Matthew, Caveat: these are suggestions from an anthropologist, follow at your own risk ;0) I think the answer to your question is really dependant on your focus. If the bulk of your work is on what has happened in the past, you will have to really dig into the 4000 pages as would a historian. It only amounts to about 20 books if each page averages to paper page length which I speculate don't. I would then use the questionnaire to test whether your findings seem to hold. In this case I'd use a closed answer survey type device (binary fixed-response, likert scale, etc.)If you are more interested in what they say about your question, you can be a bit easier on your archives and just look for themes and issues to put into a more open ended type of questionnaire and spend a lot of time analysing and interpreting that material. Of course, if you want to compare what they did to what they think about it (or think they do) you have to really hit both sources of data. Also, in terms of careerist motivation, collecting and at least categorising more data than you use in the dissertation is good because it gives you lots to publish from. Cameron Adams Research Associate Department of Anthropology University of Kent