Second Order Latent Growth Modeling With Common Item Ordinal Responses: An Application for Psychological Tests

• In Event: Innovations in Psychometrics Modelings, and Methodologies

Abstract

Second-order latent growth modeling (SOLGM) is one of the approaches that can address some limitations of traditional Latent Growth Modeling (LGM) by using individual indicators, allowing for the observation of indicator-specific effects (e.g., the common item effect). Despite the strengths of SOLGM having been introduced in several studies, it has yet to be widely used in practice, particularly for ordinal responses. This study investigates applying the SOLGM to ordinal responses, detailing the analysis procedure in various contexts and interpreting common item effects within a longitudinal psychological assessment. The findings indicate that SOLGM with common item factors is an effective strategy for analyzing repeatedly measured ordinal data and can provide additional information for interpreting change over time in a psychological assessment.

Authors

• Sook Hyun Park, University of Texas at Austin

• Tiffany A. Whittaker, University of Texas at Austin

• Nayoung Jang, Sookmyung Women's University