You need to be able to calculate the difference between group means (do you have the groups means?) and divide by the standard error of the difference between the means (I'm guessing that's what SEM stands for -- the M threw me off). That's your basic T-test.
The process basically factors variability out of the difference between the group means to see if there's still a real difference.
I thought I have replied to the lase message but I couldn't see my reply so sorry for th delay I need this help ASAP.
I have two groups which I'd like to compare I have lost raw data but I still have MEAN of each group and STANDARD DEVIATION (AND STANDARD ERROR as I know how maney samples I had in each group)