Allan et al., 2021 ; Stiles et al., 2020 ), and higher stress and anxiety (Cavalhieri & Wilcox, 2022 ; Williams et al, 2018 ). Hence, we hypothesized that different types of classism would account for unique variance in predicting attitudes toward mental health services, anxiety, stress, and well-being. We also hypothesized that the different classisms would be significant predictors of the outcome variables beyond subjective social status and perceived overall discrimination.

Participants

A priori power analysis for a multiple regression indicated that at least 114 subjects would be necessary to identify an effect size of 0.10, maintaining a power of 0.80 and an alpha of 0.05. A total of 143 people completed the survey, and 21 people (14.68%) were removed from further analyses as they did not answer at least one attention checking correctly (e.g., “please select strongly disagree”). The final sample consisted of 122 participants, with an average age of 36.25 ( SD  = 11.18). All participants lived in the USA at the time of data collection, and 12.3% identified as immigrants (i.e., had moved to the USA at some point during their lives). Participants’ information on social class standing, wealth, and racial identity can be found in Table ​ Table1 1 .

Demographic information

Demographic Questionnaire

Demographic questionnaire included questions on age, gender, race, education, employment, wealth, debt, and perceived social class growing up.

Classism was operationalized with the Perceived Classism Experience Scale (PCES; Cavalhieri & Chwalisz, 2020 ). The PCES is a measure of perceived classism, in which the 18 items are scored in a 5-point Likert-type scale. Higher scores indicate higher perceived experiences of classism. The PCES is grounded on the SCWM-R, and as such, its subscales are proposed to reflect three types of classism: downward, upward, and lateral. For the purpose of this study, the three subscale scores were used to operationalize classisms (i.e., downward, upward, and lateral). In their original study, Cavalhieri and Chwalisz ( 2020 ) found high internal reliability for all subscales (ω downward  = 0.92, ω upward  = 0.92, ω lateral  = 0.89). The authors also reported significant correlations between the subscales and self-rated health, anxiety, stress, and negative life satisfaction, providing supportive evidence for the scale’s criterion validity. The Cronbach’s alpha for the three subscale items in the current study were high, downward, α = 0.93, upward, α = 0.95, and lateral, α = 0.91.

Stress was measured with the Perceived Stress Scale (PSS; Cohen et al., 1983 ). The PSS is a measure how much people perceive their lives to be stressful, with higher scores indicating greater distress. The PSS is a widely used measure to assess and operationalize psychological distress, with strong supportive evidence of its criterion validity. The 14 items are scored on a 5-point Likert-type scale. In their original study, the authors found high internal consistency for the PSS (α = 0.85). An example of an item would be “In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?” The Cronbach’s alpha for this sample was 0.81.

Anxiety was measured with the Penn State Worry Questionnaire (PSWQ; Meyer et al., 1990 ).The PSWQ is a 16-item scale, developed to measure the trait of worry. The PSWQ is one of the most widely used measures of overall worry and anxiety in the psychological literature, with strong supportive evidence of criterion validity. Scores range from 16 to 80, with higher scores indicating greater overall worry. In their original study, Meyer and colleagues (1990) found excellent internal consistency for the PSWQ (α = 0.93). An example of an item would be “When I am under pressure I worry a lot.” The Cronbach’s alpha for this sample was 0.89.

Well-being was operationalized with the Oxford Happiness Questionnaire (OHQ; Hills & Argyle, 2002 ). The OHQ is a unidimensional measure of personal happiness and psychological well-being, with items scored on a six-point Likert-type scale. We have operationalized well-being as having higher scores on the OHQ. Scores are obtained by summing all items, and higher scores indicate greater psychological well-being. Hills and Argyle ( 2002 ) found excellent internal consistency for the OHQ in their original study (α = 0.91). The authors provided evidence of criterion validity, as their scale was positively correlated with extraversion, life-satisfaction, and self-esteem. An example of an item would be “I am well satisfied about everything in my life.” The Cronbach’s alpha for this sample was 0.89.

Attitudes Toward Mental Health

Attitudes toward mental health services was measured with the Mental Health Seeking Attitudes Scale (MHSAS; Hammer et al., 2018 ). The MHSAS is a 9-item bipolar scale, developed to measure people’s attitudes to seeking help from a mental health professional. Based on the Theory of Planned Behavior, the MHSAS’ items are purported to reflect attitudes toward seeking mental health services, and the authors found incremental variance was accounted by the MHSAS in comparison to previous instruments on attitudes toward mental health. The MHSAS is scored by adding item scores and dividing it by the total number of answered items. The authors found excellent internal consistency for the MHSAS on their original study (α = 0.94), found their scale was significantly correlated to other existing measures of attitudes toward mental health services, and that their scale accounted for unique variance in help-seeking intention in comparison to these existing scales. The Cronbach’s alpha for this sample was 0.88.

Overall Discrimination

Overall discrimination was assessed with the Everyday Discrimination Scale-Revised (EDS-R; Stucky et al., 2011 ). The EDS-R is a short 5-items scale that is purported to measure daily perceived discrimination. Stucky and colleagues ( 2011 ) conducted an item analysis of the EDS using item-response theory, to provide cross-validated evidence of its dimensionality. The authors found that the revised version of the EDS was a significant predictor of negative health outcomes and represented a unidimensional set of items. The scale was also significantly associated with overt discrimination measures and depressive symptoms. They found adequate reliability of the measure ( r  = 0.84), after removing items with local dependency issues. The EDS-R is scored by transforming summed scores into IRT/factor scaled scores (a conversion table is provided by the authors in their original study). Scaled scores have a mean of 20 and standard deviation of 10. An example item for the EDS-R is “You are treated with less respect than others.”

Data Collection

Following approval from the IRB, participants were recruited from a crowdsourcing platform. The participants completed an online survey, in which all scales (with the exception of the demographic questionnaire, which was presented last) were randomized to control for order effects. Multiple validity indicators were embedded in the survey to prevent bot responses and inattentive answers (Chmielewski & Kucker, 2020 ). The survey was set up to prevent multiple submissions from a single respondent (i.e., “ballot box stuffing”), and was embedded with a reCAPTCHA code to flag potential bot responses (Kennedy et al., 2020 ). A third-party provider was also employed (CloudResearch; Litman et al., 2017 ) to track and block suspicious geolocations, duplicate IPs, and to vet participants who passed engagement and attention measures, to ensure high quality data.

Data Analysis

A multivariate multiple regression (MMR) was run to test our hypotheses on SPSS. As the MMR partials out the variance of all predictors, it is possible to test the unique contribution to the prediction for each type of classism. As such, an MMR was run (opposed to a linear regression), to test the differential impact of downward, upward, and lateral classism on the outcome variables. The outcome variables were stress, anxiety, well-being, and attitudes toward mental health. Age, subjective social status, and gender (0 = male, 1 = female) were controlled for in our analyses. Demographic variables (i.e., age, gender, and subjective social status), overall discrimination, and classisms (downward, upward, and lateral) were entered as predictors.

To investigate how well different types of classism predict mental health symptoms and attitudes toward mental health services, a multivariate multiple regression (MMR) was computed. All tolerance scores were above 0.2, and all VIF scores were below 10, indicating there was no evidence of multicolinearity issues (Tabachnick & Fidell, 2013 ). Assumptions of normality and homoscedasticity were also met, based on the evaluation of Q-Q plots and scatterplots. Two multivariate outliers were removed (Mahalanobis Distance > 22.21, df  = 5, p  < 0.001). Correlations, mean, and standard deviations for all variables can be found on Table ​ Table2. 2 . There were no significant differences between men and women (all identified as cisgender on this sample), downward classism, t (120) = 0.91, p  = 0.364, lateral classism, t (120) = 1.46, p  = 0.145, overall discrimination t (120) = 1.51, p  = 132, attitudes toward mental health, t (120) = -0.896, p  = 0.372, well-being, t (120) = -1.26, p  = 0.210, anxiety, t (120) = -0.101, p  = 0.919, and stress, t (120) = 0.770, p  = 0.443. The only exception was upward classism, t (120) = 2.04, p  = 0.04, with participants who identified as men having higher scores.

Correlation, mean scores, and standard deviation of all variables

N  = 122. Perceived Classism Experiences Scale, Downward Classism Subscale a , Upward Classism Subscale b , and Lateral Classism Subscale c . d Everyday Discrimination Scale – Scaled Scores. e Perceived Stress Scale. f Penn State Worry Questionnaire. g Oxford Happiness Questionnaire. h Mental Health Seeking Attitudes Scale * p  ≤ .05. ** p  ≤ .01. ^the EDS scores are scaled, with a fixed mean of 20 and a standard deviation of 10

We hypothesized that different types of classism (i.e., downward, upward, and lateral) would be significantly related to stress, well-being, anxiety, and attitudes toward mental health care, above and beyond overall discrimination individually, explaining unique variance on the outcome variables. The overall multivariate model was significant for subjective social status ( λ  = 0.835, p  < 0.001, η p 2  = 0.16), downward classism ( λ  = 0.885, p  < 0.001, η p 2  = 0.11), upward classism ( λ  = 0.914, p  = 0.04, η p 2  = 0.09), and lateral classism ( λ  = 0.744, p  < 0.001, η p 2  = 0.26), but not for gender ( λ  = 0.982, p  = 0.74), age (λ = 0.941, p  = 0.15), or overall discrimination (λ = 0.973, p  = 0.56). A summary of the MMR analyses can be found on Table ​ Table3. 3 . Univariate results indicated that subjective social status was related to well-being, and negatively related to stress and anxiety, but not related to attitudes toward mental health care. Furthermore, the univariate results for the different types of classism were significant. Downward classism was significantly associated with all outcome variables (stress, anxiety, well-being, and attitudes toward mental health care), indicating people with higher scores on downward classism had more severe psychological symptoms and more negative attitudes toward mental health care. Both upward and lateral classism were significantly associated with stress, anxiety, and well-being, but not to attitudes toward mental health care, indicating people with higher scores on the upward and lateral classism subscales had more severe mental health symptoms, although it did not account for any variance on participants’ attitudes toward mental health care. As the results from MMR analyses partials out all other contributors, different types of classism (i.e., downward, upward, and lateral) were significant and unique predictors of the psychological outcomes measured. Based on partial eta-squared scores ( η p 2 ), downward classism accounted for 8% variance of stress, 5% variance of anxiety, 4% variance of well-being, and 6% variance of attitudes toward mental health care. Upward classism accounted for 3% variance of stress, 6% of anxiety, and 6% of well-being, whereas lateral classism accounted for the largest variance on the outcome variables, accounting for 16% of stress, 22% of anxiety, and 13% of well-being. Taken together, our hypotheses were partially supported. These preliminary results contribute to the evidence that different types of classism have a unique impact on psychological outcomes, which warrants further scholarly attention.

Summary of multivariate multiple regression on stress, anxiety, well-being, and attitudes toward mental health

N  = 120. a Man = 0; woman = 1. b Perceived Stress Scale. c Penn State Worry Questionnaire. d Oxford Happiness Questionnaire. e Mental Health Seeking Attitudes Scale. * p  ≤ .05, ** p  ≤ .01

The purpose of this study was to investigate whether different types of classism had a negative impact on psychological symptoms and attitudes toward mental health services in an adult community sample. Our results suggest there is a differential impact of downward, upward, and lateral classism on overall psychological functioning. Our findings were congruent with extant literature on the adverse impact of classism on mental health (Cavalhieri & Wilcox, 2022 ; Choi & Miller, 2018 ; Kim & Allan, 2021 ). Overall, we expected different types of classism to be unique predictors of mental health symptoms, which was supported by our findings, as greater experiences of classism were associated with worse mental health outcomes and a reduced likelihood of seeking mental health services.

To our knowledge, this is the first preliminary empirical study that specifically investigates the adverse effects of different types of classism, as proposed by the SCWM-R. Our novel findings corroborate with the theoretical tenets of the Social Class Worldview Model-Revised (Liu, 2011a ; Liu & Cavalhieri, 2022 ; Noonan & Liu, 2022 ), providing supportive evidence of the differential and potentially additive effect of different types of classism. In our sample, downward classism was significantly associated with all outcome variables (i.e., stress, anxiety, well-being, attitudes toward mental health care). Downward classism was found not only to contribute to more severe psychological distress, but also appeared to prevent people from seeking psychological services. This particular finding is congruent with existing literature on classism, as a significant stressor that impacts access to mental health services (Choi & Miller, 2018 ) and increases distress (Garriott et al., 2021 ; Kim & Allan, 2021 ).

However, the relationship between classism and mental health appears to be complex and non-linear. In our study, both lateral and upward classism were associated with stress, anxiety, and well-being, but were not related to one’s attitudes toward mental health services. Although one’s positionality on the social class hierarchy impacts the type of classism they experience, and consequently serves as a potential stressor (Noonan & Liu, 2022 ), the impact appears to be significant different. Upward and lateral classism appear to impact overall mental health, which is corroborated by previous literature (Kuhlmann, 2020 ; Romm et al., 2020 ; Stiles et al., 2020 )—however, participants’ perceived higher social status likely serves as a buffer to the adverse effects of classism, given the increased access to resources and lower barriers to political and social engagement (Ettman et al., 2020 ).

Of particular note, lateral classism appeared to be a significant stressor for our sample. The magnitude of the effects of lateral classism was markedly higher in comparison to upward and downward classism. Our findings appear to suggest that people who compare themselves to others in their own social class group (i.e., lateral classism) experience more negative psychological outcomes. Categorizing people in social class groups appears to be independent from objective social class indicators (Eshelman & Rottinghaus, 2015 ; Kraus et al., 2017 ) and appears to be associated with social comparison, which is a subjective experience. Racialized assumptions of what the poor and the rich look like, maintained by White Supremacy ideals, leads to contextually based experiences of classism (Liu, 2017 ). Therefore, people compare themselves to others they perceive to be in their social class group in attempt to maintain homeostasis on their social class worldview. In these comparisons, people are reminded they do not “fit in,” leading to increased distress (Noonan & Liu, 2022 ).

Limitations

Several limitations on this study must be noted. Generalizability concerns must be noted, as participants were recruited online—and people without internet who would likely experience social class discrimination were not included in our sample, furthering the digital divide for people with limited resources. Furthermore, our sample had a small number of Latino/a/x folks, which may hinder the applicability of our results to this particular segment of the population. Our study also had a cross-sectional design, which prevents us from determining a causal relationship of long-term effects of different classisms. Although downward, upward, and lateral classism were significant predictors of psychological outcomes, other types of discrimination (e.g., sexual orientation, disability, race) also impact people’s psychological functioning, and future research should attend to how the intersection of marginalized identities (e.g., classist racism ) may compound to cause worse psychological outcomes.

Implications for Research and Practice

Our preliminary findings imply that classism is a significant stressor to overall mental health, and people have different experiences of classism depending on their positionality on the social class hierarchy. Our findings appear to support the theoretical propositions of the SCWM-R, particularly how the web of classism experiences can shape one’s social class worldview (Garrison et al., 2022 ; Liu, 2011a , b ; Noonan & Liu, 2022 ). By separating the types of classism and understanding how each may have a differential psychological impact, counselors may be better positioned to better explore and conceptualize clients through a social class lens (Liu & Cavalhieri, 2022 ). Although these are promising results, it is paramount to replicate it in future research. By understanding how downward classism is a uniquely different experience compared to lateral or upward classism may allow counselors to explore how status, wealth, and social comparison serve as a potential buffer or hindering effect in regards to one’s psychological health. As such, counselors can provide specific outreach and preventive care to vulnerable communities by knowing how they are impacted by social class expectations, as well as to develop collective plans to fight against classism.

Based on our preliminary findings that lateral classism specifically (i.e., comparing oneself to your own social class group and coming out wanting) contributes to more negative psychological outcomes, working to support a client’s narrative that reduces stigma around seeking mental health services may be helpful in client outreach and retention. In social class comparisons, people create a narrative that they do not “fit in” due to perceived differences, which leads to increased distress (Noonan & Liu, 2022 ). Creating an environment that normalizes seeking mental health therapy within social class groups may provide opportunities to challenge lateral comparisons of mental health support and reduce distress related to mental health comparison in the client’s outside social relationships.

Classism appears to be a multidimensional construct, with a significant impact on one’s mental health and overall well-being. However, the effects of classism may be covert and gradual, and counselors can gently challenge classist assumptions present on client’s discourse (e.g., meritocracy, protestant work ethic, “pulling yourself up by your bootstraps”), so clients can more critically engage with their own inter- and intra-personal experiences and notice how classism has impacted their lived experiences. By helping clients connect some of their present concerns (e.g., anxiety, stress) to their own experiences of classism may be an important path to counteract the adverse effects of classism, as conscientization of how oppressive systems impact us can be liberating by itself (Mosley et al., 2021 ).

Furthermore, classism is not a static experience, but dynamic and contextually dependent. Understanding classism solely through a distal lens (i.e., resource and prestige, such as which group has more or less resources) appears to obscure the complexities of the phenomenon. Although the experiences of different types of classism cannot be equated, as counselors, it is important to attend to how the entire web of classist-related assumptions impacts our clients. Helping our clients develop a Social Class and Classism Consciousness (Liu, 2011a , b ) may help clients better understand the class-related constraints imposed on them. By challenging clients’ rigid beliefs in regard to status and inequality, counselors may help unpack assumptions stemming from their own economic cultures. To specifically address clients’ experiences of classism, counselors can inquire about experiences within their economic culture (i.e., norms and rules from clients’ own social class group)—for example, what was expected of them as they were growing up? Was the client expected to value hard work to the detriment of their own mental health? Did they value the belief that merit was the only factor involved in succeeding in life (i.e., meritocracy)? By helping clients name their experiences, a renewed sense of agency is possible, which can in turn alleviate distress. Nevertheless, it is important to note that classist myths and ideological frames are built around white privileges and worldviews and are maintained by white supremacy (Liu & Cavalhieri, 2022 ). These ideologies and myths would likely fail to explain the experiences of people of color, as these classist assumptions (e.g., meritocracy, upward mobility bias, protestant work ethic) do not address the effects of race and racism. As such, the class-related experiences of people of color would likely be different, which underscores the importance of helping clients develop a sense of social class and classism consciousness (Liu, 2011a , b ).

Declarations

The authors declare no competing interests.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Allan BA, Garriott PO, Keene CN. Outcomes of social class and classism in first- and continuing-generation college students. Journal of Counseling Psychology. 2016; 63 (4):487–496. doi: 10.1037/cou0000160. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Allan, B. A., Garriott, P., Ko, S.-J. “S.”, Sterling, H. M., & Case, A. S. (2021). Classism, work volition, life satisfaction, and academic satisfaction in college students: A longitudinal study. Journal of Diversity in Higher Education . Advance online publication. 10.1037/dhe0000221
• American Psychological Association. (2019). Guidelines for psychological practice for people with low-income and economic marginalization . Retrieved from www.apa.org/about/policy/guidelines-lowincome.pdf [ PubMed ]
• Assari S, Preiser B, Lankarani MM, Caldwell CH. Subjective socioeconomic status moderates the association between discrimination and depression in African American youth. Brain Sciences. 2018; 8 :71. doi: 10.3390/brainsci8040071. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bellet C. The McMansion effect: Top size inequality, house satisfaction and home improvement in U.S. suburbs. SSRN Electronic Journal. 2019 doi: 10.2139/ssrn.3378131. [ CrossRef ] [ Google Scholar ]
• Brockmann H, Drews W, Torpey J. A class for itself? On the worldviews of the new tech elite. PLoS ONE. 2021; 16 (1):e0244071. doi: 10.1371/journal.pone.0244071. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cavalhieri KE, Chwalisz K. Development and initial validation of the Perceived Classism Experiences Scale. The Counseling Psychologist. 2020; 48 (3):310–341. doi: 10.1177/0011000019899395. [ CrossRef ] [ Google Scholar ]
• Cavalhieri KE, Wilcox MM. The compounded effects of classism and racism on mental health outcomes for African Americans. Journal of Counseling Psychology. 2022; 69 (1):111–120. doi: 10.1037/cou0000561. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cavalhieri, K. (2021). Economic insecurity as a risk factor during the COVID-19 pandemic. Journal of Health Disparities Research and Practice, 14 (1), 79–93. Available at: https://digitalscholarship.unlv.edu/jhdrp/vol14/iss1/8
• Chmielewski M, Kucker SC. An MTurk crisis? Shifts in data quality and the impact on study results. Social Psychological and Personality Science. 2020; 11 (4):464–473. doi: 10.1177/1948550619875149. [ CrossRef ] [ Google Scholar ]
• Choi N-Y, Miller MJ. Social class, classism, stigma, and college students’ attitudes toward counseling. The Counseling Psychologist. 2018; 46 (6):761–785. doi: 10.1177/0011000018796789. [ CrossRef ] [ Google Scholar ]
• Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. Journal of Health and Social Behavior. 1983; 24 (4):385–396. doi: 10.2307/2136404. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cook JM, Clark M, Wojcik K, Nair D, Baillargeon T, Kowalik E. A 17-year systematic content analysis of social class and socioeconomic status in two counseling journals. Counseling Outcome Research and Evaluation. 2019; 11 (2):104–118. doi: 10.1080/21501378.2019.1647409. [ CrossRef ] [ Google Scholar ]
• Diniz E, Castro P, Bousfield A, Bernardes SF. Classism and dehumanization in chronic pain: A qualitative study of nurses’ inferences about women of different socio-economic status. British Journal of Health Psychology. 2020; 25 (1):152–170. doi: 10.1111/bjhp.12399. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Duffy, R. D., Kim, H. J., Boren, S., Pendleton, L., & Perez, G. (2021). Lifetime experiences of economic constraints and marginalization among incoming college students: A latent profile analysis. Journal of Diversity in Higher Education. Advance Online Publication. 10.1037/dhe0000344
• Eshelman AJ, Rottinghaus PJ. Viewing adolescents’ career futures through the lenses of socioeconomic status and social class. The Career Development Quarterly. 2015; 63 (4):320–332. doi: 10.1002/cdq.12031. [ CrossRef ] [ Google Scholar ]
• Ettman CK, Cohen GG, Galea S. Is wealth associated with depressive symptoms in the United States? Annals of Epidemiology. 2020; 43 :25–31. doi: 10.1016/j.annepidem.2020.02.001. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Garriott, P. O., Ko, S.-J., Grant, S. B., Jessen, M., & Allan, B. A. (2021). When race and class collide: Classism and social-emotional experiences of first-generation college students. Journal of College Student Retention: Research, Theory, & Practice. Advance Online Publication. 10.1177/1521025121995483
• Garrison, Y., Park, S., Yeung, C. W., Li, Z., Ho, Y. C. S., & Chang-Tran, J. (2023). The social class worldviews of Chinese international students in the United States. Journal of International Students, 13 (1), 40–58. 10.32674/jis.v13i1.4013
• Gaskin DJ, Headen AE, White-Means SI. Racial disparities in health and wealth: The effects of slavery and past discrimination. The Review of Black Political Economy. 2005; 32 (3–4):95–110. [ Google Scholar ]
• Hammer JH, Parent MC, Spiker DA. Mental Help Seeking Attitudes Scale (MHSAS): Development, reliability, validity, and comparison with the ATSPPH-SF and IASMHS-PO. Journal of Counseling Psychology. 2018; 65 (1):74–85. doi: 10.1037/cou0000248. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hills P, Argyle M. The Oxford Happiness Questionnaire: A compact scale for the measurement of psychological well-being. Personality and Individual Differences. 2002; 33 (7):1073–1082. doi: 10.1016/S0191-8869(01)00213-6. [ CrossRef ] [ Google Scholar ]
• Keister LA, Moller S. Wealth inequality in the United States. Annual Review of Sociology. 2000; 26 (1):63–81. doi: 10.1146/annurev.soc.26.1.63. [ CrossRef ] [ Google Scholar ]
• Kennedy R, Clifford S, Burleigh T, Waggoner PD, Jewell R, Winter NJG. The shape of and solutions to the MTurk quality crisis. Political Science Research and Methods. 2020; 8 (4):614–629. doi: 10.1017/psrm.2020.6. [ CrossRef ] [ Google Scholar ]
• Kim T, Allan BA. Examining classism and critical consciousness within psychology of working theory. Journal of Career Assessment. 2021; 29 (4):644–660. doi: 10.1177/1069072721998418. [ CrossRef ] [ Google Scholar ]
• Kraus MW, Park JW, Tan JJX. Signs of social class: The experience of economic inequality in everyday life. Perspectives on Psychological Science. 2017; 12 (3):422–435. doi: 10.1177/1745691616673192. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kuhlmann D. Coveting your neighbour’s house: Understanding the positional nature of residential satisfaction. Housing Studies. 2020; 35 (6):1142–1162. doi: 10.1080/02673037.2019.1651832. [ CrossRef ] [ Google Scholar ]
• Lau MY, Cho RJ, Chang JJ, Huang J. Measurement and methodological issues in social class research: A call for theorization and study. In: Liu WM, editor. The Oxford Handbook of Social Class in Counseling. 1. Oxford University Press; 2013. pp. 59–78. [ Google Scholar ]
• Litman L, Robinson J, Abberbock T. TurkPrime.com: A versatile crowdsourcing data acquisition platform for the behavioral sciences. Behavior Research Methods. 2017; 49 (2):433–442. doi: 10.3758/s13428-016-0727-z. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Liu WM. Social class and classism in the helping professions: Research, theory, and practice. Sage Publications; 2011. [ Google Scholar ]
• Liu WM. Developing a social class and classism consciousness. In: Altmaier EM, Hansen JC, editors. The Oxford Handbook of Counseling Psychology. Oxford University Press; 2011. pp. 326–345. [ Google Scholar ]
• Liu, W. M. (2017). White male power and privilege: The relationship between White supremacy and social class. Journal of Counseling Psychology, 64 (4), 349–358. 10.1037/cou0000227
• Liu WM, Ali SR, Soleck G, Hopps J, Dunston K, Pickett T., Jr Using social class in counseling psychology research. Journal of Counseling Psychology. 2004; 51 (1):3–18. doi: 10.1037/0022-0167.51.1.3. [ CrossRef ] [ Google Scholar ]
• Liu, W. M., & Cavalhieri, K. E. (2022). Internalized classism. In J. E. Estrellado, L. S. Felipe, & J. E. Celestial (Eds.), Clinical Interventions for Internalized Oppression (pp. 269–293). Solana Beach: Cognella.
• Meyer TJ, Miller ML, Metzger RL, Borkovec TD. Development and validation of the Penn State Worry Questionnaire. Behaviour Research and Therapy. 1990; 28 (6):487–495. doi: 10.1016/0005-7967(90)90135-6. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Mosley DV, Hargons CN, Meiller C, Angyal B, Wheeler P, Davis C, Stevens-Watkins D. Critical consciousness of anti-Black racism: A practical model to prevent and resist racial trauma. Journal of Counseling Psychology. 2021; 68 (1):1–16. doi: 10.1037/cou0000430. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Noonan AE, Liu WM. Psychology and the social class worldview: A narrative-based introduction. Routledge; 2022. [ Google Scholar ]
• Romm KF, Berry CM, Alvis LM. How the rich get riskier: Parenting and higher-SES emerging adults’ risk behaviors. Journal of Adult Development. 2020; 27 :281–293. doi: 10.1007/s10804-020-09345-1. [ CrossRef ] [ Google Scholar ]
• Smith L. Psychotherapy, classism, and the poor: Conspicuous by their absence. American Psychologist. 2005; 60 (7):687. doi: 10.1037/0003-066X.60.7.687. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Stiles K, Lee SS, Luthar SS. When parents seek perfection: Implications for psychological functioning among teens at high-achieving schools. Journal of Child and Family Studies. 2020; 29 :3117–3128. doi: 10.1007/s10826-020-01828-9. [ CrossRef ] [ Google Scholar ]
• Strand PJ. Inheriting inequality: Wealth, race, and the laws of succession. Oregon Law Review. 2010; 89 :453–504. [ Google Scholar ]
• Stucky BD, Gottfredson NC, Panter AT, Daye CE, Allen WR, Wightman LF. An item factor analysis and item response theory-based revision of the Everyday Discrimination Scale. Cultural Diversity and Ethnic Minority Psychology. 2011; 17 (2):175–185. doi: 10.1037/a0023356. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tabachnick BG, Fidell LS. Using multivariate statistics. 6. Pearson; 2013. [ Google Scholar ]
• Williams K, Lund TJ, Liang B, Mousseau AD, Spencer R. Associations between stress, psychosomatic complaints, and parental criticism among affluent adolescent girls. Journal of Child and Family Studies. 2018; 27 :1384–1393. doi: 10.1007/s10826-017-0991-2. [ CrossRef ] [ Google Scholar ]
• Zitelmann R. Upward classism: Prejudice and stereotyping against the wealthy. Economic Affairs. 2020; 40 :162–179. doi: 10.1111/ecaf.12407. [ CrossRef ] [ Google Scholar ]

ORIGINAL RESEARCH article

Entrenched inequalities class, gender and ethnic differences in educational and occupational attainment in england.

• Department of Sociology and Cathie Marsh Institute for Social Research, School of Social Sciences, Manchester University, Manchester, United Kingdom

Research in social stratification tends to focus on class differences in educational and occupational attainment, with particular attention to primary and secondary effects in the former, and class reproduction in the latter, domain. Research in ethnic studies tends to focus, however, on ethnic penalty or premium. Many studies have been conducted in each tradition on specific issues but little research is available that examines class, gender and ethnic effects simultaneously or in tandem with contextual effects, let alone on the whole trajectory from compulsory schooling, through further and higher education, to labor market position. Using data from the Longitudinal Study of Young People in England, this paper shows pronounced class differences but remarkable gender progress in each of the educational domains. With regard to ethnicity, people from minority ethnic heritages had lower GCSE scores due to poorer family conditions but achieved higher transition rates to A-Level study, higher university enrollment and, for some groups, greater attendance at elite universities, resulting in an overall higher rate of degree-level attainment than did whites. One might expect members of ethnic minority backgrounds to fare equally well in their earlier careers in the labor market, but only to find them more vulnerable to unemployment, less likely to have earnings, and more disadvantaged in terms of disposable incomes.

Introduction

The aim of this paper is to study the educational and occupational achievement of members of second-generation ethnic minority groups in England, whether they are subject to similar class effects as those from the majority group, and whether there are specific ethnic penalties in their educational trajectory from compulsory schooling to higher education and furthermore in their early careers in the labor market. Sociologists have conducted many studies on how family origins affect children’s educational and occupational attainment in Britain. Most of the studies are focused on educational attainment in compulsory schooling and progression to A-Level study given the prior academic performance. Yet, little research is available that combines insights from both social mobility and ethnic studies traditions to examine the entire educational trajectories from compulsory schooling through A-Level studies to higher education, and furthermore into the labour-market position after completion of education, and to interrogate the underlying socio-economic-cultural factors at the individual and community levels in terms of parental class, gender and ethnicity on the one hand, and school-level deprivation and diversity on the other, that shape the trajectories. This paper seeks to make a contribution to knowledge in this respect.

The paper is structured as follows. In the next section, we give a brief account of the sociological analyses on educational and occupational attainment, with particular attention to research on primary and secondary effects, and on ethnic penalty and premia. We show that while many studies have examined the class effects in broad terms on the transition to A-Level studies, no research is currently available that links family class, gender and ethnicity, and also contextual influences to pupils’ performances and transitions in the entire educational journey and moves further afield into labour-market positions. After that, we present data and analyses. The paper will conclude with some discussion.

Literature Review

Sociologists concerned with social inequality have conducted much research on educational and social mobility. They wish to find out how family condition in terms of parental class, education and income either singularly or in combination with other ascribed characteristics such as gender and ethnicity affects people’s opportunities and outcomes in educational and occupational attainment. Yet, as Li and Heath (2016) point out, whilst sharing the same goal of investigating social inequality, mainstream sociologists and ethnic studies scholars have largely traveled on separate tracks, with the former concerned with class effects and the latter with ethnic penalties (and, more recently, ethnic premia, see Heath and Brinbaum, 2014 ).

As education plays a pivotal role in increasing people’s human capital, broadening intellectual horizons and serving as a passport to the labor market, it is a major arena of class competition, academic debate and policy-making. The classical modernization theory proposes that with economic development and growing government provision of educational services, achievement between children from different social class origins will become increasingly equal and family influences will gradually pale into insignificance. The increasing influx of visible ethnic minority groups into Britain has posed a serious challenge to the theory: can it explain the process of educational stratification for immigrants’ children equally well as it does for the majority population? Here the first task is to test whether the theory can really explain the patterns and trends of educational attainment for the mainstream (majority) population, and the second task is to see how well it explains the educational attainment for the second-generation ethnic minority groups. How big an effect does origin class have on children’s attainment? Do class differences in children’s educational attainment stay constant or become stronger or weaker over time with greater government provision of educational services? Does the class position of immigrant families play an equally protective role in their children’s education as that of the majority families? Do ethnic minority children from advantaged class backgrounds suffer a “perverse fluidity” and experience excessive downward mobility as earlier studies found for African Americans in the United States ( Duncan, 1968 ; Hout, 1984 )? Or do immigrant children show greater aspiration, resilience and determination for more education despite family disadvantages?

In a landmark study on social stratification of education, Halsey et al. (1980 : 184) show pronounced class differences in education and increasing differentials at higher levels of educational attainment. For instance, 71.9% of the men from professional and managerial “service-class” origins attended selective secondary schools as compared with only 23.7% of working-class sons, at a disparity ratio of 3.0:1. The ratios became 4.9:1, 9.6:1 and 11.2:1 at O-Level, A-Level and University attendance respectively.

Do class differences in educational attainment stay constant or do they show signs of aggravation or amelioration? Breen et al. (2009) used the pooled data from the General Household Survey (1973–1992) to study educational stratification in Britain in comparison with seven other industrial societies. Using a semi-cohort approach, the authors showed that class differences in educational attainment were being consistently reduced for men from successive birth cohorts from 1908–24 to 1955–64, and this result obtained whether one used country-specific data or with class and educational variables standardized across countries. Similar patterns of declining social inequality in educational attainment was found for women ( Breen et al., 2010 ), lending support to the modernization theory. The authors attribute this to the reduction in family resources and the government provision of educational services after the end of the Second World War, yet this is contrary to economists’ findings of declining social mobility in education ( Blanden et al., 2005 ). 1

Why do children from different classes have different educational outcomes? One theory is that their families have differential possession of resources. Bourdieu (1986) holds that middle-class families possess cultural, social and economic capitals beyond the reach of working-class families, and that it is differences in family resources that will engender differences in educational outcomes. For instance, middle-class families tend to use their superior resources to help their children’s education by creating a pro-learning family environment, practising “concerted cultivation” ( Lareau, 2003 ), moving to more expensive catchment areas where good-quality state schools are located or sending the children to private schools. Perhaps most importantly, according to Bourdieu, middle-class families equip their children with a habitus which enables them to “move in their world as a fish in water” whereas the anti-learning attitude of working-class children makes them feel like “fish out of water” in educational environments ( Bourdieu, 1990 : 108).

Bourdieu’s cultural capital theory is challenged by Goldthorpe (2007a) who finds it inherently flawed, that is, incompatible with the observed facts. In Britain as in other developed countries, working-class children have steadily increased their attendance beyond compulsory education in the last few decades. If there is a working-class habitus which instils an anti-learning attitude in them and which makes them feel like a fish out of water in school, why would their attendance rates have increased so much? The very fact of increasing attendance suggests that working-class children are not as anti-learning as the habitus theory would imply, but have an eagerness for more advanced learning if their family resources would allow them to. In an effort to provide an alternative and more viable explanation, Goldthorpe developed the “rational action theory” (RAT), also called “relative risk aversion” (RRA) theory ( Breen and Goldthorpe, 1997, 2001 ; Goldthorpe, 2000, 2007b, 2014 ; see also; Kahneman, 2011 ) to explain both the increasing working-class uptake of education at the absolute level and the constant differential with the middle class uptake at the relative level. Key in the RRA thesis is the proposition that parents in all social positions would wish their children to do at least as well as they themselves have done in terms of educational and occupational attainment and to try to avoid downward mobility. When children are faced with the need to make decisions as to whether or not to proceed to a more advanced level of study or to enter the labor market at the end of compulsory schooling, they will consult with their parents. The outcome of such consultation tends to be that working-class children with more limited socio-cultural-economic resources will usually make “realistically feasible” decisions (called “strategy from below”) whereas middle-class children, backed by superior resources, will more often than not make more ambitious decisions, even when they have similar or even lower levels of academic performance as compared with working-class children (called “strategy from above”). This tendency to exercise caution (risk aversion) in the case of working-class children and to embrace challenge (risk venture) in the case of middle-class children underlies the distinction between the primary and the secondary effects, a distinction made by Boudon (1974) . The primary effects may be of a genetic or socio-economic-cultural kind, and refer to levels of academic performance that are actually achieved by children from different class origins. It is usually the case that students from more advantaged backgrounds have higher levels of performance than do those from more disadvantaged backgrounds. The secondary effects refer, however, to the different choices that children of different class origins will tend to make in consultation with their parents at critical junctions on their educational journey from compulsory (GCSE) to post-compulsory work such as transition to A-Level and, furthermore, to undergraduate and post-graduate studies in England. Both the “realistically feasible” choices and the more “ambitious” choices are deemed rational by the actors given the circumstances in which they find themselves.

Goldthorpe and his colleagues have made several efforts to test the thesis of primary vs. secondary effects. Using the National Child Development Study of 1958 when the respondents turned 16 in 1974, and two Youth Cohort Study (YCS) datasets where the respondents were also aged 16 (in 1987 and 2002), they find that people from professional-managerial (“service-class”) families have higher scores in English and mathematics examinations than do working-class students in each of the three cohorts, which is as expected. Yet, they also find that, even at similar levels of academic performance, students from service-class families have a higher likelihood of transition into A-Level work than do working-class students, by around 15 to 20 percentage points; that there is little change over time in the class differentials from 1974 to 2002; and that secondary effects account for around one quarter to one half of the class differentials in educational attainment ( Erikson et al., 2005 ; Jackson et al., 2007 ; Goldthorpe and Jackson 2008 ). These findings lend powerful support to the rational action theory. Yet it is also the case while these are among the best research findings in this area, they only differentiate three broad origin classes without taking gender or ethnicity into consideration. Jackson (2012) tried to improve upon the situation by pooling three YCS datasets together (when students turned 16 in 1998, 2000 and 2002) and analyzing the transition rates to A-level and to university studies between different ethnic groups. The primary effects are measured by standardized scores in the public examinations of mathematics and English at GCSE, and of A-level grades, and the secondary effects by class-based transition rates given prior levels of performance. She found that most ethnic groups had lower test scores but higher transition rates than did the white majority group, which she interpreted as evidence of significant disadvantages in the primary effects but significant advantages in the secondary effects. Jackson holds that the former runs counter to claims of positive selection (ethnic premium) as proposed by scholars in prior research but the latter indicates a defensive strategy against possible discrimination at the hands of employers. Jackson’s view of the higher transition rates by ethnic minority groups as a defensive strategy makes sense in light of the systematic findings on barriers faced by ethnic minority groups in the British labor market ( Berthoud, 2000 ; Li and Heath, 2008 ; Li and Heath, 2018 ; Heath and Di Stasio, 2019 ) although to term such “defensive strategies” as an advantage seems debatable.

While early studies may have a reasonable excuse to ignore the issue of ethnicity on grounds of data limitation, the rapid increase of the visible ethnic minority composition in the population indicates that any continued adoption of an ethnic-blind approach is no longer viable. Given this, researchers have paid increasing attention to second (or multiple) generation ethnic experiences in education, access to employment and career advancement ( Heath and Brinbaum, 2014 ; Li, 2018b ; Lessard-Phillips and Li, 2017). Yet it has been difficult to accommodate the conventional class analysis approach with the ethnic studies approach. For instance, one may aptly term middle-class children’s greater educational ambition a resource-based “advantage”, because middle-class families do have superior resources of various kinds relative to working-class families, but in what sense can we call the higher transition rates by poverty-ridden ethnic minority students an “advantage”? Scholars have made a few suggestions as to why ethnic minority children who come from poorer families and who achieve lower test scores at the stage of compulsory schooling should exhibit higher transition rates to further and higher education, and posited different theses such as “positive selection” ( Borjas, 1987 ; Feliciano, 2005 ; Ichou, 2014 ), “consonant acculturation” ( Portes and Zhou, 1993 ), or “reinvigorated aspiration” ( Li, 2018a ). The positive selection thesis holds that visible ethnic minority immigrants from far-away countries (rather than from nearby countries such as the “guest workers” who moved from Turkey to West Germany after the Second World War) are not a random selection of the population in their country of origin but have exceptional qualities in terms of aspiration, ambition, determination, perseverance and resilience. 2 The first generation arriving in the receiving country will often meet with multiple handicaps due to a lack of economic capital, disrupted social capital, insufficient cultural and human capital (such as ignorance of the local labor market, low levels of education, possession of foreign qualifications unrecognized by the employers, and poor English) and other factors, and will tend to find themselves in poorly-paid jobs shunned by the mainstream population. But they are determined to survive and thrive, and will pass on their ambition, aspiration, determination and other positive qualities to their children. This thesis sounds attractive but does not explain why there is so much variation among different second-generation ethnic minority groups whose parents came from countries of similar distances to Britain. The segmented assimilation theory ( Portes et al. 2009 ), which proposes three modes of assimilation (consonant, selective and dissonant acculturations), is designed to explain the variation among the different groupings. The most successful group will, according to the theory, adopt the “consonant acculturation” strategy where professional parents and their children will learn the language and culture together and the children will obtain elite middle-class positions upon entry into the labor market, achieving full integration. The second group who adopt the “selective acculturation” strategy will be economically successful but will choose to preserve their unique cultural traditions. The third group with “dissonant acculturation” will join the ranks of the underclass. This theory sounds elegant, but does not stand rigorous empirical test, as the great majority of second-generation children do not fit neatly into any of the modes ( Waters et al., 2010 ). The thesis of reinvigorated aspiration as posited by Li (2018a) assumes that the second-generation, growing up in poor families and poor communities, will have a good understanding accrued from lived/perceived experience and parental communications that, as members of ethnic minority heritages, they are likely to experience disadvantage and discrimination in the labor market, at all processes of job application, interviewing, and gaining promotions in the career life, and therefore have to aim higher now so as not to fall too low in future (see also Carmichael and Woods, 2000 ; Connor et al., 2004 ; Modood, 2005 ; Heath and Li, 2008; Wood et al., 2009 ; Rafferty et al., 2012 ; Zwysen and Longhi, 2018 ). At the core of this thesis is the “signaling” theory ( Spence, 1973 ; Weiss, 1995 ) which assumes that competitors perceived to be in weaker positions tend to give stronger signals to avoid being ignored and to gain adequate recognition. Previous work applied the idea to analysis of degree-level attainment by the second-generation ethnic minority members in the United Kingdom but the thesis needs further and more rigorous test from the educational trajectory at different junctures to the labor market position in the different spheres to demonstrate its viability. The present analysis is devoted to this task.

To sum up, there has been much research on educational attainment in the United Kingdom but existing work is mostly limited to class effects on performance and transition to A-Level studies. Only a few studies extend to transition to university enrolment. No research has linked the family origin (including class and education), gender and ethnic effects on children’s educational and career trajectories in one go whilst at the same time controlling for other socio-economic factors at the individual and contextual levels. With regard to the last point, we may note that most mobility studies adopt an individualistic approach, yet it is well known that contextual effects play an important role in children’s education, a role keenly appreciated by parents and government decision-makers. Middle-class parents try to buy houses in catchment areas with good schools. Government offices have launched various widening-participation programs to help improve the life chances of children in deprived areas. Yet government analyses tend to focus on indicators of local-area deprivation without looking at parental socio-economic conditions ( Social Mobility and Child Poverty Commission, 2015 ; Social Mobility and Child Poverty Commission, 2016 ; see also Friedman and Macmillan, 2017 ) just as academics tend to focus on individual attributes without taking considering contextual effects. Thus academic and government research efforts rely on different data sources and have not been able to form a meaningful dialogue, with the former being susceptible to the “atomistic” fallacy and the latter to the “ecological” fallacy ( Robinson, 1951 ; Li et al., 2005 ). The present analysis is fortunate in being able to draw data from both personal and contextual perspectives and we hope to ameliorate the situation by including not only respondents’ and their families’ demographic and socio-cultural attributes that have been demonstrated to have an important bearing on primary and secondary effects, but also school-level indicators of family poverty and ethnic diversity. The former refers to the proportion of students being eligible to means-tested free school mean (FSM) and the latter to the Herfindahl index of ethnic diversity in each of the schools that took part in the survey. With these factors in mind, the present study seeks to address the following research questions:

• How do different ethnic groups perform in their GCSE studies as compared with white children and among one another, given their parental class, education, family composition and other socio-economic, including contextual, circumstances?

• How do the different ethnic groups differ in their transitional probabilities to A-Level studies, and to university (including Russell-Group) enrolments?

• ethnic minority children have the same returns to education in the labor market as do their white peers?

Data and Methods

To address the foregoing questions, this study will use the Longitudinal Study of Young People in England (LSYPE1), also called Nest Steps (NS). The survey represents all young people aged 14 and resident in England attending maintained schools, independent schools and pupil referral units (PRU) in February 2004. It adopted a stratified, multi-stage, and random sampling design with oversamples of the major ethnic minority groups to provide sufficient ethnic samples for statistical analysis. 838 maintained schools, 52 independent schools and two PRUs were sampled. It follows their lives through seven waves annually until 2010, and then again when they were aged 25 in 2015. The initial sample size was 15,770 but at wave 4, a boost sample of 352 respondents was added, with a total size of 16,122. As with other cohort and panel studies, the NS has suffered sample attritions, with only 7,707 respondents being found in Wave 8 (age 25). The NS data can be linked with the National Pupil Database (NPD), which contains information on pupils’ examination results at each key stage, schools and colleges attended, eligibility for legally-defined and means-tested free school meal (FSM), school-level characteristics such as proportion eligible for FSM and proportions of pupils from each of the main ethnic minority groups. The data thus contain a wealth of information at the individual and school levels enabling researchers to make a detailed analysis of the primary and the secondary effects at different stages of educational career, and of the labor market position in their early working careers.

With regard to parental socio-economic position, class in terms of National Statistics Socio-economic Classification 3 (NSSeC) and education in the form of highest level of qualification will be used with the dominance approach adopted ( Erikson, 1984 ; Li and Devine, 2011 ), namely, the higher position from father or mother. For single-parent families, his or her class and education will be used as family position. As Ilie et al. (2017) show, parental class and education have better predictive power than family income. Siddiqui et al. (2019 : 82) also argue that as over half (58%) of the respondents in the survey had missing data on family income, any attempt to use existing variables to impute missing income would make subsequent analyses of income effects blighted. At a theoretical level, class, as Goldthorpe and McKnight (2006) hold, serves as a better indicator of permanent household income than wages or salaries on grounds of economic security, income stability and future prospect.

Other explanatory variables at individual and contextual levels include family composition, eligibility to free school meal, nativity, school-level deprivation in the form of proportions of students eligible for free school meal, and school-level ethnic diversity as indicated by proportions of students belonging to each of the main ethnic groups. A Herfindahl index was created on ethnic diversity for each school.

We use several outcome variables. The first of these pertains to GCSE test results taken at the end of compulsory schooling. Pupils usually take eight GCSE subjects in England. Some schools also offer students the optional short-course GCSEs which contain roughly half the learning material and count as half a GCSE. A summary score was created with A* = 8, A = 7, B = 6, C = 5, D = 4, E = 3, F = 2 and G = 1 for full GCSEs and A* = 4, A = 3.5, B = 3, C = 2.5, D = 2, E = 1.5, F = 1 and G = 0.5 for half GCSEs. The scores range from 0 to 111; with a mean score of 39.7 with standard deviation 20.7. For some of the analysis, the scores will be standardized with a mean of zero and standard deviation of unity. The other outcome variables pertain to transition rates to A-Level studies at the end of GCSE, and to university enrolment at the end of A-Level study or to elite Russell Group university attendance, and labour-market position including employment status, class position, gross weekly pay among the employed, and the ‘continuous weekly income’ for all respondents at wave 8. The analysis of both kinds of income are necessary as nearly a third of the young adults were workless, including unemployment (5.7%), full-time students (5.0%), looking after home (4.7), sick or disabled (1.7%) or inactivity for other reasons. Analyzing the “continued weekly income” from the perspectives of family class, gender and ethnicity is also important in addition to that of labour market earnings as it will allow us to see how the different social groups are being treated at the societal level. Statistical methods will be adopted as appropriate for the task at hand.

The analysis in this section will focus on the respondent’s educational and early career trajectories from ages 16 to 25. As earlier noted, we shall first analyze ethno-class differences in educational achievement before moving to occupational attainment. We examine GCSE scores at age 16, transition rates to A-Level and to university studies (including attendance at Russell Group universities). In the second part, we shall look at the employment situation and incomes.

Educational Attainment in Compulsory Schooling

The data in Table 1 show an overall view of the class, ethnic and gender differences in GCSE scores and probabilities of progression to A-Level, university and elite university studies.

TABLE 1 . Descriptive analysis of GCSE score, transition rate (%) into A-Level, university and Russel-Group (RG) university work by parental class, ethnicity and sex.

With respect to class effects, we find pronounced differences with clear gradients in each of the four domains under discussion. As noted above, the mean GCSE score for the sample is around 40 but we see that people from higher salariat (professional and managerial) families had a mean score of 55 whereas those from routine manual families only had a mean score of 24, with a difference of 31 points. The class differentials increased when we look at the transition rates to A-Level and to university studies, with the differences between the higher salariat and routine students being 41 percentage points in the former and 59 points in the latter regard. And with respect to access to the more prestigious Russell Group universities, as shown under the last column, the class differences are also striking, with over a quarter (26%) of the higher salariat children studying in Russell Group universities in contrast with a meagre two percent for those from routine families. 4

The middle part of the table shows the data on ethnic differences. As white students comprise an overwhelming majority in the sample (86%), their attainment level closely represents the mean performance in each of the four aspects. We see clear and striking differences both between white and ethnic minority students, and among the ethnic minority groupings. In each of the aspects, Chinese students showed themselves as the highest performers, followed by Indians, in clear contrast with Black Caribbeans. Children from Black African, Pakistani, Bangladeshi families had the lowest GCSE scores but higher transition rates to A-Level and university than white students.

The data on gender differences show no female disadvantage. If anything, girls outperformed boys at each stage. The data on university enrollment echo the historical profile Heath et al., 2018a : 68, Figure 4.2 ) which shows men as having a lead over women in access to higher education from the mid-1950s to mid-1990s but since then, women have caught up with and increasingly surpassed men.

The intriguing question is why students in ethnic minority groups underperform in GCSE examinations but make “bold choices” at transitions to further and higher education. If the most important determinant of academic performance and subsequent choice concerns the “class-lined inequalities of condition” ( Goldthorpe and Mills, 2004 : 223), it is understandable that ethnic minority students who come from poorer families will have lower performance. But if the secondary effects are also reliant, and even more so than the primary effects, on family resources as the “relative risk aversion” thesis would argue, why would the poorer and worse-performing ethnic minority students make even bolder choices than their more affluent and better-performing white peers rather than take a “realistically-feasible” strategy as the RRA thesis would predict? In other words, if family poverty that leads to the lower performance is regarded as “disadvantage,” how does this disadvantage in the primary effects turn around to become an “advantage” in the secondary effects? Most analyses in this regard have, as noticed above, tended to use a one-dimensional approach, with a three-way schema of parental class, and focus on contrasting performances between service- and working-class students and, in so doing, ignored ethnicity as a non-issue. Therefore, the questions that are of crucial importance for present research and that reflect the genuine concern of an increasingly diverse society were overlooked in most of the existing sociological analyses in this regard. As we take a multi-dimensional approach in the present study, we need to have a closer look at the other domains of socio-economic disadvantages that reinforce one another in their impact on ethnic minority students’ performance. Here the primary question we need to establish is: what kind of socio-economic disadvantages do members of ethnic minority heritages face?

Table 2 shows some selected family circumstances to represent social disadvantages: proportions of parents in working-class positions, with low level of or no formal education, of single-parent family type and being eligible for free school meal (FSM) which, for our sample, was equivalent to annual gross household income below £13,480 ( Hobbs and Vignoles, 2010 ). These are, we believe, best available indicators of family economic, cultural and social deprivation.

TABLE 2 . Selected family characteristics by ethnicity: proportions (%) growing up in working-class, poorly educated, single-parent households and being eligible for free school meal (FSM).

It is clear that white students have much better socio-economic resources as judged from the range of indicators under consideration. White parents are least likely to be in work-class positions (23%) but ethnic minority parents are much more likely to be in such positions, with Pakistani and Bangladeshi parents particularly disadvantaged (68 and 45% respectively). Even more pronounced are differences in parental education, with only 17% of white parents having primary level or no schooling whereas for people from Bangladeshi, Pakistani and Chinese heritages, parental low education reaches a staggering high, at 83, 60 and 56% respectively. The combination of lower class position and poor education would mean that, even without labor market discrimination and differences in family size, ethnic minorities would have much greater vulnerability to poverty. While the large amount of missing income data in the NS file, at 58% as previously noted, makes it inadvisable to construct a poverty measure, we do have solid evidence on ethnic income poverty. Using the United Kingdom Household Longitudinal Study (UKHLS), Li (2018a : 487; see also Heath et al., 2018b ) showed an ethnic poverty profile closely corresponding to the distributions to class and education position as shown in the table. The proportion of households in poverty, as defined by the United Kingdom government criteria (60% below median of the standardized household mean incomes) runs from 15, 21, 22, 25, 36, 49 to 56 percent for white, Indian, Black Caribbean, Chinese, Black African, Bangladeshi and Pakistani groups respectively. Although FSM eligibility does not fully reflect family poverty as Hobbs and Vignoles (2010) showed, we still find a close correspondence between indicators of socio-economic disadvantage (class, education, poverty) and FSM eligibility, with white students least likely and all other groups (except for Chinese) more likely to have FSM. Finally, single-parent family structure may be an indicator of inadequate family social capital crucial for the maintenance of cultural tradition ( Sakamoto et al., 2009 ), “concerted acculturation” ( Lareau, 2003 ) and emotional support ( Putnam, 2007 ). Here we find that Black Caribbeans are most likely to live in single-parent families, with 64% being “always” or ‘sometimes’ headed by single parents, followed by Black Africans (43%).

Overall, data in Table 2 show that white students do enjoy superior socio-economic-cultural resources relative to their ethnic minority peers who face multiple disadvantages. People from Bangladeshi and Pakistani origins have the poorest economic situation, next come the Chinese in terms of low parental education, with the two black groups lying in between, and Indians being closest to whites. It is probably an interplay of these and other influences such as oriental cultural tradition ( Hirschman and Wong, 1986 ) which emphasizes over-achievement and perception of pervasive disadvantages in the labor market such as shown in Li and Heath (2018) , that led to the poorer academic performance but more ambitious choices for more advanced educational studies by the ethnic minority students. We now turn to multivariate modeling on such effects.

We first look at the net effects on academic performance as demonstrated in GCSE examination results. The data are shown in Table 3 with three models. Model 1 contains family class, ethnicity and gender, our key intersectional variables. Model 2 adds respondent-level FSM eligibility and school-level proportion of students eligible for FSM. The inclusion of the two FSM variables is of both conceptual and substantive importance. Conceptually, one may expect schools with high proportions of students eligible for FSM as being highly deprived and having an unfavourable learning environment, a negative effect over and above personal poverty (own FSM). Controlling for individual and school-level FSMs can hopefully help mitigate ecological and atomistic fallacies. Substantively, while Siddiqui et al. (2019) suggest that with the availability of individual FSM data, there is no need to include family circumstances such as parental class and education, we can directly test whether parental position is still significant after controlling for both individual- and school-level types of FSM. One further consideration is that Ilie et al. (2017) recommend using two other contextual-level deprivation indices in lieu of FSM, but our prior analysis suggests little need for so doing. 5 The results in Model 2 can help us to address the questions of relative merits or otherwise of the claims from different theoretical perspectives as outlined above. Finally, Model 3 adds variables on parental education, family structure, nativity, and school-level ethnic diversity as measured by the Herfindahl index. Sociologists tend to use parental class alone as family position in addressing intergenerational educational or occupational mobility ( Halsey et al., 1980 ; Goldthorpe and Jackson, 2008 ; Breen et al., 2009 ) but increasingly there is an appreciation that parental education plays a crucial role over and above parental class in shaping children’s educational and occupational attainment when parental education is used as a “positional good.” namely, in a relative rather than absolute sense ( Bukodi et al., 2014 ; Li, 2018a ). As the cohort members in the present study are of the same age, there is no need to produce relative measures of parental education. Finally, as students are nested in schools and as schools differ in the levels of socio-economic deprivation and ethnic diversity, multilevel regression techniques are used, with school-level FSM and Herfindahl diversity serving as level-2 covariates.

TABLE 3 . Random coefficient models on GCSE scores by socio-economic attributes.

The data in Model 1 of Table 3 show powerful class and some ethnic and gender effects net of one another. Students from higher salariat families have, controlling for ethnicity and gender, 20.6 scores higher than those from routine families. We noticed in Table 1 that Pakistani and Bangladeshi students had lower mean GCSE scores than white students and, from Table 2 , we also saw that their family class and education positions were much lower than those of whites. Yet, here, we find that their performance is significantly higher than that of white pupils, suggesting that it was their lower parental class that suppressed the achievement. With similar family positions, Pakistani and Bangladeshi students would perform equally well as, or probably better than, their white peers. Girls, on average, outperformed boys even when parental class and ethnicity are held constant.

As people eligible for FSM tend to be from poor households, we would expect them to have, other things being equal, lower levels of academic performance, which is shown as true. They have six scores lower on average. Furthermore, we find that school-level FSM also have a net and substantial impact on students’ performance. With an overall FSM at around 14%, an increase of ten percentage points of school-level FSM would, other things being equal, lower a student’s performance by around four scores. As most of the ethnic minority students except Indians and Chinese were more likely to be in receipt of FSM, controlling for individual and school level FSM have placed them on higher (net) performance scores than white students.

Finally, in Model 3, we find that parental education, family structure, nativity and school-level ethnic diversity all play an important role. People with degree-level parents have, other things being equal, 15 scores higher than those whose parents have only primary level of education or no formal schooling. People growing up in lone-parent families, whether “sometimes” or “always” lone-parent, also had lower scores. Yet, those who were foreign born but who arrived in the United Kingdom at a young age achieved higher scores than did the others, by three points on average, possibly reflecting the “positive selection” effect due to the recency of immigration and their parental higher qualifications. 6 School-level ethnic diversity also has a positive impact on students’ achievement.

An interesting and important point is that, after controlling all these individual and contextual factors, we still find highly significant effects of parental class and ethnicity. Combining the findings from Tables 1 – 3 , we may say that most ethnic minority students had lower performance scores due to the multiple handicaps arising from “inequalities of condition” inherent in their family position and, yet, if they had had comparable parental socio-economic conditions to those found in white families, they may well have obtained similar, or even better, results. Only Black Caribbean students might have fared worse.

Transition to A-Level Studies

We now move to the choices made by the young people to follow A-Level studies. Most existing work on primary and secondary effects have focused on this, with the secondary effects gleaned from differences between salariat- and working-class children. Our analysis in Table 4 follows the structure of Table 3 , with Model 1 focused on intersectional effects, Model 2 adding prior levels of achievement to assess secondary effects, and Model 3 further controlling for other individual and contextual factors. The data in Table 4 show average marginal effects (AME) from logit models, with logit coefficients transformed to proportions, or transition rates, to A-Level work.

TABLE 4 . Average marginal effects (AME) from logit models on transition into A-Level work by socio-economic attributes.

The data in Model 1 shows the expected class differentials. Ethnic and gender status being equal, those from higher salariat families were 45 percentage points more likely to choose A-Level studies than those from routine families. Most people from ethnic minority backgrounds are also significantly more likely to choose A-Level studies than the white majority, holding constant family class position. As ethnic parents have lower class positions than whites, controlling for class boosted their transition rates as compared with the raw figures shown in Table 1 . Girls are significantly more likely to choose A-Level studies than boys.

The crucial findings are shown in Models 2 and 3 where academic performance and other personal and contextual attributes are taken into account. It is surprising that parental class loses its significance altogether. Chinese students have very high GCSE scores, but once prior performance is controlled for, they are not significantly more likely to opt for A-Level studies. The overall pattern in Model 2 is echoed in Model 3 when the other factors are controlled for. The most salient feature that emerges from the findings under the two models is the lack of significant parental class effects. One reason for the difference in the findings as shown here and those by Goldthorpe and colleagues as cited above may be due to the number of class categories used: a seven-class schema is used here but a three-class schema used in their analyses; another reason may be due to the inclusion of ethnicity, gender and other covariates here, making the analysis more complicated, diluting the impacts of class. To further ascertain why the discrepancy emerged, further analysis was conducted, with a three-way schema for parental class, and with GCSE scores normalized with a mean of zero and standard deviation of unity, which is the same framework as adopted in prior analysis ( Erikson et al., 2005 ; Jackson et al., 2007 ; Goldthorpe and Jackson, 2008 ; Jackson, 2012 ); Jackson, 2013 .

The data in Figure 1 shows clear class differences in the primary effects, with students from salariat families having much higher scores than those from working-class families, which closely resembles previous findings by other scholars using other datasets. Yet, controlling for prior attainment, the differences in the transition rates, or the secondary effects, for children from the three classes as shown in the S-shaped curves are quite indiscernible. Does this contradict the predictions of the rational action theory that middle-class children will tend to make more ambitious choices and working-class children more realistically-feasible choices? Probably not. If we compare the historical trends on transition rates between the NDCS (born in 1958 and reaching age 16 in 1974) and the 2001 YCS data as shown in Goldthorpe and Jackson (2008 , Figures 3.1 and 3.2 ), we can see that the secondary effects were being reduced from earlier to later time points, suggesting that all children were becoming more likely to continue with A-Level studies. Our NS children’s transition time occurred in around 2006, even later than in the YCS2001 data, hence the class differences may be expected to be even smaller than shown in the YCS2001. From this perspective, we may say that even if primary effects remain, the strength of secondary effects may well decline or shift to more advanced levels, and this explanation would be consistent with Goldthorpe’s critique of Bourdieu’s cultural capital (habitus) theory, and with the “maximum maintained inequality” (MMI) and the “effectively maintained inequality” (EMI) theses by Raftery and Hout (1993) ; Lucas (2001) .

FIGURE 1 . Graphical representation of regression of transition to A-level work on academic performance.

Another feature in this regard that merits further consideration pertains to the possibility that the secondary effects may not cover the whole range of performance but only emerge at a particular performance level. Jackson et al. (2007 : 218) state: “It would seem reasonable to suppose that students who perform very poorly in their examinations at 16 will have a low probability of going on to A-levels and that those who perform very well will have a high probability almost regardless of their class origins, while it is at intermediate levels of performance that the scope for secondary effects to operate is largest.” We can have a closer look to see whether this proposition is verifiable in our data.

The data in Table 5 are organized for this purpose. Academic performances (GCSE scores) are divided into three bands: low, middle and high. In the last row of the table, we find that the transition rates for A-Level studies under the three bands are 18, 56 and 93 percent. Thus those in the high band of achievement are around 5 times as likely to make the decision to go on to A-Level studies as those in the low band. Do we find class differentials only among the middle-band achievers but not among the high and the low achievers? Surprisingly, we do not. The first three rows under “All” show little class difference among the low and the mid, but significant class differences among the high, achievers. A very high proportion of high-achievers from all class origins choose to move to A-Level studies and working-class high-performers have a higher rate than salariat low- or mid-performers. But a close look still shows that, among the high performers, working- and intermediate-class children have a significantly lower rate than salariat children, at 86, 91, 94% respectively. Thus, our data show a pattern of secondary effects only among the high-achievers rather than among the intermediate performers as Jackson et al. (2007) have expected.

TABLE 5 . Transition rate (%) into A-levels work by family class, ethnicity, sex and bands of GCSE scores.

Since we are also concerned with ethno-gender differences, further analysis is conducted on ethno-class-gender effects on children’s performance and transition probabilities, with results listed in the lower part of the table. Here we find that the RRA predictions mainly apply to the high-achieving white students. For both men and women in the majority group, there are clear and significant class differences among the high achievers. For ethnic minorities, however, it is academic performance rather than parental class position that plays a more decisive role. It is noted here that even at the low level of performance, ethnic men and women are more likely to make the transition than their white peers. Yet it is also the case that among ethnic minority women in the low band, class differences exist, with working-class girls being 22 percentage points behind their salariat counterparts in the transition rates (30 and 52% respectively), which constitutes a statistically significant difference. Further analysis shows that all low-performing working-class girls from ethnic minority heritages apart from Black Africans (no Chinese girls were in this category) had low transition rates, at 30, 26, 22, 29 percent for Black Caribbean, Indian, Pakistani and Bangladeshi groups although they were still more likely to opt for A-Level studies than their white counterparts from salariat families.

Overall, our analysis has enhanced the application of the rational action theory with regard to the class-ethno-gender specificity rather than showing encompassing support. With this mind, we move on to the transition to university including Russell Group universities.

Transition to University

Table 6 shows the transition rates to university (Models 1 and 2) and to Russell Group (RG) universities. Models 1 and 3 show the intersectional effects and Models 2 and 4 show full effects akin to Model 3 in Table 4 . The data in Model 1 on access to university are similar to those in Model 1 on transition to A-Level studies, showing pronounced class and clear ethno-gender effects. The only notable differences between the patterns shown here and those revealed previously on transition to A-Level studies are that family class and ethnicity effects are even more pronounced here on access to university, suggesting that the higher the level of educational attendance, the more important the family class position and that white working-class children are being left further behind. With regard to the secondary effects, we need to take into account prior performance but there is no clear guidance as to what can effectively serve as such an indicator: one could use GCSE scores, number of A-C grades, or having achieved five or more A-C grades at GCSE or equivalent including English and Mathematics. After some careful comparison, we decided to adopt the last of these as it is an important and quite commonly used indicator. 51% of white as compared with 35% Black Caribbean and 39% of Pakistani students achieved this, with Chinese (78%) and Indians (61%) being in the lead, and Black African (45%) and Bangladeshi (43%) students being in the middle. In addition, the other personal and contextual variables as previously used are included in the model for as covariates.

TABLE 6 . Average marginal effects (AME) from logit models on access to university and to Russell-Group (RG) universities.

The data in Model 2 shows that achieving five or more GCSE A-C grades including English and Mathematics is of crucial importance in securing a place in university. Other things being equal, those students with this level of achievement have a transition rate being 35 percentage points higher than those without this attainment. Parental education has a positive effect but coming from single-parent family has a negative effect. School-level poverty (in terms of percentage FSM eligibility) and ethnic diversity have the effect as expected. Controlling for these, we find that ethnic effects were little changed but class effects declined sharply. Yet, these declines notwithstanding, it is still the case that those from salariat families are more likely to be enrolled in university by around 10–15 percentage points, and those from intermediate families by around five points, than working-class students. The class advantage as shown here echoes what Goldthorpe and colleagues observed for transitions to A-Level study, and the pattern again renders support to relative risk aversion thesis.

The main features of access to university are largely echoed in access to Russell Group universities, albeit with weaker strengths due to the small numbers involved. As Bangladeshi students tend to face more disadvantages in terms of parental class and primary attainment, they are found to have a higher probability of accessing Russell Group universities when prior conditions are held constant, in contrast to Black Caribbean students.

Labor Market Position

Having looked at the educational trajectory in some detail, we move to the respondents’ labor market situation in wave 8 when they were aged 25. In the preceding analysis, we found that ethnic minority students, with the exception of Chinese and Indians, performed less well than did white students in the primary effects but better in the second effects. The first result arose chiefly from family disadvantages and the second result obtained in spite of family poverty. A question that would lend itself in this regard is: did their aspiration, determination and efforts pay off? In other words, did ethnic minority students obtain occupational and earnings’ positon commensurate with their human capital investment? How well did they fare in their earlier career life as compared with their white peers?

Table 7 shows the main characteristics of the respondents’ human capital and labor market positions at wave 8. The data cover percentage with a degree, labour-market position, and gross and net weekly incomes by ethnicity. 7 Labor market position is a combination of employment status and class position with four categories: salariat and non-salariat among the employed, and unemployed and inactive among the workless. Gross weekly pay is payment from the main job for those in employment, with the workless including the unemployed. full-time students, looking after home and sick and disabled having no earnings from the labor market. 36 of the respondents reported abnormally high earnings (over £100 per hour) and these are omitted from analysis following the government instructions in the collection of earnings data (see Labour Force Survey, 2015 : 384). It is clear that people of ethnic minority heritages are well educated and have a higher likelihood of having a degree-level qualification than do the majority, with those from Black African, Indian and Chinese heritages having a probability nearly twice as high. It is noteworthy in this regard that even those from Pakistani, Bangladeshi and Black Caribbean origins who grew up in poverty-ridden homes outperform whites in gaining a degree qualification.

TABLE 7 . Education, labor market position and income (£) by ethnicity (N = 7,707).

With such a high educational profile, we would have reason to expect ethnic minority groups to make similarly impressive progress in the labor market positions. Unlikely their parents, they do not have language problems and their social capital is similar to that of white students. Yet, when we turn our gaze to employment and income situation, we are disappointed. The educational attainment by the ethnic minorities did not have the returns as expected. Every minority group were more likely to be unemployed, with the two black groups and Pakistanis being nearly twice as likely as whites to face unemployment, and that in spite of the higher educational qualifications. For those lucky enough to have a job, the chances of securing a “nice” job (in professional-managerial salariat position) are not too bad, although they may still be regarded as being disadvantaged if educational attainment is taken into account. For instance, 50% of Black Africans and 25% of whites had degree-level education but the salariat occupancy of the former is only slightly higher than that of the latter (45 vs. 35%). What is of even greater concern is the fact that, despite the higher levels of educational qualifications and of somewhat similar levels of occupational attainment (for those with a job), the two black groups and the two Muslim groups (Pakistani and Bangladeshi) have notably lower gross weekly earnings, and the “continuous weekly income” for the cohort member and partner is much lower for all ethnic minorities than for whites, suggesting lower returns to education and labor market position and greater economic disadvantages for the ethnic minorities.

Finally, we take a look at the two kinds of income data: gross weekly earnings and continuous weekly income. For the former, we use the Heckman regression method as the earnings depend on being employed. For the selection part, we use limiting long-term illness as the “identifying” variable in addition to other variables that are also used in the regression part. As the probit coefficients predicting whether earnings’ data are actually observed are not intuitive, we have transformed into percentages using the average marginal effects. Thus the first two columns in Table 8 refer to the avoidance of worklessness and the last two columns to the earnings differentials conditional on employment. Under both selection and regression parts, we use two models. Model 1 includes family class, ethnicity and gender, and Model 2 includes marital status, number of dependent children, and parental and own education.

TABLE 8 . Average marginal effects (AME) on avoidance of worklessness (%) and gross weekly earnings (£) conditional on employment based on Heckman’s model.

Looking firstly at the joint effects of worklessness in the selection part, we find that parental class exerts a powerful influence, with those from higher salariat families being 26.4 percentage points more likely to be in employment than those from routine manual families, other things being equal, with clear class gradients. Holding constant family class, all ethnic minority groups were less likely to be in employment, with Black Caribbean and Pakistani respondents being nine and ten percentage points less likely than whites to be employed. Under Model 2 when the other covariates are taken into account, we find, as expected, highly salient effects of own education and fairly noticeable parental educational effects, but parental class effects are much reduced. Yet, interestingly, controlling for education brought the ethnic penalties into much sharper relief, with those of Black African, Indian, Pakistani and Bangladeshi heritages being significantly more likely to face worklessness than whites, and the magnitude ranged between 11 and 18 percentage points higher. 8

For those fortunate enough to be in employment, family class still plays a highly important role, and Black Caribbeans and female respondents receive much less gross weekly pay, with Indians and Chinese having significantly more gross weekly earnings. When the other factors are taken into account, family class effects are sharply reduced. Black Caribbean’s penalty remains at a similar level although Indians’ and Chinese premiums are much reduced. People’s own education plays a very important role. Demographic attributes like gender, marital status and number of dependent children play a more salient role in terms of the amount of earnings than the probability of being in employment, other things being equal. 9

As around 12 percent of the respondents are married or partnered 10 who are expected to share economic weal and woe, and as those not in employment may have other sources of income, we now turn to the “continuous weekly income,” that is, incomes from all sources, which is a good measure of the overall economic well-being of our respondents. The data, obtained from OLS analysis, are shown in Table 9 with four models. Model 1 contains our main variables on parental class, ethnicity and gender, Model 2 adds personal attributes on marital status, number of children and health condition (in terms of GHQ12), 11 Model 3 further adds parental and own education and, finally in Model 4, we add respondents’ own class position differentiating salariat, non-salariat and workless.

TABLE 9 . OLS regression of weakly take-home income (£).

The data in Table 9 show marked ethnic disadvantages. Firstly, we find that, under Model 1, parental class exerts a huge impact on people’s income, with those from higher salariat families having over £100 per week than those from routine families, a difference similar to that found by Laurison and Friedman (2016) . After taking parental class into consideration, we find that ethnic minorities have much lower incomes, ranging from 56 to 81 pounds less than whites. As ethnic minorities’ parental class are generally in low positions, controlling for parental class makes little impact on respondents’ income differentials, which is clearly shown when we compare the findings under model 1 with those under the last column of Table 7 . As our respondents were still young in wave 8, most of them were unmarried and only a small portion of them had children or health issues, controlling for these factors does not change the patterns very much. In model 3 where we further control for parental and own education, we find that educational qualifications make a big difference and that, as a result, parental class effect is almost halved. In model 4, we further control for respondents’ own class position. Here we find that, as expected, people in salariat positions have higher weekly incomes than do the workless (unemployed + inactive). Yet, it is also important to note that, if we compare the figures from models 1–4, we find that, as more variables are controlled for, parental class effects are progressively reduced whereas ethnic effects are actually increased. For instance, respondents from higher salariat families are found to have £102.8 more weekly income in model 1 than do those from routine families, holding constant ethnicity and gender effects, but when the other factors are taken into account in model 4, the class differential is reduced to £51.5. If we look at Black Africans’ income, we find that they have, given parental class and gender status, £76.3 less per week in model 1 than do white respondents but when all other factors are taken into account in model 4, their income differentials becomes larger, at £94.8 less. People prefer to “compare like with like,” but the more like the personal and other characteristics we compare, the more unlike the take-home incomes between the ethnic minority and the majority groups we find.

Discussion and Conclusion

This paper has sought to contribute to scholarship on socio-ethno differences in British society. Most existing analyses on primary and secondary effects have confined their efforts to a three-way parental class effects on GCSE scores and transition to A-Level studies. Using the Longitudinal Study of Young Persons in England (LSYPE1, also known as Next Steps, NS), the present study has used a more elaborated seven-class NSSEC schema, and addressed class, ethnicity and gender effects simultaneously whilst controlling for parental education, family structure, economic situation (in terms of FSM eligibility) and contextual (school) level ethnic diversity and deprivation. We analyzed the socio-ethno differences not only in the primary and secondary effects during compulsory schooling, but in transition to university and to elite Russell Group universities too; and, furthermore, we linked the educational trajectory to labor market position and income profiles at age 25. Previous analyses in this area tend to focus on one or another specific aspect ( Strand, 2007 ; Anders, 2012 ; Croll and Attwood, 2013 ; Anders, 2017 ; Ilie et al., 2017 ; Siddiqui et al., 2019 ; and those by Goldthorpe and his colleagues as noted above), but the present study has sought to provide a more systematic and comprehensive perspective.

The main findings can be summarized as follows. Firstly, there are pronounced parental class effects in all aspects under investigation: ranging from GCSE scores, transition rates to A-Level, university and elite (Russell Group) university studies, obtaining degrees, avoidance of worklessness to gross weekly earnings and continuous weekly take-home income. As ethnic minority groups come from disadvantaged families in terms of parental class, education and incomes, they tend to perform less well in school but are more likely to opt for A-Level and higher education studies, providing further evidence to the validity of the thesis of “reinvigorated aspirations” ( Li, 2018a ). Their attendance at elite universities is, on the whole, still lower than that of the white students, echoing previous findings by Boliver (2013) .

The mainstream sociological analyses on primary and secondary effects have focused on parental class differences in academic performance at GCSE, and in transition rates to A-Level studies conditional on prior attainment. With respect to the secondary effects, the rational action theory expects the parental class effects to manifest themselves at lower levels of achievement or, more specifically, at the intermediate level. Most research in this respect has adopted a three-way class and ignored ethnicity and other factors. The present analysis has adopted a framework with a more elaborate class schema, with more explanatory variables and a greater coverage of analytical scope. Our analysis is not limited to testing the validity of the rational action theory concerning primary and secondary effects although we did find some support for the theory. Our findings in this regard are both substantively grounded and culturally fine-tuned.

The determination, ambition and aspiration of the young people from ethnic minority heritages were clearly shown in the choices they made with respect to transition to higher education. All members of ethnic minority groups were more likely to attend university and to hold a degree at age 25 that whites. Only Black Caribbeans were significantly less likely to attend elite Russell Groups universities.

All this suggests, as Li and Heath (2016) posit, a generally level playing ground of the educational system in Britain. Where ethnic minorities lag behind, such as in GCSE performance, it is mainly due to inequality of condition such as family and school deprivation rather than inequality of opportunity. They made laudable efforts in spite of family hardships, aimed higher and attained better educational qualifications. Given this, we might expect them to fare at least equally well in the labor market. Yet, to our dismay, we found that in spite of their better qualifications, they were more likely to face unemployment and inactivity, and had markedly lower weekly incomes even though among those lucky enough to be in employment, they were not too much disadvantaged (only Black Caribbeans were making significantly lower earnings). They started lower, worked harder, achieved well in education but were not fully rewarded in the labor market. 12

Overall, we found persisting class effects and entrenched ethnic inequalities in British society. The first-generation immigrants may have been positively selected but they had to face the harsh reality in the labor market upon arrival in the United Kingdom, resulting in having depressed class positions and economic hardships. They may have passed their aspiration, determination and resilience to their children who, as we have seen, started from pervasive family poverty but made determined efforts at decision points, and achieved remarkable progress in educational attainment. Yet, in spite of all this, they still found themselves in greater worklessness resulting in lower incomes. The former Prime Minister Therese May (2017) said that continued ethnic disadvantages must be “explained or changed.” The analysis in this paper has sought to explain the entrenched ethnic disadvantages in British society, and our evidence calls for greater efforts by policy-makers, employers and wider society to adopt more decisive and more effective measures that can eliminate labour market discrimination against ethnic minorities, for social justice and for national prosperity.

Data Availability Statement

Publicly available datasets were analyzed in this study. This data can be found here: discover.ukdataservice.ac.uk/series/?sn=2000030 .

Author Contributions

The author confirms being the sole contributor of this work and has approved it for publication.

Conflict of Interest

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

1 Further analysis using the same data as Breen and colleagues used shows that if one focuses on degree-level education, the differences between children from the service class and those from the manual working-class families enlarged from 12.9 to 21.6 percentage points from the oldest to the youngest cohort, which would lend support to the declining mobility thesis by Blanden and her colleagues (2005). The two aspects are not in contradiction: a reduction of class differences at the lower levels of education was going hand in hand with an increase of class differences at the higher (degree or above) levels of education. As more people were attending the lower levels of education, this would lead to an overall reduction in educational inequalities but this does not prevent a deepening of class differences at the higher levels of education.

2 It is generally recognized that immigrants are positively selected in that they tend to have higher levels of education than do their fellow citizens in the country of origin. It would have been a nice idea to test whether our ethnic minority respondents’ parents were positively selected and, if so, by how much. In order to do this, we need information on their parents’ country of birth and time of arrival to the United Kingdom, but neither variable was available in the datasets. Even if the variables were available, it would still be impossible to test this idea fully because there is no information on the average education in all the countries and for all the years concerning immigrants to the United Kingdom. The Understanding Society (USoc) has data on father’s and mother’s country of birth (pacob macob) and respondents’ year of arrival to the United Kingdom (yr2uk4). In the 10 waves of the USoc, 148,337 people were interviewed, including 3,704 Indians. Excluding those with missing information on pacob and yr2uk4, there are 2,227 Indians in the file. From 1952 to 2017, there were Indians coming to the United Kingdom every year and their father’s country of birth includes India, Pakistan, Bangladesh, Italy, Sri Lanka, Kenya, Uganda and South Africa, Jamaica and other countries. But even using the best education data source currently available ( http://www.barrolee.com/ ) would fail to provide the relevant information in most of the time-year-country combinations for Indians, let alone for all other immigrant groups. For instance, there was no information on average education in China before the 1980s (see Barro and Lee, 2010 ; Barro and Lee, 2013 : 197) although Chinese immigrants started to arrive to the United Kingdom from 1946 in the USoc file. I wish to thank one of the reviewers for alerting me to this potentiality although I have been thinking about how to improve on this for years. Perhaps a better approach is to compare the immigrant’s parental class in the origin country with the white’s parental class in the United Kingdom. If immigrant fathers’ class position is similar to white fathers’ class, we would have reasons to believe that they are positively selected, as they tend to come from poor countries with low levels of socio-economic development. Research in this respect does support this idea Li, 2020 ).

3 Prior analysis showed little difference between “routine” and “never worked and long-term unemployed” categories in parental class, thus the two categories were combined to produce a seven-way parental class variable.

4 There are also data on access to Oxford and Cambridge Universities. Further analysis shows that 3.76% of respondents from higher salariat families attended these universities as compared with 0.22% from routine families, a disparity ratio of 17.

5 The two contextual variables recommended by Ilie et al. (2017) are idaci (“income deprivation affecting children index”) and imd (“index of multiple deprivation”). Analyses were conducted including the two variables, rescaled to range from 0 to 100, on top of the variables already in Model 3. The coefficients were rather weak: 0.08 and 0.02 respectively, the latter being non-significant and the coefficients for the other variables in Model 3 being little affected. Given this, the two variables were not included in the model.

6 Further analysis shows that around 5% of the sample were foreign born and came to the United Kingdom as children. Yet, among the foreign born, parental education is more stratified, with 23.8% of parents having degrees or higher, as compared with 18.4% of the United Kingdom born; yet the proportions having only primary or no education were also higher among the foreign born than the United Kingdom born, at 43.4 and 19.1% respectively. The positive selection effect is particularly strong among foreign born Chinese and white parents with 45 and 32% having degree-level education.

7 For brevity, we do not present parental raw class effects on respondent’s education, class and incomes here but will include the effects in the modeling. We have conducted the analysis and found salient effects in each of the domains. For instance, 44% of higher salariat children had degrees as compared with 13% from routine families. Similarly, 53% of the former held salariat positions as compared with 17% of the latter, and differences in “continuous weekly income” amounted to £108 (£343 for the former versus £235 for the latter).

8 An important question in this respect is whether ethnic minorities have equal returns to education in terms of employment opportunities, hence having earnings. Further analysis shows that at the degree level, the two black groups, Indians and Pakistanis were significantly less likely than whites to have a job; at the sub-degree level, Chinese were significantly behind whites; at A-Levels, the three South Asian groups were significantly behind; at the O-Levels, Chinese were significantly behind; and for those with only primary or no formal qualifications, Indians and Chinese are significantly behind. These findings are obtained with all other factors in the models held constant.

9 Again, a relevant question that poses itself is whether there are equal returns of education to earnings. Here, the significant effects are as follows: at the degree level, whites have £46 more than Black Caribbeans; at the sub-degree level whites have £83 and £286 more than Indians and Chinese respectively; at the A-Levels, whites have £93 more than Black Caribbeans, but £167 and £196 less than Indians and Chinese respectively; and at the O-Levels, whites make £70 and £218 more than Pakistani and Chinese respondents respectively, holding constant all other factors in the models.

10 The percentages of respondents who are married or partnered at the age of 25 are 11, 14, 30 and 25 for Whites, Indians, Pakistanis and Bangladeshis respectively. Seven percent of the mixed, six percent of Black Africans are also married. Only two percent of the Black Caribbeans and no one from the Chinese origins are found married.

11 Using information of “limiting long-term illness” does not change the main patterns of the other variables.

12 Even at age 25, 4.5% of the sample were still in education, with Chinese women and Black African men and women being much more so than others, at 29, 12, 12 percent respectively.

Anders, J. (2012). The link between household income, university applications and university attendance. Fisc. Stud. 33 (2), 185–210. doi:10.1111/j.1475-5890.2012.00158.x

CrossRef Full Text | Google Scholar

Anders, J. (2017). The influence of socioeconomic status on changes in young people’s expectations of applying to university. Oxf. Rev. Educ. 43 (4), 381–401. doi:10.1080/03054985.2017.1329722

Barro, R., and Lee, J. W. (2010). A new data set of educational attainment in the world, 1950–2010. NBER Working Paper, 15902 . Cambridge, MA: National Bureau of Economic Research , 1–47.

Google Scholar

Barro, R., and Lee, J. W. (2013). A new data set of educational attainment in the world, 1950–2010. J. Dev. Econ. 104, 184–198. doi:10.1016/j.jdeveco.2012.10.001

Berthoud, R. (2000). Ethnic employment penalties in Britain. J. Ethnic Migrat. Stud. 26, 389–416. doi:10.1080/713680490

Blanden, J., Gregg, P., and Machin, S. (2005). “Educational inequality and intergenerational mobility,” in What’s the good of education? The economics of education in the UK . Editors S. Machin, and A. Vignoles (New Jersey: Princeton University Press ), 99–114.

Boliver, V. (2013). How fair is access to more prestigious UK universities? Br. J. Sociol. 64 (2), 344–364. doi:10.1111/1468-4446.12021

PubMed Abstract | CrossRef Full Text | Google Scholar

Borjas, G. J. (1987). Self-selection and the earnings of immigrants. Am. Econ. Rev. 77 (4), 531–553.

Boudon, R. (1974). Education, opportunity and social inequality, changing prospects in Western Europe . New York, NY: Wiley .

Bourdieu, P. (1990). Distinction . London: Routledge & Kegan Paul .

Bourdieu, P. (1986). “The forms of capital,” in Handbook of theory and research for the sociology of education . Editor J. G. Richardson (New York: Greenwood ), 241–258.

Breen, R., and Goldthorpe, J. H. (1997). Explaining educational differentials: towards a formal rational action theory. Ration. Soc. 9 (3), 275–305. doi:10.1177/104346397009003002

Breen, R., and Goldthorpe, J. H. (2001). Class, mobility and merit. Eur. Socio Rev. 17 (2), 81–101. doi:10.1093/esr/17.2.81

Breen, R., Luijkx, R., Müller, W., and Pollak, R. (2009). Nonpersistent inequality in educational attainment. Evidence from eight European countries. American Journal of Sociology 114: 1475–1521. doi:10.1086/595951

Breen, R., Luijkx, R., Müller, W., and Pollak, R. (2010). Long-term trends in educational inequality in Europe: class inequalities and gender differences. Eur. Socio Rev. 26 (1), 31–48. doi:10.1093/esr/jcp001

Bukodi, E., Erikson, R., and Goldthorpe, J. H. (2014). The effects of social origins and cognitive ability on educational attainment: evidence from Britain and Sweden. Acta Sociol. 57, 293–310. doi:10.1177/0001699314543803

Carmichael, F., and Woods, R. (2000). Ethnic penalties in unemployment and occupational attainment: evidence for Britain. Int. Rev. Appl. Econ. 14 (1), 71–98. doi:10.1080/026921700101498

Connor, H., Tyers, C., Modood, T., and Hillage, J. (2004). Why the difference? a closer look at higher education minority ethnic students and graduates . London: Department of Education and Skills .

Croll, P., and Attwood, G. (2013). Participation in higher education: aspirations, attainment and social background. Br. J. Educ. Stud. 61 (2), 187–202. doi:10.1080/00071005.2013.787386

Duncan, O. D. (1968). “Inheritance of poverty or inheritance of race,” in On understanding poverty: perspectives from the social sciences . Editor D. P Moynihan (New York: Basic Books ), 85–110.

Erikson, R. (1984). Social class of men, women and families. Sociology . 18, 500–514. doi:10.1177/0038038584018004003

Erikson, R., Goldthorpe, J., Jackson, M., Yaish, M., and Cox, D. (2005). On class differentials in educational attainment. Proc. Natl. Acad. Sci. U.S.A. 102 (27), 9730–9733. doi:10.1073/pnas.0502433102

Feliciano, C. (2005). Does selective migration matter? Explaining ethnic disparities in educational attainment among immigrants’ children. Int. Migrat. Rev. 39 (4), 841–871. doi:10.1111/j.1747-7379.2005.tb00291.x

Friedman, S., and Macmillan, L. (2017). Is London really the engine-room? migration, opportunity hoarding and regional social mobility in the UK. Natl. Inst. Econ. Rev. 240 (1), 58–72. doi:10.1177/002795011724000114

Goldthorpe, J. H., and Jackson, M. (2008). “Education-based meritocracy: the barriers to its realisation,” in Social class: how does it work? . Editors A. Lareau, and D. Conley (New York: Russell Sage Foundation ), 93–117.

Goldthorpe, J. H., and McKnight, A. (2006). “The economic basis of social class,” in Mobility and inequality: Frontiers of research in Sociology and economics . Editors S. L. Morgan, D. Grusky, and G. S. Fields (Redwood City, CA: Stanford University Press ), 109–136.

Goldthorpe, J. H., and Mills, C. (2004). “Trends in intergenerational class mobility in Britain in the late Twentieth Century,” in Social Mobility in Europe . Editor R. (Oxford: Oxford University Press) . 195–224.

Goldthorpe, J. H. (2000). On Sociology: numbers, narratives, and the integration of research and theory . Oxford: Oxford University Press .

Goldthorpe, J. H. (2007a). “Cultural capital”: some critical observations. Sociologica 2 (2), 1–23. doi:10.2383/24755

Goldthorpe, J. H. (2007b). On Sociology . Stanford, CA: Stanford University Press , Vol. 2, 365.

Goldthorpe, J. H. (2014). The role of education in intergenerational social mobility: problems from empirical research in sociology and some theoretical pointers from economics. Ration. Soc. 26 (3), 265–289. doi:10.1177/1043463113519068

Heath, A., and Brinbaum, Y. (2014). Unequal Attainments: ethnic educational inequalities in ten Western countries . Oxford: Oxford University Press .

Heath, A., and Li, Y. (2008). Period, life-cycle and generational effects on ethnic minority success in the labour market. Kölner Zeitschrift für Soziologie und Sozialpsychologie. 48, 277–306.

Heath, A., Garratt, E., Kashyap, R., Li, Y., and Richards, L. (2018a). Social progress in Britain . Oxford: Oxford University Press .

Heath, A., Li, Y., and Woerner-Powell, T. (2018b). Trapped in poverty?: a study of transient and persisting factors for muslim disadvantages in the UK. Comparative Islamic Studies. 11 (5), 205–233.

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 labour market. Br. J. Sociol. 70, 1774. doi:10.1111/1468-4446.12676

Halsey, A. H., Heath, A. F., and Ridge, J. M. (1980). Origins and destinations: family, class, and education in modern Britain Oxford: Clarendon Press .

Hirschman, C., and Wong, M. G. (1986). The extraordinary educational attainments of Asian Americans: a search for historical evidence and explanations. Soc. Forces . 65, 1–27. doi:10.1093/sf/65.1.1

Hobbs, G., and Vignoles, A. (2010). Is children’s free school meals eligibility a good proxy for family income? Br. Educ. Res. J. 36, 673–690. doi:10.1080/01411920903083111

Hout, M. (1984). Status, autonomy, and training in occupational mobility. Am. J. Sociol. 89 (6), 1379–1409. doi:10.1086/228020

Ichou, M. (2014). Who they were there: immigrants’ educational selectivity and their children’s educational attainment. Eur. Socio Rev. 30 (6), 750–765. doi:10.1093/esr/jcu071

Ilie, S., Sutherland, A., and Vignoles, A. (2017). ‘Revisiting free school meal eligibility as a proxy for pupil socio‐economic deprivation’. Br. Educ. Res. J. 43 (2), 253–274. doi:10.1002/berj.3260

Jackson, M. (2012). Bold choices: how ethnic inequalities in educational attainment are suppressed. Oxf. Rev. Educ. 38 (2), 189–208. doi:10.1080/03054985.2012.676249

Jackson, M. (2013). “Social background and educational transitions in England,” in Determined to succeed? Performance, choice and education . Editor M. Jackson (Stanford, CA: Stanford University Press ), 253–279.

Jackson, M., Erikson, R., Goldthorpe, J. H., and Yaish, M. (2007). Primary and secondary effects in class differentials in educational attainment: the transition to a level courses in England and Wales. Acta Sociol. 50, 211–229. doi:10.1177/0001699307080926

Kahneman, D. (2011). Think fast and slow . London: Allen Lane .

Labour Force Survey (2015). LFS user guide variable details. http://doc.ukdataservice.ac.uk/doc/7842/mrdoc/pdf/lfs_user_guide_vol3_variabledetails1992-2002.pdf

Lareau, A. (2003). Unequal childhoods: class, race, and family life . Berkeley, CA: University of California Press .

Laurison, D., and Friedman, S. (2016). The class pay gap in britain’s higher professional and managerial occupations. Am. Socio. Rev. 81 (4), 668–695. doi:10.1177/0003122416653602

Lessard-Phillips, L., and Li, Y. (2017). Social stratification of education by ethnic minority groups over generations in the UK. Social Inclusion. 5 (1), 45–54.

Li, Y. (2018a). Against the odds? Educational attainment and labour market position of the second generation minority ethnic members in the UK. Ethnicities 18 (4), 471–495. doi:10.1177/1468796818777546

Li, Y. (2018b). Integration journey: the social mobility trajectory of ethnic minority groups in Britain. Social Inclusion 6 (3), 270–281. doi:10.17645/si.v6i3.1542

Li, Y. (2020). Social progress: social mobility of ethnic minorities in Britain in the last fifty years (1972–2019)—A report for the commission on race and ethnic disparities. London: The Cabinet Office.

Li, Y., and Devine, F. (2011). Is social mobility really declining? Intergenerational class mobility in Britain in the 1990s and the 2000s. Socio. Res. Online 16 (3), 28–41. http://www.socresonline.org.uk/16/3/4.html . doi:10.5153/sro.2424

Li, Y., and Heath, A. (2008). Ethnic minority men in British labour market (1972–2005). Int. J. Sociol. Soc. Pol. 28 (5–6), 231–244. doi:10.1108/01443330810881277

Li, Y., and Heath, A. (2016). Class matters: a study of minority and majority social mobility in Britain, 1982-2011. Am. J. Sociol. 122 (1), 162–200. doi:10.1086/686696

Li, Y., and Heath, A. (2018). Persisting disadvantages: a study of labour market dynamics of ethnic unemployment and earnings in the UK (2009–2015). J. Ethnic Minor. Stud. 46 (5), 857–878. doi:10.1080/1369183x.2018.1539241

Li, Y., Pickles, A., and Savage, M. (2005). Social capital and social trust in Britain. Eur. Socio Rev. 21 (2), 109–123. doi:10.1093/esr/jci007

Lucas, S. R. (2001). Effectively maintained inequality: education transitions, track mobility, and social background effects. Am. J. Sociol. 106, 1642–1690. doi:10.1086/321300

May, T. (2017). Prime Minister launches world-leading project on impact of ethnicity on everyday life. United Kingdom: Gov Retrieved from: www.gov.uk/government/news/prime-minister-launches-worldleading-project-on-impact-of-ethnicity-on-everyday-life (Accessed November 12, 2017)

Modood, T. (2005). “The educational attainments of ethnic minorities in Britain,” in Ethnicity, Social mobility and public policy: comparing the US and UK . Editors G. C. Loury, T. Modood, and S. M. Teles (Cambridge: Cambridge University Press ), 288–308.

Portes, A., and Zhou, M. (1993). The new second generation: segmented assimilation and its variants among post-1965 immigrant youth. Ann. Am. Acad. Pol. Soc. Sci. 530, 74–96.

Portes, A., Fernández-Kelly, P., and Haller, W. (2009). The adaptation of the immigrant second generation: theoretical overview and recent evidence. J. Ethnic Migrat. Stud. 35 (7), 1077–1104. doi:10.1080/13691830903006127

Putnam, R. D. (2007). E. pluribus unum : diversity and community in the twenty-first century, the 2006 johan skytte prize lecture. Scand. Polit. Stud. 30 (2), 137–174. doi:10.1111/j.1467-9477.2007.00176.x

Raftery, A. E., and Hout, M. (1993). Maximally maintained inequality: expansion, reform, and opportunity in Irish education: 1921–75. Sociol. Educ. 66 (1), 41–62. doi:10.2307/2112784

Rafferty, A. (2012). Ethnic penalties in graduate level over-education, unemployment and wages: evidence from Britain. Work. Employ. Soc. 26 (6), 987–1006. doi:10.1177/0950017012458021

Robinson, W. S. (1951). The logical structure of analytical induction. Am. Socio. Rev. 16, 812–818. doi:10.2307/2087508

Sakamoto, A., Goyette, K. A., and Kim, C. (2009). Socioeconomic attainments of Asian Americans. Annu. Rev. Sociol. 35, 255–276. doi:10.1146/annurev-soc-070308-115958

Siddiqui, N., Bolivia, V., and Gorard, S. (2019). Reliability of longitudinal social surveys of access to higher education: the case of next Steps in England. Soc. Incl. 7 (1), 80–89. doi:10.17645/si.v7i1.1631

Social MobilityChild Poverty Commission (2015). Bridging the social divide. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/408405/Bridging_the_Social_Divide_Report.pdf

Social MobilityChild Poverty Commission (2016). The social mobility index. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/496103/Social_Mobility_Index.pdf

Spence, M. (1973). Job market signalling. Q. J. Econ. 87 (3), 355–374. doi:10.2307/1882010

Strand, S. (2007). Minority ethnic pupils in the longitudinal study of young people in England. Department for Children, Schools and Families , DCSF Research Report RR-002

Waters, M., Tran, V., Kasinitz, P., and Mollenkopf, J. (2010). Segmented assimilation revisited: types of acculturation and socioeconomic mobility in young adulthood. Ethn. Racial Stud. 33 (7), 1168–1193. doi:10.1080/01419871003624076

Weiss, A. (1995). Human capital vs. signalling explanations of wages. J. Econ. Perspect. 9, 133–154. doi:10.1257/jep.9.4.133

Wood, M., Hales, H., Purdon, S., Sejersen, T., and Hayllar, O. (2009). A test for racial discrimination in recruitment practice in British cities. Leeds: Corporate Document Services . DWP Research Report 607. doi:10.5149/9780807878118_wood

Zwysen, W., and Longhi, S. (2018). Employment and earning differences in the early career of ethnic minority British graduates: the importance of university career, parental background and area characteristics. J. Ethnic Migrat. Stud. 44 (1), 154–172. doi:10.1080/1369183x.2017.1338559

Keywords: class, ethnicity, gender, educational attainment, labor market position, England

Citation: Li Y (2021) Entrenched Inequalities? Class, Gender and Ethnic Differences in Educational and Occupational Attainment in England. Front. Sociol. 5:601035. doi: 10.3389/fsoc.2020.601035

Received: 31 August 2020; Accepted: 22 December 2020; Published: 28 January 2021.

Reviewed by:

Copyright © 2021 Li. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Yaojun Li, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

Class Differences and Impact on Student Access and Outcomes

2020, The SAGE Encyclopedia of Higher Education

The study of class differences and social mobility remain areas of fascination for sociologists. Debates con- cerning how class plays out in education, specifically higher education, have focused on many different areas from the effects of poverty, first-in-family (first generation) status, government efforts to widen participation, as well as entitlement, resilience, competition, and intergenerational histories. Regardless of the focus, there exist great disparities for those entering higher education, which both highlight pervasive inequality and show the power of class to influence opportunity and life chances. As Muriel Egerton and A. H. Halsey have written, three significant aspects shaped conversations regarding access to higher education over the 20th century. These include, first, a period of significant expansion; second, a reduction in gender inequality; and, third, little to no reduction in relative social class inequality. This entry highlights the main historic trends and theoretical approaches in the research on social class. This foundational knowledge illustrates how theorists have come to think about class and how those approach- es inform current research in higher education. The entry then discusses research on the influence of class in students’ higher education experiences and postgraduate outcomes and potential future directions for re- search on class differences and higher education.

Related Papers

This chapter investigates the continuing problem of social class in the age of mass higher education. After a brief overview of the history of social class in higher education, it maps out the contemporary landscape focusing on, but also looking beyond, the statistics. The main part of the chapter highlights the actual experiences of working class students in UK higher education, drawing on data from two ESRC projects of classed experiences of higher education in the 21st century (Reay et al 2005; Reay et al 2010). The massification and democratisation of higher education has led to the admittance of far more working class students, but it has also resulted in a steeply hierarchical and stratified system with working class students, for the most part, clustered in the low status, poorly resourced, institutions (Crozier et al 2009; Stich 2012).

Alistair G Ross

Critical Studies in Education

Maree Martinussen

This article explores the role of affect in addressing the advantage conventionally accorded to high socioeconomic status (SES) in higher education and how this advantage plays out for students from low SES backgrounds. Positioned as the 'other' to an assumed norm, the capacities of these students can be considered the 'wrong' capacities, such that privilege prevails. Drawing on interview data from a project undertaken in Australia with female postgraduate students from low SES backgrounds, we bring a pluralised affective capacities approach to bear. We argue that thinking class (dis)advantage with affect has considerable political potential. Affect emerges as a key site through which the normative and transformative capacities of the classed subject emerge. By attuning to affective dissonance, responsivity and capacities, we aim to challenge the advantage afforded high socioeconomic status in higher education. We demonstrate how a focus on affective relations can create more complex constructions of 'advantage' and 'participation' in higher education, disrupting deficit models.

Studies in the Education of Adults

Social class is a major determining factor of people’s life chances. Much sociologically based research shows that socio-economic position is still one of the best predictors of who will achieve success, prosperity, and social status, and in particular who will enjoy the highest levels of educational outcomes. Survey data and qualitative studies alike also confirm that many people continue to see class as a feature of everyday life, in ways that are also connected with their understanding of learning and its possibilities. Yet, despite its continued significance in people’s lives, class has virtually disappeared from modern adult learning research. The paper concludes that in an era when governments across Europe are setting about the dismantling of social support and collective protection produced by social democracy and trade unionism, class analysis presents an important means of understanding change and changing understanding.

Benjamin Brundu-Gonzalez

We report the results of a study of LSE home undergraduate students which addresses the significance of social class background in shaping a range of student outcomes. We explore how class background and other sociodemographic variables affect access (who gets in), study choice (who studies what), attainment (how students perform in summative assessment), and satisfaction (how students rate their programme). We show that parental class background plays a major role across all the dimensions and is a major force shaping LSE undergraduate student outcomes. This is evident from observing raw bivariate associations and remains true when we report linear regression models controlling for numerous other socio-demographic and institutional factors. We also demonstrate powerful intersectional associations, especially with race, and also with declared disability status. Our results underscore the need to take social class seriously in the analysis of the undergraduate experience, both in analytical and in policy terms.

British Journal of Sociology of education

Kathleen Lynch

Universities in Transition: Foregrounding Social Contexts of Knowledge in the First Year Experience, edited by Heather Brook, Deane Fergie, Michael Maeorg & Dee Michell

Dee Michell

In this chapter we introduce the term ‘classism’ into the higher education debate in Australia. By ‘classism’ we mean the tendency to construct people from low socio-economic status (SES) backgrounds as inherently deficient according to prevailing normative values. Using an analysis of the Bradley Review, we show that low SES students are constructed as inherently lacking in aspirations in current policy discourse and are regarded as ‘needier’ higher education students in comparison with their higher SES peers. This construction, we argue, is an example of classism, and therefore we suggest that adding ‘classism’ to existing understandings of disadvantage will help to raise awareness of discrimination as well as formulate best practice in higher education.

‘One of the most pessimistic statistics about modern Britain is the lack of improvement in social mobility over recent decades, but one of the most optimistic facts is the way in which higher education can wipe out prior educational advantage’. David Willetts, (www.bis.gov.uk, 2011:6) We have entered a political era of a coalition government where ‘realpolitik’ is the rhetoric of politicians looking to remain electorally palatable. David Willett’s language again represents the potential of higher education to assist social mobility however the historical realism of the graph suggests otherwise! The purpose of this report will be therefore to explore representations and realisms of social class in HE and ask whether Widening Participation (WP) can ever overcome the hegemony of a two tiered HE system, ideologically entrenched in meritocratic principles and where ‘elitism is built into the very fabric of HE’ (Reay et al., 2005: 163

Louise Archer

Abstract At a time when the rhetoric of the new Labour Government in the UK is celebrating the expansion of higher education, the widening of access and the essential fairness of the meritocratic principle, this paper draws on data that challenges and problematises such comforting notions and argues that the welcome expansion of higher education has been accompanied by a deepening of social stratification within HE.

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

RELATED PAPERS

David James

Tallulah Eyres

Berenice Scandone

Paul Wakeling

Nida Denson

Handbook on Promoting Social Justice in Education

Philip Woodward

Pedagogy, Culture and Society

Carole Leathwood

On Education

Colin J Samson , Chris Cunningham

Australian Journal of Education

Sociological Review

Jacqueline Davies

Higher Education

Rosemary Deem

British journal of sociology of …

Panagiotis Sotiris

Open Access Publishing Group

Educational Researcher, 43, 196-200

Nida Denson , Mark Rubin , David Zyngier , Sue Kilpatrick

British Journal of Sociology of Education

Barbara Merrill

Rihab Ghaoui

National Institute Economic Review

Anna Vignoles , Oscar Marcenaro

British Educational Research Journal

Gill Crozier

fatma ayyad

European Journal of Education

The Radical Teacher

John Alberti

RELATED TOPICS

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

• Corpus ID: 226366493

Social Class Differences in Students’ Experiences during the COVID-19 Pandemic

• Krista M. Soria , Bonnie Horgos
• Published 17 September 2020
• Sociology, Education

Figures and Tables from this paper

17 Citations

Educational inequality and covid‑19 pandemic: relationship between the family socio-economic status and student experience of remote learning, gendered aspects of undergraduate student experience in the time of covid-19: a case study of nazarbayev university, comparing college students’ capacities for resiliency before and during the covid-19 pandemic, student support as social network: exploring non-traditional student experiences of academic and wellbeing support during the covid-19 pandemic, socioeconomic inequalities in the incidence of covid-19 in barcelona students, risk and protective factors of college students’ psychological well-being during the covid-19 pandemic: emotional stability, mental health, and household resources, what do our students think perceptions of transitioning to remote learning during the pandemic at land-grant universities, impact of covid-19 on medical education: perspectives from students, perceived scarcity of job opportunities and job search: an evolutionary life history perspective, role of social determinants in anxiety and depression symptoms during covid-19: a longitudinal study of adults in north carolina and massachusetts.

• Highly Influenced

6 References

Development and validity of a 2-item screen to identify families at risk for food insecurity, the patient health questionnaire-2: validity of a two-item depression screener, anxiety disorders in primary care: prevalence, impairment, comorbidity, and detection, related papers.

Showing 1 through 3 of 0 Related Papers

Identities in Context: How Social Class Shapes Inequalities in Education

• First Online: 01 November 2019

Cite this chapter

• Matthew J. Easterbrook 2 ,
• Ian R. Hadden 2 &
• Marlon Nieuwenhuis 3  

3894 Accesses

10 Citations

Educational inequalities between social classes are large and persistent in the UK. Students from economically disadvantaged backgrounds have much lower attainment and engage less with education than their peers from advantaged backgrounds. Although structural factors contribute significantly to these inequalities, social psychological processes also play a crucial but less visible role. We draw on the social identity approach to propose a new model of how social and cultural factors in the local educational context shape the meaning of people’s social class identities in ways that create and sustain inequalities. Our identities-in-context model brings into focus educational contexts in which lower-class people are expected to perform badly, are not well represented in high-status educational roles or institutions, and are negatively disposed toward education. We argue that, for lower-class people, these contexts ignite a sense of social identity threat and incompatibility between their background and doing well in education. These, in turn, lead to poorer educational outcomes. We propose ways in which our model can be used to inform social psychological interventions that aim to reduce educational inequalities between social classes.

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

Access this chapter

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• 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

Similar content being viewed by others

Researching Class and Higher Education

Education, Social Class and Marxist Theory

Can children break the cycle of disadvantage structure and agency in the transmission of education across generations.

Although there are important differences between the different indicators of socioeconomic status and social class, here we use the term lower class as a general term indicating lower socioeconomic status and lower social class. We do this in order to be consistent in our writing throughout the chapter and to avoid getting side-tracked by technical discussions that could detract from our main focus of educational inequalities.

Andrews, J., Robinson, D., & Hutchinson, J. (2017). Closing the gap? Trends in educational attainment and disadvantage. Education Policy Institute. Retrieved from: http://dera.ioe.ac.uk/29787/1/Closing-the-Gap_EPI.pdf

Arulampalam, W., Naylor, R. A., & Smith, J. P. (2005). Effects of in-class variation and student rank on the probability of withdrawal: Cross-section and time-series analysis for UK university students. Economics of Education Review, 24 (3), 251–262. https://doi.org/10.1016/j.econedurev.2004.05.007

Article   Google Scholar  

Autin, F., Batruch, A., & Butera, F. (2015). Social justice in education: How the function of selection in educational institutions predicts support for (non)egalitarian assessment practices. Frontiers in Psychology, 6 , 1–13. https://doi.org/10.3389/fpsyg.2015.00707

Bandura, A., & Caprara, G. V. (1996). Multifaceted impact of self-efficacy beliefs on academic functioning. Child Development, 67 (3), 1206–1222.

Article   PubMed   Google Scholar  

Batruch, A., Autin, F., & Butera, F. (2017). Re-establishing the social-class order: Restorative reactions against high-achieving, low-SES pupils. Journal of Social Issues, 73 (1), 42–60. https://doi.org/10.1111/josi.12203

Bennett, M., & Sani, F. (2008). Children’s subjective identification with social groups: A group-reference effect approach. British Journal of Developmental Psychology, 26 (3), 381–387. https://doi.org/10.1348/026151007X246268

Borman, G. D., Grigg, J., Rozek, C. S., Hanselman, P., & Dewey, N. A. (2018). Self-affirmation effects are producted by school context, student engagement with the intervention, and time: Lessons from a district-wide implementation. Psychological Science, 29 , 1773–1784. https://doi.org/10.1177/0956797618784016

Bourdieu, P. (1974). The school as a conservative force: Scholastic and cultural inequalities. Contemporary Research in the Sociology of Education, 32 , 32–36.

Google Scholar  

Bourdieu, P. (1984). Distinction: A social critique of the judgement of taste . Cambridge, MA: Harvard University Press.

Bourdieu, P. (1985). The social space and the genesis of groups.pdf. Theory and Society, 14 (6), 723–744. https://doi.org/10.2307/657373

Bourdieu, P., & Passeron, J.-C. (1977). Education, society and culture . Beverly Hills, CA: Sage Publications.

Breda, T., Jouini, E., & Napp, C. (2018). Societal inequalities amplify gender gaps in math. Science, 359 , 1219–1220. https://doi.org/10.1126/science.aar2307

Browman, A. S., Destin, M., Carswell, K. L., & Svoboda, R. C. (2017). Perceptions of socioeconomic mobility influence academic persistence among low socioeconomic status students. Journal of Experimental Social Psychology, 72 , 45–52. https://doi.org/10.1016/j.jesp.2017.03.006

Brown, A., & Dittmar, H. (2005). Think “thin” and feel bad: The role of appearance schema activation, attention level, and thin–ideal internalization for young women’s responses to ultra–thin media ideals. Journal of Social and Clinical Psychology, 24 (8), 1088–1113. https://doi.org/10.1521/jscp.2005.24.8.1088

Cohen, G. L., Garcia, J., Apfel, N., & Master, A. (2006). Reducing the racial achievement gap: A social-psychological intervention. Science, 313 (5791), 1307–1310. https://doi.org/10.1126/science.1128317

Cohen, G. L., & Sherman, D. K. (2014). The psychology of change: Self-affirmation and social psychological intervention. Annual Review of Psychology, 65 , 333–371. https://doi.org/10.1146/annurev-psych-010213-115137

Croizet, J., & Claire, T. (1998). Extending the concept of stereotype threat to social class: The intellectual underperformance of students from low socioeconomic backgrounds. Personality and Social Psychology Bulletin, 24 (6), 588–594. https://doi.org/10.1177/0146167298246003

Dasgupta, N. (2011). Ingroup experts and peers as social vaccines who inoculate the self-concept: The stereotype inoculation model. Psychological Inquiry, 22 (4), 231–246. https://doi.org/10.1080/1047840X.2011.607313

Department for Education. (2015). GCSE and equivalent attainment by pupil characteristics, 2013 to 2014 (Revised) . London: Crown Copyright.

Désert, M., Préaux, M., & Jund, R. (2009). So young and already victims of stereotype threat: Socio-economic status and performance of 6 to 9 years old children on Raven’s progressive matrices. European Journal of Psychology of Education, 24 (2), 207–218. http://link.springer.com/article/10.1007/BF03173012

Diermeier, M., Goecke, H., & Niehues, J. (2017). Impact of inequality-related media coverage on the concerns of the citzens . DICE Discussion Paper, No. 258. Düsseldorf: Düsseldorf Institute for Competition Economics (DICE).

Durante, F., & Fiske, S. T. (2017). How social-class stereotypes maintain inequality. Current Opinion in Psychology, 18 , 43–48. https://doi.org/10.1016/j.copsyc.2017.07.033

Article   PubMed   PubMed Central   Google Scholar  

Easterbrook, M. J. (2018). Psychological barriers faced by first-generation students in the UK . Manuscript in preparation.

Easterbrook, M. J., Kuppens, T., & Manstead, A. S. R. (2015). The education effect: Higher educational qualifications are robustly associated with beneficial personal and socio-political outcomes. Social Indicators Research, 126 (3), 1261–1298. https://doi.org/10.1007/s11205-015-0946-1

Easterbrook, M. J., Nieuwenhuis, M., Fox, K., Harris, P. R., & Banerjee, R. A. (2018). Social psychological processes and academic performance . Manuscript in preparation, University of Sussex, UK.

Education Endowment Foundation. (2017). The attainment gap. Retrieved from https://educationendowmentfoundation.org.uk/public/files/Annual_Reports/EEF_Attainment_Gap_Report_2018.pdf .

Ellemers, N., van Knippenberg, A., De Vries, N., & Wilke, H. (1988). Social identification and permeability of group boundaries. European Journal of Social Psychology, 18 (6), 497–513. https://doi.org/10.1002/ejsp.2420180604

Ellemers, N., & van Laar, C. (2010). Individual mobility. In The SAGE handbook of prejudice, stereotyping and discrimination (pp. 561–576). London: SAGE Publications Ltd.. https://doi.org/10.4135/9781446200919.n34

Chapter   Google Scholar  

Elmore, K. C., & Oyserman, D. (2012). If “we” can succeed, “I” can too: Identity-based motivation and gender in the classroom. Contemporary Educational Psychology, 37 (3), 176–185. https://doi.org/10.1016/j.cedpsych.2011.05.003

Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7 (2), 117–140. https://doi.org/10.1177/001872675400700202

Flore, P. C., & Wicherts, J. M. (2015). Does stereotype threat influence performance of girls in stereotyped domains? A meta-analysis. Journal of School Psychology, 53 (1), 25–44. https://doi.org/10.1016/j.jsp.2014.10.002

Gibson, D. E., & Cordova, D. I. (1999). Women’s and men’s role models: The importance of exemplars. In A. J. Murrell, F. J. Crosby, & R. J. Ely (Eds.), Mentoring dilemmas: Developmental relationships within multicultural organizations. Applied social research (pp. 121–141). London: Lawrence Erlbaum Associates.

Good, C., Aronson, J., & Inzlicht, M. (2003). Improving adolescents’ standardized test performance: An intervention to reduce the effects of stereotype threat. Journal of Applied Developmental Psychology, 24 (6), 645–662. https://doi.org/10.1016/j.appdev.2003.09.002

Goudeau, S., & Croizet, J. C. (2017). Hidden advantages and disadvantages of social class: How classroom settings reproduce social inequality by staging unfair comparison. Psychological Science, 28 (2), 162–170. https://doi.org/10.1177/0956797616676600

Hadden, I., Easterbrook, M. J., Nieuwenhuis, M., Fox, K., & Dolan, P. (2019). Overcoming stereotype threat in schools: Self-affirmation reduces the socioeconomic attainment gap in England. British Journal of Educational Psychology . https://doi.org/10.111/bjep.12291

Hanselman, P., Bruch, S. K., Gamoran, A., & Borman, G. D. (2014). Threat in context: School moderation of the impact of social identity threat on racial/ethnic achievement gaps. Sociology of Education, 87 (2), 106–124. https://doi.org/10.1177/0038040714525970

Harackiewicz, J. M., & Priniski, S. J. (2018). Improving student outcomes in higher education: The science of targeted intervention. Annual Review of Psychology, 69 (1), 409–435. https://doi.org/10.1146/annurev-psych-122216-011725

Heckman, J. J. (2011). The economics of inequality: The value of early childhood education. American Educator, 35 , 31–36. http://files.eric.ed.gov/fulltext/EJ920516.pdf

Hernandez, D., Rana, S., Rao, A., & Usselman, M. (2017). Dismantling stereotypes about Latinos in STEM. Hispanic Journal of Behavioral Sciences, 39 (4), 436–451. https://doi.org/10.1177/0739986317731100

Huat See, B., & Gorard, S. (2015). The role of parents in young people’s education—A critical review of the causal evidence. Oxford Review of Education, 41 (3), 346–366. https://doi.org/10.1080/03054985.2015.1031648

Hutchinson, J., Dunford, J., & Treadaway, M. (2016). Divergent pathways: The disadvantage gap, accountability and the pupil premium . London, UK: Education Policy Institute.

Heiserman, N., Simpson, B. (2017). Higher inequality increases the gap in the perceived merit of the rich and poor. Social Psychology Quarterly, 80 (3), 243–253. https://doi.org/10.1177/0190272517711919

Inzlicht, M., & Ben-zeev, T. (2000). A threatening intellectual environment: Why females are susceptible to experiencing problem-solving deficits in the presence of males. Psychological Science, 11 (5), 365–371.

Iyer, A., Jetten, J., Tsivrikos, D., Postmes, T., & Haslam, S. A. (2009). The more (and the more compatible) the merrier: Multiple group memberships and identity compatibility as predictors of adjustment after life transitions. The British Journal of Social Psychology, 48 , 707–733. https://doi.org/10.1348/014466608X397628

Iyer, A., Zhang, A., Jetten, J., Hao, Z., & Cui, L. (2017). The promise of a better group future: Cognitive alternatives increase students’ self-efficacy and academic performance. British Journal of Social Psychology, 56 (4), 750–765. https://doi.org/10.1111/bjso.12201

Jackman, M. R. (1994). The velvet glove: Paternalism and conflict in gender, class, and race relations. In Contemporary sociology . Berkeley, CA: University of California Press.

Jerrim, J., Chmielewski, A. K., & Parker, P. (2015). Socioeconomic inequality in access to high-status colleges: A cross-country comparison. Research in Social Stratification and Mobility, 42 , 20–32. https://doi.org/10.1016/j.rssm.2015.06.003

Jones, O. (2011). Chavs: The demonization of the working class . London: Verso.

Jetten, J. (2018). The wealth paradox: Prosperity and opposition to immigration. European Journal of Social Psychology. Advance online publication. https://doi.org/10.1002/ejsp.2552

Kuppens, T., Spears, R., Manstead, A. S. R., Spruyt, B., & Easterbrook, M. J. (2017). Educationism and the irony of meritocracy: Negative attitudes of higher educated people towards the less educated. Journal of Experimental Social Psychology, 76 , 429–447. https://doi.org/10.1016/j.jesp.2017.11.001

Leyens, J.-P., Desert, M., Croizet, J.-C., & Darcis, C. (2000). Stereotype threat: Are lower status and history of stigmatization preconditions of stereotype threat? Personality and Social Psychology Bulletin, 26 (10), 1189–1199. https://doi.org/10.1177/0146167200262002

Lindqvist, A., Björklund, F., & Bäckström, M. (2017). The perception of the poor: Capturing stereotype content with different measures. Nordic Psychology, 69 (4), 231–247. https://doi.org/10.1080/19012276.2016.1270774

Loughnan, S., Haslam, N., Sutton, R. M., & Spencer, B. (2013). Dehumanization and social class: Animality in the stereotypes of “White Trash,” “Chavs,” and “Bogans.”. Social Psychology, 45 (1), 54–61. https://doi.org/10.1027/1864-9335/a000159

Machin, S., & Vignoles, A. (2005). Educational inequality: The widening socio-economic gap. Fiscal studies, 25 , 107–128. https://doi.org/10.111/j.1475-5890.2004.tb00099.x

Malligan, K., Moretti, E., & Oreopoulos, P. (2004). Does education improve citizenship? Evidence from the United States and the United Kingdom. Journal of Public Economics, 88 , 1677–1695. https://doi.org/10.1016/j.jpubeco.2003.10.005

Markus, H. R., & Nurius, P. (1986). Possible selves. American Psychologist, 41 (9), 954–969. https://doi.org/10.1037/0003-066X.41.9.954

Nieuwenhuis, M., Manstead, A. S. R., Easterbrook, M. J. (2019). Accounting for unequal access to higher education: The role of social identity factors . Group Processes and Intergroup Relations, 22(3), 371–389. https://doi.org/10.1177/1368430219829824

Oyserman, D., Bybee, D., & Terry, K. (2006). Possible selves and academic outcomes: How and when possible selves impel action. Journal of Personality and Social Psychology, 91 (1), 188–204. https://doi.org/10.1037/0022-3514.91.1.188

Oyserman, D., & James, L. (2011). Possible identities. In S. J. Schwartz, K. Luyckx, & V. L. Vignoles (Eds.), Handbook of identity theory and research (pp. 117–145). New York: Springer. https://doi.org/10.1007/978-1-4419-7988-9

Rubin, M. (2012). Working-class students need more friends at university: A cautionary note for Australia’s higher education equity initiative. Higher Education Research and Development, 31 (3), 431–433. https://doi.org/10.1080/07294360.2012.689246

Schweinle, A., & Mims, G. A. (2009). Mathematics self-efficacy: Stereotype threat versus resilience. Social Psychology of Education, 12 (4), 501–514. https://doi.org/10.1007/s11218-009-9094-2

Shanks, T. R. W., & Destin, M. (2009). Parental expectations and educational outcomes for young African American adults: Do household assets matter? Race and Social Problems, 1 , 27–35. https://doi.org/10.1007/s12552-009-9001-7

Shutts, K., Brey, E. L., Dornbusch, L. A., Slywotzky, N., & Olson, K. R. (2016). Children use wealth cues to evaluate others. PLoS One, 11 (3), 1–21. https://doi.org/10.1371/journal.pone.0149360

Spencer, B., & Castano, E. (2007). Social class is dead. Long live social class! Stereotype threat among low socioeconomic status individuals. Social Justice Research, 20 (4), 418–432. https://doi.org/10.1007/s11211-007-0047-7

Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual performance of African Americans. Journal of Personality and Social Psychology, 69 (5), 797–811.

Stephens, N. M., Fryberg, S. A., Markus, H. R., Johnson, C. S., & Covarrubias, R. (2012). Unseen disadvantage: How American universities’ focus on independence undermines the academic performance of first-generation college students. Journal of Personality and Social Psychology, 102 (6), 1178–1197. https://doi.org/10.1037/a0027143

Stephens, N. M., Markus, H. R., & Phillips, L. T. (2014). Social class culture cycles: How three gateway contexts shape selves and fuel inequality. Annual Review of Psychology, 65 (1), 611–634. https://doi.org/10.1146/annurev-psych-010213-115143

Stephens, N. M., Townsend, S. S. M., Markus, H. R., & Phillips, L. T. (2012). A cultural mismatch: Independent cultural norms produce greater increases in cortisol and more negative emotions among first-generation college students. Journal of Experimental Social Psychology, 48 (6), 1389–1393. https://doi.org/10.1016/j.jesp.2012.07.008

Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey, CA: Brooks/Cole.

Te Wang, M., & Sheikh-Khalil, S. (2014). Does parental involvement matter for student achievement and mental health in high school? Child Development, 85 (2), 610–625. https://doi.org/10.1111/cdev.12153

Turner, J. C. (2006). Social influence . Belmont, CA: ThomsonBrooks/Cole Publishing.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. (1987). Rediscovering the social group: A self-categorization theory . Oxford, UK: Basil Blackwell Ltd..

Tylson, A., & Maniam, S. (2016). Behind Trump’s victory: Divisions by race, gender, education. Pew Research Centre. Retrieved from http://www.pewresearch.org/fact-tank/2016/11/09/behind-trumps-victory-divisions-by-race-gender-education/

UCAS. (2017). End of cycle report 2017. UCAS Analysis and Research. Retrieved from https://www.ucas.com/data-and-analysis/ucas-undergraduate-releases/ucas-undergraduate-analysis-reports/2017-end-cycle-report

Van Laar, C., Derks, B., Ellemers, N., & Bleeker, D. (2010). Valuing social identity: Consequences for motivation and performance in low-status groups. Journal of Social Issues, 66 (3), 602–617. https://doi.org/10.1111/j.1540-4560.2010.01665.x

Walton, G. M. (2014). The new science of wise psychological interventions. Current Directions in Psychological Science, 23 (1), 73–82. https://doi.org/10.1177/0963721413512856

Walton, G. M., & Spencer, S. J. (2009). Intellectual ability of negatively stereotyped students. Psychological Science, 20 (9), 1132–1139. https://doi.org/10.1111/j.1467-9280.2009.02417.x

Yeager, D. S., & Walton, G. M. (2011). Social-psychological interventions in education: They’re not magic. Review of Educational Research, 81 (2), 267–301. https://doi.org/10.3102/0034654311405999

Zhang, A. (2018). New findings on key factors influencing the UK’s referendum on leaving the EU. World Development, 102 , 304–314. https://doi.org/10.1016/j.worlddev.2017.07.017

Zhan, M. (2006). Assets, parental expectations and involvement, and children’s educational performance. Children and Youth Services Review, 28 (8), 961–975. https://doi.org/10.1016/j.childyouth.2005.10.008

Download references

Acknowledgements

We would like to thank Tony Manstead for his helpful comments on a previous version of the chapter.

Author information

Authors and affiliations.

University of Sussex, Brighton, UK

Matthew J. Easterbrook & Ian R. Hadden

University of Twente, Enschede, The Netherlands

Marlon Nieuwenhuis

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Matthew J. Easterbrook .

Editor information

Editors and affiliations.

School of Psychology, University of Queensland, St. Lucia, QLD, Australia

Jolanda Jetten  & Kim Peters  & 

Rights and permissions

Reprints and permissions

Copyright information

© 2019 Springer Nature Switzerland AG

About this chapter

Easterbrook, M.J., Hadden, I.R., Nieuwenhuis, M. (2019). Identities in Context: How Social Class Shapes Inequalities in Education. In: Jetten, J., Peters, K. (eds) The Social Psychology of Inequality. Springer, Cham. https://doi.org/10.1007/978-3-030-28856-3_7

Download citation

DOI : https://doi.org/10.1007/978-3-030-28856-3_7

Published : 01 November 2019

Publisher Name : Springer, Cham

Print ISBN : 978-3-030-28855-6

Online ISBN : 978-3-030-28856-3

eBook Packages : Behavioral Science and Psychology Behavioral Science and Psychology (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

• Utility Menu

Jennifer L. Hochschild

H.l. jayne professor of government, and professor of african and african american studies.

Center for Government and International Studies 1737 Cambridge Street, Cambridge, MA 02138

Social Class in Public Schools

Date published:.

Running head:  SOCIAL CLASS IN PUBLIC SCHOOLS

Departments of Government and Afro-American Studies

Harvard University

Journal of Social Issues,  November 2003

*Correspondence for this article should be addressed to: Jennifer Hochschild, Government Department, Littauer Center, North Harvard Yard, Harvard University, Cambridge MA 02138, E-mail: [ [email protected] ].  This article derives from (Hochschild & Scovronick, 2003).  My thanks to Nate Scovronick for all his work on our joint venture, and to Elizabeth Cole and Joan Ostrove for their engagement and encouragement. 

Contact Information, Not for Publication:

Phone: (617) 496-0181

Fax: (617) 495-0438.

This article shows the pattern of socioeconomic class differences in schooling outcomes and indicates some of the causes for those differences that lie within the public realm. Those causes include “nested inequalities” across boundaries of states, school districts, schools within a district, classes within a school, and sometimes separation within a class.  Urban public schools demonstrate a particular set of problems that generate differential schooling outcomes by economic class.  The article also demonstrates ways in which class biases are closely entwined with racial and ethnic inequities.  It concludes with the broad outlines of what would be necessary to reduce class (and racial) disparities in American public schools. 

The American dream will succeed or fail in the 21 st century in direct proportion to our commitment to educate every person in the United States of America.

-- President Bill Clinton, 1995 (Clinton, 1995: 617)  

There is no greater test of our national responsibility than the quality

of the education we provide.

                       --Democratic presidential candidate Al Gore, 2000 (Gore, 2000)

Both parties have been talking about education for quite a while.  It’s time to come together to get it done, so that we can truthfully say in America: No child will be left behind.

-- President George W. Bush, 2001 (Bush, 2001)

That presidents and candidates were all saying the same thing is no coincidence. They were echoing what the American public said in survey after survey throughout the past decade: education is “the most important problem facing the nation” (e.g. CNN/ U.S.A. Today , 2000; Gallup Organization, 2000), or “most important in [my] vote for president” (e.g. ABC News, 2000; Kaiser Family Foundation & Harvard University School of Public Health, 2000; see also Wilgren, 2000).  In the election of November 2000, fourteen states offered 24 measures about K-12 schooling for citizens to vote on directly. The Economist lectured Britain’s former subjects that the next American president “will have to get to grips with… the public education system.  This is America’s last best chance to tackle” what it called the “failure” of public schooling ("And Now, Mr. President...", 2000: 27).

Citizens, politicians, and journalists are correct, at least about the importance of schools.  Education largely and increasingly determines an individual’s job choice and income (Danziger & Reed, 1999, p.16). It more and more determines whom one will marry (Kalmijn, 1991; Mare, 1991). It has more impact than any other factor, possibly excepting wealth, on whether one participates in politics, what one believes politically, and how much political influence one has (Verba, 2001; Verba, Schlozman, & Brady, 1995). It is the arena in which the United States has sought to overcome racial domination and class hierarchy, to turn immigrants into Americans, to turn children into responsible citizens, to create and maintain our democracy (Cremin, 1988; Gutmann, 1987; Kluger, 1975; Spring, 2000; Tyack, 1974).               

In many ways public schools in the United States have responded to these aspirations.  Compared with a few decades ago, dropout rates have declined (National Center for Education Statistics, 2002, tables 108, 109); children with disabilities are in school buildings rather than institutions that could be described as “human warehouses” (Braddock & Parish, 2001; McDonnell, McLaughlin, & Morison, 1997; National Center for Education Statistics, 2002: tables 53, 110); resources are more equally distributed ( Education Week, 2002;   Reed, 2001; Rothstein, 2000); black children are not required by law to attend inferior schools for fewer hours a day and shorter school years than white children (Orfield, 1978; Salomone, 1986; Tushnet, 1987); overall achievement scores are up (National Center for Education Statistics, 2002, tables 112, 115, 124, 125).  Most importantly perhaps, the gap in nationally-recognized achievement test scores between students with poorly- and well-educated parents has declined since the 1970s (National Center for Education Statistics, 2002, tables 112, 124). 

Yet this progress has met limits.  Hispanics drop out much more frequently than others, as do poor students and students in large urban schools (Driscoll, 1999; Rumberger & Thomas, 2000; Hauser, Simmons, & Pager, 2001; National Center for Education Statistics, 2002, tables 107, 108).  Achievement scores have changed little in the 1990s; the gaps between black and white achievement, and between the scores of the highest and lowest achievers, have remained static or even risen over that decade (National Center for Education Statistics, 2002, tables 112, 113, 124, 125). Disadvantaged children continue to score roughly ten percent below the national average on NAEP tests while advantaged children score several percent above (author’s calculations from data in National Center for Education Statistics, 2000c).  Some urban schools seem to teach very little despite teachers’ and students’ valiant efforts (Anyon, 1997; Education Week , 1998; Hayward, 2000; Henig, Hula, Orr, & Pedescleaux, 1999). 

Most importantly, adults’ life chances depend increasingly on attaining higher education, but the number of young adults completing college has stalled since the 1970s and class background is as important as ever in determining who attends and finishes college (Ellwood & Kane, 2000; Kane, 2001). Over three-quarters of well-off young adults go straight from high school to college, compared with half of poor youth. Well-off students are also more likely to go to a four-year rather than a two-year college (Card & Lemieux, 2001; National Center for Education Statistics, 1994).

               This article shows the pattern of class differences in schooling outcomes and indicates some of the causes for those differences that lie within the public realm.  It also points out the implications of the fact that the poor in the United States are disproportionately African Americans or recent immigrants; class biases are closely entwined with racial and ethnic inequities.  I conclude with the broad outlines of what would be necessary to reduce, even if we can never eliminate, class (and racial) disparities in American public schools.

Nested Inequalities

Disparities in schooling outcomes can be understood as two deeply embedded patterns of inequality.  The first is a system of nested inequalities affecting all students.  It begins with states. Children in Iowa, New Jersey, Massachusetts, and North Dakota have more than a 50 percent likelihood of enrolling in college by age nineteen, but children in Florida, Arizona, Alaska, and Nevada have less than a 30 percent chance (Hodgkinson, 1999, figure 2).  Fewer than three percent of students in Wisconsin, North Dakota, and Iowa drop out of school; more than seven percent do in Louisiana, Arizona, Georgia, New Mexico and Nevada (National Center for Education Statistics, 2002, table 105). 

Overall, about a third of the variation in students’ achievement is determined by what state they live in (Murray, Evans, & Schwab, 1998; National Center for Education Statistics, 2000d, p. 40-42).  But inequalities within a state can be just as severe.  Connecticut provides unusually detailed evidence on this point. The district that spends the most provides almost twice as much per student as the district that spends the least.  There are over 150 times more poor students in the poorest town than in the richest town; some districts have no minority students whereas in others virtually all students are non-Anglos; in some districts all students speak English at home whereas in others up to two-thirds of the students speak some other language with their families (Connecticut Conference of Municipalities, 1997, p. vii-ix).   These disparities correspond to equally great differences in educational outcomes across districts:  “On the Connecticut Mastery Test, the best performing municipality has scores nearly three times as high as the lowest scoring community…. In the worst-performing municipality, 49% of the class of 1995 dropped out during the four years before graduation; in the best performing community the drop-out rate was 0%.  The rate of graduates who continue their education beyond high school ranges from less than 50% to 98%” (Connecticut Conference of Municipalities, 1997, p. viii).  Many, although not all, of the indicators of class-based advantage or disadvantage correlate highly with the differences in educational outcomes.  Districts with a lot of poor students have lower average test scores and higher dropout rates; districts with a lot of minority students, or a lot whose native language is not English, also have lower average test scores.  (These districts are often the same.)  The highest spending districts report high test scores, and some of the lowest spending districts report the lowest test scores, although the pattern in the middle-wealth districts is less clear (Connecticut Conference of Municipalities, 1997, p. 5, 10-12).

Third, schools vary within districts.  In Yonkers, New York, for example, schools in the northern and eastern section were built relatively recently and have beautiful grounds and excellent facilities; schools in the southwestern section were built in some cases a century ago, with tiny playgrounds of cracked and slanted cement (or none at all) and dismal laboratories and libraries.  It is not difficult to figure out the racial or ethnic and class composition of the students in these schools (Hochschild & Danielson, 1998).  Yonkers is not alone. In New York City, funding for regular students in elementary schools varied by as much as $10,000 per student in the late 1990s; per capita operating funds were particularly low in schools with many poor or immigrant students. In some New York City grade schools all of the teachers are certified, and in some the pupil/teacher ratio is well below ten; in others, only two out of five teachers are certified or the ratio of students to teachers is well over 20. Schools with a lot of poor students or limited English speakers had significantly fewer certified teachers and higher student/teacher ratios.  In some New York City schools, all of the students perform at least at the fiftieth percentile in reading tests, but in others barely one-seventh do (data on New York City are in Iatarola & Stiefel, 2003; see also Hertert, 1995; Rothstein, 2000).

Finally, children’s schooling varies even within a school.  Almost all high schools, many middle schools, and some elementary schools sort students by measured ability; well-off children, who are disproportionately white and Asian, almost always dominate the high tracks (Argys, Rees, & Brewer, 1996; Lucas, 1999; Mickelson & Heath, 1999).  Students with disabilities or with limited English proficiency are not likely to be in high tracks regardless of their talents (August & Hakuta, 1998; McDonnell, et al., 1997). Students shunted into low-ability classes or nonacademic tracks frequently end up with poorly- or inappropriately-trained teachers, few resources, trivial curricula, and no accountability (Heubert & Hauser, 1999; Ingersoll, 2002).

Thus every student sits at the center of at least four nested structures of inequality and separation – states, districts, schools, and classes.  Well-off or white and Asian parents usually manage to ensure that their children obtain the benefits of this structure; poor and non-Asian minority parents have a much harder time doing so (Mollenkopf, Zeltzer-Zubida, Holdaway, Kasinitz, & Waters, 2002). As a result, the United States has not witnessed the full equality of educational opportunity between classes that one would expect from all the reforms since the 1960s and from Americans’ commitments to equality of opportunity (Hochschild, 1995).  Analysts talk about “speed bumps on the road to meritocracy” (Hout, 1997, title) or “(re)emerging inequality in the opportunity structure going into the 21 st century,” (Biblarz & Raftery, 1999, p. 249) or an “increase [in the] relative importance of social background for college entry” (Lucas, 1996, p. 511). Details vary in these analyses but the pattern is clear: the progress our nation made toward equal opportunity in schooling up until the 1980s has stopped and perhaps even reversed (see also Acemoglu & Pischke, 2001; Biblarz, 2000; Ishida, 1993).

Failing Inner City Schools

“We have kids without teachers, teachers without classrooms, and a district without a clue. The system is broken.  Students and teachers are a forgotten priority here,” says the president of the Los Angeles teachers union (White, 1999, p. 3).  City schools like these demonstrate the other deeply embedded pattern of class disparities in schooling.  Disastrous schools affect only a minority of children, but them very seriously; “for years it was like storming the Bastille everyday,” reports one urban teacher (Olson, 1998, p. 1).

As there is with the system of nested inequalities, there is plenty of evidence pointing to the disproportionate failures of urban schools. More than twice as many students attend high-poverty schools in urban than in nonurban districts (Department of Housing and Urban Development, 1998, p. 16-17), but in some states, urban districts spend less per pupil than do nonurban districts (National Center For Education Statistics, 2002, indicator 56). Urban districts have larger classes and contain the largest schools  ( Education Week, 1998, p. 19;  National Center for Education Statistics, 2001, table A). Compared with suburban districts, teachers in city schools are less likely to be certified or to have studied in the areas that they teach, and more likely to leave before the end of the school year.  In some years and for some subjects, it is hard to find any teachers at all to fill slots in urban schools ( Education Week , 1998 , p. 16-17). Urban schools are more likely to have inadequate buildings, classrooms, and technology ( Education Week , 1998, p. 21; General Accounting Office, 1995).  They suffer from much more administrative and behavioral turmoil and have a higher level of disruption, violence, and anxiety about safety ( Education Week , 1998 , p. 18-19).  All of the big districts with high dropout rates are in large cities ( Education Week , 1998, p. 13; Hochschild & Scovronick, 2003, p. 25-27; 61-63, 78-80, 84-87; National Center for Education Statistics, 2001, p. table 16).

From an educator’s perspective, interactions among these characteristics can be overwhelming. The San Diego school district, for example, offered each of twenty failing schools $16,500 of extra funds in 1998.  An evaluation of one such school then called for a full-time nurse, a full-time counselor, a parent room, a pre-kindergarten program, an adult literacy program, and an end to assigning teachers by seniority (a union regulation; Reinhard, 1998).  Ninety percent of the children in this school are poor, 40 percent have limited English proficiency, many move frequently. Two of the twenty teachers are out on “stress disability,” and a third are brand new.  In the face of these substantial challenges, the principal claims that “we’ve pulled together, and we’re going to do the best we can” (Reinhard, 1998, p. 15).  But her chances of success seem slim, and the children in her school will probably have little chance to pursue their dreams or to share meaningfully in the responsibilities of democratic citizenship.

            Thus the worst-off students and schools have a completely different educational experience from the best-off, with predictably different outcomes. Here is where class and race are most tightly entwined, since in the 100 largest school districts, almost 70 percent of the students are non-Anglo (compared with 40 percent of students nationally), and over half are poor or near-poor (compared with fewer than 40 percent nationally) (National Center for Education Statistics, 2001, p. 4-5).  And the interactions among race and class are becoming tighter: during the 1970s and 1980s, “the gap in the quality of schools that blacks and whites attend has widened … due entirely to a worsening in the relative quality of schools located in poor, inner-city areas and in schools that are less than 20% white” (Cook & Evans, 2000, p. 747).  In fact, black students in nonurban schools actually did better during this period, even while black students in urban schools did worse (Cook & Evans, 2000).  Similarly, during the 1990s, the least accomplished quarter of fourth grade readers lost ground in NAEP tests, while the most accomplished improved their test scores.  The top scorers were mostly white, the low scorers were disproportionately black and Latino boys in poor urban schools (Zernike, 2001).

Increasing Inequality among Communities

Both patterns of class disparity – nested inequalities among all students, and utterly disastrous schooling for a few students – are made worse by the fact that socioeconomic separation across the society is growing, even as racial and ethnic separation is slowly declining (Abramson, Tobin, & VanderGroot, 1995; Rusk, 2002).  Residential separation between well-off and poor Americans declined in the 1950s and 1960s; by 1970, the typical affluent American lived in a neighborhood where two-fifths of the residents were also affluent.  But residential separation rose as the wealthy moved to outer suburbs, so that by 1990, the typical affluent American lived in a neighborhood where over half of the neighbors were also affluent (Massey, 1996, p. 396-399).  Conversely, the proportion of poor people living in poor neighborhoods in inner cities increased from 55 to 69 percent over the same twenty years. From 1970 to 1990, every one of the 48 largest cities, from the poorest in comparison to its suburbs (Hartford) to the wealthiest compared with its suburbs (Greensboro, North Carolina), became poorer in relation to its suburbs (Madden, 2000, p. 3-7). The very poorest Americans have become even more concentrated; the 100 largest cities’ share of the nation’s welfare recipients grew from almost 48 percent in 1994 to over 58 percent in 1999 (Allen & Kirby, 2000). 

Not surprisingly given these demographic changes, in the decade after 1982 economic disparities between school districts rose, whether measured by household income, poverty rates, or rates of housing vacancy (Ho, 1999). In my view, if leaders of the American system of public schools truly sought to promote equal opportunity, they would enact policies to offset these growing disparities.  And sometimes they do, as we have seen in the context of efforts to promote desegregation of schools, to equalize funding across wealthy and poor districts, and to improve test scores of poorly-achieving students. But too many policies have the effect if not the intent of reinforcing inequality and helping to maintain acute deprivation, as I demonstrate in the next section.

Policies That Reinforce Inequality

               Poor children bring many problems to school that more affluent children usually avoid, all of which affect their readiness to learn and their ability to take advantage of what they are taught. These problems include poor health and nutrition, greater family instability, more frequent moves, less safe communities, fewer books and educational resources in the home or neighborhood, a greater likelihood of having parents or other caretakers who have little formal education and/or speak little English, and anxieties about racial or ethnic discrimination (Anyon, 1997; Brooks-Gunn, Duncan, Klebanov, & Sealand, 1993; Garfinkel, Hochschild, & McLanahan, 1996; Pogue, 2000).  If policy-makers seek to reduce class disparities, they must attend to these problems, for which the educational system cannot be blamed.  Nevertheless, public schools could do much more than they do to offset the harms that poor students bring to school. In particular, three features of schooling correspond to the system of nested inequalities and worsen the disadvantages of poor urban schools, thereby reinforcing social class inequities. They are financial inequality across states and districts, disparities in the quality of teaching across districts, schools, and classrooms, and excessive ability grouping and unequal curricular offerings across schools and classrooms.

Financial Inequities : The nation as a whole spent about $7,080 per student in 2001.  Controlling for regional cost differences, the most generous states were New Jersey at $9,360, New York at $8,860, Connecticut at $8,800, and Wisconsin at $8,740.  The most abstemious states, with the same controls, were Utah at an astounding $4,580, Arizona at $5,010, and California at $5,600.  In six states, virtually all of the students attended school in districts with per pupil expenditures at or above the U.S. average; in an additional six states, six percent or fewer of the students enjoyed similar levels of resources (all data in Education Week , 2002, p. 86-87).   

            Befitting a structure of nested inequalities, disparities in funding across districts within a state may be almost as great. Some states have very slim bands of inequality; Delaware, Florida, Iowa, and West Virginia show less than eight percent variation among all districts around the average-spending district. But in other states the variation around the average-spending district is huge -- 33 percent in Alaska and close to 20 percent in Vermont, Montana, North Dakota, and New Hampshire (the measure here is the coefficient of variation; data are adjusted to control for local cost differences and weighted for student needs [defined as poverty and special education]; all data in Education Week, 2002, p. 88-89).   In New York and New Jersey, disparities between the schools in the top and bottom deciles of funding grew dramatically in the two decades after 1973-74 (Schneier, 2001, p. 229).  That trajectory was reversed in New Jersey in the 1990s; we do not know how typical these two states are.

            In my view, more money is not all that is needed to improve schooling outcomes for poor children, and I, like others, have observed schools with few resources doing a fantastic job of teaching poor children.  But usually more money is necessary if not sufficient to provide better schooling; it enables preschool, smaller classes, better libraries and labs, higher-paid teachers, newer textbooks, art and music classes, professional development, and all the things that contribute to improved educational outcomes.  It is unlikely that a parent chooses to move to a lower-spending district if she can afford to live in a higher-spending district, and districts never vie to spend less in the endless disputes in state legislatures over funding formulas.  So money matters, although how and how much needs more careful consideration than we can give it here (for more on school finance reform and its effects, see Burtless, 1996; Hochschild & Scovronick, 2003; Ladd, Chalk, & Hansen, 1999; Ladd & Hansen, 1999; Ludwig & Bassi, 1999).

            Quality of Teaching:  The evidence is clear on the positive effects of good teachers and the harm that can be done by bad ones; in one study, elementary students taught for three years in a row by highly ineffective teachers ended up in the 45 th percentile or below on state math tests, whereas students with three particularly good teachers scored over the 85 th percentile (Sanders & Rivers, 1996; see also Bembry, Jordan, Gomez, Anderson, & Mendro, 1998; Mendro, Jordan, Gomez, Anderson, & Bembry, 1998; National Center for Education Statistics, 2000e, p. 5-7). As these studies suggest, the impact of poor teaching can be dramatic, cumulative, and difficult to reverse.

Yet students who live in poor districts, or poor students (often students of color) in a given district or school, are much more likely to be taught by less effective teachers, no matter how effectiveness is defined (Darling-Hammond, 2000; Education Trust, 2000; Puma & Drury, 2000; Rivkin, Hanushek, & Kain, 1998: 32; Wenglinsky, 2000). Schools with the highest levels of poverty and the largest proportion of minority students have twice as many new teachers as the best-off and whitest schools (Lankford, Loeb, & Wyckoff, 2002; see also National Center for Education Statistics, 2000e, p. 13-14), despite the fact that experienced teachers are more effective ( National Center for Education Statistics, 2000e, p. 13; Ogawa, Huston, & Stine, 1999, p. 661). Teachers are especially likely to leave high-poverty schools, which makes it difficult to develop a sense of community and a shared culture of learning (Recruiting New Teachers, 2000).  Some studies assert the effectiveness of state certification and licensure requirements (Darling-Hammond, 2001; Darling-Hammond, Berry, & Thoreson, 2001; for counter-arguments see Abell Foundation, 2001; Goldhaber & Brewer, 2000) -- but more noncertified teachers work in high-poverty and/or urban schools than in their wealthier or suburban counterparts (Ingersoll, 2002).  Even in the context of an overall decline in academic qualifications of new teachers over the past few decades (National Center for Education Statistics, 2000e, p. 7-10), students in poor districts are most likely to have teachers who themselves test poorly (Education Trust, 2000, p. 8).  Minority children, students in high-poverty schools, and lower-achieving classes more often have teachers who have not majored or minored in the subject they are teaching, especially in math and science (Ingersoll, 2002). These are, however, the fields for which the relationship between subject area knowledge and effectiveness has been most clearly demonstrated (Goldhaber & Brewer, 2000).

The evidence continues.  Schools, and especially classrooms, with high concentrations of poor or non-Anglo children have fewer and older computers, and less access to the internet (National Center for Education Statistics, 2000a). More generally, teachers in high-poverty or urban schools are more likely to report inadequate teaching resources ( Education Week , 1998, p. 21).  In California, the number of unqualified teachers rose dramatically in recent years, mainly in classrooms with Hispanic, disadvantaged, or low-achieving students ( CSR Research Consortium 2002; Ogawa et al., 1999; Jepson & Rivkin, 2002).

The challenges here are as analytically simple as they are politically and organizationally huge; without a large number of qualified, dedicated, experienced teachers for poor children, and classrooms with reasonable resources for those teachers to use, the odds against their participation in the American dream are almost insuperable.

Ability Grouping and Curricular Offerings:  Finally, in the fourth level of the structure of nested inequalities, students are separated by socioeconomic class as well as by measured ability into different experiences within a given school.  Arguments flourish about the causes and consequences of tracking and ability grouping, but several things seem clear. 

First, although tracking used to be racially discriminatory, by now “the claim of racial discrimination in group placement by teachers is not supported by research, once conventional indicators of merit or economic standing are accounted for” (Ferguson, 1998, p. 329). However, analysts almost universally agree that there is considerable discrimination in ability grouping on the basis of class , even controlling for achievement and other factors.  The raw facts are startling enough – almost three times as many high-income as low-income students enroll in college preparatory tracks. In more sophisticated analyses, achievement and ability (typically measured by test scores, prior placements, and teachers’ judgments) almost always show up as the chief determinants of students’ placement – but class-based factors usually come in second (Dauber, Alexander, & Entwistle, 1996; Gamoran & Mare, 1989; Jones, Vanfossen, & Ensminger, 1995; Miller, 1995, p. 237-240; National Center for Education Statistics, 2000b).

There is a more worrisome problem with the practice of ability grouping.  If achievement tests are racially biased, or if poor (especially poor black and Hispanic) children consistently receive the worst teaching and therefore learn the least, then the fact that measured prior achievement most strongly determines a student’s placement is not reassuring to those concerned about equal opportunity in schooling or diversity in classrooms.  So the congruence among poverty, minority status, and low quality of teaching becomes reinforced by ability grouping, even when it relies more on measured achievement than on teachers’ (perhaps biased) judgments or parents’ insistence.

The thorniest issues, however, present an even more severe challenge to the goal of equal opportunity: if grouping by ability harms the chances of many, even while benefiting some, its costs may be too high. The empirical literature on the effects of ability-based separation does frustratingly little to help resolve the issue of whether the costs outweigh the benefits, since careful studies show all possible combinations of results.  Some find that all grouped students can benefit (Camarena, 1990; Epple, Newlon, & Romano, 2002; Epstein & MacIver, 1992; Ferguson, 1998; Figlio & Page, 2002; Lou et al., 1996; Valli, 1990).  Others find that grouping makes little difference compared with other schooling variables, or that it reduces overall achievement levels (Gamoran, 1992; Slavin, 1990a; Slavin, 1990b). The most recent and methodologically sophisticated articles in this literature, however, find that students in high tracks benefit from grouping and students in low tracks are harmed, or at least are not helped (Argys, et al., 1996; Gamoran & Mare, 1989; Garet & DeLany, 1988; Lucas, 1999).

Two conclusions shine through this morass.  First, contradictions in the research point to differences in practice that call for a careful policy choice. Experimental studies that control for most factors affecting students’ outcomes show that, when curriculum and instructional methods are similar for all students, skill grouping by itself neither consistently helps nor harms students (e.g. Ferguson, 1998).  But studies of actual school settings usually find that students in the low groups do worse than they should, even given their presumedly lower ability (Shepard, 1992). The proper debate, then, is whether educators should seek to abolish ability grouping on the grounds that it will never be fairly done, or whether they should concentrate on ensuring a challenging curriculum, equal teaching quality, and a fair allocation of resources across groups (Exchange, 1994).

            Second, the contradictory research results imply that “decisions about grouping are preliminary and what matters most comes next: decisions about what to do with students after they are assigned to classes. Given poor instruction, neither heterogeneous nor homogeneous grouping can be effective; with excellent instruction, either may succeed” (Gamoran, 1993, p. 44; see also Ferguson, 1998; Oakes, Gamoran, & Page, 1992).  As the most influential book seeking to abolish tracking put it, “the most significant thing we found is that generally our entire sample of classes turned out to be pretty noninvolving places…. Passive activities… were dominant at all track levels…. Any statements that can be made about differences between tracks… must be seen in this context” (Oakes, 1985, p. 129).

              The deepest problem, then, is that too many students are poorly taught, and students in low ability groups – disproportionately poor students, who are disproportionately of color -- usually are the most poorly taught of all (Good, 1987; Ingersoll, 1999; Weiss, 1997).  And these failures and inequities have long-term effects:  the intensity and quality of secondary school curriculum have the greatest impact on completion of a bachelor’s degree, a far greater impact than SES, ethnicity and race, and even test scores and high school class rank (Office of Educational Research and Improvement, 1999, Executive Summary).

The issue of curriculum quality points us toward “tracking” at the level of schools or districts rather than students.  Middle schools in poor or non-Asian minority communities frequently do not offer algebra in eighth grade, even though it is essential for high-level mathematics in high school (Jones et al., 1995; Monk & Rice, 1997; Raudenbush, Fotiu, & Cheong, 1998; Spade, Columba, & Vanfossen, 1997).  Poor schools are less likely to offer advanced mathematics or science courses, Advanced Placement (AP) courses, or honors English and history courses than schools in wealthier and predominantly white communities  (Oakes et al., 1992, p. 589; National Center for Education Statistics, 1995, table A2.2b). Children of parents who have not attended college, who are disproportionately poor and nonwhite, are twice as likely to attend schools that do not offer algebra in eighth grade as children whose parents completed college (National Center for Education Research, 2000b).  

In 1999, the American Civil Liberties Union (ACLU) filed suit against the state of California, claiming that “129 California public high schools with 80,000 students do not offer any AP courses; and 333 schools offer four or fewer.  In contrast,… 144 public high schools in California offer more than 14 AP courses” (Sahagun & Weiss, 1999, p. A13). Small rural schools and schools in poor urban districts were least likely to offer AP courses, thus disadvantaging African Americans, recent Latino immigrants, and poor whites, especially since the University of California at Berkeley and UCLA weigh AP courses and their test scores heavily in admissions decisions.  The general counsel for the state’s department of education agreed that “this is a genuine equity issue and I think it will have enough political push to bring about a solution” (Bathen, 1999, p. M3). Prodded by this lawsuit, the College Board set up a program to ensure that all public high schools offer AP courses within a few years (currently 40 percent do not), and some schools are encouraging more students to take them (Viadero, 2001).

Directions for Public Policy

This is not the place to analyze in detail what ought to be done to reduce the patterns of nested inequalities and concentrated harms in public schooling; any serious policy change is enormously complicated, particularly in the diffuse and decentralized world of public schooling. Nevertheless, the outlines of the moves needed to weaken the link between social class and educational outcomes are clear.

Where it has been reasonably implemented, educating poor children with students who are more privileged, or educating them like students who are more privileged, has improved their performance and long-term chances of success (Kahlenberg, 2000; Rubinowitz & Rosenbaum, 2000).  Quality preschool, individual reading instruction, small classes in the early grades, assignment to classes with peers who take school seriously and behave in ways that enable them to learn, and consistently challenging academic courses have been shown to help disadvantaged children achieve, just as they enable middle-class children to achieve (for reviews of this extensive literature, see Hochschild & Scovronick, 2003; Puma & Drury, 2000).  Most importantly, qualified, knowledgeable teachers make a difference, as described above.  Well-off children almost always attend schools that have most of these features; poor children too frequently do not.

An honest attempt to secure a good education for poor children therefore leaves policymakers with two difficult choices.  They can send them to schools with wealthier children, or they can, as a reasonable second-best, seek to give them an education in their own neighborhoods that has the features of schooling for well-off students.  The former has proved so far to be too expensive politically, and the latter has often been too expensive financially (for histories of and evidence on school desegregation and school finance equalization efforts, see Hochschild & Scovronick, 2003).  After a decade of studying the subject, I conclude that if Americans really wanted all children to have a real chance to learn, they would

• eliminate disparities in funding across states, districts, and schools and provide extra funding for the poorest schools and districts as needed;
•  provide the resources needed to overcome the social, health-related, and physical problems that poor children disproportionately bring into schools;
• redistribute the teaching staff and enhance the quality, training, and deployment of all teachers;
• implement clear standards for higher-order learning, with appropriate supports, and hold schools and educators as well as students accountable;
• eliminate the forms of ability grouping with no demonstrated benefits and ensure that all schools and classrooms offer stimulating and difficult curricula;
• redraw district and neighborhood assignment lines to ensure a broad mix of students across economic strata (and races or ethnicities) within a school.

The worst urban schools would be reconstituted or shut down, and the children in them dispersed among schools with a much higher proportion of middle class students.

Moving poor children into more affluent schools is not a panacea. When poor families move from deeply poor neighborhoods into communities with very little poverty, the children typically have more behavioral problems in school, even though their test scores improve (Ludwig, Ladd, & Duncan, 2001).  African American students also report more racism among their new classmates and neighbors, and worry about holding their own socially in their new environment  (Rubinowitz & Rosenbaum, 2000).  These social and emotional difficulties warrant concern, but they pale beside the much larger problem of racial and class isolation; I think it would be a sign of enormous progress if our chief problem was encouraging poor and well-off children in the same school and classroom to engage with each other more effectively.

Similarly, improving the quality of schooling in impoverished schools is extraordinarily difficult.  Educators within a school develop a culture, and some urban schools have developed a culture of failure (Payne, 1997).  In others, educators focus more on workplace concerns, racially based frustrations, a search for power in their community, or other issues of real importance but remote from a focus on teaching and learning (Henig et al., 1999; Orr, 1999; Rich, 1996).  These problems similarly warrant concern, but probably most urban schools suffer more from the less exotic problems of insufficient resources, lower quality teaching, and students’ needs for intensive instruction.  In any case, it would be worth finding out.

Public schools are essential to enable Americans to succeed, but schools are also the arena in which some children first fail.  Failure there almost certainly guarantees failure from then on.  Americans would like to believe that failure results from lack of individual merit and effort; in reality, failure in school too closely tracks structures of racial and class inequality.  American schools too often reinforce rather than contend against those structures; that is understandable but not acceptable. 

ABC News. (2000). ABC News poll. available at R-POLL.  http://80-web.lexis-nexis.com.ezp2 .

harvard.edu/universe/form/academic/s_roper.html?_m=9e0dd1592e1cabf7b934c7bfeac3ff4d&wchp=dGLbVzb-lSlAl&_md5=c1429a7e6bbd77f5fe21d00a93d0757e

Abell Foundation. (2001). Teacher certification reconsidered: Stumbling for quality . Baltimore MD: Abell Foundation.

Abramson, A., Tobin, M., & VanderGroot, M. (1995). The changing geography of metropolitan opportunity. Housing Policy Debate, 6 (1), 45-72.

Acemoglu, D., & Pischke, J.-S. (2001). Changes in the wage structure, family income, and children's education. European Economic Review, 45 (4-6), 890-904.

Allen, K., & Kirby, M. (2000). Unfinished business: Why cities matter to welfare reform . Washington D.C.: Brookings Institution Press.

Anyon, J. (1997). Ghetto schooling: A political economy of urban educational reform . New York: Teachers College Press.

Argys, L., Rees, D., & Brewer, D. (1996). Detracking America's schools: Equity at zero cost? Journal of Policy Analysis and Management, 15 (4), 623-645.

August, D., & Hakuta, K. (Eds.). (1998). Educating language-minority children . Washington D.C.: National Academy Press.

Bathen, S. (1999, October 17). Education: The deeper inequality behind the AP-course suit. Los Angeles Times, pp. M1, M3.

Bembry, K., Jordan, H., Gomez, E., Anderson, M., & Mendro, R. (1998). Policy implications of long-term teacher effects on student achievement . Dallas TX: Dallas Public Schools.

Biblarz, T. (2000, November 13). Data on 'the pathology of matriarchy' ". Personal communication with the author.

Biblarz, T., & Raftery, A. (1999). Family structure, educational attainment and socioeconomic success: Rethinking the 'pathology of matriarchy'. American Journal of Sociology, 105 (2), 321-365.

Braddock, D. L., & Parish, S. L. (2001).  An institutional history of disability.  In G. L. Albrecht, K. D. Seelman, & M. Bury (Eds).  Handbook of disability studies (pp. 11-68).  Thousand Oaks, CA:  Sage Publications.

Brooks-Gunn, J., Duncan, G., Klebanov, P. & Sealand, N. (1993) Do neighborhoods influence child and adolescent development? American Journal of Sociology, 99 (2), 353-395.

Burtless, G. (Ed.). (1996). Does money matter?: The effect of school resources  on student achievement and adult succcess . Washington D.C.: Brookings Institution Press.

Bush, G. W. (2001). Remarks by the president in submitting education plan to congress . Washington D.C.: Department of State International Information Programs,  http://usinfo.state.gov/usa/schools/bush23.htm .

Camarena, M. (1990). Following the right track: A comparison of tracking practices in public and Catholic schools. In R. Page & L. Valli (Eds.), Curriculum differentiation: Interpretive studies in U.S. Secondary schools (pp. 159-182). Albany NY: SUNY Press.

Card, D., & Lemieux, T. (2001). Dropout and enrollment trends in the postwar period: What went wrong in the 1970s? In J. Gruber (Ed.), Risky behavior among youth: An economic analysis (pp. 438-482). Chicago IL: University of Chicago Press.

Clinton, W. (1995). Remarks honoring Franklin D. Roosevelt in Warm Springs, Georgia, April 12. Weekly Compilation of Presidential Documents, 31 (15), 614-618.

CNN/ U.S.A. Today . (2000). Gallup, CNN, U.S.A. Today poll. available at R-POLL http://80-web.lexis-nexis.com.ezp2.harvard.edu/universe/form/academic/

s_roper.html?_m=9e0dd1592e1cabf7b934c7bfeac3ff4d&wchp=dGLbVzblSlAl&_md5=c1429a7e6bbd77f5fe21d00a93d0757e

Connecticut Conference of Municipalities. (1997). Education-related disparities in Connecticut . New Haven CT: Connecticut Conference of Municipalities.

Cook, M., & Evans, W. (2000). Families or schools? Explaining the convergence in white and black academic performance. Journal of Labor Economics, 18 (4), 729-754.

Cremin, L. (1988). American education: The metropolitan experience, 1876-1980 . New York: HarperCollins.

Danziger, S., & Reed, D. (1999). Winners and losers : The era of inequality continues. Brookings Review, 17, 14-17.

Darling-Hammond, L. (2000). Teacher quality and student achievement: A review of state policy evidence. Education Policy Analysis Archives, 8 (1). epaa.asu.edu/epaa/v8n1 [Available 2003, February 21].

Darling-Hammond, L. (2001). The research and rhetoric on teacher certification: A response to 'Teacher certification reconsidered' . Stanford CA: Stanford University, School of Education.

Darling-Hammond, L., Berry, B., & Thoreson, A. (2001). Does teacher certification matter? Evaluating the evidence. Educational Evaluation and Policy Analysis, 23 (1), 57-77.

Dauber, S., Alexander, K., & Entwistle, D. (1996). Tracking and transitions through the middle grades: Channeling educational trajectories. Sociology of Education, 69 (4), 290-307.

Department of Housing and Urban Development. (1998). The state of the cities 1998 . Washington D.C.: U.S. Department of Housing and Urban Development.

Driscoll, A. (1999). Risk of high school dropout among immigrant and native Hispanic youth. International Migration Review, 33 (4), 857-875.

Education Trust. (2000). Honor in the boxcar . Washington D.C.: Education Trust.

Education Week . (1998). Quality counts '98: The urban challenge . Washington D.C.: Education Week and Pew Charitable Trusts.

Education Week . (2002). Quality counts 2002: Building blocks for success . Washington D.C.: Education Week and Pew Charitable Trusts.

Ellwood, D., & Kane, T. (2000). Who is getting a college education?  Family background and the growing gaps in enrollment. In S. Danziger & J. Waldfogel (Eds.), Securing the future: Investing in children from birth to college (pp. 283-324). New York: Russell Sage Foundation Press.

Epple, D., Newlon, E., & Romano, R. (2002). Ability tracking, school competition, and the distribution of educational benefits. Journal of Public Economics, 83( 1), 1-48.

Epstein, J., & MacIver, D. (1992). Opportunities to learn: Effects on eighth graders of curriculum offerings and instructional approaches (34). Baltimore MD: Johns Hopkins University, Center for Research on Elementary and Middle Schools.

Exchange (1994). Exchange between Maureen Hallinan and Jeannie Oakes. Sociology of Education, 67 (2), 79-91.

Ferguson, R. (1998). Can schools narrow the black-white test score gap? In C. Jencks & M. Phillips (Eds.), The black-white test score gap (pp. 318-374). Washington D.C.: Brookings Institution Press.

Figlio, D., & Page, M. (2002). School choice and the distributional effects of ability tracking: Does separation increase inequality? Journal of Urban Economics, 51 (3), 497-514.

Gallup Organization. (2000). Gallup poll.available at R-POLL. http://80-web.lexis-nexis.com

.ezp2.harvard.edu/universe/form/academic/s_roper.html?_m=9e0dd1592e1cabf7b934c7bfeac3ff4d&wchp=dGLbVzblSlAl&_md5=c1429a7e6bbd77f5fe21d00a93d0757e

Gamoran, A. (1992). The variable effects of high school tracking. American Sociological Review, 57 (6), 812-828.

Gamoran, A. (1993). Is ability grouping equitable? Education Digest, 58 (7), 44-46.

Gamoran, A., & Mare, R. (1989). Secondary school tracking and educational inequality: Compensation, reinforcement, or neutrality? American Journal of Sociology, 94 (5), 1146-1183.

Garet, M., & DeLany, B. (1988). Students, courses, and stratification. Sociology of Education, 61 (2), 61-77.

Garfinkel, I., Hochschild, J., & McLanahan, S. (Eds.). (1996). Social policies for children . Washington D.C.: Brookings Institution Press.

General Accounting Office. (1995). School facilities: America's schools not designed or equipped for 21st century (GAO/HEHS-95-95;). Washington  D.C.: U.S. General Accounting Office.

Goldhaber, D., & Brewer, D. (2000). Does teacher certification matter? Educational Evaluation and Policy Analysis, 22 (2), 129-145.

Good, T. (1987). Two decades of research on teacher expectations: Findings and future directions. Journal of Teacher Education, 38 (4), 32-47.

Gore, A. (2000) no title. www.algore.com/education/edu_agenda1.html (accessed Aug. 8, 2001)

Gutmann, A. (1987). Democratic education . Princeton NJ: Princeton University Press.

Hauser, R., Simmons, S., & Pager, D. (2001). High school dropout, race-ethnicity, and social background from the 1970s to the 1990s . Madison WI: University of Wisconsin at Madison, Department of Sociology.

Hayward, C. (2000). De-facing power . New York: Cambridge University Press.

Henig, J., Hula, R., Orr, M., & Pedescleaux, D. (1999). The color of school reform: Race, politics, and the challenge of urban education . Princeton NJ: Princeton University Press.

Hertert, L. (1995). Does equal funding for districts mean equal funding for classroom students? Evidence from California. In L. Picus & J. Wattenbarger (Eds.), Where does the money go? Resource allocations in elementary and secondary schools (pp. 71-84). Thousand Oaks CA: Corwin.

Heubert, J., & Hauser, R. (Eds.). (1999). High stakes: Testing for tracking, promotion, and graduation . Washington D.C.: National Academy Press.

Ho, A. (1999). Did school finance reforms achieve better equity? Ames Iowa: Iowa State University, Department of Political Science.

Hochschild, J. (1995). Facing up to the American dream: Race, class, and the soul of the nation . Princeton NJ: Princeton University Press.

Hochschild, J., & Danielson, M. (1998). Can we desegregate public schools and subsidized housing?  Lessons from the sorry history of Yonkers, New York. In C. Stone (Ed.), Changing urban education (pp. 23-44). Lawrence: University of Kansas Press.

Hochschild, J., & Scovronick, N. (2003). The American dream and the public schools . New York: Oxford University Press.

Hodgkinson, H. (1999). All one system: A second look . Washington D.C.: Institute for Education Leadership; National Center for Public Policy and Higher Education.

Hout, M. (1997). Speed bumps on the road to meritocracy . Berkeley CA: University of California at Berkeley, Department of Sociology.

Iatarola, P., & Stiefel, L. (2003). Intradistrict equity of public education resources and performance. Economics of Education Review, 22 (1), 69-78.

Ingersoll, R. (1999). The problem of underqualified teachers in American secondary schools. Educational Researcher, 28 (2), 26-37.

Ingersoll, R. (2002). Out-of-field teaching, educational inequality, and the organization of schools: An exploratory analysis . Seattle: University of Washington, Center for the Study of Teaching and Policy.

Ishida, H. (1993). Social mobility in contemporary Japan . Stanford CA: Stanford University Press.

Jones, J., Vanfossen, B., & Ensminger, M. (1995). Individual and organizational predictors of high school track placement. Sociology of Education, 68 (4), 287-300.

Kahlenberg, R. (2000). All together now: The case for the economic integration of the public schools . Washington D. C.: Brookings Institution Press.

Kaiser Family Foundation, & Harvard University School of Public Health. (2000). Post-election survey: The public and the health care agenda for the new administration and Congress. 73dps89s.pdf [Available: 2003, February 21]

Kalmijn, M. (1991). Shifting boundaries: Trends in religious and educational homogamy. American Sociological Review, 56 (6), 786-800.

Kane, T. (2001). College going and inequality: A literature review . New York: Russell Sage Foundation.

Kluger, R. (1975). Simple justice: The history of Brown v. Board of Education and black America's struggle for equality . New York: Knopf.

Ladd, H., Chalk, R., & Hansen, J. (1999). Equity and adequacy in education finance . Washington D.C.: National Academy Press.

Ladd, H., & Hansen, J. (1999). Making money matter: Financing America's schools . Washington D.C.: National Academy Press.

Lankford, H., Loeb, S., & Wyckoff, J. (2002). Teacher sorting and the plight of urban schools: A descriptive analysis. Educational Evaluation and Policy Analysis, 24 (1), 37-62.

Lou, Y., Abrami, P., Spence, J., Poulson, C., Chambers, B., & D'Apollonia, S. (1996). Within-class grouping: A meta-analysis. Review of Educational Research, 66 (4), 423-458.

Lucas, S. (1996). Selective attrition in a newly hostile regime: The case of 1980 sophomores. Social Forces, 75 (2), 511-533.

Lucas, S. (1999). Tracking inequality: Stratification and mobility in American high schools. New York: Teachers College Press.

Ludwig, J., & Bassi, L. (1999). The puzzling case of school resources and student achievement. Educational Evaluation and Policy Analysis, 21 (4), 385-403.

Ludwig, J., Ladd, H., & Duncan, G. (2001). Urban poverty and educational outcomes. In W. Gale & J. Pack (Eds.), Brookings-Wharton papers on urban affairs (pp. 147-201). Washington D.C: Brookings Institution Press.

Madden, J. (2000). Changes in income inequality within U.S. metropolitan areas . Kalamazoo MI: Upjohn Institute for Employment Research.

Mare, R. (1991). Five decades of educational assortative mating. American Sociological Review, 56 (1), 15-32.

Massey, D. (1996). The age of extremes: Concentrated affluence and poverty in the twenty-first century. Demography, 33 (4), 395-412.

McDonnell, L., McLaughlin, M., & Morison, P. (Eds.). (1997). Educating one and all: Students with disabilities and standards-based reform . Washington, D.C.: National Academy Press.

Mendro, R., Jordan, H., Gomez, E., Anderson, M., & Bembry, K. (1998). An application of multiple linear regression in determining longitudinal teacher effectiveness . Dallas TX: Dallas Public Schools.

Mickelson, R., & Heath, D. (1999). The effects of segregation on African American high school seniors' academic achievement. Journal of Negro Education, 68 (4), 566-586.

Miller, L. S. (1995). An American imperative: Accelerating minority educational advancement . New Haven CT: Yale University Press.

Mollenkopf, J., Zeltzer-Zubida, A., Holdaway, J., Kasinitz, P., & Waters, M. (2002). Chutes and ladders: Educational attainment among young second generation and native New Yorkers . New York: Center for Urban Research, City University of New York.

Monk, D., & Rice, J. (1997). The distribution of mathematics and science teachers across and within secondary schools. Educational Policy, 11 (4), 479-498.

Murray, S., Evans, W., & Schwab, R. (1998). Education-finance reform and the distribution of education resources. American Economic Review, 88 (4), 789-812.

National Center for Education Statistics. (1994). National educational longitudinal study of 1988: Third follow-up . Washington D.C.: U.S. Department of Education.

National Center for Education Statistics. (1995). National education longitudinal study of 1988: Trends among high school seniors, 1972-1992 . Washington, D.C.: U.S. Department of Education.

National Center for Education Statistics. (2000a). Internet access in U.S. public schools and classrooms: 1994-99 . Washington D.C.: U.S. Department of Education.

National Center for Education Statistics. (2000b). Mapping the road to college: First-generation students' math track, planning strategies, and context of support . Washington D.C.: U.S. Department of Education.

National Center for Education Statistics. (2000c). NAEP 1999: Trends in academic progress . Washington D.C.: U.S. Department of Education.

National Center for Education Statistics. (2000d). School-level correlates of academic achievement . Washington D.C.: U.S. Department of Education.

National Center for Education Statistics (2000e). Monitoring school quality: an indicators report . Washington D.C.: U.S. Department of Education.

National Center for Education Statistics. (2001). Characteristics of the 100 largest public elementary and secondary school districts in the United States: 1999-2000 . Washington D.C.: U.S. Department of Education.

National Center For Education Statistics. (2002). Condition of education 2002 . Washington D.C.: U.S. Department of Education.

National Center for Education Statistics. (2002). Digest of education statistics 2001 . Washington D.C.: U.S. Department of Education.

Oakes, J. (1985). Keeping track: How schools structure inequality . New Haven CT: Yale University Press.

Oakes, J., Gamoran, A., & Page, R. (1992). Curriculum differentiation: Opportunities, outcomes, and meanings. In P. Jackson (Ed.), Handbook of research on curriculum (pp. 570-608). New York: Macmillan

Office of Educational Research and Improvement. (1999). Answers in the toolbox: Academic intensity, attendance patterns, and bachelor's degree attainment . Washington D.C.: U.S.  Department of Education. www.ed.gov/pubs/Toolbox [Available: 2003, February 21]

Ogawa, R., Huston, D., & Stine, D. (1999). California's class-size reduction initiative: Differences in teacher experience and qualifications across schools. Educational Policy, 13 (5), 659-673.

Olson, L. (1998, March 25). Failing schools challenge accountability goals. Education Week, pp. 1, 14.

Orfield, G. (1978). Must we bus? Segregated schools and national policy . Washington, D.C.: Brookings Institution Press.

Orr, M. (1999). Black social capital: The politics of school reform in Baltimore, 1986-1998 . Lawrence KS: University Press of Kansas.

Payne, C. (1997). 'I don't want your nasty pot of gold': Urban school climate revisited. Evanston IL: Northwestern University, Department of Sociology.

Pogue, T. (2000). No silver bullet: Questions and data on factors affecting educational achievement . Finance Project. Available: www.financeproject.org/achievement.htm [2001, February 6].

Puma, M., & Drury, D. (2000). Exploring new directions: Title I in the year 2000 . Washington D.C.: National School Boards Association.

Raudenbush, S., Fotiu, R., & Cheong, Y. (1998). Inequality of access to educational resources: A national report card for eighth-grade math. Educational Evaluation and Policy Analysis, 20 (4), 253-267.

Recruiting New Teachers. (2000). The urban teacher challenge . Available: www.rnt.org/quick/new.html [2001, January 4].

Reed, D. (2001). On equal terms: The constitutional politics of educational opportunity . Princeton NJ: Princeton University Press.

Reinhard, B. (1998, March 25). In troubled schools, policy and reality collide. Education Week, p. 15.

Rich, W. (1996). Black mayors and school politics: The failure of reform in Detroit, Gary, and Newark . New York: Garland.

Rivkin, S., Hanushek, E., & Kain, J. (1998). Teachers, schools, and academic achievement . Cambridge MA: National Bureau of Economic Research.

Rothstein, R. (2000). Equalizing education resources on behalf of disadvantaged children. In R. Kahlenberg (Ed.), A notion at risk: Preserving public education as an engine for social mobility (pp. 31-92). New York: Century Foundation Press.

Rubinowitz, L., & Rosenbaum, J. (2000). Crossing the class and color lines: From public housing to white suburbia . Chicago IL: University of Chicago Press.

Rumberger, R., & Thomas, S. (2000). The distribution of dropout and turnover rates among urban and suburban high schools. Sociology of Education, 73 (1), 39-67.

Rusk, D. (2002). Trends in school segregation. In R. Kahlenberg (Ed.), Divided we fail: Coming together through public school choice (pp. 61-85). New York: Century Foundation Press.

Sahagun, L., & Weiss, K. (1999, July 28). Bias suit targets schools without advanced classes. Los Angeles Times, pp. A1, A13.

Salomone, R. C. (1986).  Equal education under law:  Legal rights and federal policy in the post-Brown era.  NY:  St Martin's Press.

Sanders, W., & Rivers, J. (1996). Cumulative and residual effects of teachers on future student academic achievement . Knoxville TN: University of Tennessee, Value-Added Research and Assessment Center.

Schneier, E. (2001). The politics of local education. In J. Stonecash (Ed.), Governing New York State (4th ed., pp. 214-240). Albany NY: SUNY Press.

Shepard, L. (1992, May 31). Studies support both sides in school debate. Sunday Camera , pp. 38.

Slavin, R. (1990a). Ability grouping in secondary schools: A response to Hallinan. Review of Educational Research, 60 (3), 505-507.

Slavin, R. (1990b). Achievement effects of ability grouping in secondary schools: A best-evidence synthesis. Review of Educational Research, 60 (3), 471-499.

Spade, J., Columba, L., & Vanfossen, B. (1997). Tracking in mathematics and science: Courses and course-selection procedures. Sociology of Education, 70 (2), 108-127.

Spring, J. (2000). The American school 1642 - 2000 . New York: WCB/McGraw-Hill.

Tushnet, M. V. (1987).  The NAACP's legal strategy against segregated education, 1925-1950 .  Chapel Hill, NC:  The University of North Carolina Press.

Tyack, D. (1974). The one best system: A history of American urban education . Cambridge MA: Harvard University Press.

Valli, L. (1990). A curriculum of effort: Tracking students in a Catholic high school. In R. Page & L. Valli (Eds.), Curriculum differentiation: Interpretive studies in U.S. secondary schools (pp. 45-65). Albany NY: SUNY Press.

Verba, S. (2001). Political equality: What is it? Why do we want it? Cambridge MA: Harvard University, Department of Government.

Verba, S., Schlozman, K. L., & Brady, H. (1995). Voice and equality: Civic voluntarism in American politics . Cambridge MA: Harvard University Press.

Viadero, D. (2001, April 25). AP program assumes larger role. Education Week .

Weiss, I. (1997). The status of science and mathematics teaching in the United States: Comparing teacher views and classroom practice to national standards . University of Wisconsin, Wisconsin Center for Education Research. Available: http://www.wcer.wisc.edu/nise/publications/briefs/vol%5F1%5Fno%5F3/index... [ July 6, 2000].

Wenglinsky, H. (2000). How teaching matters: Bringing the classroom back into discussions of teacher quality . Princeton NJ: Educational Testing Service.

White, K. (1999, October 20). L.A. Board names CEO with broad powers. Education Week, p. 3.

Wilgren, J. (2000, February 16). Poll finds education is chief concern of likely voters. New York Times, p. A14.

Zernike, K. (2001, April 7). Gap between best and worst widens on U.S. reading test. New York Times, pp. A1, 10.

Author Biography

Jennifer Hochschild is a Professor of Government at Harvard University, with a joint appointment in the Department of Afro-American Studies.  She received a B.A from Oberlin College in 1971 and a Ph.D. from Yale University in 1979.   She is the author of Facing Up to the American Dream: Race, Class, and the Soul of the Nation (Princeton University Press, 1995); The New American Dilemma: Liberal Democracy and School Desegregation (Yale University Press, 1984); What's Fair: American Beliefs about Distributive Justice ( Harvard University Press, 1981) and a co-author of Equalities (Harvard University Press, 1981).  She is a co-editor of Social Policies for Children (Brookings Institution Press, 1995).  Prof. Hochschild is a Fellow of the American Academy of Arts and Sciences, a former vice-president of the American Political Science Association, a member of the Board of Trustees of the Russell Sage Foundation, and a member of the Board of Overseers of the General Social Survey.  She has received fellowships or awards from the Guggenheim Foundation, the American Council of Learned Societies, the American Philosophical Society, the Spencer Foundation, the American Political Science Association, the Princeton University Research Board, and other organizations.  She has also served as a consultant or expert witness in several school desegregation cases, most recently the on-going case of Yonkers Board of Education v. New York State .

Race/Class Conflict and Urban Financial Threat

Do Facts Matter?: Information and Misinformation in American Politics

Outsiders No More?: Models of Immigrant Political Incorporation

Creating a New Racial Order: How Immigration, Multiracialism, Genomics, and the Young Can Remake Race in America

Bringing Outsiders In: Transatlantic Perspectives on Immigrant Political Incorporation

The American Dream and the Public Schools

The New American Dilemma

• Search Menu

Sign in through your institution

• Advance articles
• Author Guidelines
• Submission Site
• Open Access
• Why Submit?
• About Social Forces
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Self-Archiving Policy
• Dispatch Dates
• Journals on Oxford Academic
• Books on Oxford Academic

• < Previous

Social Class, Gender, and Contemporary Parenting Standards in the United States: Evidence from a National Survey Experiment

• Article contents
• Figures & tables
• Supplementary Data

Patrick Ishizuka, Social Class, Gender, and Contemporary Parenting Standards in the United States: Evidence from a National Survey Experiment, Social Forces , Volume 98, Issue 1, September 2019, Pages 31–58, https://doi.org/10.1093/sf/soy107

• Permissions Icon Permissions

Social scientists have documented a substantial increase in both mothers’ and fathers’ time spent with children since the 1960s in the United States. Yet parenting behaviors remain deeply divided by social class and gender, with important implications for the reproduction of inequality. To understand rising parental investments in children and persistent class and gender differences in parenting, popular accounts and academic studies have pointed to an apparent cultural shift toward norms of time-intensive, child-centered parenting, particularly for mothers and among middle-class parents. However, prior research has produced inconclusive evidence relating to social class, gender, and contemporary parenting norms. Using data from an original vignette survey experiment conducted with a nationally representative sample of more than 3,600 parents, this study examines cultural norms related to parenting elementary school-aged children, considering how both social class and gender shape views about good parenting. Results indicate that parents of different social classes express remarkably similar support for intensive mothering and fathering across a range of situations, whether sons or daughters are involved. These findings suggest that cultural norms of child-centered, time-intensive mothering and fathering are now pervasive, pointing to high contemporary standards for parental investments in children.

Personal account

• Sign in with email/username & password
• Get email alerts
• Save searches
• Purchase content
• Activate your purchase/trial code
• Add your ORCID iD

Institutional access

Sign in with a library card.

• Sign in with username/password
• Recommend to your librarian
• Institutional account management
• Get help with access

Access to content on Oxford Academic is often provided through institutional subscriptions and purchases. If you are a member of an institution with an active account, you may be able to access content in one of the following ways:

IP based access

Typically, access is provided across an institutional network to a range of IP addresses. This authentication occurs automatically, and it is not possible to sign out of an IP authenticated account.

Choose this option to get remote access when outside your institution. Shibboleth/Open Athens technology is used to provide single sign-on between your institution’s website and Oxford Academic.

• Click Sign in through your institution.
• Select your institution from the list provided, which will take you to your institution's website to sign in.
• When on the institution site, please use the credentials provided by your institution. Do not use an Oxford Academic personal account.
• Following successful sign in, you will be returned to Oxford Academic.

If your institution is not listed or you cannot sign in to your institution’s website, please contact your librarian or administrator.

Enter your library card number to sign in. If you cannot sign in, please contact your librarian.

Society Members

Society member access to a journal is achieved in one of the following ways:

Sign in through society site

Many societies offer single sign-on between the society website and Oxford Academic. If you see ‘Sign in through society site’ in the sign in pane within a journal:

• Click Sign in through society site.
• When on the society site, please use the credentials provided by that society. Do not use an Oxford Academic personal account.

If you do not have a society account or have forgotten your username or password, please contact your society.

Sign in using a personal account

Some societies use Oxford Academic personal accounts to provide access to their members. See below.

A personal account can be used to get email alerts, save searches, purchase content, and activate subscriptions.

Some societies use Oxford Academic personal accounts to provide access to their members.

Viewing your signed in accounts

Click the account icon in the top right to:

• View your signed in personal account and access account management features.
• View the institutional accounts that are providing access.

Signed in but can't access content

Oxford Academic is home to a wide variety of products. The institutional subscription may not cover the content that you are trying to access. If you believe you should have access to that content, please contact your librarian.

For librarians and administrators, your personal account also provides access to institutional account management. Here you will find options to view and activate subscriptions, manage institutional settings and access options, access usage statistics, and more.

Short-term Access

To purchase short-term access, please sign in to your personal account above.

Don't already have a personal account? Register

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1534-7605
• Print ISSN 0037-7732
• Copyright © 2024 University of North Carolina Chapel Hill
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Rights and permissions
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.