Pub. date: 2009 | Online Pub. Date: October 05, 2009 | DOI: 10.4135/9780857020994 | Print ISBN: 9781412930918 | Online ISBN: 9780857020994| Publisher:SAGE Publications LtdAbout this handbook
Chapter 6: Factor Analysis
Robert C. MacCallum
Factor analysis Researchers in psychology routinely observe that variables of interest are intercorrelated in sample data. Correlations among measured variables (MVs) may be due to a number of phenomena, including direct causation, indirect causation, or joint dependence on other variables. Although it may be relatively straightforwardtoexplainasimple correlation between two variables, accounting for an array of correlations among a substantial number of variables is much more difficult. Given a set of p MVs, the observed inter-correlations among them comprise a complex set of information, and the investigator seeks to understand and account for this information in a simple and meaningful way. Factor analysis models and methods provide a framework for addressing this problem. The fundamental premise of factor analysis is that there exist latent variables (LVs) that influence the MVs. An LV, or a factor in factor analysis, is a hypothetical construct that is not directly measured. LVs are a p ...