I'll just lightly comment on some of these tips (thanks for the PM Quercus):
1) While it's likely that a study in a non-peer-reviewed journal is inferior research (not a certainty), the converse that a study published in a peer-reviewed journal is superior (or sufficient) research is not necessarily the case. The "peer-review" process is only as good as the peers involved in the process. For instance, although Medicine and Science in Sport and Exercise (the official journal of the American College of Sports Medicine) is a peer-reviewed journal, the quality of the studies it publishes varies widely from excellent to garbage in terms of the data being used to support the conclusions written in the articles themselves.
Peer-review is a good indicator of "better" research, but it alone, is not sufficient to give blanket approval for quality.
2) Not all randomized placebo-controlled, double blind studies are created equal. Again, I'm simply pointing out that just because the paper claims to be a randomized placebo-controlled, double blind trial, it doesn't mean that the study is necessarily of high quality. A well run randomized control trial can lead to very strong evidence for or against something. If you're really interested in finding out how to evaluate one, there's an entire research group that studies this called the CONSORT group (
http://www.consort-statement.org). Their guidelines for assessing the quality of randomized control trials is widely used around the world as a checklist for ensuring at least adequate reporting of randomized control trials. Again, just because it passes the CONSORT checklist, still doesn't mean it's good research--but at least you'll have all the information you need to make that decision