0

Is covariance a measure of effect size? Since correlation is such a measure, and correlation is normalized covariance, I would think so, but I'd like to know for sure.

Julian A.
  • 223
  • 2
  • 5
  • "Effect" of what? To be understood, this term needs a context. What is your null hypothesis and what is the alternative? – whuber Aug 08 '15 at 15:18
  • Effect sizes are frequently about differences in means (e.g. the mean difference between treatment and control). See this question, [Non-transitivity of correlation: correlations between gender and brain size and between brain size and IQ, but no correlation between gender and IQ](http://stats.stackexchange.com/q/131065/1036), for discussion on why correlations/covariances typically say nothing about the means. – Andy W Aug 08 '15 at 16:06

0 Answers0