matrixStrucTest: Tests of Matrix Structure for Construct Validation

Tests for block-diagonal structure in symmetric matrices (e.g. correlation matrices) under the null hypothesis of exchangeable off-diagonal elements. As described in Segal et al. (2019), these tests can be useful for construct validation either by themselves or as a complement to confirmatory factor analysis. Monte Carlo methods are used to approximate the permutation p-value with Hubert's Gamma (Hubert, 1976) and a t-statistic. This package also implements the chi-squared statistic described by Steiger (1980). Please see Segal, et al. (2019) <doi:10.1007/s11336-018-9647-4> for more information.

Getting started

Package details

AuthorBrian D. Segal [aut, cre]
MaintainerBrian D. Segal <>
LicenseGPL (>= 3)
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the matrixStrucTest package in your browser

Any scripts or data that you put into this service are public.

matrixStrucTest documentation built on July 18, 2019, 9:06 a.m.