iOS app Android app More

Featuring fresh takes and real-time analysis from HuffPost's signature lineup of contributors
John Merrow

GET UPDATES FROM John Merrow

With Testing, Where Do We Go From Here?

Posted: 07/19/11 09:03 AM ET

As always, remember that John's book The Influence of Teachers is for sale at Amazon.

Forget cheating on tests for a minute and think about the concept of 'teaching to the test.' Just what does that mean? The usual line (which I have used myself) goes something like this: "It's OK if it's a good test," and that may be correct. Unfortunately, most of the tests that I have seen are not 'good' tests.

Think about teaching students to write, and then testing their skills. Clear writing is important. Employers want to hire people who can write clearly, accurately and well -- but learning to write takes time and requires rewriting and more rewriting, under the guidance of a good teacher. There are no shortcuts. However, our obsession with numbers subverts both teaching and learning. Teachers are told that their students must be able to pass bubble tests and write a lot of short so-called essays (usually one or two paragraphs!) There's no time for reflection or rewriting.

Instead, students are drilled in the 'constructed response' process: write a declarative statement and then add three or four details to support a statement, such as: "I always use sun block when I go to the beach." And so they follow the formula they've been given and produce something like: "I always wear sun block when I go to the beach because too much sun can cause cancer, and because too much sun will make me all wrinkled when I get old, and because cancer can kill you. My mother makes me use sun block too."

That 'essay' would get a passing score because the student supported his statement in four ways. The teacher (or machine?) grading the 'essay' could simply count the supporting reasons. Everybody -- teachers, principal, superintendent and school board -- would pat themselves on the back, but is Microsoft, GE or Hilton likely to offer someone who's been trained to write that way a job?

That's what we are doing to our children. It's only slightly hyperbolic to say that we are lying to our kids.

Cracking down on cheaters -- which we should do -- won't fix our problem. Think about it this way: You are sitting in your living room when drops of water begin falling on your head. Clearly, you have a problem. If you move your chair, have you solved it? After all, you no longer have water falling on your head.

Of course not, because the problem persists, although now the water is falling on your living room rug. Suppose you get a large pot and place it where it can catch the falling water? Have you solved the problem? Of course not, because you still have a leak somewhere.

You get the point. I think it's time for those of us who are attacking bubble testing and the intense pressure to 'produce' to back off and ask, "Where do we go from here?"

Unfortunately, we haven't asked and answered that question in the past. Subverting the testing system is an old story that we don't seem to learn much from. Remember Austin, Texas, where most of the school board was implicated in test score deception? How about that small town in Connecticut with its 'miraculous' test score gains a few years ago? Not miracles, just plain old cheating.

Sometimes the system aids and abets the deception, as in Florida, where a loophole in the state law allowed districts to counsel low-performing students to drop out to go into GED programs. By law, the districts didn't have to count these kids as dropouts as long as they suggested the GED alternative, no matter that no one had to follow up to see if the kids actually enrolled.

How about the so-called 'Texas Miracle" that turned out to be the 'Texas Mirage?' Houston had great test scores, and Superintendent Rod Paige eventually became U.S. Secretary of Education. Then we learned that an inordinate number of low-performing 8th graders were simply being held back, often for more than one year, because high-stakes testing didn't begin until 9th grade. Some find the seeds of No Child Left Behind in that misadventure.

Atlanta may actually be the proverbial tip of the cheating iceberg because evidence that suggests major cheating has also occurred in D.C., Pennsylvania, Florida, Houston, Baltimore, Los Angeles and elsewhere.

Some consultants, test security companies and even the test makers themselves are licking their chops right now, expecting to make a lot of money designing what they will claim will be better defenses against cheating, because 'firewalls,' 'fail-safe' steps, 'erasure detection software', and other 'technical fixes' are a big part of the conversation. In fact, Education Secretary Arne Duncan told the Atlanta Journal Constitution:

The technical fix is very simple, and they need to put that in place. The job for a new superintendent coming in after a crisis is to rebuild public confidence with absolute integrity, transparency.

I respectfully disagree, because cheating is not the real problem; it's a symptom of a larger problem, and the solution is not simple. Not by a long shot.

The problem in Atlanta, in D.C., and wherever else cheating is occurring proves Campbell's Law, which states "The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor."

Live by the test, die by the test.

We rely too heavily on the scores of relatively simple (and relatively cheap) machine-scored 'bubble' tests as the measure of educational accomplishment, and that invites deception, cheating and criminal behavior.

So where do we go from here? Well, we aren't going to 'get rid of testing,' that's for sure. Anyone who wants to throw out that bath water ought to recall the New Orleans high school valedictorian that could not pass the Louisiana state graduation test, despite being given multiple opportunities!

Nor is it enough to endorse "multiple measures" of achievement. It's more complicated. We have to ask ourselves what we want young people to be able to do upon graduation and figure out how to teach and encourage those behaviors. Then -- and only then -- do we figure out ways to measure them.

What if we were to ask large employers like Michael Dell, Steve Ballmer of Microsoft, Carol Bartz of Yahoo, the heads of Hilton, Hyatt, Avis and Hertz, Wendy Kopp of Teach for America, Steve Jobs, Jeffrey Immelt of GE, the provosts of some major universities, top advertising agencies and so on what they look for in potential employees? What would they say?

Or maybe you hire people for your company. What do you look for?

Life is not all about work, of course, so we ought to ask what we want our youth to be: good parents, concerned citizens, informed voters, discerning consumers, and so on.

Then let's figure out what sort of school-based experiences teach or sharpen those skills and attributes. My hunch is that group activities and project-based learning will figure prominently. I think we will be reminded of the truth of the late Ted Sizer's observation that "Less is more."

Tests drive public education right now. But what should be driving the enterprise are agreed-upon goals that come from the real world.

Where do we go from here? That's up to us, isn't it?

 

Follow John Merrow on Twitter: www.twitter.com/john_merrow