| Session Title: Exploratory Factor Analysis Within a Confirmatory Factor Analysis Framework (E/CFA) |
| Expert Lecture Session 608 to be held in Pacific C on Friday, Nov 4, 10:45 AM to 11:30 AM |
| Sponsored by the Quantitative Methods: Theory and Design TIG |
| Presenter(s): |
| Cristian Gugiu, Western Michigan University, crisgugiu@yahoo.com |
| Abstract: Survey development often necessitates the validation of a factor structure via a confirmatory factor analysis (CFA) following its exploration via an exploratory factor analysis (EFA). However, researchers who follow this analytical sequence typically encounter poor-fitting CFA solutions, which result in the rejection of the model. The reason for this is that unlike an EFA, a CFA requires a strong empirical basis to guide specification and evaluation of the model. Consequently, before conducting a CFA, it is important to refine the instrument further by eliminating items that do not contribute to its internal validity. Unfortunately, EFA is not well suited for this task because it does not permit model comparisons or have a direct method for assessing item misfit. To this end, the present author developed an EFA within a CFA (E/CFA) method for identifying and eliminating poor fitting items. |