During the last week, Pat has shown a 6.3 rate of response in class, with a score of 75 on the Hendrix Orientation Scale. Informal assessment yields a 82 (well within the “moderate” range) for response accuracy. In a 40-minute period Pat showed a 45 for Attentiveness Measure, but Pat’s interactions with peers score a 21 on Hemmings Interaction Index. However, that places Pat’s Optimal Utilization Index Score well under the minimum desired level of 55.
That’s one report I could fill out for Imaginary Student Pat. Or I could say:
Pat has some trouble paying attention, but Pat can still answer questions if I call on Pat. Pat’s not really distracting any of the other students, but it’s pretty clear that Pat isn’t trying and is still managing to do just enough to get by. Pat’s a good kid, just kind of bored.
In fact, the first one is composed entirely of fake data measures, but you weren’t sure, were you. It sounds more official than the second one. Because it’s data, with numbers.
The thing is, the second explanation is also packed with data. In fact, it’s the same data, but expressed in human terms rather than in numbers and technobabble. Reform-resistant teachers are often accused of being anti-data, but the problem with much of the data we’re offered by education technicians is that it is flat and meager compared to what we are used to gathering on a daily basis. To reduce and aspect of a student’s behavior, performance or existence to a single digit on some manufactured scale means I must actually flatten or simplify the data I have, throwing out plenty that is valuable.
It’s like coming up with a digital rating for a kiss. It can be done, but what I end up with will be far less descriptive, rich or thorough than a poem or a song or a description.
Data Overlords don’t like verbal packages of data because they are “messy,” but the only way to make them less messy is to throw a bunch of “extra” data out. The resulting digit score actually contains less data than the messy version― and the mess that we’ve thrown out can be hugely important. We complain a lot about what the Data Overlords decide to focus on, but what they choose to ignore is at least as large a part of the problem.
But Big Data requires digital data so that it can be crunched and spread-sheeted and used for big picture centralized measuring and planning (someday we will have to talk about Seeing Like a State, a somewhat wonky but also brilliant book about this phenomenon). And in the process, Big Data has also usurped the answer to the question, “What data are important, and which can be safely ignored?”
In fact, the shift to digital data is about a shift in audience ― schools are no longer expected to be accountable to local taxpayers and parents, but to some larger government or corporate entity. If I give Pat’s parents that digital data report, their first question will be some version of, “What does that mean in plain English.” And then I’ll give them the second explanation, which will have far more data in a far more useful context.
Do not buy the idea that teachers do not gather data. We have never done anything except gather data. What we haven’t done is gather bad flattened data selected according to the instructions of functionaries who are far away from the intersection of the rubber and the road. And just because it’s a digit, that doesn’t mean it’s good data ― in fact, it may be exactly the opposite.
Originally posted at Curmudgucation.