1. Use the Conducting a Literature Review handout to help convert your research hypothesis into a searchable question. At this point, you just need to have a general idea of search terms to scope out how much is out there on your topic.
2. If the frameworks in the handout don't fit your research, here are a few other frameworks that might help you break down your question into searchable concepts.
3. Do a quick search in PubMed (or another appropriate database such as CINAHL, PsycINFO, Scopus) and Prospero to see if somebody already has a similar review published or in progress. If there is one, can you justify why your review is different enough to be needed. Adjust your question if needed.
Should this be a systematic review?
4. There are many types of literature reviews. Looking at your review type can help you figure out how to formulate your question, what you will need to search, and how to present your results. (Grant & Booth, 2009; Sutton et. al. 2019)
5. Working backwards from reporting guidelines, many found at Equator Network, can help you keep track of the things you need to consider when looking at different study types e.g.
6. For other options Systematic Review Toolbox
7. This is also a good time to start planning your data extraction tables/forms
*Tricco, Andrea C, Lillie, Erin, Zarin, Wasifa et al. (25 more authors) (2018) PRISMA Extension for Scoping Reviews (PRISMA-ScR) : Checklist and Explanation. Annals of Internal Medicine. pp. 467-473.http://eprints.whiterose.ac.uk/136633/ (just the checklist is downloadable below)
8. Age, conditions, outcomes, control, type of study, minimum numbers, published, peer-reviewed, language, dates, etc., are all possible criteria that you may use to decide which articles you want or don't want to include in your review.
9. Be clear about context, show your reasoning
10. Needs to be written out for Methods section
11. Use a combination of keywords and subject headings to develop your search strategy. This PICO Research Guide can help you break down your concepts and search effectively with Boolean logic, or use the Conducting a Literature Search handout. Translate your search separately for each database. Syntax, fields, subject headings, etc. are different for each database.
12. Document strategies used, subject headings and keywords.
13. Indicate different databases used and anything specific about that database, such as dates covered or used.
14. What about grey literature? Have you looked at any reports, books, preprints, or done citation searching? (check out Other Resources tab)
Having performed your searches, managing your results efficiently will save you a lot of time.
15. Create an account in the databases and systems you are using to search; the Library website, PubMed, EBSCO, ProQuest, and Scopus all allow you to set up a free account to save your search sets, collections of articles, and even search strategies. This will allow you to repeat a search if something happens to the results, as well as check if there are new articles as you write up and just before submission.
16. For a systematic review, be sure to put all the citations you find into your citation management database. Even if you can quickly tell an article is off topic from the title, it should not be removed until you do a proper Title/Abstract review with multiple reviewers.
17. It is also important to have the correct numbers when you add the remove duplicates step to a PRISMA flow chart.
20. Usually Title and abstract review first, then whole paper. In a systematic review, you need to check for inter-rater reliability
21. Avoid bias - don’t think about what you want to prove with your hypothesis
22. Is there enough for a systematic review? Should it be a scoping review?
This does not have to be a thorough review of every paper in a scoping review, just keep in mind quality and bias when reading papers that will be included. Even in a regular literature review for a paper, keeping in mind the questions in a CASP checklist will make sure you use good research.
23. Need good quality articles before starting data extraction. Use the tool of your choice to assess quality and/or bias of the articles you are considering.
Pull out the information you want to compare from each paper.
24. Use a checklist like STROBE to consider appropriate data elements for question (or find an appropriate reporting checklist at EQUATOR network
25. Enter data in a spreadsheet
26. Check for new articles in databases - rerun the search you've saved to make sure nothing new has turned up.
28. Try to avoid bias
29. Use the Writing Up section of the Sage Research Methods Project Planner to choose the type of paper you are writing, and learn more about how to structure your paper. Or, if you are writing in hopes of journal publication, the PLOS Writing Center has some good articles.
There are also a couple of online books that can help: