There are a couple of ways you can tell if a journal is peer-reviewed:
If it's online, go to the journal homepage and check under About This Journal. Often the brief description of the journal will note that it is peer-reviewed or refereed or will list the Editors or Editorial Board.
Go to the database Ulrich's and do a Title Keyword search for the journal. If it is peer-reviewed or refereed, the title will have a little umpire shirt symbol by it.
BE CAREFUL! A journal can be refereed/peer-reviewed and still have non-peer reviewed articles in it. Generally, if the article is an editorial, brief news item or short communication, it's not been through the full peer-review process. Databases like Web of Knowledge will let you restrict your search only to articles (and not editorials, conference proceedings, etc).
They both come out once every month. They're both in English. Both published in the United States. Both of them are "factual".They both have pictures. They even cover some of the same topics.
The difference is that one--Journal of Geography--is peer-reviewed, whereas National Geographic is a popular-press title.
Peer review is scientists' and other scholars' best effort to publish accurate information. Each article has been submitted by a researcher, and then reviewed by other scholars in the same field to ensure that it is sound science. What they are looking for is that:
The methodology has been fully described (and the study can thus be replicated by another researcher)
There are no obvious errors of calculation or statistical analysis
Crucially: The findings support the conclusions. That is, do the results of the research support what the researcher has said about them? The classic problematic example is a scientist claiming that hair growth causes time to pass: The correlation is clearly not causation.
It isn't a perfect system: Scientists make errors (or commit fraud) as often as any other human being and sometimes bad articles slip through. But in general, peer-review ensures that many trained eyes have seen an article before it appears in print.
Peer-reviewed journals are generally considered "primary source" material: When a new scientific discovery is made, a peer-reviewed journal is often--but not always--the first place it appears.
Popular and trade publications are not peer-reviewed, they are simply edited. That does not mean they are any less potentially truthful or informative--most popular and trade publications take pride in careful fact-checking.* But when the topic is scientific research, the information is generally "secondary": It has already appeared elsewhere (usually in a peer-reviewed journal) and has now been "digested" for a broader audience.
Peer-reviewed journals will always identify themselves as such. If you want to verify that a journal is peer-reviewed, check Ulrich's Periodical Directory (Reference Z6941 U75 vols.1-4 2014).
Google Scholar takes Google's PageRank algorithm and runs it on a pre-selected set of tables-of-contents and metadata from a preselected set of scholarly journals and papers. These are largely--though not entirely--peer-reviewed. It is a much better option than a regular Google search for scholarly information.
Thomson Reuters Web of Science (formerly ISI Web of Knowledge) is today's premier research platform, helping you quickly find, analyze, and share information in the sciences, social sciences, arts, and humanities. You get integrated access to high-quality literature through a unified platform that links a wide variety of content and search terms together, creating one common vocabulary and one seamless search.