Skip to main content

How to Set Up Data Extraction in Rayyan

  • Updated

Before you begin Make sure screening conflicts are resolved before adding articles to Data Extraction — this avoids having to add articles again later. Create your extraction questions before your team starts extracting. See How to Create and Organize Your Extraction Questions.

Data Extraction is the stage where you capture structured data from your included articles — study design, population, intervention, outcomes, and any other variables relevant to your review. Each article is presented alongside your extraction form, and collaborators answer each question you have defined.

Access Data Extraction by clicking the Data extraction tab in the top navigation of your review.

How to add articles to Data Extraction

  1. Go to the Data extraction tab.
  2. Click Add articles in the toolbar.
  3. Choose your source — Included, Maybe, or Labelled — from either the Screening or Full-text screening stage.
  4. If you choose Labelled, select which label to use. Rayyan shows how many articles carry each label.
  5. Confirm the selection. Articles are added to the extraction table.
Best practice Add articles after all screening conflicts are resolved. If you include new references in Screening after Data Extraction has started, click Add articles again — Rayyan adds only the newly eligible ones and skips articles already in the stage.

How to extract data

Click any article in the extraction table to open it. The extraction form shows all your questions organized into groups — answer each one using the response type defined. If a PDF is attached, it opens in a side-by-side view so you can read the full text while answering questions without switching between tabs.

Note In collaborative reviews on Individual plans, each Collaborator requires a paid seat to access the Data Extraction stage.

Tracking progress

Use the Filter by status dropdown in the extraction table to filter articles by progress. The page header updates to reflect the active filter — for example, Showing 2 / 3 Not Started Articles.

  • All Data — all articles added to this stage
  • Not Started — no extraction has begun
  • Uncompleted — some answers filled but not all required fields
  • Completed — all required fields answered. Expand to filter by Completed by 1, Completed by 2, etc. to see articles finished by a specific number of reviewers.
Note Rayyan's Auto Extraction feature — part of ResearchPilot™ — can automatically populate extraction answers from full-text PDFs. Available on institutional plans with ResearchPilot™ enabled. See Auto Extraction (ResearchPilot™).

Continue to: How to Create and Organize Your Extraction Questions

Related articles

Comments

0 comments

Article is closed for comments.