This test assumes equal variance of the two samples, and it is one tailed - that is yor hypothesis is that one side will be greater than the other, not just different, and the results at least appear to be greater, say, the means seem to suppot your hypothesis.

For the two tailed test, double the p value. For the one tailed in which the means didn't come out as expected, subtract the given p value from one for your p value, and in that case things aren't looking too good for you.

Send a message to the author, Rob Burgess via web-form

Visit the homepage of Rob Burgess