Skip to main content
CABI Digital Library logo

Data extraction

Data extraction

Copyright ©2025 CABI

Abstract image of dots in a series of lines

Data extraction in evidence synthesis involves systematically collecting relevant information from included studies using a standardized template or form.

At least two reviewers generally perform this process independently to ensure accuracy and minimize errors. The extracted data is recorded in tables or spreadsheets, allowing for easy comparison and synthesis across studies.

Data extraction is a critical step, as it provides the foundation for analysis, synthesis, critical appraisal and ultimately answering the research question.

Level descriptors

Expand all

Understands the importance of accurate data extraction and piloting to increase confidence in data extraction.

Understands that there are different types of data (e.g. qualitative, quantitative and mixed methods) requiring different approaches to data extraction and analysis.

Follows instructions to extract data from eligible articles using pre-defined data extraction tools.

Aware of different tools to record extracted data and has used one or more of these.

Confidently extracts data from eligible articles, requesting help from team members where needed.

Adapts existing data extraction templates, or creates new templates, to enable extraction of data as defined in protocols.

Contributes towards protocol definition regarding data extraction.

Manages the data extraction process, coordinating the team and checking for a suitable level of agreement in consensus checking. Arbitrates disagreements and seeks advice from a more senior topic expert where needed.

Design and/or delivers training on data extraction and analysis within evidence synthesis projects.

Leads on drafting data extraction section of the protocol.

Designs ways to improve and standardize data handling across your organization and wider afield.

Keeps up to date with, and evaluates emerging methods for data extraction including machine learning and AI for data extraction.

Leads and promotes the development of data extraction methods on a national and international basis.

Feedback

Contribute your ideas to the skills framework to help it grow and improve based on your experience.

Get involved