When this article came out, I read it and thought, "This is an excellent article. Everyone should read it, preferably first in a qualitative research course, then every time they write a research article." Then I put it on the shelf.
Just under two years later, I received comments on a qualitative research article I had submitted to a journal. And among the editor's comments was an exhortation to revise and deepen my methods section along the lines of the Smagorinsky article. How embarrassing - but the revision turned out to be extremely rewarding as I applied the comments to the draft article, producing a much more solid, responsible, and well-argued piece.
So what is the article - Smagorinsky's article - about? Smagorinsky, who coedited RTE for seven years, notes something about the research reports he has had to evaluate as an editor and as a reviewer. Specifically, their method sections are vague: they "often lack sufficient detail to make any results that follow from the analytic method trustworthy" (p.389). And that's a disaster, since "If I don't know pretty clearly how the researcher is conducting the study, then it doesn't matter much to me what the results are because I have no idea how they were produced" (p.393). Smagorinsky likens the situation to
reading of a wonderful dish and being told how to prepare it as follows:
First, select all ingredients that could conceivably go in the dish. Review them carefully, then pick the ones you want to use and put the rest back in the pantry, perhaps saving them for another meal that you will prepare later. Then reconsider the ingredients you’ve selected and decide which are most important. Do this again just to make sure. Then mix the important ones together and give it a taste, adding other ingredients as necessary. Put them in cookware, heat, and serve. (p.393)
The first time I read this passage, I laughed out loud. The second time, I smiled sheepishly, because I knew that I had done exactly what Smagorinsky had complained about. It's not that readers have to be able to replicate the study in terms of "conducting an identical study with identical results" - but they should have a decent idea of what the researcher did (p.394). The researcher should be able to show her work.
Based on this principle, Smagorinsky describes the basics: data collection, data reduction, data analysis, data coding, the context of the investigation. In all of these, Smagorinsky helps us to better understand how to develop arguments. He's not trying to reduce methodological pluralism here, but he is trying to help us develop a strong argument by connecting methods to evidence in principled ways. Most importantly, perhaps, is the issue of how the methods section can serve as the conceptual epicenter - the place where the articles major sections become aligned in terms of theory, methods, results, and analysis (p.405). His critical advice: write the methods section first.
Reading the article the first time was valuable. Applying it during the second read? Fantastic. See what you think of it: My revised article will be published in Written Communication in October.
This is a great article. You know where I've found it invaluable? In preparing IRB protocols.
ReplyDeleteWhen you can clearly articulate your qualitative research methods, you can clearly articulate to an outside audience (e.g., an IRB review board or the subjects of your study) what you'll collect, how you'll collect it, how you'll treat the data once it's collected, what it means for ongoing analysis, how you'll code it, who the audience is for dissemination of results, etc.
Good stuff.