How fourth- and eighth-graders perform on math and reading tests relative to their peers, by race/ethnicity and poverty level, 2003

Eighth-grade math a Eighth-grade reading a Fourth-grade math Fourth-grade reading
African American -0.712*** -0.545*** -0.663*** -0.532***
(0.014) (0.014) (0.012) (0.012)
Hispanic English language learner -1.032*** -1.079*** -0.839*** -0.900***
(0.041) (0.044) (0.022) (0.025)
Hispanic non-English language learner -0.302*** -0.265*** -0.287*** -0.291***
(0.015) (0.018) (0.022) (0.020)
Asian English language learner -0.588*** -0.917*** -0.287*** -0.601***
(0.084) (0.089) (0.070) (0.065)
Asian non-English language learner 0.168*** 0.047 0.268*** 0.086***
(0.033) (0.029) (0.041) (0.032)
Native American or other -0.310*** -0.349*** -0.317*** -0.321***
(0.051) (0.044) (0.023) (0.027)
Eligible for reduced-price lunch -0.260*** -0.207*** -0.304*** -0.313***
(0.013) (0.015) (0.013) (0.014)
Eligible for free lunch -0.460*** -0.414*** -0.514*** -0.487***
(0.013) (0.011) (0.011) (0.010)

Notes: The table shows regression estimates measured in standard deviations—how students in different minority groups scored relative to white peers (top panel) and how students who were eligible for reduced-price lunch or free lunch scored relative to students who were not eligible (bottom panel). *** p<0.01

Notes on data controls: a. Eighth-grade estimates for race/ethnicity include a control variable for parents’ education in addition to eligibility for free or reduced-price lunch (FRPL); fourth-grade estimates for race/ethnicity include a control for eligibility for FRPL, but do not include parents’ education because it is not available. Eighth-grade estimates of FRPL coefficients include a control variable for parents' education in addition to race/ethnicity; fourth-grade estimates of FRPL coefficients include a control for race/ethnicity, but do not include the parents' education variable.

Source: National Assessment of Educational Progress microdata, 2003

View the underlying data on epi.org.