TESTS OF SIGNIFICANCE IN TWO‐WAY DESIGNS (MIXED MODEL) WITH DICHOTOMOUS DATA

1 May 1971

journal article
Published by Wiley in British Journal of Mathematical and Statistical Psychology

Vol. 24 (1) , 111-116
https://doi.org/10.1111/j.2044-8317.1971.tb00453.x

Abstract

Computer simulation experiments are performed to investigate the applicability of different tests of treatment and interaction effects in mixed two‐way designs with dichotomous data. The main results are that the F test can be used with good approximation as a test of equal treatment effects, except for strongly asymmetric populations, and that the F test may be used with some caution as a test of no interaction.

Keywords

TREATMENT
MODEL
MIXED
GOOD
ASYMMETRIC
CAUTION
DICHOTOMOUS DATA
TWO WAY DESIGNS
INTERACTION

This publication has 0 references indexed in Scilit: