Selecting relevant literature for your research question/problem and your inclusion/exclusion criteria is often a big job. We call it screening. It is wise to do this as efficiently as possible.

You have likely collected literature from several different databases, and you must now go through this literature. Some of the articles may appear in several databases, so you will find several identical articles when you gather the literature. We call these duplicates. There are different ways of working with these:

You can go through the results from the various databases individually
You can merge your search results and use a program to remove duplicates, for instance EndNote or Zotero.

See the next page 6.2. Archiving results and removing duplicates for tips on how to do this.

You perform screening in two steps. First, you do a rough classification (Step 1) where a lot of literature is discarded. Then, you perform a fine classification of the most relevant sources (Step 2).

Step 1: Go through titles and abstracts
Check the title and abstract against your inclusion/exclusion criteria. If you see from the title or abstract of the article that it does not fit with your criteria, you exclude it. If you are unsure, it’s better to keep the article for a later assessment of the full text. Often, most of your results will be excluded here. The references that are included in Step 1 are extracted in full text.

Step 2: Read the entire article (full text)
Does it fit with your inclusion/exclusion criteria? If you discover underway when reading the article that it should be excluded, you don’t have to finish reading it—but you do have to provide a reason for exclusion, for instance that the article is about the wrong population, wrong intervention, or wrong research design. This is the final part of the screening, and the articles that are included in Step 2 are included in your assignment.

Tips!

Ideally, there should be at least two people involved in the screening process. To maintain objectivity, it is smart to screen independently of one another (blinded). If you are two people, you can use Covidence for screening, see more in 6.3. Screening tools (Covidence and ASReview) in your screening.
If you are two people screening independently of one another, discuss disagreements when you are finished screening. If you are unable to reach a conclusion, consult a third party, for instance your supervisor.

Don’t forget to note down how many publications are reduced for each step of the screening process. You are encouraged to use a PRISMA Flow Diagram to document the screening. This is a way of describing how you have handled the results of your literature search. In the PRISMA-form, you note the number of studies identified in the above method “identified via…”. The top form on the website is most commonly used for literature assignments. If you find articles in other ways than through searching for literature, you can use the second-top form.

It is now that you choose what you will keep and what you will discard. After all, you are on the hunt for the findings that satisfy your predefined requirements.

se på fangsten_5. philipp-deus-CS1lesKltIE-unsplash.jpg

Photo by Philipp Deus on Unsplash