Question

This is a tough one. I suggest you use Lee Becker’s pooled variance program (use the hyperlink from a few slides back). Let’s say you have a study looking at the DV “Time to help”. Your IV is whether potential helpers did not see (condition #1) or saw (condition #2) someone else helping. The mean time to help for those in condition #1 was 58 seconds (variance was 2.81). The mean time to help for those in condition #2 was 55 seconds (variance was 3.06). First, tell me what Cohen’s d (pooled) is. Second, tell me if the effect size is meaningful in terms or magnitude!

Answer 1

Cohen's d indicates the standardized difference between two means to measure the significance of an effect. (For example: Drug A has a higher rate of recovery for Covid-19 as compared to Drug B). The 'd' represents the standard deviation between the two groups.  

The standard deviation (square root of the variance) would be:

Group 1: 1.67

Group 2: 1.89

To calculate the d, you subtract the mean of group 1 and group 2 and divide it by the average standard deviation of both groups.

The average SD (SD of group 1 + SD of group 2 / 2 ) is 1.78

58- 53/ 1.78 = 1.68

The size effect interpretations according to Cohen are:

small effect : 0 to .20

medium effect: .20 to .50

large effect: anything greater than .50

Going by this interpretation, the results from the illustration listed indicates a large effect. However, the statistical significance of it varies based on the size of your sample.