Identifying Latent Classes in Changes of Students’ Academic Anxiety and Testing Determinants of the Classes

• In Event: Poster Session 11
  In Poster Session: Methods for Analyzing Longitudinal Data

Abstract

The purpose of this study was to identify the number and the trajectory of latent classes based on students’ academic anxiety and test determinants of the classes such as gender, academic self-efficacy, plan for entering university, information of self, academic attention from parents, satisfaction of home life, and satisfaction of school life. To identify the latent classes, we applied semi-parametric group-based approach to the five waves of longitudinal data from the Korean Educational and Employment Panel. As a result, four latent classes emerged according to the different trajectories: the middle-level group, the low-level group, the decreasing group, and the increasing and decreasing group. Results of multinomial logit analysis showed that all variables were significant to determine the classes.

Authors

• Unkyung No, Korea University

• Sehee Hong, Korea University