NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED516588
Record Type: Non-Journal
Publication Date: 2011-Feb-3
Pages: 32
Abstractor: As Provided
Reference Count: 7
ISBN: N/A
ISSN: N/A
Factor Scores, Structure and Communality Coefficients: A Primer
Odum, Mary
Online Submission, Paper presented at the Annual Meeting of the Southwest Educational Research Association (San Antonio, TX, Feb 2-5, 2011)
(Purpose) The purpose of this paper is to present an easy-to-understand primer on three important concepts of factor analysis: Factor scores, structure coefficients, and communality coefficients. Given that statistical analyses are a part of a global general linear model (GLM), and utilize weights as an integral part of analyses (Thompson, 2006; Thompson, 2004), terminology used for factor analytic procedures are analogous to terminology in other GLM analyses. Therefore, transfer of ideas beyond factor analysis to various other GLM analyses is anticipated. (Methodology) An introductory overview of meanings and applications of factor scores, structure coefficients, and communality coefficients is presented. Additionally, four methods for calculating factor scores are compared: (1) The Anderson-Rubin method (Anderson & Rubin, 1956); (2) the Bartlett method (Bartlett, 1937); (3) the regression method (Gorsuch, 1983); and (4) the Thompson method (Thompson, 1993). Step-by-step instructions are provided for utilizing these four methods, with heuristic examples. (Recommendations) The paper concludes with recommended reporting practices and a recap of key ideas on factor scores, structure coefficients, and communality coefficients. An appendix contains: SPSS Syntax to Execute Factor Score Calculation with Regression, Bartlett, Anderson-Rubin and Thompson Methods. (Contains 12 tables.)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A