I hope that this thread will remain open until Vic Smith has posted the second of the two pieces on Pardon, since I am interested in reading all the 'research' on him that is available.

I found one other reference on JSTOR, an obituary written by Jim Carroll and Pat Mackenzie. I don't think it added anything new, though it expressed their admiration in appropriate terms. It does acknowledge that some of Pardon's repertoire came via records played on a wind up gramophone, not necessarily contradicting the assertion heard all the songs he sang at family gatherings, of course. (Hillery lists the songs in the repertoire that Pardon also had on old 75 RPM records.) I won't say more about the review except to say it was Folk Music Journal Vol 7 No 2

I raised and Brian Peters (rightly, albeit with a tad of sarcasm) followed up on aspects of the use of interviews as a research technique. I'll discuss this, and try, bearing Jeri's instruction to 'edit', to be brief.

'Interviews' can be more or less formal. They range from those intended to result in an amusing piece for a popular magazine to those intended to produce a searching inspection of the point of view of a politician, as well as being a research method used in fields ranging from marketing through to social sciences such as Marketing, Psychology and Social Science. They are a less structured way of getting information than, for example, a questionnaire. But their use presents some of the same risks and pitfalls.

There are plenty of beginner's guides to the use of interview techniques on line, setting out the pros and cons of the method as a way of obtaining accurate information (leaving aside the question of how far a person's opinions and ideas may change from time to time).

One well-known phenomenon is a tendency for people doing interviews to try to be helpful to the interviewer by telling them what they want to hear. Therefore, those training interviewers tend to emphasise the need to appear neutral to avoid biasing the outcome. I think this is probably the main point I would make in connection with research carried out by amateurs seeking to investigate people like Pardon.

To point out how subtle and complex these effects can be, the best example I can come up with is a finding I read about some time ago that when IQ tests were given to Black Americans by other Black Americans the people being tested scored higher than when they were given the same IQ test by white Americans. I think it is worth giving this example, because suggesting that an interviewee may be in some sense tailoring their comments as a result of their perceptions conscious or unconscious about the interviewer is not the same as accusing them of lying.

When it comes to looking at the output from interviews, it is only natural to consider how far the approaches and ideology of the interviewer may have affected the final output, whether this is an article (which may be an edited version) or a tape recording (though these too can be edited).

This is before we even begin to think about issues of bias and selectivity in the way the findings are reported.