Large Variation in Math Achievement for Black Students Across School Districts in the United States

Studies of Black students’ standardized test scores in the United States have been largely relegated to discourse on racial disparities in achievement at a fixed point in time (Reardon et al., 2019; Reardon, 2016; Jencks & Phillips, 2011; Downey, 2008; Neal, 2006). Often overlooked are three facts: (a) there is far more heterogeneity in academic outcomes within racial/ethnic groups than between them, (b) Black students find themselves in a diversity of educational districts nationally; and (c) a static, cross-sectional view may obscure important dynamics. Past research on “the Black-White test score gap” creates an impression of uniformly low Black achievement across school districts. But this impression is distorted if there is substantial variation over time among districts serving Black students. A focus on achievement means leaves unanswered important questions: 1) How do districts compare with respect to the learning rates of the Black students they serve? 2) Is there evidence that some districts are improving with respect to Black mean achievement? Answers to these questions may stimulate exploration of strategies for improving district practice. The question of concern in this study was whether entire districts are especially effective (or ineffective) for Black children. Some districts may be educating Black children miserably, while others are doing much better. If so, this may challenge individual- and school-level arguments for the success or challenge of Black student educational progress. A national district level analysis fills a gap in the extant literature and may have implications for the level at which policy interventions should occur. Using longitudinal data on multiple cohorts of students flowing through nearly all public school districts in the United States, I estimated district average learning rates (change within cohorts over time) and district average improvement rates (change between cohorts at a given age) in mathematics on standardized test scores for Black students nationally. I define average district level Black student academic achievement as the true average score in mathematics for students who started kindergarten in 2006.5 and were in grade 5.5 within each district. The average district level Black student improvement rate is the true average district level linear trend in mathematics across cohorts during the period 2008/2009-2014/2015 for Back students in grade 5.5. The average district level Black student learning rate is the true linear trend in mathematics via standardized test scores within cohorts over the period 2008/2009-2014/2015. That is the district average rate at which achievement means change across grades three to eight for Black students who started kindergarten in 2006.5. Results indicate substantial district level variation in these indicators. Average district-level Black student academic achievement in mathematics on standardized test scores for Black students in grade 5.5 who started kindergarten in 2006.5 was at the 4.2 grade level. The corresponding standard deviation was 0.76. 95% of districts that reported average district-level Black student academic achievement had average Black student academic achievement between the 2.7 and 5.8 grade levels. The average district-level Black student improvement rate for Black students in mathematics on standardized test scores in grade 5.5 who started kindergarten between 2001 and 2012 was 0.014. The corresponding standard deviation was 0.09. 95% of districts that reported average district-level Black student improvement rate had an average district-level Black student improvement rate between -0.166 and 0.194. The average district-level Black student learning rate for Black students in mathematics on standardized test scores between grades 3-8 who started kindergarten in 2006.5 was 0.87 (1.0 equates to 1 grade level of growth from one academic year to the next). The corresponding standard deviation was 0.15. 95% of districts that reported average district-level Black student learning rates had an average district-level Black student learning rate between 0.57 and 1.17. In approximately 11% of the districts that serve Black students and report Black student academic outcomes, Black student learning rates in mathematics exceeded the national average learning rate for all students. That fraction is around 47% for district-wide improvement in Black Student achievement. Conversely, there were only 3% of districts where Black student mean achievement in mathematics exceed the national average for all students. Black student academic success or failure in mathematics vary widely across districts in the United States. This variability may prompt a thorough exploration into how entire districts can effectively structure themselves to promote widespread academic success for Black students. REFERENCES Downey, D. B. (2008). Black/white differences in school performance: The oppositional culture. Jencks, C., & Phillips, M. (Eds.). (2011). The Black-White test score gap. Brookings Institution Press. Neal, D. (2006). Why has black–white skill convergence stopped?. Handbook of the Economics. Reardon, S. F. (2016). School segregation and racial academic achievement gaps. RSF: The Russell Sage Foundation Journal of the Social Sciences, 2(5), 34-57. Reardon, S. F., Kalogrides, D., & Shores, K. (2019). The geography of racial/ethnic test score gaps. American Journal of Sociology, 124(4), 1164-1221.