Nobody likes soliciting compliments, so it can feel awkward asking someone you’ve worked with for a testimonial to corroborate the impact of your research. As a result, most researchers put off requesting this information until they absolutely have to. By this time, the person who was familiar with their work may have moved on, retired or passed away. Some organisations stop providing testimonials at busy times, when they are over-run with requests. As a result, researchers are commonly advised to request testimonials sooner rather than later. To address this, I collect testimonials via interviews (face-to-face if possible, or by phone/Skype). I start by asking if I can meet them to find out where they're at and if there's any way my colleagues and I can further help them, and then explain that I'd also like to find out if the work we did with them previously had been a help or not. Instantly I don't feel so bad about asking for our meeting, as there is a clear benefit for them, with the opportunity for me to catch up and provide further help. However, this takes time and can create additional work, so I reserve this for the key people who I think may hold the key to corroborating my impact. Here are the steps:
Get research ethics permission to do your interviews. This is not routinely done, but I like to play safe and it is low risk so you are unlikely to meet any resistance (depending on your process however, it may take some time)
Create a consent form - you can adapt the one I've been using for one of my impact case studies here
Contact key people you think may be able to corroborate your impacts, asking: i) to catch on their progress since you were last in touch and offering them an update on some of the latest research in their area; ii) to find out if and how they used research you previously conveyed to them, explaining that you would like to record this part of your conversation and if they are up for it you'll send them a consent form
I organise interviews to last an hour, with the first half for catch up and updates, and the second half for the interview (for people you are regularly in touch with, you can skip straight to the second half and organise a 30 minute interview). Consider sending a plain English summary of some of your latest research or bringing this along with you if meeting face-to-face
Before the interview segment, make sure you get informed to consent to record, explaining what will happen with the recording, giving them a counter-signed copy to keep and keeping a copy yourself. Make sure you and they are clear about whether they are providing a personal testimony or one on behalf of their organization (typically the latter as you will want to be able to quote their position in the organization to underscore the legitimacy of what they have said about your research).
During the interview segment, ask four questions: i) have there been any significant benefits arising from your research; ii) how far-reaching are those benefits; iii) to what extent can these benefits be attributed to your research versus other factors; and iv) did anything negative happen as a result of your research? If the answer to the first question is that there have been no significant benefits, then skip the next two questions and check if anything negative happened as a result of your work. This last question is important as a robust research impact evaluation should always aim to consider both positive and negative impacts, in case there is anything you can do to mitigate things that went wrong or learn for future practice. You don't need to be a social scientist or an experience interviewer to do this. You just need to understand the concepts of significance, reach and attribution, and so be able to recognise if a person has given you an answer that tells you an impact is significant, far-reaching and linked to your research. If the answer doesn't do this, then you keep probing until you are clear that there is genuinely nothing of significance to report
Get your recording transcribed, and highlight the sections that are most relevant
Email the highlighted transcript to your interviewee, asking them to write a letter (on their letterhead) based on the highlighted sections. The transcript should make it significantly quicker and easier for them to do this
You may want to get a concise, quotable summary of the key points made in the testimony, that you can quote in a case study without using up too much space. You can ask for this, or you can propose your own summary based on their testimonial and ask them to amend this so they are happy with it, for inclusion in their letter
Testimonials are normally be provided on letter-headed paper, but it is worth noting that 743 case studies in the UK’s Research Excellence Framework in 2014 (REF2014) submitted email testimonials. If necessary auditors may follow up emails with those who sent them to verify their legitimacy, so you had better hope that they still answer to this email address if you want the evidence to withstand scrutiny.
In REF2014 there were three levels of redaction you could apply for, to enable sensitive material to be reviewed, including redacting sentences from your case study prior to publication, not publishing the entire case study publically or requesting in advance that only certain reviewers be allowed to see the case study (e.g. those with Home Office clearance for research pertaining to national security). This can include the redaction of names, positions and organisations or entire quotes if necessary to protect the identity of those who provide testimonials.
In conclusion, interviews can be an efficient way for busy people to engage with the process. There are three key benefits to interviewing people for testimonials.
Interviews are a great way to probe for impacts that are difficult to articulate in words, such as changes in attitude or wellbeing, or cultural change, enabling the researcher and interviewee to explore the questions creatively to reach quotable conclusions that would not have been offered via the written word if you just requested a testimonial via email.
Interviews give the researcher the opportunity to ask open questions to explore the possibility of impacts that they had not planned for or expected. The danger of taking a highly structured approach, asking only for anticipated impacts, is that people do not report the additional benefits you are not looking for (and these additional benefits may be significant). Open questions during interviews provide the space for people to think about and explore the breadth of impacts arising from your research.
Although you should only submit impacts on letter-headed paper to REF2021 the method contains a fail-safe where you have the consent form and recording to justify quoting from the interview even if they never send the letter (I would always contact them to check if this is okay, whilst continuing to prompt for the letter).