Let Us Do Better
I watched Congress this week vote to sell off all of our browsing histories. It is a horrifying thing to observe and the Philosopher and I need to sort out what exactly we can do at least on our personal devices. There are many resource lists available and I’m particularly grateful to the Library Freedom Project and Alison Macrina for giving me clear places to start.
But over the weekend at ACRL, I was appalled to encounter several program descriptions that celebrated grabbing all of the student data that we can with the proclaimed goal of showing library impact. On the heels of the incredibly-invasive ACRL app that I wrote about before (please note, I talked several peers who said things such as WTF this thing wants my passwords and contacts — I was not alone) I was already primed for some serious frustration and this did not help.
Librarians: we must do better than this.
Some of the points that I have settled on thus far:
- I speak regularly with academic librarians who are frustrated that our institutions continue to boil researcher impact down to a single number (journal impact factor, h-index, etc) to hire, award grants, give tenure, etc. And yet we are openly praising invasive data gathering about our students in an attempt to demonstrate that their asking if we own a book or walking in to buy a coffee correlates to academic success.
- We’re grabbing this data from a position of power and forcing our students to actively opt-out. Assuming that opt-out is even possible, of course, some students will, but most will not understand that they can or that they might want to. It’s one more decision they have to make in a day full of fraught choices and “swiping my ID card to make the librarian happy” is not one I want to put on them. We are not a clothing store trying to send them coupons, but this is very similar behavior.
- Many of us are creeped out by the insidious data mining that means anything I look at on Amazon follows me around via my browser ads for the next two months. Historically librarians have fought strongly against government monitoring and freedom to read. So why are we treating our students as mindless drones who should be grateful to fork over all of their data?
- I’m at a public institution. Data we capture is FOIA-able. Our university may be able to deny the FOIA but if someone knows it exists, they can ask for it. A standing statement around here is that if we don’t want it on the cover of the Tribune...
- We’re flailing around grabbing data because it might be useful. This always seems to come with vague outcomes of “showing value” or “student success” –which any evidence-based practice medical librarian will tell you does not make a good PICO question. We have few protocols, access procedures, or policies —and sometimes we haven’t even formulated questions. In a previous project at MPOW, we tried to figure out what data everyone was capturing and what reports it fed into, primarily from a workload/duplication of effort perspective. It was an enormous project that led to realizing the giant piles of data that various library departments had. We also found quite a number of reports that were regularly run but didn’t seem to go anywhere. We’re still working on a deletion practice.
- Without clear policies and procedures and particularly access practices, we are at the mercy of the most vindictive among us. If we’re gathering granular building access data, is this available to a student employee trying to locate where their significant other is or is not? Are we absolutely sure?
- Eric Hellman talks about leaking catalog searches to Amazon and in 2015 there was a NISO meeting on patron privacy and library vendors. We know we have to argue with our vendors to try to assure good privacy procedures and data handling. If we start gathering more student data, it is not be a long leap until our vendors require it in their contracts. Recently, I was delighted when Browzine told us that they could not provide individual user-level data and that they do not capture it. If only that were true about all of our vendors.
- It is unlikely that we will get statistically significant information about our populations through many of the measures I have heard suggested. And our actions in making captured data anonymous or sharing it appropriately needs some work as well. I have seen multiple examples in our literature and presentations where indirect identifiers meant I could quickly and easily get to a single individual or very small groups–particularly of underrepresented and therefore potentially more vulnerable patrons.
Quantitative data assists us in understanding the use and happenings in an academic library. I do not contest that. Part of helping my department head understand why I’m drowning is giving her the number of consultations I’m doing; the classes I’m teaching; the research projects I am engaged with; the endless committee list; etc. Resource use counts helps us to inform collection development, as does ILL numbers and turnaways. If we’re seeing a spike or a major decline in usage of one of our research guides or at the reference desk, it’s good to have some information that may help us determine why. Attendance at workshops informs and can give us ideas as to what topics are resonating with our campuses.
But a blithe assumption that our students do not care or should not be allowed to care about their academic behavioral privacy and tossing aside our professional code of ethics is not behavior I can condone. I keep coming back to the Ian Malcolm quote from Jurassic Park “Your scientists were so preoccupied with whether or not they could, they didn’t stop to think if they should.”