Testing | Viewpoint

We Don’t Care About the Tests; We Care About Life Outcomes

• By Trent Batson
• 08/04/10

In an article in the New York Times by David Leonhardt on July 27, 2010, we learn that longitudinal research over thirty years revealed the long-term benefit of a good kindergarten experience--better pay over the long term, better personal health, a more successful life. In response to this surprising and hopeful result, one of the researchers, Raj Chetty, observed that testing in school had missed this phenomenon and is reported to have made the statement, “We don’t really care about test scores. We care about adult outcomes.” From this, I derived the title of this article, words that perhaps should be inscribed on ivory tower porticos across the land. And all of this implies, essentially, that tests are irrelevant because they show us so little or perhaps nothing about actual life outcomes. If tests, as they are administered now, miss the kindergarten story, what else do they miss?

To find out more about testing and how we might improve our practices, I visited a parallel universe. How I got there will remain a secret.

I emerged into this other universe right on the steps of a courthouse. I heard voices inside, so I pulled open the front door, went down the hall and into what appeared to be a courtroom, but instead of seeing a high bench and a judge in robes, I saw desks set in rows and people writing at the desks. Police officers were scattered about the room.

This seemed so incongruous to me, I asked one of the officers what was going on. “You’ll have to ask that guy over there.” He pointed to a middle-aged man who stood at the side of the room watching the people writing at their desks. The man glanced at his watch. I walked over and told the man that the officer had sent me to ask about what was going on. He responded, “Wait a few minutes and the prisoners will be finished.”

“Prisoners?” I asked. He looked at me quizzically. “Of course,” he answered.

Once we were out of the room and walking down the hall, I asked him what was going on in the room. “Well, tests, of course, what else?”

“You are testing prisoners?” At that, he looked at me and said, “Yeeeaaaah, what universe are you from?”

“Ok, you test prisoners. Uh, I know I’m being dense here, but why do you test prisoners?”

“You don’t know? Are you playing dumb?” I still looked hopefully for an answer. “Ok, you need me to say it, I guess. As you must know, the test is to determine who is guilty and who is not. You know--if they pass the test, they’re not guilty.”

I was speechless.

“But what about evidence?” He looked at me for a second. “Evidence?” he asked, looking puzzled. “Yeah, you know, where the person was when the crime was committed, fingerprints, testimony, DNA, previous record, and all that.”

“We don’t do any of that stuff: The tests are what determine innocence and guilt. We’ve been doing it this way for hundreds of years. We compare suspects with each other, and those who do the best test are not guilty. The others we send to jail; the worse they do on the test, the longer their term in jail.”

“But, what if you know that one of the suspects was seen at the crime scene?”

“We round him up just like the others.”

“But you don’t think that guy was more likely to be guilty?”

“Evidence is too hard to interpret. It takes too much time and we don’t know that it’s reliable. So, we ignore it. Besides the tests keep people honest: “The fear of being sent to jail keeps society in line.”

Back in my home universe, I thought hard about the tests I had given in my own classes. I thought how I had always felt uneasy about the practice.

End of bad drama. It is in fact time to question why we test, because now we don’t have to rely so much, or perhaps at all, on testing. Students can in fact bring forth evidence of their work that shows much more about their abilities and achievements, their growth, and their discoveries than any test.

And it is time to ask, also, what outcomes are we looking for from education? Shouldn’t they in fact be life outcomes? Why are our “outcomes” now only about how well a student does in courses? Does that predict success in life? Is a snapshot of a process actually revealing of anything beyond skill at test taking?

Wouldn’t it be better if students were evaluated on their evidence of learning, and wouldn’t that kind of evaluation have more to do with real life? Because the key questions are: What about outcomes for students 5 years after graduation? Success in getting a job or in getting into graduate school? Or founding a company? Those are outcomes.

If, in my bad drama, testing suspects in a crime as is done in that parallel universe seems absurd, consider if our own use of testing in an educational setting makes any more sense.

The most basic fact about our learning technologies now is that the only limits on what we can do in education are the limits of our imagination, our courage, our trust, our habits, and our expectations. The education emperor has no clothes. Just look and see. Even in this universe, that must be obvious.