500N - 388 is not insignificant. There is a method of measuring the margin of error of a sample of a population, and the margin of error is a little under 5%. When you're talking of an effect in the order of 270% then there can be no overlap between the range of the effect and the range of the control, meaning that the number is statistically significant. I would have thought that someone who studied Psych did some statistics as part of the course and would know about sample sizes and margins of error.
Last edited by PTT; 22nd Jan 2013 at 21:16.