We have just conducted a study comparing cognitive development of low-birthweight

(premature) and normal-birthweight babies at one year of age. Using a score of my own devising, I found the sample means of the two groups to be 25 and 30, respectively, with a pooled standard deviation (s) of 8. There were 20 subjects in each group. If we assume that the true means and standard deviations have been estimated exactly, what was the a priori probability (the probability before the experiment was conducted) that this study would find a significant difference?