Second-order confirmatory factor analysis on the academic motivation scale using MIIV-2SLS estimator

Testing the validity and reliability of psychological measurement tools is important to ensure its quality and feasibility. Second-Order Confirmatory Factor Analysis is often used to test the internal structure validity in complex psychological variables, for example, Academic Motivation. However, choosing the right estimation method is crucial to ensure the accuracy of the model. In this case, observed variables that do not follow the multivariate normal distribution, have misspecification, and exhibit heteroscedasticity are common issues. The measurement tool examined in this study is the Short Version of Academic Motivation Scale that has been adapted into Bahasa Indonesia (AMS-ISV). A total of 344 data were collected from undergraduate students in Surabaya, Indonesia. Since the observed variable data did not meet the assumptions of multivariate normality, a suitable estimation method was required. This study aims to estimate parameters for Second-Order Confirmatory Factor Analysis (SO-CFA) using the Model Implied Instrumental Variables Two-Stage Least Squares (MIIV-2SLS) estimator for the AMS-ISV. The results of this study indicate that the AMS-ISV is valid, as the p-values of the Sargan test were all nonsignificant. The AMS-ISV also has a reliability coefficient greater than 0.7, indicating that the instrument is reliable for measuring students’ academic motivation in Indonesia. These findings suggest that the application of MIIV-2SLS estimator has good performance on observed variables that do not follow multivariate normal distributions.