Cohen's d is an effect size that describes the difference between two means in terms of which metric?

Cohen's d is an effect size that describes the difference between two means in terms of which metric?

• A. Standard deviation
• B. Standard error
• C. Pooled variance
• D. Median

Answer: (A)

Cohen's d expresses how big the difference between two means is relative to how spread out the data are. The denominator is the standard deviation (typically the pooled standard deviation across the two groups), which captures the typical variability within the groups. By dividing the mean difference by this variability, the result becomes unitless and comparable across different measures or studies. Think of it this way: a larger mean gap with little overlap in values (small spread) yields a large d, while the same gap with lots of overlap (large spread) yields a smaller d. The standard error would scale by sample size and reflects precision of the estimated means, not the inherent variability of the scores, so it isn’t the right denominator. The variance is the squared spread, and while the standard deviation is its square root, the standard deviation is the quantity used in Cohen’s d. The median is not a measure of dispersion, so it doesn’t fit here.

Cohen's d expresses how big the difference between two means is relative to how spread out the data are. The denominator is the standard deviation (typically the pooled standard deviation across the two groups), which captures the typical variability within the groups. By dividing the mean difference by this variability, the result becomes unitless and comparable across different measures or studies.

Think of it this way: a larger mean gap with little overlap in values (small spread) yields a large d, while the same gap with lots of overlap (large spread) yields a smaller d. The standard error would scale by sample size and reflects precision of the estimated means, not the inherent variability of the scores, so it isn’t the right denominator. The variance is the squared spread, and while the standard deviation is its square root, the standard deviation is the quantity used in Cohen’s d. The median is not a measure of dispersion, so it doesn’t fit here.