Measuring Student Involvement: A Comparison of Classical Test Theory and Item Response Theory in the Construction of Scales from Student Surveys

Abstract

This study compares the psychometric utility of Classical Test Theory (CTT) and Item Response Theory (IRT) for scale construction with data from higher education student surveys. Using 2008 Your First College Year (YFCY) survey data from the Cooperative Institutional Research Program at the Higher Education Research Institute at UCLA, two scales are built and tested—one measuring social involvement and one measuring academic involvement. Findings indicate that although both CTT and IRT can be used to obtain the same information about the extent to which scale items tap into the latent trait being measured, the two measurement theories provide very different pictures of scale precision. On the whole, IRT provides much richer information about measurement precision as well as a clearer roadmap for scale improvement. The findings support the use of IRT for scale construction and survey development in higher education.