Effect Sizes

Effect sizes are measures of the strength of effects. In terms of the training course example, the effect size measures how much the training program improved test scores.

Cohen’s d is a measure of effect size that is discussed in the tutorial.

The basic formula for d is: wpe1.jpg (1249 bytes)

This corresponds to the differences between means divided by the standard deviation.  

Cohen offered the following rough guidelines to help you interpret effect sizes: a small ES = .2, a medium ES = .5, and a large ES = .8 and above.

Return to Sampling Exercise 1