Literature Reviews: Select Articles To Include

Selecting articles to read from a results list is an art, not a science.  Expect that you will be repeating the steps of searching and selecting articles several times; it is unlikely you will capture all needed materials in one search.

Do not expect to find any articles that cover your exact topic.  Expect to find articles that explain or investigate one or two aspects of your topic. Your job as the reviewer is to find articles that cover all aspects of the topic and combine their findings into a cohesive literature review.

One possible method:

1. Conduct a keyword search and locate a few relevant looking articles
2. Obtain the full text and read the articles.
3. After reading these articles, consider: which of my initial questions about this topic have been answered? What new or additional questions have I developed? Can I see any patterns or gaps starting to emerge in the literature?
4. Use these new questions and patterns/gaps to help develop additional keywords to add to your search, to investigate another aspect of this topic.
5. Repeat this process until you have answered your initial questions, answered new questions developed while reading, and can identify patterns/commonalities in the work, or can identify that there are few commonalities.

Another method is to begin by appraising abstracts only, not the full text of the article or other item.

If an abstract meets the inclusion criteria and should be included, mark it as INCLUDE. If it does not meet the criteria, mark it as EXCLUDE. If you are not sure, mark it as MAYBE.

Obtain the full text of the items marked "INCLUDE"

1. Has your topic been studied a lot (over 200 articles a year on this topic)? 

2. How has it been investigated before? Which techniques or laboratory procedures? What was measured and using what scales or procedures? Can you characterize the results of the prior investigations? Are they uniform or do they vary widely?

3. When was this topic studied, and do my selected articles represent relevant time frames? Where has this topic been studied, and do my selected articles represent relevant locations?

4. If the topic I am investigating affects a lot of people-- do I have representative information about a variety of people? 

5. How does my research project fit into the existing research? Do I offer a new perspective; develop a new technique; replicate the results of a study; answer questions that other researchers developed; fill in gaps in knowledge; etc.?