Cognitive and noncognitive skills gaps between high-SES and low-SES children using household income as a proxy for socioeconomic status (SES), under unadjusted and fully adjusted models

Reading Math Self-control  (reported by teachers) Approaches to learning (reported by teachers) Self-control (reported by parents) Approaches to learning (reported by parents)
1 (unadjusted) 4 (fully adjusted) 1 (unadjusted) 4 (fully adjusted) 1 (unadjusted) 4 (fully adjusted) 1 (unadjusted) 4 (fully adjusted) 1 (unadjusted) 4 (fully adjusted) 1 (unadjusted) 4 (fully adjusted)
Gap in 1998 1.090*** 0.384*** 1.308*** 0.443*** 0.419*** 0.119** 0.603*** 0.325*** 0.443*** 0.272*** 0.436*** 0.073
(0.042) (0.058) (0.041) (0.060) (0.045) (0.050) (0.044) (0.049) (0.045) (0.051) (0.044) (0.052)
Change in gap by 2010 -0.127** -0.006 -0.230*** -0.060 0.049 0.228*** -0.128** 0.008 0.044 0.106 0.032 0.051
(0.060) (0.084) (0.059) (0.082) (0.066) (0.081) (0.064) (0.079) (0.065) (0.084) (0.064) (0.080)
Controls
Demographics No Yes No Yes No Yes No Yes No Yes No Yes
Education and engagement No Yes No Yes No Yes No Yes No Yes No Yes
Parental expectations No Yes No Yes No Yes No Yes No Yes No Yes
School fixed effects No Yes No Yes No Yes No Yes No Yes No Yes
Observations 28,650 26,050 29,560 26,890 27,550 25,080 29,110 26,460 28,170 27,220 28,190 27,240
Adjusted R2 0.103 0.276 0.143 0.321 0.023 0.174 0.036 0.199 0.019 0.079 0.019 0.226

Notes: Model 1 uses the full sample; Model 4 uses the complete cases sample. Robust standard errors are in parentheses. For statistical significance, *** denotes p < 0.01, ** denotes p < 0.05, and * denotes p < 0.1. The number of observations is rounded to the nearest multiple of 10.

Source: EPI analysis of ECLS-K, kindergarten classes of 1998–1999 and 2010–2011 (National Center for Education Statistics)

View the underlying data on epi.org.