Further Considerations in Applications of Factor Analysis

Problems of communality estimation, number of factors, rotation, and computation of factor scores are reviewed with implications for counseling psychology research. Issues in communality estimation include use of reliabilities, highest correlation, and squared multiple correlations. Methods of deciding on the number of factors include the Kaiser criterion, scree test, parallel analyses, and others. Reasons for rotation of factor matrices are discussed, and suggestions are made concerning oblique versus orthogonal, Quartimax versus Varimax, and other rotational decisions. Use of three types of factor scoring procedures is discussed. Suggestions also are made concerning data essential to meaningful reporting of a factor analysis. (PsycINFO Database Record (c) 2006 APA, all rights reserved).