POLYMAT-C: a comprehensive SPSS program for computing the polychoric correlation matrix

• Published in: Behavior research methods 2015-09, Vol.47 (3), p.884-889
• Main Authors: Lorenzo-Seva, Urbano, Ferrando, Pere J.
• Format: Article
• Language: English
• Subjects: Behavioral Science and Psychology; Cognitive Psychology; Factor Analysis, Statistical; Humans; Psychology; Software; Statistics as Topic - methods
• ISSN: 1554-3528; 1554-3528
• DOI: 10.3758/s13428-014-0511-x

We provide a free noncommercial SPSS program that implements procedures for (a) obtaining the polychoric correlation matrix between a set of ordered categorical measures, so that it can be used as input for the SPSS factor analysis (FA) program; (b) testing the null hypothesis of zero population correlation for each element of the matrix by using appropriate simulation procedures; (c) obtaining valid and accurate confidence intervals via bootstrap resampling for those correlations found to be significant; and (d) performing, if necessary, a smoothing procedure that makes the matrix amenable to any FA estimation procedure. For the main purpose (a), the program uses a robust unified procedure that allows four different types of estimates to be obtained at the user’s choice. Overall, we hope the program will be a very useful tool for the applied researcher, not only because it provides an appropriate input matrix for FA, but also because it allows the researcher to carefully check the appropriateness of the matrix for this purpose. The SPSS syntax, a short manual, and data files related to this article are available as Supplemental materials that are available for download with this article.