Home » Uncategorized

Develop a national test before deciding what to teach is backward

Submitted by on Monday, 22 June 2009 2 Comments

U.S. Education Secretary Arne Duncan’s plan to start developing a national test before anyone’s even come up with national standards is far worse that putting the cart before the horse.

It’s more like hitching up Secretariat before we’ve even decided what we want the cart to haul. Before we even know if we need Secretariat, for that matter.

Technically speaking, we already have a thoroughbred of a testing program called the National Assessment of Educational Progress. Problem is, it’s a little too blue-blooded, and not just for American students.

It’s simply too hard. It tests students on everything we would possibly want them to know in an ideal world, as opposed to everything they need to learn to succeed in this world.

NAEP, though, is a bur under Duncan’s saddle. Why do some states perform so well under No Child Left Behind but flail on NAEP, he asked. It’s because states get to set up their own testing systems under NCLB.

He’s right about that, but the problem goes beyond the content of the tests. It goes to the dozens of tiny ways states can game the system, the easiest of which is to manipulate racial categories students are placed in.

Duncan’s solution: Spend $350 million to develop a uniform national test. The test needs to be developed first, Duncan said, because that’s where the financial “heavy lift” will be. Developing the standards will be easy, he said.

Maybe he’s so used to evaluations driving virtually every second of classroom time these days that it didn’t even occur to him that maybe we should decide what kids should learn before we determine on what they’ll be tested.

“Having real high standards is important, but behind that, I think in this country we have too many bad tests,” Duncan told The Washington Post. “If we’re going to have world-class international standards, we need to have world-class evaluations behind them.”

That must have been a slip of the tongue, because he got it right: The tests should be behind the standards, not before them.

It  will be a mammoth undertaking to get educators from 46 states – South Carolina, Texas, Alaska and Missouri are boycotting – to agree about what children should learn and when. Points as simple as whether “into” is a kindergarten or first-grade word could spur endless discussion.

But it’s hard to fathom how it would be easier to develop the test before those discussions even take place.

Copyright 2009 Debra Legg. All rights reserved.

Similar Posts:

    None Found

Popularity: 1% [?]


  • Brian said:

    This seems like it could work to me. A few things seem to come through from reading the story and Duncan’s remarks to the governors:

    1. The $350 million will go to states as grants for developing better tests.

    2. The U.S. Department of Education will decide which state proposals are “better”, but one factor is that they have to go beyond multiple choice.

    3. The feds aren’t going to wait until educrats and politicians in 46 states come to a consensus on standards before the agency starts handing out the money.

    Several educators tell me that a big problem with the current system is that the “standards” read like a list of Trivial Pursuit questions than a real outline of knowledge. Consequently, the tests derived from those standards tend to be multiple choice.

    If Duncan uses these grants to encourage better forms of tests, it will force the standards away from trivia lists. Once we have better testing frameworks, we can modify the content of the tests to reflect any set of guidelines we settle on.

    An iterative process where test improvements lead to standard improvements which then lead to further test improvements is the kind of iterative process that is common to many successful creative projects.

    To put it in writing terms, I usually find it more productive to write a rough draft quickly, read through it looking for improvements and then rewrite (repeating the whole process several times) instead of trying to write a finished draft on the first go around.

    Not only do I end up with a better final draft, but many of the intermediate drafts are better than the version I would have come up with trying to get everything perfect in a single pass.

    Waiting for 46 states to agree on a universal set of standards seems like a sure way to fail.

  • Debra said:

    I’ll give you your point on Trivial Pursuit in a heartbeat, Brian.

    A classic example is what we went through this year, which was nine months of drill and kill. I’m hearing and reading this is not uncommon across the country. Memorize the flash cards. Write the words repeatedly – the same two (or, toward the end of the year, three) words every night. Review the accumulated flash cards from the entire year. The result, at least with my kid, was mind-numbing boredom. And he’s in kindergarten.

    For some reason, in the final month of school they’ve decided to back off on the drill. They’ve done more writing and drawing projects. The teacher has talked about sea life, and Big Guy chatters excitedly every day about what he’s learned. “Do you know a star fish can grow back its arm if it loses it?”

    This happened to dovetail with a book a friend gave the guys recently. The book’s about sea creatures and probably is written on a third- or fourth-grade level. I haven’t run an analysis on it, but based on sentence, paragraph and word length, that’s my best guess.

    We’ve read it every night for a week. Just a few pages at a time, with Big Guy reading the words he knows, sounding out other simpler words and me filling in the blanks. He’s spellbound.

    Fast forward to last night, when it was Boots’ turn to pick a book. He chose a story about Thomas the Tank Engine and an aquarium exhibit. When Big Guy came to the words such as “fish” and “shark,” he read them. There you go: Two words in a week, not a single flash card involved, actual learning as a bonus.

    If Duncan’s process can get us away from what happened the first nine months of the year and toward what’s happened in the past month, I’d be all for it. But as long as we’re putting the test first, I’m skeptical