How Much Teachers Know and How Much It Matters in Class : Analyzing Three Rounds of Subject-Specific Test Score Data of Indonesian Students and Teachers

de Ree, Joppe

World Bank, Washington, DC

关键词: SKILLS;    PRIMARY TEACHER;    BETTER TEACHERS;    PUBLIC PRIMARY SCHOOLS;    UNIVERSITY DEGREE;   

DOI  :  10.1596/1813-9450-7556
RP-ID  :  WPS7556

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

Improving the quality of education isone of today's main challenges for governments in thedeveloping world. Based on a unique matchedstudent-to-teacher panel data set on test scores this paperpresents two empirical results for Indonesia. First, throughdetailed inspection of teacher-level responses to testquestions, the paper concludes that subject matter knowledgeof primary school teachers in Indonesia is low on averageand that a 1.0, but also a 2.0 standard deviation increasein teachers' subject matter knowledge seem to beachievable medium-term goals for education policy making inIndonesia. Second, the paper presents the results of threetypes of value-added regressions, a (standard) levelspecification, a school fixed-effects specification, and aflexible student-teacher fixed-effects specification. Thestudent-teacher fixed-effects approach estimates theparameters of a value-added model using test score variationwithin each student-teacher pair across three differentsubjects, mathematics, science and Indonesian language. Theresults suggest that a 1.0 (and 2.0) standard deviationincrease in teachers' subject matter knowledgeacross-the-board can yield increases in student achievementby 0.25 (and 0.50) student-level standard deviations by thetime students complete the six-year primary school cycle.