Managing anonymity and confidentiality in social research: the case of visual data in Community research – a research paper written by Graham Crow and Rose Willes (ESRC National Centre for Research Methods NCRM Working Paper Series 08/08) writes about the potential clashes of revelation due to the visual means of research.
I am interested in this as my Questionnaire is deemed anonymous (it asks for no signatures, printed names, and, though it asks who may have advised the responder to study a PgCert it states in the question not to give their name but their role in the University). However; I had asked, in the last question, for the responder to draw something that sums up what the PgCert means to them. Can this be construed as an identifier? Can the responder’s handwriting be construed as an identifier? My belief is that it might not – yes; sitting on the fence a bit – if the responder’s answers were to be typed up / transcribed then there would be no difference (everyone stepsons would be typed in the same font) to identify. But the image can not be homogenised in the same way.
I could collate the images in a separate location so that they have no ‘tag’ to their responders’ answers but does this take away the power and translatability of the image? Surely the context would have been diluted, the results of the questionnaire’s are brought into disrepute perhaps if the images are separated in that they become simply visual responses whose context can only be guessed at without the supporting answers.
I do not wish to identify the responders; the answers could be construed by the University as antagonistic, revealing beliefs that are not shared by line managers (or staff with more responsibilities), I need to have the trust of the responder to be able to get as honest an answer as they can provide in a safe environment (not putting them at risk of reprisals).
This paper has made me aware of the many pitfalls I could fall into with regard to making assumptions about identifiers, how we are able to glean identities from the style of a drawing (Picasso, Durer, Michaelangelo (though please don’t assume that the images drawn on the questionnaires are in any way comparable to the genius’ mentioned above!)), how locations of responders can be revealed with photographs, maps, social media data.
This is a dense subject which requires further reading…