Using Parallel Analysis to Determine the Numbers of Factors in Multilevel Exploratory Factor Analysis

• In Event: Poster Session 14
  In Poster Session: Advanced Issues in Multilevel Factor Analysis

Abstract

This study focuses on the method of multilevel exploratory factor analysis (MEFA) that is an extension of the traditional EFA to multilevel data. A key issue in the application of MEFA is to determine the optimal number of factors underlying the items. Prior research showed that the commonly used goodness-of-fit indices (e.g., RMSEA, CFI, TLI, and SRMR) and information criteria (e.g., AIC, BIC, and SBIC) are inadequate in terms of identifying the correct number of factors at the cluster level. Parallel Analysis is a preferable method for factor enumeration. However, it has only been applied in single-level EFA. Hence, the purpose of this study is to extend Parallel Analysis to MEFA and compare the performance of various eigenvalue computation techniques.

Authors

• Yuhong (Melissa) Ji, Texas A&M University - College Station

• Wen Luo, Texas A&M University - College Station

• Oiman Kwok, Texas A&M University - College Station

• Yanyun Yang, Florida State University