Gain insight into the concept of effect size, represented both with Cohen's d and the graph provided. Cohen's d, which is a measure of effect size, lets you know the number of standard deviations between the population mean (as specified by the null hypothesis) and the sample mean (our evidence). Referring to Cohen's d, an effect size of .2 is small, that of .5 is medium, and an effect size of .8 would be considered large.

In the graph at the bottom of the card, the solid red normal distribution shows the expected distribution of scores, based upon the null hypothesis. The dashed curve presents the distribution of sample means (based upon the null hypothesis and sample size). The sample mean appears along the x-axis as a solid circle. According to the null hypothesis, the sample mean should occur somewhere in the middle of the distribution of sample means.

The more extreme the sample mean, the more likely we are to reject the null hypothesis as a poor fit for the evidence. With Hypothesis Testing, we evaluate whether we consider the null hypothesis is believable. The smaller the p value, the less believable we find the null hypothesis.

The effect size, however, lets us know whether or not the difference is a worthwhile one. A treatment that results in a statistically significant difference, but where the effect is very small, is not particularly interesting. Preferable is a statistically significant result with a desirable effect size.