Doing data analysis with documents of life (i.e., documents not generated in or for research) is challenging under the best of circumstances. This type of material was created for some purpose other than to answer your research question, was made by people about whom you probably have limited information, and may make little to no explicit mention of the thing that you’re trying to learn about. There are many different ways to interpret it, and unlike human subjects, documents won’t helpfully guide you as to which aspects of what they’re telling you are most important (though in all fairness, human subjects often don’t do this, either). The vast array of possible interpretations of a documentary dataset can be overwhelming to contend with, and there’s no way to sneak past that analytic Mt. Everest. Taking a shortcut by having an overly prescriptive research question and framework before the research starts can easily make you miss what’s most important and relevant.
Meanwhile, if you’re a novice researcher, you’re also more than likely beset with impostor syndrome. This sinister malady plays mean-spirited tricks on your brain, making you doubt the viability of every insight and inference. A lot of the chat we hear and read about impostor syndrome focuses on what it does to researchers’ well-being and on aspects of the research which happen after analysis: writing, publication, dissemination, and sometimes teaching. What sometimes gets left out, though, is the havoc that it can wreak on the analysis process itself. In order to produce an analysis, you have to believe that you have standing to be the analyst. This is one of the biggest challenges of documents of life research. You can’t ask a document a follow-up question for clarification or further elaboration. What’s on the page is what you have to work with, and if what’s there doesn’t speak explicitly and incontrovertably to your research question (which it almost certainly doesn’t), then you have no choice but to insert yourself pretty unabashedly into the interpretative process and trust that you know what you’re doing.
Of course all researchers do this – no research does itself, and all sociologists are interpreters and analysts of their data. What’s helpful about many other types of data, though, is that researcher-elicited data was intended to answer research questions, for which reason it tends to be more cooperative in making itself useful for that purpose. What novice documentary researchers will sometimes (often? probably pretty much always) do is try to get documents to behave like interview transcripts. In other words, we want the data to speak for itself – to speak around us – so that we don’t have to adulterate it with our interpretations, which seem self-evidently to render our findings epistemically illegitimate. Of course we know that this isn’t actually how it works. If you ask a documentary researcher at any level whether they have to interpret their data and draw inferences through a historical lens and an analytic framework, they’ll unhesitatingly tell you that obviously they do. The intrusion of impostor syndrome in analysis, though, makes us reticent to actually do this, leaving us immobilized with self-equivocation and our research at a standstill.
So, Card-Carrying Impostor Syndrome-Infected Novice Documents of Life Researchers: you know stuff. Your interpretation doesn’t contaminate your research, and yes, it is possible to make epistemically justified claims from documents of life. You can – and in fact must – make arguments with your data. What you say in your analysis and what the documents themselves say do not have to be different articulations of exactly the same thing, and you’re not doing something wrong if they aren’t. On the contrary, once you get comfortable inhabiting your place in the interpretative process, you’ll find your data telling you all sorts of fascinating and unexpected things. So give yourself permission to be an active presence in the process, and make yourself at home.
Image credit: Clearwater Public Library System