NEPS technical report for mathematics: Scaling and linking results of starting cohort 1 for ten-year-old children (wave 11)

The National Educational Panel Study (NEPS) aims at investigating the development of competencies across the whole life span and designs tests for assessing these different competence domains. To evaluate the quality of the competence tests, a wide range of analyses based on item response theory (IRT) were performed. This paper describes the data and scaling procedure for the mathematical competence test for 10-year-old children of Starting Cohort 1 (newborns). The mathematics test was conducted as a computerized multistage test (MST) with 26 items in total. Each student was given 16 items in one of three stages that represent different content areas as well as different cognitive components and use different response formats. The test was administered to 1,507 students. A partial-credit model was used for scaling the data. Item fit statistics, differential item functioning, Raschhomogeneity, and the test´s dimensionality were evaluated to ensure the quality of the test. The results show that the test exhibited a rather low reliability (EAP/PV reliability = .59), but the items satisfactorily fitted the model. Furthermore, comparable measurements could be confirmed for different subgroups. Limitation of the test is that some items in the difficult stage had to be excluded due to a small number of valid responses. Therefore, some difficult items were missing for the estimation in the upper ability range. Overall, the results revealed predominantly satisfactory to good psychometric properties of the mathematics test, thus supporting the estimation of an acceptable mathematics competence score. Besides the scaling results, this paper also describes the data available in the Scientific Use File and provides the R syntax for scaling the data.