Preparing your REF2021 ICS – Section 5. Sources to corroborate the impact


In Section 5. Corroborating Sources you should list up to 10 sources that can be used to verify your impact claims that are described in Section 4. Details of Impact .

Corroborating evidence has two purposes:

    • to confirm the connection between your research and impacts claimed (eg. “Dr Harrington’s research was a key factor in our decision to redesign our training programme.”), and/or
    • to corroborate a specific impact claim which demonstrates the nature and reach and significance of a change (“implementing the recommendations from Dr Harrington’s research has lead to a 50% increase in attendance”).

Remember that your case study should present all the information the assessor needs to score the full reach and significance of your impact within the 5 pages of the template. The corroborating information will not be routinely accessed by assessors, so if there are some powerful quotes or significant figures make sure you include this information in the narrative in Section 4.

If your corroborating evidence contains commercially sensitive or classified information, please refer to the guidance on confidentiality and security clearance for impact case studies.

See below for more specific guidance around providing corroborating sources for REF2021 Impact Case Studies.

Referencing in Section 5. Corroborating Sources

Each item listed in Section 5. Corroborating Sources should be identified as E.1, E.2 etc and cross-referenced between the Section 4 narrative and the list in Section 5.

All evidence should be uploaded and available in the relevant Project Evidence Vault in VV Impact Tracker. These items should be given a title that starts with the relevant ‘E.’ reference, eg. E.1 Annual Report from ABC Foundation, E.2 NHS Patient Booklet etc.

Grouping multiple evidence items as a single source

You may group multiple items of evidence into a single source to corroborate the impact where this is an appropriate way of presenting related evidence items.

For example, you may have a number of feedback responses or several different letters that support a claim for impact. In these cases, you should group the items together in a single PDF document with a contents page at the front to identify and briefly describe each item in the grouped source. Items should be referenced and stored on VV Impact Tracker in the same way as individual sources.

Examples of corroborating sources

Corroborating sources can take a wide range of forms, for instance:

  • Weblinks that demonstrate the use of your research in public debate, in the media, in commercial enterprises
  • External report citing your research, published by collaborator, project partner or 3rd party 
  • Visitor feedback, social media responses from public engagement events and activities
  • Confidential reports or documents (there are special arrangements in place for handling these)
  • Contact details of users, beneficiaries etc who could be contacted by members of the REF Team to corroborate claims
  • Testimonials from project partners regarding the contribution of your research to the impact claimed

If you would like more specific examples, please take a look at this list of examples of corroborating sources from high scoring impact case studies submitted to REF2014 for each of the main panels.

A quick note about testimonies

All key impact claims should be independently verifiable. Sometimes a testimonial from a beneficiary can provide particularly compelling evidence. Research England have produced a short piece of guidance to share with people who are asked to provide testimonies.

There is no particular format or template for testimonials and the content will be dictated by the claims made your case study. It is worth working with your beneficiary to agree the content of the testimonial to ensure the best possible outcome.

Ideally you should arrange a conversation with your beneficiary to discuss the nature, reach and significance of the impact that your research has had for them. You may have some specific questions, but an open conversation may elicit powerful quotable phrases or additional impacts you had not considered. With the prior agreement of the other party, notes taken during these conversations can form the basis of a draft testimonial letter which can be sent to the beneficiary for approval.  This approach will both capture content for your case study and provide subsequent corroborating information.

There is also no requirement for the testimonial to be on headed paper. An email from the relevant person that shows their email address and relationship to the project (ie. work email, job title and company / employer in their signature) is sufficient.

Sources of further information

The REF Team publish their responses to queries to Frequently Asked Questions about REF2021 guidance here.

There is a helpful list of further examples of impact types and indicators in this extract from the REF2021 Panel Criteria and Working Methods document: Annex A Examples of impacts and indicators 


Return to REF2021 ICS Contents.