The final project for my Digital Humanities course asks students to create a data review exploring a research question of interest. Part of the source data must come from the Longitudinal Religious Congregations and Membership File, but other data sources can be drawn on for support as well. The problem for me, as it often is, is narrowing my research interests. At first, the plan was to evaluate what scholars and digital humanists even mean by ‘data’ and what counts as ‘data’, but this seemed to close to my comfort zone — more humanities than digital — and I wanted to challenge myself a bit. In the long-run, I have decided that I will tie in some commentary on data, but more to provide some ethos for myself than to be the main example of my data review.
Once I started actually researching datasets (and there are thousands accessible to UA students through the library), the need to narrow my interests only intensified. I thought I’d made progress in deciding to move away from ‘data’ generally, but it was one step forward, two steps back if you will. Checking out the Religious Congregations and Membership File sent me down a rabbit hole called Christian Science, which (to oversimplify my quick glance-over of the website) seems to be the belief that physical and mental healing can be achieved with faith and spirituality (I mean it’s just singing to be an e.g., though not for this project).
After a solid 45 minutes exploring, I found myself looking into datasets on alternative medicine (meditation, yoga, deep breathing, tai chi, special diets, relaxation, etc.) and then organ transplants, and then ER visits by patient characteristics and then deaths caused by COVID until July of this year. One rabbit hole after another and no closer to getting the actual thing done.
From all this I have gathered a few insights; my interest is in health and religion and I need to find datasets that complement these two categories. That means a few more rabbit holes will have to be investigated before I can move forward with this project. I have the meat of my argument — data does not speak for itself — but now I need to find the datasets to demonstrate that. This could be done with data sets that claim to show the same thing but show it differently (2 datasets evaluating alternative medicine but one lists yoga and the other doesn’t or a COVID death count that has different numbers because ‘death by COVID’ took some time to define). These sorts of examples would demonstrate a conflict in comparison. In other words, it would provide evidence that data doesn’t speak for itself, but the data curators do the talking.
This is all grand and fun, but how do I tie in the original dataset on church membership? That, my friends, is the question. I think the most difficult part of this project will not necessarily be narrowing my research question (keeping it broad leaves me some wiggle room), but finding data that can be connected in a meaningful way . . . In other words; how can I speak for these data sets in a way that is useful and meaningful for me in this specific moment?