Inequality from the Get Go

Abstract

IN THE MAY 2000 column, I reported on the U.S. Department of Education's study of the characteristics of students entering kindergarten. Not surprisingly, the data showed large differences across ethnic groups (favoring whites and Asians) and socioeconomic status (the higher the SES, the better). These data have been reanalyzed and interpreted in two recent books. Valerie Lee and David Burkam of the University of Michigan are the authors of the first book, Inequality at the Starting Gate, published by the Economic Policy Institute in 2002. The second book is An Uneven Start: Indicators of Inequality in School Readiness, by Richard Coley of the Educational Testing Service, whose Policy Information Center published the report. The reports concentrate on different aspects of inequality. Coley's report presents a lot of information cast in terms of the percentage of kindergartners who do certain things, broken out by ethnicity and SES and divided into quintiles. The index of SES combines the educational level, occupation, and income of parents or guardians. Some of the skill differences are relatively small. Virtually all high- SES children, for instance, can recognize numbers and shapes. For low- SES children, the smallest percentage who can recognize numbers and shapes for any ethnic group is 82% for blacks. On other tests, the differences are large, both among the SES quintiles and across ethnicities within the quintiles. For example, 77% of high-SES kindergartners understand relative size, but only 31% of low-SES children do. Within the high-SES quintile, 82% of whites understand relative size, compared to 60% of Hispanics. In the bottom SES quintile, 51% of white students understand relative size, compared to 26% of blacks. Similarly, 48% of high-SES Asian kindergartners understand ordinal sequence, compared to 21% of high-SES black kindergartners. In the lowest quintile, the figures are 15% of Asians, compared to 4% of blacks and Hispanics. Lee and Burkam report achievement in terms of scaled scores. Most of the graphs they present resemble a staircase moving upward with SES. Differences among ethnicities are presented in standard deviation units. These analyses show that low-SES black children are .56 SDs below the national average in reading and .68 SDs below average in math. For low-SES Hispanics, the figures are .69 SDs and .71 SDs respectively. Some of these differences probably arise in part because only 27% of low-SES children and just 14.6% of low/middle-SES children attended Head Start. (As seen in Figure 1, 65% of the high-SES and 52% of the high/middle-SES children attended center-based preschool, but only 20% of the low-SES children did.) Then, too, low-SES children owned an average of 38 books, while high-SES children owned 108 (Figure 2). Eighty-five percent of high-SES families had a computer in the home, compared to 29% of low-SES families. The school-readiness deficits that children bring with them to kindergarten are compounded by the schools they attend, something I did not get into in my May 2000 column. The problems begin with the environments of the schools. Black, Hispanic, and children of other races are more likely than their white counterparts to attend public schools located in the neighborhoods characterized by problem conditions, Lee and Burkam write. Those problem conditions include heavy traffic, drug use, gangs, crime, unoccupied buildings, trash, and graffiti. Low-SES children enter kindergarten classes that are larger than those attended by middle- and upper-class children, and those classes are presided over by less well-trained, less experienced teachers who feel less shared responsibility for the school and feel less that they are part of a professional community. The schools that low-SES children attend also do less to prepare them for the transition to first grade. All of these data are interesting, but, sadly, not surprising and just a wee bit frustrating. …