Stat question

Here's our results again... Group Statistics Std. Error sex of respondent N Mean Std. Deviation Mean # of times used heroin, female 123 10.3496 12.78189 1.15250 last 30 days male 127 11.5118 13.87007 1.23077 Independent Samples Test t-test for Equality of Means 95% Confidence Interval of the Mean Std. Error Difference df Sig. (2-tailed) Difference Difference Lower Upper # of times used heroin, .688 248 492 -1.16222 1.68835 -4.48755 2.16311 last 30 days Say we were a bit, well a lot suspicious of research, and we had trouble trusting the results of this one study with its one sample. Suppose we spent a ton money and drew 1000,000 random samples of the same sample sizes used in this study. What is the probability we would observe a difference between our group means of at least (an absolute value of) 1.16 ? (-1.16 is our observed sample outcome but we are using a two-tailed test.)