Friday, September 21, 2007

Projects, Promises, and Bank Shots

There's a great pair of posts up over at Bardiac's dealing with frustration over repeated overpromising (and selective guru loyalty) by administration. They're worth checking out.

Via Evil HR Lady, I found a nifty post on mandatory diversity training, and why companies (and colleges) keep doing it even though it has never been shown to work.

The common thread, I think, is that much of what administration deals with has to be dealt with indirectly. When you look at an indirect measure literally, it looks asinine or even insane. (Sometimes, of course, it is.) But some of it makes sense when seen as prevention of some other disaster.

As the post on diversity training explains well, the point of diversity training isn't to sensitize employees to diversity. Anybody with any teaching experience at all can tell you that herding a hundred people into an auditorium for mandatory consciousness-raising for ninety minutes won't work. It's terrible pedagogy, and virtually designed to fail; it's also insulting. If the point of the workshops were to change attitudes and/or behavior, those would be valid objections. But that's not the point of the workshops. The point of the workshops is to be able to answer a legal complaint alleging bias with “we take these issues seriously. See, we run mandatory workshops on them for all employees!” It's about defusing potential liability.

(Admittedly, this implies a shockingly low opinion of the judicial system. But that's another post altogether.)

If deposed, a manager can say “we provide x number of hours of training.” As with credit hours, what gets measured is seat time. Changed behavior and/or attitudes are devilishly hard to quantify, but seat time is remarkably easy. If somebody alleges, say, racism, and can prove some kind of different treatment at something (which is sort of like proving that the sun rose in the East), the burden shifts to the college to show that it isn't racist. (The presumption of innocence is remarkably weak in this area of the law.) You can't prove a negative, so the college has to use proxy measures. (Quick – prove you're not thinking about a polar bear!) Seat time in diversity seminars counts as a proxy measure. If the discrimination laws were more intelligently written and enforced – say, dispense with the requirement to prove a negative -- we could dispense with these Potemkin rituals. But they aren't, so we can't. If we did, we'd lose every case, whether it had any merit or not.

Bardiac's case is a little worse. In that case, the administration promises every so often that it will tie resources to outcomes assessment. It doesn't – that's a stupid, because unkeepable, promise to make – and over time, the faculty figure out that getting caught up in this year's guru isn't worth it.

In that case, I'd say her administration is doing a lousy job of conveying the point of outcomes assessment. It's also making promises it should know it won't be able to keep, which is a great way to burn credibility. (In a tenure-based setting with an enrollment-driven budget, most resource decisions are dictated by circumstance. Discretionary money is maddeningly rare.)

The point of outcomes assessment is twofold: to help students become more successful, and to prevent the NCLB/standardization movement from taking hold in higher ed. It's sort of like the PG/PG-13/R movie ratings system – by voluntarily adopting its own rating system, the movie industry is able to argue that legislated systems (which could be far more restrictive) aren't necessary. If we can develop a reasonably passable internally-generated assessment system, the argument goes, we could avoid having to administer, say, the GRE to every senior.

I'm no more a fan of the 'guru of the month' movement than Bardiac is. A smarter administration than hers would stick to a single style over time, and refrain from making promises it can't keep. Anybody with any social science background can tell you that you don't need to assess every single student to assess the success of a curriculum. Do representative samples; keep data over time; look at the macro trends. And for goodness' sake, don't burn out your best people over this. But to the extent that we can report, with some level of truth, that we're systematically identifying areas of weakness and trying to do something about them, we can (I hope) head off boneheaded standardized testing.

One of the habits of mind I've had to learn while deaning is what pool players call the 'bank shot.' Unfortunately, many managers either don't understand the bank shot and mistake the expedient for the point, or can't communicate it in a way that faculty will understand or respect. (Worse: in the case of the diversity training, if we communicated clearly that the workshop wasn't to be taken literally, we'd defeat its usefulness as a proxy. So we play dumb.) So faculty get herded into auditoriums to hear highly-paid consultants tell them not to be racist, or less-highly-paid administrators make promises that nobody expects to be kept, and wonder why otherwise-intelligent people suddenly drop fifty IQ points when we move into administration. We don't (at least some of us); we're just playing a different game. In my perfect world, most of these measures wouldn't be necessary. But unilateral disarmament doesn't bring peace, and ignoring legal or political dangers doesn't make them go away. I don't like sensitivity training either. But I dislike even more the possible consequences of not doing it.