Testing for regression homogeneity under variance heterogeneity

Beverly J. Dretzke, Joel R. Levin, Ronald C. Serlin

Research output: Contribution to journalArticlepeer-review

28 Scopus citations

Abstract

One of the assumptions underlying the F test of parallelism of 2 or more regression lines is that the within-group residual variances are homogeneous. In the present study, a 2-group Monte Carlo investigation examined the effect of violating this assumption for F, a large-sample chi-square approximation (U0), and an approximate F test (F). In terms of Type I error probabilities, the standard F test performed acceptably well as long as sample sizes were equal. This was not true when sample sizes were unequal, with F proving to be clearly superior. The pattern of results parallel exactly what is known about the robustness of the F test when testing for mean differences in the presence of unequal variances. (9 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).

Original languageEnglish (US)
Pages (from-to)376-383
Number of pages8
JournalPsychological Bulletin
Volume91
Issue number2
DOIs
StatePublished - Mar 1 1982

Keywords

  • F test of parallelism & variance heterogeneity, testing for regression homogeneity

Fingerprint Dive into the research topics of 'Testing for regression homogeneity under variance heterogeneity'. Together they form a unique fingerprint.

Cite this