Linking student outcome variables with school process variables: multilevel confirmatory factor analysis takes precedence

Abstract

In the current study, we investigated the relationship of constructed measures for school process variables with distinct student outcome variables, including one measure for student achievement and one psychosocial outcome variable. Two different approaches were used to develop these school process measures: standard confirmatory factor analysis plus aggregation on the one hand (i.e., the traditional approach) and multilevel confirmatory factor analysis on the other hand (i.e., the alternative approach). The main research objective involved the comparison between these two approaches. It was found that the alternative approach could yield different conclusions in comparison with the more traditional approach. Therefore, these results underline the importance of using the most appropriate approach to develop measures for school process variables, which is the alternative approach that explicitly takes into account the hierarchical nature of the data collected on school processes.