Measuring Effect Size

To analyzing effect size, Cohen’s d is calculated to measure the degree of separation between two distributions.  Gravetter and Wallnau (2013) stated that “a measure of effect size is intended to provide a measurement of the absolute magnitude of a treatment effect, independent of the size of the sample(s) being used” (p. 262).  The Cohen’s d is a direct and simple technique for measuring effect size.  Often used as a standardized measure of mean difference, Cohen’s d is computed as the mean difference divided by the pooled standard deviation.  In general, when a hypothesis test shows a statistically significant effect, a report of the effect size should be provided for the research study.

The formula of Cohen’s d calculation reveals that Cohen’s d measures the number of standard deviations for which two distributions separate from each other, therefore d = 1 tells that the means of two sample measurements are separated by one standard deviation.  For example, a Cohen’s d of .90 indicates that the means of a measure between two treatment conditions are different by 90% of a standard deviation.  According to the guidelines in Table 8.2 (p. 264), a value of d = .90 is considered a large effect.


Gravetter, F. J. & Wallnau, L. B. (2013). Statistics for the behavioral sciences (9th ed.). Belmont, CA: Wadsworth.

No comments:

Post a Comment