Wednesday, December 19, 2012

Trust

It is precisely because you intend to disappoint your readers that you must win their trust. You do this in the methods section. Basically, a methods section tells the reader what you did to arrive at your result—especially how you gathered your data—in terms that strengthen your credibility. You must appear to be someone who understands the difficulty of knowing the particular truths you are going to claim to know. And of course someone who has done what it takes to overcome that difficulty.

At this point, I always invoke the preface to Goffman's Asylums. It won't satisfy all methodologist, of course, but it is admirable in the economy with which it assures the reader that the essays in that book provide a trustworthy picture of life in a "total institution". After reading it, one feels that the author is qualified to hold an opinion on these things, at least as such things went in 1961. He explains what kind of access he had to the institution, how long he studied it, and what sorts of prejudices might have misled him. That last factor, an awareness of the sources of error in his analysis, is very important to building trust. It's not that we expect a work to be free of bias. We simply want to know about as many sources of bias as possible. We will then read the analysis critically, with our own knowledge (and its unavoidable bias) in mind.

Tell the reader how closely you've examined the phenomenon your paper is about. How many weeks, days, hours did you spend in contact with it? How many people did you speak to or survey? What sorts of questions did you ask? What documents did you examine? What doors were closed to you? What doors were eagerly opened? The more specific you are, the more likely your reader is to trust you. (I will assume you are going to be telling the truth.) Keep the reader's natural skepticism in mind. If you interviewed only a handful of people, tell the reader why, and explain that you interviewed them very thoroughly. If they were all men, or all women, or all bosses, or all employees, explain that choice and show that you know this limits the sort of conclusions you can draw.

Anyone who has studied something carefully has the credibility they need to speak about it. The main thing is to know exactly what you have qualified yourself to say. Explain those qualifications to the reader, and then maintain their trust throughout the analysis by not making claims that demand greater credibility than you've established.

No comments: