School Size and Student Achievement in TIMSS 2003

A large strand of literature within different social sciences of education deals with the effects of school size on student achievement. Much of the existing research seems to indicate that up to a point, small school size is associated with higher student achievement, with optimal school sizes of some 600-900 students. However, evidence is mostly confined to the United States. Using TIMSS data, this paper will look at the association between school size and student achievement within different countries to identify whether internationally there is such a thing as an optimal school size. Due to data limitations, however, the aim of this paper cannot be to identify causal effects of school size on student performance, but rather to draw a first picture of the kind of relationship found between school size and student performance in the different countries. Although TIMSS 2003 targeted and yielded representative samples of students both at the eighth grade and at the fourth grade, we will restrict our analysis to eighth grade students for the purpose of this paper. A second important issue with respect to school size that the paper will look at is the question: If there is a benefit attached to a certain school size, who profits most? Here, we specifically focus on groups of students that differ with respect to socioeconomic status and with respect to immigrant status. To analyze at the association between school size and student achievement within different countries, we employ student-level micro data from the Trends in International Mathematics and Science Study (TIMSS) 2003. TIMSS 2003 is an extensive international student achievement tests conducted by the International Association for the Evaluation of Educational Achievement (IEA), an independent cooperation of national research institutes and governmental research agencies. All participating countries received the same test items, so that the ensuing measures of the educational 3 performance of students in math and science are directly comparable across countries.T 1 T Furthermore, by using representative sampling methods to draw random samples of schools, the test provides representative samples of students in each participating country. The target populations of TIMSS 2003 were all students at the end of their eighth and fourth years of formal schooling in the participating countries. TIMSS 2003 was constructed as a curriculum-valid test, which ensures greater validity of international comparisons of student achievement. To ensure comparability with the former TIMSS achievement tests, the formal definition for the grade …