Skip to Main Content

Writing your thesis and conducting a literature review

Screening, evaluating and recording

Now you’ve finalised your search strings, carried out searches of the relevant resources and identified papers to review, you now need to screen the papers for relevancy and quality. This involves several stages, starting with de-duplication through to the quality appraisal check.

One thing you will notice is that a number of your articles will be duplicates, as subject databases and indexes commonly hold a few of the same sources. This can be an arduous process to manage manually.

A way to identify duplicates is to export your references, including abstracts, into a spreadsheet such as Excel. Export your search results from the various databases/indexes you have used, as a.csv file, then import them into your spreadsheet of choice. Once exported, you can use the spreadsheet sort options to identify the duplicates and keep track of which references were found in which database(s).

Another option is to use reference management software, such as Mendeley, to carry out the de-duplication. You can sort your results by author, title, source, etc., to identify any duplicates.

Whichever option you choose, you must document the process and the number of duplicates removed. Note: if you wish to export your search results from Mendeley into Excel, you can use JabRef to do this:

  • Export your results from Mendeley as a BibTex (*.bib) file.
  • Open BibTex file in JabRef by going to File>Open Database and select the BibTex file you have saved from Mendeley.
  • To export the file from JabRef go to File>Export and save as Open Office CSV (*.csv) file.
  • Open the exported .csv file and save as an Excel file.

After removing the duplicates, the next step is to scan the titles and abstracts of the remaining papers against the inclusion/exclusion criteria you have set. During this process you will remove any papers that do not meet the inclusion criteria and any that meet the exclusion criteria, documenting the number of papers removed.

As you do not need the full text of the papers, you might find this process easier to manage using an Excel spreadsheet or a reference management software package such as Mendeley. Most indexes and databases allow you to export your results, including the abstracts, to a .csv file which can be saved as an Excel file.

The number of papers left are likely to be significantly reduced after removing the duplicates and those that did not meet the inclusion/exclusion criteria in the title and abstract screening stage. For the next step you will need to ensure that you have full text copies of all the remaining papers, including those located through your snowballing/ad-hoc searches. If they are not available through the institutional subscriptions or open access, then put through interlibrary loan requests.

Once you have the full text of all the papers, you follow the same process as you did for the title and abstract screening stage. Using your inclusion/exclusion criteria, you need to carefully read the full text of each paper before deciding whether it should be included in your final analysis. Again, you must document the number of papers removed but this time you should include the basic details of the studies excluded, as well as the reasons for their exclusion.

The final step in the screening process is the quality appraisal. You need to ensure that the studies used in your final analysis are of the highest standard. During this process you will use a checklist to assess the quality and rigour of the research carried out. You must document the number of papers removed, as well as the reasons for their exclusion.

There are several checklists available, including: