How low-, middle- and high-scoring eighth-graders perform on reading tests relative to their peers, by gender, race/ethnicity, and socioeconomic status, 2003 and 2013

Place in the distribution of test scores
Variable Lowest third Middle third Upper third Lower third Middle third Upper third
Female Student 0.228*** 0.216*** 0.226*** 0.197*** 0.206*** 0.256***
(0.011) (0.013) (0.039) (0.011) (0.011) (0.037)
Student race/ethnicity
Black -0.479*** -0.450*** -0.405*** -0.432*** -0.412*** -0.364***
(0.019) (0.020) (0.058) (0.025) (0.022) (0.062)
Hispanic ELL -1.079*** -0.910*** -0.610** -1.026*** -0.928*** -0.817***
(0.052) (0.077) (0.254) (0.058) (0.060) (0.145)
Hispanic non-ELL -0.198*** -0.174*** -0.126 -0.108*** -0.118*** -0.112*
(0.024) (0.024) (0.102) (0.026) (0.022) (0.067)
Asian ELL -0.957*** -0.768*** -0.505** -0.952*** -0.836*** -0.664***
(0.153) (0.120) (0.237) (0.160) (0.117) (0.258)
Asian non-ELL 0.039 0.083* 0.114 0.192*** 0.225*** 0.272**
(0.053) (0.043) (0.134) (0.040) (0.042) (0.128)
Native American or other -0.312*** -0.253*** -0.138 -0.182*** -0.125*** -0.003
(0.042) (0.045) (0.129) (0.032) (0.035) (0.103)
Student socioeconomic status
Parents have some college education 0.255*** 0.243*** 0.234*** 0.259*** 0.243*** 0.218***
(0.015) (0.017) (0.060) (0.019) (0.019) (0.060)
Parents have completed college 0.305*** 0.312*** 0.317*** 0.297*** 0.310*** 0.328***
(0.013) (0.013) (0.055) (0.016) (0.016) (0.049)
Eligible for reduced-price lunch -0.108*** -0.084*** -0.052 -0.144*** -0.152*** -0.174*
(0.021) (0.020) (0.086) (0.027) (0.030) (0.100)
Eligible for free lunch -0.226*** -0.213*** -0.191*** -0.264*** -0.264*** -0.267***
(0.016) (0.019) (0.047) (0.017) (0.015) (0.045)
Control variables
1. Individual education plan (special education student)
Yes Yes Yes Yes Yes Yes
2. Percent of students in school who are black or Hispanic, and percent who are eligible for free or reduced-price lunch
Yes Yes Yes Yes Yes Yes
3. State fixed effects
No No No No No No

Note: The table shows regression estimates measured in standard deviations—how female students scored relative to male students, how students in different minority groups scored relative to white peers, how students who were eligible for reduced-price lunch or free lunch scored relative to students who were not eligible, and how students whose parents have some college or who have completed college scored relative to peers whose parents have no college experience. The regression estimates are provided for groups of students at different points of the performance distribution. *** p<0.01, ** p<0.05, * p<0.1

Source: EPI analysis of National Assessment of Educational Progress microdata, 2003 and 2013

View the underlying data on epi.org.