Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

Conditional ANOVA for Mean Differences When Population Variances Are Unknown

Stephen Olejnik
The Journal of Experimental Education
Vol. 55, No. 3 (Spring, 1987), pp. 141-148
Published by: Taylor & Francis, Ltd.
Stable URL: http://www.jstor.org/stable/20151687
Page Count: 8
  • Download ($45.00)
  • Cite this Item
Conditional ANOVA for Mean Differences When Population Variances Are Unknown
Preview not available

Abstract

The comparison of population variances as a preliminary test to the ANOVA for unequal sample sizes was examined. To develop the sampling distribution of the conditional F ratio for comparing means, data were generated using computer simulation methodology for several conditions of varying sample sizes, raw score distributions, and population variances. The results indicated that because of low statistical power, preliminary tests for variance equality were considered of marginal value. Even when the significance level for the preliminary test was increased to .10, Type II error rates were unacceptably high. The difference between the conditional F distribution for mean differences and the theoretical F distribution was greatest when sample sizes were small and unequal.

Page Thumbnails

  • Thumbnail: Page 
141
    141
  • Thumbnail: Page 
142
    142
  • Thumbnail: Page 
143
    143
  • Thumbnail: Page 
144
    144
  • Thumbnail: Page 
145
    145
  • Thumbnail: Page 
146
    146
  • Thumbnail: Page 
147
    147
  • Thumbnail: Page 
148
    148