«How Educational Inequality Develops George Farkas, Population Research Institute, Pennsylvania State University ...»
National Poverty Center Working Paper Series
How Educational Inequality Develops
George Farkas, Population Research Institute, Pennsylvania State University
This paper is available online at the National Poverty Center Working Paper Series index at:
Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the view of the National Poverty Center or any sponsoring agency.
HOW EDUCATIONAL INEQUALITY DEVELOPS 1George Farkas Department of Sociology and Population Research Institute Pennsylvania State University Gfarkas@pop.psu.edu Draft, 1/27/06 Presented at the conference: The Colors of Poverty: Why Racial and Ethnic Disparities Exist, University of Michigan, September 15 – 16, 2005.
I am grateful to David Harris, Ann Lin, Steven Morgan and Jeff Morenoff for their comments on an earlier draft of this paper.
Academic achievement – the credentials of schooling completed and degrees attained, as well as the skills and capabilities associated with these credentials - is an important determinant of socioeconomic success. Few if any personal characteristics are more strongly and positively related to an individual’s later occupational attainment, employment, earnings, home ownership, health, and other measures of a successful life.
In addition, as the U.S. and other national economies have evolved, technological innovation and globalization have advanced, and labor union strength has declined, the economic return to academic achievement has increased. Thus, for example, in inflationadjusted, 1999 dollars, the average U.S. male high school dropout earned $13.61/hour in 1973, and $9.78/hour in 1999, a decline of 28 percent. By contrast, the earnings of workers with an advanced degree (beyond college) increased by more than 20 percent during this time period (Krueger, 2003: 4). A given educational achievement gap between two individuals leads to a larger earnings gap today than it did in the past.
This trend has been particularly disadvantageous for race/ethnic groups such as African-Americans and Latinos, whose academic achievement has historically lagged behind that of whites. At the same time, this trend has benefited Asians, whose academic achievement has equaled and in some areas surpassed that of whites. The great importance of racial/ethnic academic achievement gaps for understanding racial/ethnic earnings gaps is illustrated by the finding that the earnings gap between AfricanAmerican and white men can be fully explained by a calculation that accounts for, among other variables, the educational credentials (years of schooling completed) and cognitive skills (test score) gaps between these groups. Here, the portion (40%) of the Black-White hourly wage gap accounted for by the cognitive skills gap is four times the size of the portion (10%) accounted for by the credentials gap (Farkas and Vicknair, 1996, Table 1).
Thus, the study of racial/ethnic gaps in academic achievement - test scores and credentials - is central to understanding poverty and income differentials across these groups in America today.
Approximately fifty years after the Brown decision, forty years after the Coleman Report, and during a period of intense discussion of No Child Left Behind, concern with these gaps is hardly new. What is new is recent evidence on the sources of these disparities in the early lives and school careers of children. This new understanding of the early development of race/ethnic inequalities in educational achievement is the focus of this chapter. I will discuss the theoretical framework and data that have been used to address these issues, the questions that have been asked, and the answers that have emerged. The goal is scientific knowledge leading to programs and policies capable of narrowing the gaps.
Theoretical Framework A variety of theoretical perspectives, each with its own disciplinary tradition, have been employed to understand how race/ethnicity and poverty affect children’s schooling outcomes. However, while each of these perspectives has its own particular emphasis, they are essentially all grounded in a concern for the resources that parents have available for parenting, and the life stressors that may make successful parenting difficult to achieve. These resources and stressors include those stemming from educational, social, economic, cultural, and psychological factors. In general, families with higher social status (measured by education, income, and occupation) and those embedded in stronger networks of social relationships have more resources available for parenting, experience fewer social stressors, and can use their resources to ameliorate the effects of stressors.
By contrast, families closer to the bottom of status hierarchies, and those embedded in fewer and/or weaker networks of social relationships, have fewer resources for parenting and dealing with stress, while experiencing more powerful negative stressors in their daily lives.
Family economic status is a common beginning point for analyses in this tradition. Economists tend to emphasize the parental economic resources that are available to be invested in the human capital of their children (Becker 1981).
Psychologists and sociologists, as well as some more eclectic economists, have also focused on income, but have been more interested in its absence – families in poverty – as a determinant of the daily life stressors that make parenting difficult (McLoyd 1990;
Duncan and Brooks-Gunn 1997; Mayer 1997).
A second resource is whether or not the child’s parents are a co-resident married couple. Single parent households not only tend to have fewer financial resources than two-parent households, but they also have half the amount of parental time and other parental non-monetary resources (such as social contacts) available for care-giving. It is therefore not surprising that, after statistically adjusting for the effects of other variables, children raised in single-parent households tend to have less positive outcomes than children from two-parent families (McLanahan and Sandefur 1994; McLanahan 1997;
Carlson and Corcoran 2001; McLanahan, Donahue, and Haskins 2005).
A third resource is the educational level of parents, and related to this, parents’ cognitive skill level. Well-educated parents tend to have greater financial resources to be used in parenting and coping with the stresses of daily life. They also tend to have a greater focus on promoting the cognitive and educational attainment of their children, as well as the skills and knowledge necessary to do so (Lareau 2003). Further, their children likely inherit above-average cognitive ability from them. Essentially all empirical studies have found that parents’ educational performance and attainment are strongly and positively associated with the performance and attainment of their children (Jencks et al.
1979; Duncan and Brooks-Gunn 1997; Entwisle, Alexander, and Olson 1997; Jencks and Phillips 1998; Duncan and Magnuson 2005).
Thus, whether or not both parents are present in the household, as well as the levels of parental education and income, have been found to be the primary determinants of children’s outcomes. When there are two parents, and their educational, income, and occupational levels are high, a rich set of resources – financial, cognitive, social, emotional, and psychological – are available to achieve high educational aspirations for the child. By contrast, single parent households in which parental education and income are low have far fewer resources that can be devoted to parenting, and at the same time are more exposed to negative stressors from everyday life. They are thereby much less able to prepare their children for success in school.
In addition, parental household structure, education, income, and occupation are correlated with other variables that tend to magnify their effects. Thus, single parents who are high school dropouts and have low earnings are also more likely to be teenage parents, to suffer from inadequate health care, to be depressed and to have other psychological, behavioral, and health-related problems, to reside in unsafe neighborhoods, and to send their children to substandard schools. When these and related risk factors cumulate, the results for children can be devastating.
The factors just described are relatively exogenous to parenting itself. They provide the setting within which parents’ time and effort can be employed to help their children develop those cognitive skills and behavioral habits that lead to success in school and subsequent employment. However, different theoretical traditions have emphasized different aspects of parenting as being more or less crucial to children’s development.
Perhaps the most widely-used perspective builds on the work of Bradley and Caldwell (1984a, b) to conceptualize and measure parenting along three dimensions – the learning, physical, and emotional warmth environments provided in the home. An additional perspective investigates parental social support, depression, and active behavioral coping (Moos et al. 1986). The home learning and emotional warmth environments have been found to exert positive effects on children’s cognitive development, and the learning environment and parental active behavioral coping have been found to reduce child externalizing behavior problems (Phillips et al. 1998; Duncan, Brooks-Gunn, and Klebanov 1994; Farkas and Beron 2004; Brooks-Gunn and Markman 2005). These and related variables have also been found to fully account for the effects of poverty on child outcomes (Guo and Harris 2000, and the papers cited above).
Variants of this perspective typically focus on one or another detailed aspect of parenting. One such perspective, derived from developmental psychology and research on the early development of literacy skills, stresses parental oral language and its use with children. Hart and Risley (1995, 1999) showed that, for children between 12 and 36 months of age, better educated parents converse far more with them, using a much richer vocabulary, than do less well-educated parents. As a consequence, by 36 months of age the children of the better educated parents had developed far more extensive vocabularies than those raised by less educated parents. These social class differences in vocabulary are maintained as the children age from 3 to 13 years of age (Farkas and Beron 2004). It has also been shown that the child’s preschool oral vocabulary, as well as other preschool oral language skills, are the primary determinants of reading success in early elementary school (Whitehurst and Lonigan 2002; NICHD Early Child Care Research Network 2005; Brooks-Gunn and Markman 2005). In addition, early elementary school success – along both cognitive and behavioral dimensions - has schooling effects that carry through to the end of high school, and beyond (Ensminger and Slusarcik 1992; Alexander, Entwisle, and Horsey 1997; Entwisle, Alexander and Olson 1997; Duncan et al. 1998;
McLeod and Kaiser 2004).
Research has also shown that higher parental social class status, particularly higher levels of parental education, is positively associated with higher levels of early mathematics skills by their children (Downey et al. 2004; Duncan and Magnuson 2005;
Farkas and Hibel forthcoming). Whether or not this is a result of specific instruction provided by better-educated parents has yet to be ascertained.
In a different, but related, research tradition, Lareau (2003) has reported on how higher SES parents schedule their children for a great many activities, and in general focus on the “concerted cultivation” of their children’s skills, abilities, and behaviors.
(For supporting quantitative evidence, see Farkas and Hibel, forthcoming.) By contrast, Lareau observes that lower SES parents view their role as permitting their children to “accomplish natural growth,” which involves far less stimulation of specific, schoolrelevant skills and behaviors. This connects to the cultural capital research tradition, where higher SES parents transfer school-relevant skills, habits, and styles to their children (Swidler 1986; Lamont and Lareau 1988; Farkas et al. 1990; Farkas 1996;
Lamont and Small, this volume).
Finally, another related research tradition demonstrates that the skills and behaviors that higher SES and two parent families inculcate in their preschool children transfer almost immediately to placement into higher ability groups in elementary school, which in turn raises the student’s school engagement and academic performance, setting her or his academic trajectory on a steeper growth slope (Alexander, Entwisle, and Dauber 1993; Pallas et al. 1994; Marks 2000; Carbonaro 2005; Tach and Farkas, forthcoming). Once ability and curriculum group placement are in place to reinforce the higher academic performance and school engagement of more advantaged students, the gap between their performance and that of less advantaged students continues to increase as students move to higher schooling levels (Kerckhoff and Glennie 1999).