Andy Brunning of Chemistry site Compound Interest has put together this guide on warning signs for bad science. Not every point is relevant all of the time – there are a lot of fields where control groups are impossible for example, or where even the best studies have tiny sample sizes because that is all we have to go on.
The thing that makes science stand out from other endeavors is its self-correcting nature. Bad research will sometimes be picked up prior to publication, but just as often it slips through, only for the problems to be found later. Once the problems are identified, however, scientific researchers usually take them into account and move on. Sadly, this is not always the case with popular reporting of science.
It is painfully common to see people referring to studies that were debunked years ago as if they were the latest, if not final, word. That's where infographics like this really come in handy. If kept in mind these signs might help non-scientists do a first screen to keep out much of the bad science, or non-science masquerading as science. Which is great, because it leaves more room for the good stuff. We aren't kidding when we say We Love Science, but it might be more accurate to say We Love Good Science. There's a lot of bad science out there, either through deliberate fraud or genuine mistakes. There is also a lot of bad science reporting, making out perfectly good research to be something it is not.
by Stephen Luntz
http://imgur.com/Q2htxvN