Systematic Reviews (2+ people)

Screening may sound a little daunting, like a medical procedure, but it's something that you're likely to need to do because it's rare to do a search for a topic and only get a few results. It's more likely, especially when you're working on longer assignments, dissertations, and theses, that you'll still have hundreds (if not thousands) of results in your final list from the database.

The databases can only base your results on what you've asked them to look for and exclude (so your keywords that have to be present; date range; population groups; language; peer reviewed etc.) so it's not unusual to have an article in your results list that does contain your key terms but the overall context of the article is not relevant to your research - so you need to 'screen' out those articles.

Whatever level of study you're engaged in, the process is the same. 

Step 1 - Look at the article title and the abstract

In those two locations you should have enough information to help you decide whether it's worth your while to read the article in full or not. The abstract should be able to confirm whether the article is based on primary research or not; what type of research was conducted and how; and what the conclusions of the research were.

The abstract is a distillation of the article's content; its purpose is to hit the high points to convince you to read it so it's a good guide as to whether it contains useful information or not. Sometimes there will be records that don't have an abstract so you will need to go through to the full text of the article in order to screen it out / in to your list of papers that you'll be discussing in your research, but most records will have an abstract either within the record itself or, sometimes in Library Plus, there'll be a link out to another database where the abstract is held.

You should be able to remove a lot of articles from your list of results in this title / abstract stage of the process.

Step 2 - Read the full-text of the articles you think are relevant

Don't be surprised when, as you read (or speed search) an article you sometimes decide it isn't relevant after all. It's okay to exclude an article at this point if the context isn't correct, but if you're using the PRISMA flow diagram and / or working as part of a review team, you will need to justify why you're excluding that / those articles after giving up your time to read through them.