Abstract
This thoughtful and stimulating article reflects on the central question posed by the Coleman Report: What role do schools play in promoting equality of opportunity? The Coleman Report relied on analysis of variance and regression analysis, but over the past 50 years, social scientists have developed new tools to address this question. The counterfactual revolution has made us aware of the bias emerging from advantaged families selecting into better schools, tracks, teachers, and peers, and it offers experiments and natural experiments to examine the causal effects of schools, and schools’ instructional and organizational resources, on student outcomes. Although still mostly restricted to test scores, the outcomes examined are, auspiciously, expanding to other domains, including children’s health and well-being and longer-term outcomes such as employment and earnings.
Approaches using a counterfactual formulation include, for example, assignment of students to schools via lottery, random allocation of teachers, comparisons in learning between the summer and the school year, and variation across time and state in universal preschool and kindergarten provision (for an excellent review, see Raudenbush and Eschmann 2015). Although these approaches are far from perfect—important concerns include generalizability, partial versus general equilibrium outcomes, and adequate functional form—they hold promise to accumulate evidence about which dimensions of the educational environment matter most and for which children. In what follows, I both depart from Downey and Condron (DC) about what the focus of research using the counterfactual approach should be and wholeheartedly agree with them about the need to consider socioeconomic and ethno-racial gaps before children start school.
DC focus on the summer learning perspective, which compares learning during the school year and the summer recess among elementary school children. Studies using this creative natural experiment offer two findings: All children learn more during the school year than during the summer, and poor children benefit more than wealthy children from being in school during the academic year. This compensatory effect of school exposure likely emerges because the household environments that poor children experience during the summer are less educationally enriching than those experienced by their wealthy peers.
It is important to understand the contours and implications of these findings (as of any finding using a counterfactual approach!). As DC indicate, the relevant counterfactual here is “What would inequality look like if schools did not exist?” rather than “How well would a particular student perform if they attended school A versus school B?” But a “world without schools” is not a meaningful counterfactual for the vast majority of elementary school children in the United States and other rich countries. Thus, the summer learning approach is only relevant in two distinct settings: for considering a policy extending the U.S. school year into the summer months and in some developing countries where elementary school enrollment is not yet universal.
I submit, however, that in other settings, the question about “school A versus school B”—or, more generally, the effect of schools and school resources, such as teachers’ characteristics, class size, and curriculum, on advantaged versus disadvantaged children—is indeed the critical question. This was in fact the question that the Coleman Report attempted to address. This is also largely the question to which the reproduction approach responds, particularly Schooling in Capitalist America (Bowles and Gintis 1976). The core of this critique is not that working-class children learn less in school than elite children but rather that children are sorted into different tracks, teachers, (hidden) curricula, or schools that socialize them into different jobs based on their social class of origin, thus reproducing unequal social structures. As research using counterfactual approaches accumulates, we have a great opportunity to understand which kinds of early school resources matter most for disadvantaged children and—given political will—foster the compensatory role of schools.
There is good reason to focus attention on the early educational career. A growing body of research from neurobiology and psychology shows that the skills children require for learning are shaped by early experiences, starting in the prenatal period. As an important study recaps, “virtually every aspect of early human development is affected by the environment and experiences encountered in cumulative fashion, beginning in the prenatal period and extending into the early childhood years” (Shonkoff and Phillips 2000:6). Given the plasticity of the human brain early in life and the hierarchical and interdependent manner in which skills develop, learning is in a very real sense a process that starts at conception (Knudsen et al. 2006).
This understanding of early human development has direct consequences for inequality of opportunity. Disadvantaged children are much more likely to be exposed to environmental insults that shape early skills, including violence, pollution, and stress, among many others, and their families are less likely to compensate for the effects of these exposures (Bernardi 2014; Torche 2016). As a result, inequalities in skills critical for learning are quite advanced even before children enter the formal educational system, and most of the socioeconomic gaps in achievement in adolescence can be tracked back to differences already present at school entry (Bradbury et al. 2015).
DC suggest that “the public needs to understand that socioeconomic achievement gaps form primarily before formal schooling.” I wholeheartedly agree. But I would like to push this further. The new findings about early human development invite a novel understanding of the role of schools in reducing the influence of the accidents of birth. If the learning process starts at conception, then the early life course should be a legitimate domain of educational policy. There is low-hanging fruit in this respect in the United States, a laggard compared to other rich countries on policies such as universal preschool education, paid parental leave, and yes, reducing inequality among households with children. Compensatory policies that provide educational resources to disadvantaged children are necessary, but starting earlier than age five might yield the most substantial payoffs.