Your protocol should have included a description of how you'll pull out - or extract - data from the screened articles. Examples of datapoints could be article titles, study IDs, date published, etc.
Please note each data point should match word for word.
Standardize it!
Make a bulleted list or create check boxes so that the answers are standardized. This will make the extraction process easier and will eliminate conflicts.
It's also recommended to test out your extraction template.
Extract a few!
Choose five or so articles to test out the extraction template individually. Then come back as a team to discuss conflicts (vagueness, needing to remove or add data points, etc).
PRISMA has a 2020 checklist which provides reporting guidelines for documenting and reporting the data extraction process and results.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
