How tracking can raise the test scores of high-ability minority students

This study examines the impact of achievement-based “tracking” in a large school district. The district in question required schools to create a separate class in fourth or fifth grade if they enrolled at least one gifted student (as identified by an IQ test). However, since most schools had only five or six gifted kids per grade, the bulk of the seats in these newly created classes were filled by the non-gifted students with the highest scores on the previous year’s standardized tests. This allowed the authors to estimate the effect of participating in a so-called Gifted and High Achieving (GHA) class using a “regression discontinuity” model.

Based on this approach, the authors arrive at two main findings: First, placement in a GHA class boosts the reading and math scores of high-achieving black and Hispanic students by roughly half of one standard deviation, but has no impact on white students. Second, creating a new GHA class has no impact on the achievement of other students at a school, including those who just miss the cutoff for admission. Importantly, the benefits of GHA admission seem to be driven by race as opposed to socioeconomic status. They are also slightly larger for minority boys than minority girls (especially in reading).

As the authors note, these differences between groups aren’t easily explained by factors like teacher quality, peer composition, and the “match” between student ability and the level of instruction—each of which might be expected to benefit all GHA students (or none). This interpretation is largely confirmed by a direct examination of these factors, which suggests that differences in teacher quality between GHA and non-GHA classes do not explain the achievement gains experienced by minority students in the former, and that differences in peer composition explain just 10 percent of these gains.

In light of these findings, the authors hypothesize that higher-ability minority students face obstacles in the regular classroom environment—such as lower teacher expectations and negative peer pressure—that cause them to underperform relative to their potential, and that some of these obstacles are reduced or eliminated in a GHA class. To support this hypothesis, they use data from an IQ-like test called the Nagliari Nonverbal Ability Test (NNAT) to show that minority students have lower achievement scores than white students with similar cognitive ability. For example, black students’ third-grade achievement scores are between 0.2 and 0.45 standard deviations below those of white students with similar NNAT scores, suggesting an “underachievement gap” similar in magnitude to the boost these students receive from a GHA classroom.

In a previous study conducted in the same school district, the authors also found that switching from a giftedness identification system based on teacher nominations to an automated process based on NNAT scores led to increases of 80 percent and 130 percent, respectively, in the numbers of black and Hispanic students identified as gifted, suggesting that teachers were either unaware of these students’ cognitive ability or unwilling to overlook their comparatively low achievement. Based on this finding, the authors argue that high-achieving minority students may benefit from GHA classes in part because teachers are more likely to recognize their potential in a GHA context.

Citing the ethnographic research on minority underperformance, the authors also suggest that the pressure to avoid "acting white" by achieving at a high level may be reduced for minority students in a GHA class, where all students are labeled as gifted or high-achieving. As evidence for this hypothesis, they cite lower rates of unexcused absences and suspensions among minority students in GHA classes.

Both of these explanations are plausible (if difficult to prove). But regardless of the true explanation, the fact that high-achieving minorities see such clear benefits from a policy of achievement-based “tracking” should give die-hard proponents of “equity” pause. After all, it’s one thing to prioritize those at the bottom, but it’s something else entirely to hold back those who could achieve escape velocity because they are short on company.

Perhaps it’s time we found a better word for encouraging these students to embrace their academic potential than “tracking.”

SOURCE: David Card and Laura Giuliano, "Can Tracking Raise the Test Scores of High-Ability Minority Students?," NBER (March 2016).

David Griffith
David Griffith is a Senior Research and Policy Associate at the Thomas B. Fordham Institute.