Profile Analysis of Multivariate Data: A Brief Introduction to the profileR Package
Okan Bulut, C. Desjardins
Abstract
Profile analysis is a multivariate statistical technique, which is the equivalent of multivariate analysis of variance (MANOVA) for repeated measures. This technique is widely used by researchers in education, psychology, and medicine for the non-orthogonal decomposition of observed scores into level and pattern effects. A suite of procedures for decomposing observed scores into level and pattern effects and statistical techniques utilizing these effects exists for the R programming language in the profileR package (Bulut & Desjardins, 2018). This package includes routines to perform criterion-related profile analysis, profile analysis via multidimensional scaling, moderated profile analysis, profile analysis by group, and a within-person factor model to derive score profiles. This article showcases several of these methods, illustrating their applications with various data sets included with the package. The profileR package is geared towards researchers in the social sciences and medicine, with limited familiarity with R, and aims to lower the entry to using these methods for this audience.