The purpose of a methods section is to get the reader to trust your data. You will need this trust because your data will, hopefully, constitute some of sort of challenge for either the currently accepted theories or the currently dominant practices in your area. That is, the reader is prepared to learn something from your analysis, and this will require a change in the way they see the world or what they propose to do about it. But don’t imagine your reader is a naively open-minded pushover. They came by their current theories and practices honestly, and they will only be won over when presented with solid results.
Your methods section is a preparatory argument for the solidity of your results. We might say that it preempts a certain kind of criticism. The more of a challenge your analysis poses for the current theory, the more skeptical your reader will be about the quality of your data. So you must explain how carefully you collected it and how scrupulously you analyzed it.
If you’ve done onsite ethnography, you must say how many hours, days, weeks, months or years you spent in the field. You must explain what sort of access you had to the people and facilities you studied. You must say which doors were open to you and which doors were closed. You must account for the choices you made when following leads. You must also explain why you chose a particular part of a particular organization to study.
If you conducted interviews you must explain how they were structured. How did you decide who to interview? Did anyone you wanted to talk to turn you down? Did anyone ask to be interviewed? Did you reject any of the interviews after doing them? How many interviews did you do? How long did they last? Did you tape them and did you transcribe them? How were they coded? Did you have anyone to help you with this? All of these questions need to have straightforward answers. Your ability to provide this information is in itself part of your credibility.
Or you may have used more quantitative data. Here the survey instruments and sampling methods you have used must be explained in a way that displays awareness of the importance of accurate measurement. Remember that you are claiming that your sample allows you to model, not just your data, but the real world that the data brings you into contact with. Your methods section must display an awareness of the standards that are operative in your discipline, and your care in collecting and storing your data.
The unifying theme of your methods section is your own competence as a researcher. Remember to present your methods to a reader who is familiar with the theory you have used to frame your study. Your methods must be considered credible by the people who see the world in the terms provided by the theory. If you are using a method that is unfamiliar in your reader’s research community, or even outright discredited there, you are giving yourself an almost impossible rhetorical task for an ordinary methods section. You might consider writing a methods paper that addresses the unfamiliarity or controversy directly. Your reader will normally understand how your method is normally applied, and will have an opinion about how useful it is in studying the phenomenon you are interested in. It is in that context that you will elaborate upon, and perhaps defend, your methodological choices.