This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Vickers, A. J. Search for Related Content PubMed Articles by Vickers, A. J.

PLEASE DON’T TALK BAD ABOUT GOOD OLD AUNT ANOVA!: A REPLY TO A.J. VICKERS’ CRITIQUE

RESPONSE

Department of Medicine,; Department of Epidemiology and; Biostatistics; Memorial Sloan Kettering Cancer; Center; New York, NY; vickersa{at}mskcc.org

I agree with many of the points raised in this letter. Indeed, most are described in the text of my article. For example, I state explicitly in the discussion that it is quite possible to report a p-value from ANOVA and then calculate a difference between groups with 95% confidence interval, and that many investigators do this. But, as I point out, calculation of the effect size is not an inherent part of conducting an ANOVA and this makes it more likely that investigators fail to report or downplay this statistic. Dr. Hartmann’s argument that it is possible to write special statistical code to calculate effect sizes from an ANOVA illustrates this point nicely. So, no, there is nothing inherently wrong with ANOVA. But yes, this is a technique that is prone to misuse, that is widely misapplied and for which there are alternatives. Hence, I recommend caution.

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Vickers, A. J. Search for Related Content PubMed Articles by Vickers, A. J.