Testing Measurement and Structural Equivalence in Different Age Groups of Children

In this study the measurement and structural equivalence was tested across two instruments, the original Dutch version of the Competentie-Beleving Schaal voor Kinderen (CBSK) and a shortened and simplified version of the same test across different ages. The CBSK is intended for children between 8 and 12 years old. The shortened version is developed for children aged between 6 and 8 years old. To use the shortened CBSK as a counterpart of the original CBSK, both instruments should be equivalent across ages. The analyses have been done on two data sets. The first consists of a sample of 75 nine years old children who responded on both instruments. The second data set consists of a sample of 951 children aged between six and nine years old and responded on the shortened CBSK and 758 children between nine and 13 years old who responded on the original CBSK. Structural equation modeling was used to test the measurement and structural equivalence. Given the small sample size of one of the groups, bootstrap methods were used to assess the fit of the models. The results show that both instruments are reasonably well comparable and exchangeable for children aged nine years old and younger. However this conclusion does not persist for responses of children aged ten years or older.