There are some very basic things you can check to evaluate the credibility of a scientist that in many ways are similar to doctors. Where did they receive their education? Are they members in good standing with professional associations? Do they work at credible institutions? Do they have a conflict of interest? And for scientists specifically, do they publish peer-reviewed material?
If you for instance look at the 'scientists' who deny global warming, you will invariably find that they are not trained in climate science, do not publish climate research in peer-reviewed publications, and do not work for credible scientific institutes. Many will also have a conflict of interest--the Heartland Institute, for instance, which is funded by fossil fuel interests, is a good example.
That doesn't mean there aren't legitimate scientific disputes. A questioning and skeptical mind is important in science, but that investigative spirit needs to take place within the framework of the scientific method and within the framework of the entire body of knowledge which already exists on an issue.
For non-scientists I think there are several relatively easy things to watch out for. Scientists in general are usually extremely careful in describing certainties. So they will say things like, "there is a high level of probability that X causes Y." They will almost never, however, talk in absolutes. Anytime you hear someone say, they know the truth, there is a good chance they are not a scientist.
Scientists, also, in my opinion, tend to be uncomfortable making any claim without providing some evidence for it. People who do so are also unlikely to be scientists. Lastly, a good scientist shouldn't ever simply dismiss legitimate scientific evidence. People who wave their hand at the entire body of work developed to support climate change as if it is some great conspiracy are NOT scientists.
Part of the problem is that in the public realm, very few actual scientists are ever heard from. Mostly, the people we hear from are commentators on the science, which adds that extra filter through which scientific information can get skewed.