Learning gains and race

In a prior post, I looked at the relationship between the Buckeye State’s value-added index scores and the state’s measure of poverty. Value-added scores are Ohio’s estimate, using a statistical method, of a school’s contribution to their students’ learning over the course of a school year. In this post, I examine the relationship between a school’s racial composition and its value-added score.

First, by selecting race as a variable that may influence school-level impacts on education, I’m not implying that kids of an African descent versus European versus Hispanic versus Asian have any inherent advantage. However, race (specifically, a school’s percentage of black students) could capture the impact of many untracked variables in the state’s education data, including the following factors:

• In Ohio, 74 percent of black children come from a single-parent family compared to 28 percent of white children;
• 58 percent of Ohio’s black children live in families where no parent has a full-time, year-round job, compared to just 27 percent of white children;
• 50 percent of Ohio’s black children come from families living below the federal poverty level ($23,000 per year for a family of four), compared to 17 percent of white children;
• 43 percent of black males (national data) have seen someone shot by age 18.

These bleak statistics surely factor into the lower achievement scores for black vis-à-vis white children. (To see the racial achievement gap in Ohio, see figures 3.5 and 3.6 in our 2012-13 Report Card analysis.)

But, does the racial composition of a school, as a percentage of black students, factor into a school’s value-added score?[1] In other words, does race make a difference with respect to whether a school can impact student learning? Chart 1 shows the relationship, showing the “gain index score” of Ohio’s schools (the vertical axis) and the percentage of black students enrolled in each school (horizontal axis). Each dot represents a school building.

Chart 1: Weak relationship between % black students and value-added

Ohio schools, 2012-13

Source: Analysis of Ohio Department of Education, School Report Card Data. Note: Schools with 0% black students are excluded. The department does not report 0% figures, nor does it report 100% percent figures (a 100% white rate obviously implies 0% black students). Schools with 95%+ students in any subgroup are reported as >95%. Correlation,  -.13. A value of 0 indicates no relationship on a scale of -1 (perfect negative relationship) to +1 (perfect positive relationship).

There are two points to be made about this chart:

1.)    It’s stark to see how few schools are racially “diverse.” The center of the chart, where schools that are more or less 50/50 black to non-black, is sparse. Meanwhile, the ends of the chart are denser with dots: schools that are primarily black are found on the right and schools that are primarily white on the left.

2.)    The relationship between the percentage of black students in a school and its value-added score is nearly non-existent. The best-fit line—displayed in red—through the points is nearly horizontal (a completely flat line suggests no relationship). The correlation is weak. In fact, there are very high percentage black schools that achieve greater-than-expected learning gains for their students (and conversely, schools that don’t achieve very big gains).

It is evident that schools can and do have an impact on student’s learning (by the state’s own metric), even if their students are from (a) greater impoverishment and (b) from underrepresented minority families. Schools—and the leaders and teachers who work in them—can make a difference in any child’s life.