Teaching to the test?

Among K-12 teachers, few things are more reviled than “teaching to the test.” In public schools, in which most students take a long battery of standardized tests to demonstrate learning, teachers report feeling pressure from principals to prep students for the tests, because school funding is often tied to student performance on those tests. Teachers complain that teaching to the test comes at the expense of students’ real learning.

Does the same logic hold at the college level? Standardized tests are relatively uncommon in postsecondary education, but professors may feel that the assessment movement is analogous to the testing movement in K-12 education: it is an artificial standard of accountability imposed by administrators and at odds with good classroom teaching.

This may be true of the assessment movement at its very worst, but it need not be so. Indeed, if faculty themselves design the test, then how can they complain about having to teach to it?

In her book Assessing Student Learning: A Common Sense Approach, Linda Suskie writes that “good assessment is teaching to the test” (p. 11). She continues:

Teaching to the test gets a bad name when tests measure something other than what we value, either because someone has told us what to assess or because our own tests measure relatively trivial learning. (p. 12)

This brief paragraph reflects Suskie’s “common sense” approach to assessment. There are four steps: faculty (1) set and articulate worthwhile course or program goals, (2) do all they can to help students attain those goals (i.e., teach), (3) find out how well students have actually attained them (assess, through a graded assignment or otherwise), and then (4) use the results of the assessment to make any necessary changes to help future students attain the goals more readily.

On Suskie’s approach, the question isn’t whether to teach to the test or not. It is, rather, whether our tests represent genuine opportunities for students to demonstrate that they have acquired the knowledge and skills our courses are meant to teach. It is also the question of whether our goals align with genuine and important knowledge, skills, and attitudes of the practitioners of our disciplines.

A good example of tests that merit being taught to are Advanced Placement exams. I was fortunate to have many excellent teachers in high school. Two of the very best were Mr. Lennon, whom I had for AP American history, and Mr. Ferrantini, who taught AP calculus. I have no doubt that the test was on their minds every class session. Mr. Lennon drilled us constantly in how to analyze documents and structure AP exam essays. In teaching to the test, though, he taught us essential historical and academic skills that would serve us well in college. And after Mr. Ferrantini’s demanding class, I was set up for success in Calc II and III in college.

Of course, high school AP teachers do not know what actual questions will be on the test. American college professors generally design their own tests. Without course goals that demand higher-order thinking, the entire course can devolve into the instructor saying, “Here’s what you need to know for the test…”

Thus there’s something to be said for a practice common in European universities, in which examinations are written by a faculty committee, and current-year instructors know that there will be a test, but they cannot know what exactly will be on the test. Or, even if they know the questions, they may not be the ones grading the exams. Faculty in such a situation have good motivation to prepare students to answer any reasonable question that could come their way. For this approach to work well, faculty also need to agree on goals and trust each other considerably.

That’s no problem, right?

Do you “teach to the test” in any sense? At what point does teaching to the test cross a line into illegitimacy? What merit to standardized test have at the college level?

Photo by stock.xchng user kd5ytx / Creative Commons licensed


Comments are closed.