Hi,
I am currently conducting a number of different one-way ANOVA's and have used eta-squared as a measure of effect size. Where I have had unequal samples sizes and heterogeneity of variance, I have used Welch's F test; however, I am unsure whether I can calculate eta-squared for Welch's F test because SPSS does not provide the sums of squares, and if not, which measure of effect size I should use instead? I am keen to use the same measure of effect size for all tests.
Thanks in advance for your thoughts on this,
Tamara
|