TEST PREP: Junk Food Instruction

Steve Peha

Testing is the most misunderstood element of education reform.

Ignoring decades of research that shows that testing actually helps kids learn, we assume it doesn’t. Ignoring more recent research, we think that non-stop test preparation improves kids’ test performance. While we may see small gains from teaching to the test, those gains fade quickly because what we’re really teaching kids is test familiarity instead of familiarity with the knowledge and skills test problems are designed to measure.

To paraphrase one of the most knowledgeable educators I know: “Test prep is like junk food: a quick boost of energy in the short run; malnutrition in the long run.”

But if we looked at the research—and here’s an excellent meta study—we’d see that it’s not testing that is the problem, it’s the way we react to it. Specifically, high-stakes testing can have adverse effects; but low- or no-stakes testing (like non-graded quizzes, for example) is very good because it stimulates active recall which improves long term memory. Even better than that, is self-testing where kids take tests they make for themselves.

So let’s not throw our babies out with the bathwater or drown ourselves in the tub with a deluge of misconceptions.

In hundreds of schools where I’ve worked, and hundreds more that I know of, near non-stop test prep dominates the day. In all my years, I have never heard a single teacher or principal or district office leader say that this was what they wanted to do, or thought was good to do, or even felt good about doing it. I think we all know it’s not smart, or that at least there are smarter approaches we could choose.

A new article backs this up strongly. This is but one of many articles I have read over the years illustrating the problems of teaching to the test, especially with regard to explicit preparation for specific question types. The author even adds a nice bit about the small things we can do that are helpful when we know we have to take tests, like teaching kids how to read test questions more carefully.

Here, in Kim Marshall’s Marshall Memo, is an excerpt from his summary of the article written for The Reading Teacher by Timothy Shanahan:

“The idea of having students practice answering test questions is ubiquitous and ineffective in raising test scores,” writes Timothy Shanahan (University of Illinois/ Chicago) in this article in The Reading Teacher. He understands the pressure to raise scores on the new generation of more-challenging ELA tests coming down the pike—PARCC, Smarter Balanced, and others. But the time-honored approach of analyzing sample test items and having students answer questions on main idea, supporting detail, providing evidence, describing a character, identifying a theme, and drawing conclusions doesn’t work, he says. “It has never worked. And it won’t work any better with the new assessments on the horizon. It’s as ineffective as pushing the elevator button multiple times to hurry it along or turning the thermostat to 90º to make a room warm up faster.”

Our own assumptions aren’t far off—we put our faith in excessive elevator button pushing and extreme thermostat raising—but not because of testing. We react poorly to testing by persistently pursuing four things that research soundly discourages:(1) Teaching to the test for extensive periods of time; (2) Significant amounts of explicit direct test preparation with sample problems; (3) Poor or no use of the data we receive from testing; and (4) The projection of adult stress onto kids as a way of pressuring them into taking testing more seriously.

Extensive test prep is not required by any federal or state law that I’m aware of. Nor have I ever seen it as a required district policy. Test preparation, as an activity unto itself, isn’t a part of education reform at all—except when we choose to make it that way. However, in many cases, teachers may feel they have no choice. Their principals may demand that they teach to the test. But principals don’t seem to like this either, so this demand is probably coming from farther up the organization.

Again, however, I’ve never met a superintendent, assistant superintendent, curriculum director, or school board member who told me they supported this approach. Sometimes I wonder where the idea came from at all. And when I do, I’m forced to conclude that we, collectively, made it up to satisfy some need we have that no one seems to be able to explain to me.

But if we can choose it, we can change it.

We’re choosing to do excessive and ineffective test preparation even though we know it’s not the best that we can do for our kids. Yet what do we say to them most often? “Just do the best you can.”