Cohen's d

Abstract Cohen's (1988) d is one of several statistics used to examine differences between means, as in the case of the difference in mean outcome between participants who received treatment X and participants who received treatment Y. This difference score, calculated in standard deviation units, falls under the broader category of effect sizes. Effect sizes provide an important complement to traditional null hypothesis statistical significance testing. The traditional p value indicates the likelihood of obtaining the observed results if the null hypothesis is true (Cohen, 1988). It does not, however, offer any information about the magnitude of the research findings (e.g., how much of a difference it makes if someone receives treatment X vs. treatment Y; Rosenthal & Rosnow, 2008). Cohen's d offers an effective way of obtaining this information from means and standard deviations calculated in primary research.