Paul Merchant writes, in the British Library’s History of Science blog:
Arriving to interview one scientist for his first recording session, I quickly learnt that the family dog had been put in the garden because it was felt that, as an ‘old, stuffy’ British Library historian (as they imagined me to be), I would be easily alarmed by animals. Seeing that I differed from the above in some way, they let it back in.
I wondered… Has an experiment been arranged, whereby, controlling for other variables, an interviewee is asked the same questions by different interviewers, and answers inspected for their deviation from each other?
As it happens, I am mildly alarmed by dogs.