Ok that's been helpful, thanks for your answer!  ... View more

Thank you!  ... View more

@SJ12 wrote: That's exactly it, I want to do a further step. Let me know if you have suggestions!  Hard to make suggestions when we don't know where you are going with this. ... View more

Hi, the data set was provided in the question. Here is it again:    data test; input id foodtype $ Nova $ Subnova $ kcals; datalines; 001 butter 1 oils 10 001 fish 2 protein 15 001 banana 3 fruit 10 001 cherry 4 fruit 25 002 burger 2 frozen 5 002 pumpkin 2 other 4 002 carrot 3 veg 6 003 apple 1 fruit 100 003 tahini 4   spread   60 004 .   .    .   0 005 burger  2  frozen  13 005 ceral  3 grain 56 005 milk  1  prod 35 005 vinegar 4  Other  100 006 .    .   .   0 006 vinegar 4  Other  100 ; run;     This is an example of what I want (see below). I want to have the % of consumption of "Nova 4'' among all participants. I thought I mentioned this already, but maybe it's not clear since I'm new to SAS. Let me know if that's the case. Thanks          ... View more

@PaigeMiller , I really wish you could convince toxicologists and especially clinical pathologists about #4. I've spent the last 15 years trying to convince folks that taking the average of several animals' percent change from baseline doesn't result in a number that applies to the population the way they think it does..   SteveDenham. ... View more