Appendices TableB.5: Regression estimates of student-level heterogeneity in the relationship between second- and sixth-grade skill levels. SKILLG6 Reading Spelling Math Panel I: Parental education (Middle=ref.) SKILLG2 x Low-educated .015 -.124*** .003 (.036) (.036) (.030) SKILLG2 x High-educated -.066*** -.051** -.085*** (.021) (.024) (.017) Panel II: Sex (Male=ref.) SKILLG2 xSex -.007 .017 .033** (.016) (.018) (.013) Panel III: Migration background (No=ref.) SKILLG2 x Western migrant -.073** -.083** -.103*** (.030) (.035) (.025) SKILLG2 x Non-western migrant -.011 -.072*** -.011 (.023) (.023) (.018) Note. SKILLG2 and SKILLG6 refer to the standardized skill scores in grades two and six, respectively. We control for parental education, age, sex, migration background, parental structure, family size, grade retention, and a COVID dummy. We also include a year and school FE. Robust standard errors are displayed in parentheses. *** p< .01, ** p< .05, * p< .1. 204
RkJQdWJsaXNoZXIy MjY0ODMw