A Brief Thought on Final Exams

The long-awaited Part Four of the Beethoven series is finally in the works, but in the meantime, I wanted to talk a little about final exams.

In many subjects - particularly science and math - the culminating assessment of student learning is a final exam.  Students are meant to demonstrate their understanding of the material by answering a massive collection of multiple choice and free response questions.  In principle, this all seems like a good idea: after all, if the subject is really important enough that students should take the class in the first place, we need to make sure students know the material at the end of the course.

The problem is, there's no reason to believe that final exams (or tests in general) really do assess student knowledge in a meaningful way.  What do I mean?  Well, what is learning?  I would argue that, for learning to really have occurred, students should retain the skills and knowledge from that learning well beyond the end of the course.  What good is it to memorize the various organelles in a cell if two weeks after the test you won't remember them?  Why do we test students on their ability to remember formulas that they will later be able to look up?  Why do we force students to test alone when in their real lives they'll almost always be working with others (collaboration is called "cheating" on a final exam)?

My point, however, is not to raise those questions, but this one.  How many students from your most recent class in which you had a final would pass that final if they took it again today?  Would you pass your high school Chemistry final?  Physics?  Biology?  Could you succeed on an Algebra test?  Geometry?  US History?

What values do we show when we design assessments that test knowledge that is not actually important or useful?  Why do we make students demonstrate their learning by taking tests they will not be able to pass even months - let alone years - later?  What good does that do?

Now consider the alternative: project based assessments, collaborative design challenges, even portfolios.  I know mathematics and science teachers in particular might find those foreign to their subjects, but they meet a criterion that seems important to me.  Students who can succeed at completing a Physics design challenge in high school will likely be able to complete that same challenge with the same level - if not greater - success later on in their lives.  Why is that?  Because that design challenge is testing the stuff we do care about, like collaboration, the ability to find and filter information, and, yes, physics content knowledge, but in a way that actually matters instead of in abstract and meaningless questions.

As I completed my own creative writing course this summer, I reflected on the wonder of teaching skills-based, instead of content-based courses.  Assessment can be about growth, and can be designed in such a way that students could succeed equally well twenty years from now as they did this summer.  But the "content-based" excuse for bad assessment is just that, an excuse.