A New Look at the Evaluation of Sociological Theories in International Large-Scale Educational Assessments

• In Event: Measuring Student Home Background in Large Cross-National Studies: Conceptual and Methodological Issues

Abstract

The paper uses a mixed approach of theory generating and theory testing to evaluate theoretical models of economic, cultural, and social capital. The data stem from two large-scale international assessments: PIRLS 2006 and PISA 2009. In line with the mixed approach employed, the method is a combination of exploratory factor analysis (EFA) and structural equation models (SEM) known as exploratory structural equation modeling (ESEM). The main advantage of using this method is that an EFA measurement model with rotations can be used in a SEM, instead of a confirmatory factor analysis (CFA) measurement model. When using the EFA we have the advantage of being able to relax the zero loading restriction, allowing variables to load on different factors simultaneously. The SEM then provides model fit indices, yields measurement invariance tests across countries, and corrects factor scores for measurement error. The results provide statistical criteria to evaluate the fit of theories with the empirical data (i.e., more reliably and validly). The paper contributes to the discussion on the advantages and disadvantages of using these methods when evaluating sociological theories in the context of international large-scale assessments.

Authors

• Daniel H. Caro, IEA Data Processing and Research Center

• Andres Sandoval-Hernandez, International Association for the Evaluation of Educational Achievement