Assessment without sunshine is difficult.

This idea that external measurement regimes are bad is interesting as it seems to me that we already have them. The GRE subject exams, for example, define a core of information that students need to know before they enter graduate school. Likewise, our department has an advisory board that is composed of employers who give us feedback on how our students are doing in the “real world” of employment. We rely on these indicators heavily to plan curriculum, so that we meet the needs of our students and those who employ them.

Assessment is also a way of being accountable for what happens in your classroom and it opens the door to create a real, integrated curriculum. It forces everyone to talk about what they expect the students to achieve in their class and how that fits into the student’s whole development through their major. This is a worthwhile goal and if we don’t find way to articulate our definition of what students should know, we will be essentially inviting others to come in and define that for us. Perhaps we deserve what we get (in terms of external review) if we are not voluntarily proactive about assessment.

I use standardized tests heavily in reviewing my own program but I have the advantage of training a group in a highly technical area where my professional organization has defined a core set of competencies based on work practice surveys. My students also take state and national licensing exams. I know I’m doing a good job with the curriculum when my students score well on their exams. But the fact that people are fighting to employ my grads helps me feel confident that I’m succeeding in the areas the exams don’t test – the kinesthetic and affective domains. I wonder if there is anything similar for the social sciences and humanities.

The Honors exams at Swarthmore are an interesting thing, but probably not a good model. They’re more a local cultural peculiarity (well, an anglophiliac idea as well, so not entirely local) rather than something that could be broadly emulated. I think you could only believe that they would be a good model for measuring outcomes in general as long as you believed in a single, clear canon.

Which come to think of it is a major concern I have about the people most enthusiastic for strong imposed accountability standards: they tend to be people who not only think there is a single clear canon in most scholarly fields, they don’t even think they have to make an argument on behalf of that canon, that those in the know are clear about what it is, and everyone else is some sort of naughty postmodern relativist.

So: no assessment, period. Sorry.

(Or just the bare minimum of jumping through hoops so the institution stays accredited.)