Chapter 3. Primary School Skill Development: from First to Sixth Grade skills, parental characteristics (e.g., parental education), and investments (e.g., schooling). Therefore, we model the component as consisting of initial skills, parental education (P), schooling (S), and other factors that might influence the skill production (Z): SKILLGt+p =SKILLGt +g(SKILLGt,P,S,Z) (3.2) We reshape the abovementioned equation into a regression model: SKILLisc,G6 =β0 +β1SKILLisc,G<6 +β2Pisc,G6 + 1,248X s=1 Ss,G6ϑ+Z′isc,G6φ+ 2021X c=2019 δc,G6γ +εisc,G6 (3.3) where SKILLisc,G6 represents the standardized sixth-grade skill score of student i of school s in cohort c,whereas SKILLisc,G<6 is its counterpart in an earlier grade. In most of our analyses, this earlier grade is the first grade for OLS analyses and mid-second grade for IV analyses. We include a measure of parental education (Pisc,G6), distinguishing students from low-, middle-, and high-educated parents. Furthermore, we add school fixed effects (Ss,G6), which control for sixth-grade between-school differences with respect to school characteristics (e.g., size, denomination, urbanity, etc.) and additional school-specific factors, such as the learning climate created by schools and the school policy on basic skills.10 10The data contain information of students of 1,248 schools. We use the school where the student was administered by the enrolment on 1 October in grade six. 56
RkJQdWJsaXNoZXIy MjY0ODMw