In the journal databases you have the option to tick a box to limit your search results to peer reviewed articles.
Peer review is an editorial process. It is important to readers because it checks the quality and credibility of an author’s work before it is published. It is meant to guard against things like irrelevant conclusions, unjustifiable arguments and biased opinions. Ultimately it is a way of determining whether or not a piece of research fits with or adds to the body of knowledge on a subject.
Traditionally when a journal editor receives an article submission they send it to be peer reviewed by two or three experts on the same subject. For the sake of partiality the author and the reviewers tend to be kept anonymous from one another. The reviewed copies are returned with suggestions for changes but articles can be rejected outright.
It is important to remember that just because an article is peer reviewed it does not mean that its findings are correct. You should always approach all content analytically. If you find fault with an article or fail, for example, to replicate its findings this does not necessarily mean that peer review has failed. This is what happened in the famous case of the Reinhart and Rogoff paper ‘Growth in a Time of Debt’, which hit international headlines in 2013 when a student discovered errors in their work.
Reinhart and Rogoff’s paper is of additional interest as an example because it had not been peer reviewed. Articles that have not been peer reviewed are not entirely without merit nor are they automatically untrustworthy. However, non-peer reviewed articles have not been subjected to the same pre-publication standards as peer reviewed articles so they ought to be referred to with more care. Citing non-peer reviewed articles demands a greater level of judgment and in some cases is not recommended.