Advice for New Teachers: You Say "Teaching to the Test" Like It's a Bad Thing

In the realm of higher education--or whatever it is I do--SLO's are the new black.  "Student Learning Outcomes" have been all the rage for several years.  In a nutshell, SLO's state what students should be able to do upon completion of a course.  In a math class, for example, an SLO might state that, upon completion of the course, students will be able to solve problems using the Pythagorean theorem.  Instructors then collect exam data to see how many students correctly answer questions involving said theorem.  In theory SLO's thus provide ways for teachers and other interested parties to measure the effectiveness of instruction.

As you can imagine, many teachers hate them.

At a meeting last week, our dean presented responses to a survey wherein she had asked faculty how they used SLO data.  Several respondents indicated that, not to put too fine a point on it, they didn't.  Use the data, that is.  Among those who did indicate that they used the data, though, many suggested the data helped them focus their instruction better so as to improve SLO results.  "Of course," some teachers at the meeting tsked, "teaching to the test."

Well, I should hope so!

Among teachers, the knee-jerk reaction to any suggestion of TTTT will often consist of shocked indignation.  Tests are often seen as a necessary evil, at best--a bureaucratic requirement that primarily serves as an obstacle to the real mission of teaching, which only philistines fail to recognize is an art of the highest order--and anyone who would betray his or her art to kowtow to bean-counting quantifiers of the ultimately unquantifiable should immediately turn in his chalk and elbow patches and pursue a more suitable career like certified public accountantcy.  I, too, used to shake my head at the idea that I would ever sell myself or my students so short by teaching to the test.  And then I thought about it.

Next week, my students will take their first significant exam of the semester: They will write a short illustration paragraph (a paragraph in which the writer states a main point and then supports it with facts and examples).  They will be graded on how well they fulfill the requirements of such a paragraph: Do they have a topic sentence? Do they provide sufficient evidence to support their points? Is the paragraph well-organized, etc.  Over the years, I have developed certain strategies to help students do well on the exam; I have devised specific instructions on how to construct the desired paragraph.  And I have suffered serious anxiety about whether I was becoming a TTTT teacher.

But then I realized something: In this class, one of our goals--our desired outcomes--is that students be able to write illustration paragraphs.  The test allows us to see whether students have, in fact, mastered this skill.  So, by teaching to the test, I merely teach to the outcome that we ultimately desire.

It seems to me, then, that if teachers are concerned about teaching to the test, the problem lies not in the instructional philosophy, but in the design of the tests.  If you design tests that actually require the students to demonstrate the knowledge that you set out to teach them, then you should feel no shame about teaching to the test.  Indeed, you should do nothing else.

**************************
Publication note: Keen-eyed (not to say obsessive) Sloppists will note a change in the title of this occasional series, which was previously called, "Advice to Young Teachers."  Certain readers (well, DOS) took exception to the title.  Primarily, I suspect, this was due to the thought that, if I were presenting myself as a some sort of "elder statesman," that would make certain other people (well, DOS) practically antediluvian.  Hence, articles in this occasional series will from now on be titled, "Advice to NEW Teachers."  We hope you are satisfied!