Hi all,

Has anyone had any experience of contacting their research participants after writing up to check if it's ok to use a quotation from an interview? To give some background, I conducted interviews with civil servants who would only agree to participate if I later showed them a draft of my thesis so they had an opportunity to say if they felt they were being misrepresented or if they didn't want a particular quotation to be used. (There's also a methodological justification for this called respondent validation where basically the validity of your argument is increased by having the support of your research participants, but that's another issue).

I keep putting it off because I dread them all turning round and saying I can't use their stuff because it's too sensitive or paints the department in a bad light. Has anyone had experience of participants changing their mind about being involved, refusing permission to use material, or anything similar? And if so, how did you manage the situation?

I don't do this but have had a good few discussions about it. I think the thing is, if you do research without respondent validation then you have to acknowledge that what you have found in bound in time, i.e., not static. So I think what you'd need to do is take them the quotes, assure them of complete anonymity and remove details which could be used to identify them and then acknowledge in the thesis that these comments should be considered in context and may not be static...

Then start praying!!!

A

Hi,

Respondent validation does increase the quality of your researchand is a good thing to do. Nevertheless, you will find respondents tend to think they did not say what they actually said, not necessarily criticising your interpretation, but assuming the transcription is not accurate.

Depending on what you promised them that is what you will have to honour. At the end of the day, you should remove the quotes that compromise anonomity and so on, but if you remove the quotes where 'the department looks bad', well, you will be biasing your research...

I do not do respondent validantion altogether, less problems. My advice would be do what you promised, but nothing more.