Types of Articles
Also called primary sources or original research articles. The works, ideas, or observations presented in these articles are those of the authors, not a summary of others' works.
Empirical articles are easy to identify by looking for a Method section that includes details needed to replicate the study (descriptions of the participants, the measures used in the study, experimental procedures, etc.).
Empirical reports by government-type agencies
Many government and non-governmental agencies collect data and distribute summary reports. For example, every other year, the Centers for Disease Control and Prevention distribute the Youth Risk Behavior Survey to high school students in the United States. They publish a summary of the data, including statistical analysis of the responses and evaluation of trends over time. The Rand Corporation and The Pew Research Center are both think tanks that conduct their own data collection and analysis and provide reports on their findings. Some of these reports will clearly look like research reports with tables of contents, others (example) will look more like basic websites. However, if you look closer, you will see that they provide a clear description of their methods, which you wouldn't see at an unreliable website. These will typically include a recommended citation somewhere (example) and may even include a doi (in this example see the right-hand side under "Document Details").
Non-Empirical Research Articles
Aka. secondary research articles. NOTE: These should still be found in peer-reviewed journals.
Review articles summarize all of the existing literature on a subject. These are a great starting point for research. Review articles can be used to figuring out what has and has not been done and where there is consensus or disagreement. They are also useful if you have a research question but don't know what methods are typically used to test it.
For topics that have had a lot of research conducted, review articles often pick up wherever the most recent review on the topic left off. When this is the case, review article authors will typically refer readers to the previous review articles.
Theoretical articles introduce a new theory. Pure theoretical articles are pretty rare in modern psychology
In addition to being very informative in and of themselves, meta-analyses can be great resources for finding empirical articles on a topic.
Meta-analyses statistically summarize all of the existing literature on a particular subject. In order to compare apples to apples, they often have a pretty limited scope. For instance, a meta-analysis of the effects of sleep on memory may only include experimental studies (where participants were assigned a certain number of hours to sleep). They would also need to only include studies that used the same operational definition for memory because recall memory for a list of words may be affected by sleep in a different way than recognition memory for a series of pictures. Including both kinds of studies could make it look like there's no significant effect, even if there actually is for word recall (but not picture recognition).
A meta-analysis is similar to a review article, but instead of just talking about what prior studies found, they actually combine all of the statistics to find out if, overall, there is a statistically significant effect. Because both styles are useful and are very similar in terms of the work involved in writing the paper, many meta-analyses are published as "A Review and Meta-Analysis" which combines both.
Unlike other secondary sources, meta-analyses will include a Method section. However, instead of describing the participants in the study, the authors will describe the process by which they selected, combined, and analyzed the existing literature. They often include the specific search terms and the different search engines they used in order to get a comprehensive list of all of the studies on the topic. From there, they will describe how they selected which articles would be part of the meta-analysis and which ones were dropped (due to scope, operational definitions, questionable methods, sample size/characteristics, etc.). It will also include a Results section with the statistical analyses that were conducted.
When evaluating the quality of a book, look at the references. Does it have the number of references you would expect to see from a full-length book? If not, it's probably not a reliable source. Most academic books will have references at the end of each chapter (e.g. handbooks) or in a bibliography at the end of the book that is several pages long (other academic books, textbooks). Lay-oriented books might have "suggested reading" or even a short Works Cited section, but a high-quality academic book's reference section will probably be at least 15 pages long.
Several subdisciplines in psychology regularly publish handbooks (e.g. The Handbook of Child Psychology and Developmental Science, The APA Handbook of Sport and Exercise Psychology, and The Wiley Handbook of Human Computer Interaction). The chapters in these handbooks are very similar to review articles in design and scope. These chapters have been written by top researchers on the given topic, and they have been reviewed by one or more editors who is also a leader in the broad field the handbook covers. Handbooks are on-par with review articles in reliability, but they are often more outdated, even if they were published the same year. Although the process of writing, reviewing, revising, and publishing a review paper can take quite a while, it is still faster than most book publishing. Handbooks also publish new editions much less frequently than new review articles come out. They're a great resource, especially for background information on symptoms, prevalence, risk factors, etc., but it's important to make sure you're also including more recent publications.
Lay in this respect means a non-scientist. Lay-oriented books are the ones you would find in a typical bookstore's non-fiction section. Some of them will include references and will be good sources of information, but they haven't undergone any kind of peer-review process, so they are likely not appropriate for inclusion in your paper. As the label suggests, the target audience is a lay person, so the focus is on making the content easy to understand rather than on 100% accuracy. Some authors (e.g. Robert Sapolsky, Franz de Waal, Oliver Sacks) are good at doing both, but most lay-oriented books are too inaccurate to be reliable sources for a research paper. You also want to take a critical eye to the biography of the author (beyond just what is on the book cover). If you have an "M.D." or "Ph.D." after your name, you can often get away with publishing books way outside of your area of expertise. Reliable authors will also be publishing peer-reviewed research in the same topic area. When in doubt, ask your professor! We're happy to look at the source and help you decide whether or not it's appropriate.
Textbooks are great resources for the very earliest stages of research, but that's as far as they should go. You should NEVER cite a textbook in a research paper. If the textbook has information that you want to include, you should look at the in-text citation in the textbook, look up the original source in the References section at the end of the textbook, and cite that original source.
These sources are not peer-reviewed or scholarly sources. They may be published by legitimate organizations, including universities, but they have not undergone the peer-review process that ensures the reliability of the research. Some of these sources will provide links to the original articles that were published in scholarly journals, and others will list the original article title, authors, universities, journal title, etc. that you can use to find the original article.
Some of these sources will include a brief bibliography at the end of the page/article. You can use this to access more reliable sources, but this does not mean that the page/article itself is an acceptable source.
Especially if you use Google Scholar, you are likely to come across at least one Honor's thesis, Master's thesis, or Doctoral dissertation. These can be really useful in terms of finding other sources because the introductions to these are typically very thorough (especially Master's theses and dissertations). Use their Introductions, Discussions, and Reference sections to find other primary sources. But DO NOT use the thesis or dissertation as a source. Even though the student likely spent years working on it, and their advisors and committee members read it, it has not undergone peer-review and therefore is not a reliable source. If you really want to cite it, take a look at the other stuff that has been published by the student and/or their advisors (committee members' names are usually listed in the first few pages of the document somewhere). Usually they will have at least one peer-reviewed publication on the topic.
This includes television/radio news sources such as CNN, Fox News, and NPR and newspapers such as the Washington Post and the New York Times. These sites will often list the names of the article and/or the researchers when they report on new research. Rather than using the news source, you should search for and cite the original scholarly research article. NYT Example Original Research Article
Universities often put out press releases for research that has been published by researchers at their institution. These are often written by the university's public relations department (Example). Even if they are written by the researchers (Example), you should still look for a peer-reviewed article about the research. If you're on the university/institute webpage, you can check the researcher's webpage or CV to see if they have recently published a paper on the topic.
Many organizations publish news and information on their websites. Although the Mayo Clinic is a highly-respected hospital, their website is not a reliable source (example). The American Psychological Association often publishes news articles on its site (example), but even if doctors or researchers write the articles, they have not been vetted for accuracy.
Popular press articles
Sites like Psychology Today, National Geographic, and Scientific American publish magazines, and although the articles are often written by researchers in the field, they are not peer-reviewed. The American Psychological Association publishes a popular press magazine called Monitor on Psychology. Even though it is published by a well-respected psychology organization, this magazine is not peer-reviewed. For peer-reviewed articles published by the APA, you should look on their Journals page
Some conferences publish the abstracts for the talks and posters presented at the conference. Although these appear in scholarly journals, they have not gone through the full peer-review process that other articles have undergone. You can identify conference proceedings because they will be very short (usually just the title and/or abstract) and won't have a full article associated with them. Example