Great news! Scores are way down.

Editorial

Arkansas Democrat-Gazette

Sunday, October 9, 2005

Let us now praise Ken James, the state’s education czar, for all those lousy scores by our fourth-graders on the latest Arkansas Benchmark exams. (Those are the Arkansas-only tests we use that, even now, don’t give us any indication of how our kids compare with students across the country.)  The kids’ scores took a hard tumble from last year. In math, only half the fourth-graders had high enough scores to be deemed proficient in their studies. Down from 64 percent a year ago. In literacy, pretty much the same drop: about 50 percent scored at or above proficiency—compared with 68 percent last year. What a flop! So all we’ve got to say is: Well done, Commissioner James! 

Keep up the good work. Honest. From now on, this is a sarcasm-free editorial. A gift from us to you. Because there’s more to these scores than just scores. As they say on about two-dozen shows on ESPN2(a)U, let’s go behind the numbers. 

A little context, please:  (1) The Commish is almost wholly responsible for these plunging test scores. Why? Because he had the nerve to raise the bar, not lower it. In educanto, raising and lowering the bar has nothing to do with the price of bourbon. It’s about Setting the Cut Scores. In plain English, that means folks like Ken James, the state’s board of education and teachers decide what a student should score to make the cut and go from Basic to Proficient to Advanced. 

Advanced is, well, advanced. Proficient is grade level. Basic is not so good. Below Basic is bring-in-the-parents time. 

Cut scores on the state’s benchmark exams haven’t been raised since 1998. Until last August, when Ken James & Co. got serious about standards and—all together now—raised the bar. Which explains why a fourth-grader’s score of, say, 44 in math was considered Advanced last year but Basic this year. 

According to Julie Thompson at the state’s Department of Education, the scores this year based on last year’s cut rates would have been: 68 percent proficient and advanced in math (compared to last year’s 64 percent), and 67 percent proficient and advanced in literacy (compared to last year’s 68 percent). 

Ms. Thompson notes that "the comparison involved more than just using last year’s cut scores applied to this year’s test because it also involved mapping questions to like questions on last year’s exam in terms of content and level of knowledge tested. . . . ." 

Got that? 

In short, the test has changed—for the better. 

(2) Why raise the bar? To quote Ken James: "What we expect students to know and be able to do in 2005 is different from what it was in 1998." 

It had better be. Because the world is a far different, more competitive and economically smaller place. According to Thomas Friedman’s new book, The World Is Flat, there are at least a billion more people competing in the free world today than in 1998. And because of new technologies, these folks are now competing against each other as national borders fade. If you’re a banker in Kansas City, you’re competing with a banker in Hong Kong. Heck, if you’re a financial analyst in Hong Kong, you may be working for a bank in Kansas City. 

Arkansas’ kids better be prepared to compete against the best—from everywhere—even if they stay right here in the Land of Opportunity, we hope. 

(3) Some of us have been suspicious of Benchmark tests, and had reason to be—especially when scores spiked. That’s because it’d be so easy to lower the bar and so make the test results look, but only look, good. Before the arrival of Ken James, did the Department of Ed even tell the public what the cut scores were, and would be, in advance of the tests or results? Who knows how much the numbers were jiggered? 

Here’s a case of the opposite—raising the bar and being up front about it. Who knows, standards in education might actually mean something again one day. Imagine: honesty in education. 

(4) Commish James announced these results Wednesday, October 5th. Wanna guess when the kids took the tests? 

March. 

More than six months ago. 

By efficient contrast, last May 5,000 students here in Arkansas took part in a pilot project that used a different and better test. That test combined a test like this Benchmark exam with a standardized, nationally normed test, and the results came back in three weeks. Since everybody wants to spend, spend, spend more money on education, here’s a place where Arkansas really should spend more money: on getting a quicker turnaround on these test results. 

Spending can help—but not when you give blind salary increases across the board without regard to merit. That way lies, well, what we’ve got now—a lot of legal wrangling over how much spending is ever enough without any clear idea of what the spending is accomplishing. 

And finally:  (5) Much as we admire Ken James for raising the bar and setting higher standards, he’s also revealed a serious problem with the way tests work now, or don’t. That the bar can be raised and lowered, and the standards jimmied, makes comparisons pretty much meaningless. And that’s probably the best argument we can think of for requiring a standardized, nationally normed test for Arkansas’ students. That test could be given either separately or incorporated into the state’s Arkansas Benchmark exam as part of a single test. Think of it: Then we’d know how our kids were doing year to year—and how they stack up against kids all over the country. Then we wouldn’t have to translate all the educanto. Then we wouldn’t have to go behind the numbers. Or play games with them every few years. Then a score would be a score—as understandable and in context as 70-17.