In need of help

This is mainly directed to the research scientists out there, but any and all help will be greatly appreciated.

I’m going into my third year of grad school in mathematics (specializing in applied statistics) and I need to write my masters report. I finally thought of something that would actually be interesting and not just a 35 page book report. I figured I’d bring in some exercise physiology, nutrition, and pharmacology.

I wanted to take some studies with somewhat dubious conclusions and reanalyse the data from a statistician’s perspective. I’d try to see if any significant results might not actually be so significant or vice-versa. I figured the fine folks at T-mag and on the forum would probably be able to point me in the right direction.

Ideally, I'd like pairs of studies with seemingly contradictory results and larger data sets which might see some benefit from transformations, but i can do stuff with single studies and small samples as well. With as many years and dollars as I've paid toward the "Weider Tax" (i think Brock coined that one) I wouldn't mind seeing how they justified some of the crap they put out. Some amount of science, however misguided, had to lead to the myths we all heard and read for so many years. eg. high rep ab training, low-fat diets, steroids were created by satan, etc.

anyway, i’ve droned on long enough, thanks in advance for any assistance. i just need titles, authors, journal names, that sort of stuff. i have a gigantic university library system at my disposal.

just keeping this alive. anyone?

dude, it would be cool to see if there were any studies that claimed that a high carb, low fat diet reduced fat in the participants, and then you could maybe compare those studies to the high protein and fat, low carb participants of other studies?

hey Mustang. yeah, that’s exactly the kind of thing i was thinking about. so far the data gathering is going pretty slow, though. hours upon hours of cruising through microfiche, only to find that some jackball didn’t include his raw data in his doctoral thesis gets to be a pain. i have a few things on steroids and left-ventricular size, but it’s going to take a lot more time.