Dr. Wang completed his Ph.D. in Quantitative Psychology at the University of Kansas, specializing in latent-variable-based models for the analysis of self-reported Likert-type survey data.
"A brief discussion on factor analysis – common misconceptions and proper applications"
A thorough investigation of a research question often requires data collected on a large number of variables. Despite the benefit of carrying more information, having too many variables in a model can lead to unexpected problems (e.g., multicollinearity in regression). This talk will cover the use of factor analysis as a dimension reduction technique when dealing with many variables. The distinction between exploratory factor analysis and principal component analysis, the relationship between exploratory and confirmatory factor analysis, and the connection between regression and factor analysis (and structural equation modeling) will also be discussed.