Historically, information sciences as a field has done a decent job at keeping up with the times. The advent of the internet in libraries meant for example that librarians needed to be trained on how to use the internet, because not doing so would be neglecting the use of an increasingly pervasive, useful resource. Librarianship as a profession at national laboratories, universities, and government institutions could not afford to miss the train, and so it went as technologies became increasingly complex and prevalent in our daily lives, and in the research sphere. Librarians moved on from cataloging work and working the circulation desk and into positions that required computing skills, programming abilities, and data analysis. These sorts of skills are so far removed from what most people might conceive of as the natural responsibilities of a librarian that it may raise the question of what it really even means to be a librarian in the 21st century if practically none of the work is related to what librarians have traditionally done, but it highlights the importance of recognizing that the information professions as a whole have a lot to contend with in terms of preparing practitioners for a constantly evolving milieu.

In my second year of graduate school, I held a graduate assistantship at the Oak Ridge National Laboratory in Oak Ridge, Tennessee. I was the very first intern of a data librarian, which itself was a relatively newly established position in the Research Library. My mentor and I were given a fair bit of leeway by the supervising group leader to basically make the role whatever we wanted it to be. There were some traditional library responsibilities, which were pretty slow moving and probably the least memorable part of the entirety of that experience; I would say that instead the most memorable part was probably the chance I was given to design and deliver a series of workshops to the Research Library staff in data cleaning, analysis, and statistical methods. Because I had never taught anything before, I found the process challenging from beginning to end — designing teaching materials, exercises, the presentations, and so on — but I found that it was also an intensely rewarding experience, because it revealed to me a few important realities. First is that statistical literacy is not something that should be expected from graduates of MLIS/MSIS programs. Retrospectively, one of my biggest mistakes in designing those workshops was that I assumed a degree of foundational knowledge that simply did not exist for the majority of those I was instructing. It was not that the material itself was simply unmanageably difficult, but rather that the course was designed at a level that assumed a depth of knowledge that should not have been assumed in the first place. Certainly everyone there had the ability to understand everything that was communicated, it was only that the limited time that we had for each session coupled with the busy schedules of all involved coupled with the level of the material did not bode well for an equally beneficial experience.

I walked away from my experience convinced that something definitely needed to change vis a vis the curricula of graduate information science programs. At the University of Tennessee, where I obtained both my bachelor’s and master’s degrees, I was able to take a class on research methods which provided a review of some basic statistics, and I did take a data visualization course, but I knew that much more could probably be done in the way of course design for students whose goals are to work more intensively with data and statistical methods. Why? Because not everyone who enters an information sciences graduate program may be looking to work in traditional librarianship — see those from our own MSIS program at UTK, who go into UX design, data engineering, and beyond — and because people enter graduate programs in information sciences with varied backgrounds. Some may be well suited to the challenge of more involved quantitative-leaning coursework, and then there are those who may not come from a quantitative or STEM background and may still be well-served by the opportunity to broaden their skillset in a job market that commands a certain degree of familiarity with data analysis and literacy.

I have given serious thought to what such courses should and probably would look like, given the faculty makeup of information sciences departments across universities. Certainly, none of this is a question of the ability of faculty. My own thesis committee was composed of professors who are no strangers to scientometrics and quantitative science studies, and in general a sufficient number of information sciences faculty ought to have the ability to deliver decent courses in the field of data analysis, statistical methods, and data literacy. In my mind, this is simply a question of whether or not we choose to pay attention to the changing landscape of the information professions, not whether or not we can muster enough instructors to deliver this sort of material in the first place. It seems undeniable that the market rewards those who are able to manipulate — not be manipulated by — data, to make sense of it and make it useful to companies and institutions, and it would serve the field well to adapt accordingly.

The increasing prevalence of AI tools has signaled something troubling to me, which is the delegation of thinking and reasoning to tools by people who do not really understand those tools themselves. This is not a new problem; it is the same failure that shows up when a practitioner accepts a bibliometric ranking at face value, or interprets a p-value incorrectly, or draws causal conclusions from correlational data. In each case, the underlying issue is the same: a tool is trusted without being understood. Statistical literacy is not just about knowing how to run a regression or read a confidence interval. It is about developing the habit of interrogating outputs rather than simply accepting them, and that habit is exactly what the information professions need more of right now. The field has always adapted to new technologies. The question is whether we adapt critically or passively, and that distinction matters more than ever.