Tuesday, December 19, 2006

Offense and Defense

From the good-news-and-bad-news department: a student who alleged grading bias (in a humanities class) had her complaint dismissed when we discovered that all of the tests in the class were multiple-choice and graded by scantron. Since a scantron doesn't know one student from another, bias (in the sense alleged) was simply impossible.

(That's not to address larger issues of learning styles, etc.)

From a liability standpoint, this was a clean win. But I was disturbed and annoyed to hear that all of the exams were multiple-choice. The best strategy for avoiding a lawsuit was not the best strategy for actually tracking (or motivating, or provoking) student performance.

The logics behind lawsuit-avoidance and good teaching are different. Good teaching challenges students, including challenging their deeply-held beliefs, and it frequently treats different students differently. (Some students need to be encouraged to come out of their shells; others need to settle down and focus. Some respond to praise, some to challenge, and some to butt-kicking. Calibrating the right approach to each student is part of the challenge of teaching.) Lawsuit avoidance is about treating everyone identically, avoiding ambiguity and/or subjective judgment, and following set procedures. I like to think of the basics of lawsuit avoidance as the minima of good teaching – be fair to everybody, base grades on valid considerations, etc. -- but there's no denying that it's easier to defend a scantron grade than, say, an essay grade, even though an essay is often a better gauge of student performance than a multiple-choice test.

It's the difference between trying to win and trying not to lose, between offense and defense.

The stereotype of administrators is that we're always playing defense, and thereby always giving offense. There's a lot of truth to that – more than I'd care to admit, actually – but it's not like there's always another option at hand. The world at large doesn't defer to professional opinion in all things, and getting huffy and offended about it won't change that. Part of my job is to mediate between the academic aspirations of faculty and the actual legal and political climates of the outside world. That's why I press for things like very specific syllabi and actual use of due process for accused cheaters, which both strike many faculty as absurd. To them, I'm focusing on the wrong things, and I don't mind that they think so; I want them to play offense, not defense. If they indulge me a few obsessions, like including their grading policy for late work on their syllabi, then I can leave alone the real meat of what they're doing, confident that I can defend it if need be.

This issue has come up again in the context of outcomes assessment. Some departments are taking umbrage to the very idea of quantifying student performance independent of grades, making the argument that this is just one step towards expanding No Child Left Behind to higher education. My view, which isn't original to me, is that if we don't do something internally, something will be done to us externally; better that we come up with measures that actually make sense. If we can show that we're taking care of business, we have a shot at being left alone to do exactly that.

A centrally-defined test will look to the minimum. I don't want to focus on the minimum. I want to push students as far as they can go. If we can cover our bases enough that we can avoid national exams, then by all means, let's do it. The counterargument – that once the camel's nose is in the tent, all is lost – strikes me as a nonstarter. With college tuitions rising and the credential of a degree necessary even for jobs where you might not expect it, the public is less inclined to extend the benefit of the doubt than it once was. In the absence of internally-generated measures or procedures, we even have a hard time responding to ignorant wingnuts like David Horowitz, who claim that the scandal of higher education is too many Democrats in it. (In six years of deaning, I've never heard a student complain about political bias. I've heard a lot of student complaints, but never that. David Horowitz's total years of deaning are...let's see, carry the three...zero.) If we respond to popular discontent with “trust us,” we lose. We earn trust by marshalling facts, based on actual processes and records, and by actually addressing problems when we find them. We can't just assume trust anymore, if we ever could.

The danger, of course, is in conveying the message too well, and getting faculty thinking too much about defense. If they aren't challenging students, nothing I do matters anyway. They need to feel safe to push students, knowing I've got their back. We're getting there – trust has to be earned here, too – but it's a matter of fits and starts, rather than a smooth progression. To me, there's a world of difference between spelling out a grading policy in a syllabus and basing an entire grade on scantron tests. It's the difference between 'wearing a seatbelt' and 'never leaving the house.' Indulge me the seatbelt, and take the students as far as they will go.