The Akron (OH) Beacon Journal is a bastion of obviousness, I tell you. Check out the headline on this article:
New state tests could keep more from diplomas: Newspaper report says difficult questions designed to ensure that more high school students will fail
Wow. Who at the BJ thought this needed to be spelled out for readers? For those readers who have repeatedly flunked such an exam, I suppose.
Beginning next spring, sophomores must pass the new Ohio Graduation Test in five subjects if they're to graduate on time in 2007. The new test requires achievement at two grade levels above current graduation exams, which means thousands could fail, the newspaper said.
Three-quarters of sophomores flunked a sample version last year, and nearly one-third failed this spring after the department shortened the test and lowered the recommended score for passing.
So this new, "tougher" exit exam has already been shortened and the standards have been lowered, and two-thirds of 10th-graders can pass it. Isn't it supposed to be a test for 12th-graders? What's the uproar about, then?
Scoring mistakes will become more common as states rush to meet deadlines, said W. James Popham, professor emeritus at UCLA who ran his own testing company. "But scoring mistakes can be corrected,'' he said. "What worries me more is the harm that will be done to children because of lousy tests.''
As compared to the harm done by lousy teaching?
And testing companies are trying to program computers to score essay questions to save money. A Dayton Daily News reporter composed a deliberately nonsensical essay that one company's program awarded a perfect score and declared "effective writing.''
Finally, a valid point. Automated essay scoring programs that would catch something like this do exist, but I'm sure there are programs that don't catch this method of "gaming" the system. But a good reporter would ask why essay questions are being introduced onto these types of standardized exams, when such questions are so expensive and difficult to score.
It's because of the anti-testing types who insist that multiple-choice items are not "authentic" enough and don't measure "real" learning. Somehow, test developers who respond to those charges end up getting blamed for iffy essay scoring procedures as well. Gee, it's almost like there's no test that would please some of these people. You think?
George Madaus, a senior fellow with the National Board on Educational Testing and Public Policy at Boston College, said more attention should be paid to whether achievement tests accurately predict academic success.
When a student fails, it should raise alarms, Madaus said. Instead, "It's just dismissed by saying, 'Oh well, the student can take it again four or five times.''' Popham said he's more concerned that tests that are supposed to measure what a student learned are instead designed so some fail.
"That is wrongheaded and makes no educational sense,'' he said.
This must be a misquote. Madaus is right when he says failures should raise alarms. But surely he doesn't believe that no one should fail these types of tests. Perhaps he means such tests should not be norm-referenced with a standard set to guarantee failers, which would be correct. But the reporter should explain that.
The state adjusts the difficulty of questions so that most students get average scores and a small number get the very highest and very lowest. The idea was to identify struggling students and get them extra help.
Depending on where and how the standard is set, though, this doesn't mean a certain percentage of students are guaranteed to fail each time. It just means most of the questions are of average difficulty; thus, students who perform below average are more likely to fail. There's no way to avoid this other than to (a) put only easy items on the test, which doesn't help identify those who have trouble, or (b) keep the difficulty as is, but set the standard so low that no one fails.
Look, when a state implements an exit exam, some students are going to flunk it, because some students are either unmotivated, unintelligent, or underachievers. States should pay attention to how many flunk and why they do so. But all this hand-wringing because Ohio has created a test in which students who perform below average may flunk is ridiculous. The majority of the 10th-graders passed the exam. Why should we worry about the 12th-graders who don't?
Some students who excel in the classroom don't do well on achievement tests. At Dayton's Meadowdale High School, 18-year-old senior Tynisha Edmondson makes A's in science classes but for four years has failed to pass the science proficiency exam -- despite coming achingly close. In a retest in March, she scored 198, two points shy of passing. If she failed the test she took this month, she can't graduate and might not be able to attend Wright State University in the fall.
"I'm scared,'' Edmondson said.
The article ends with this heart-wrenching anecdote. Tynisha has taken science classes for four years, but the article doesn't mention which types of classes they are. Were they the right classes to be prepared for the science exam? Here's the new guide to the graduation exam - skip to page 11 for the Science stuff. The sample item given is rather inventive. It's certainly not pure recall; it requires students to actually think about the task and synthesize known information.
The items may very well be challenging, and rather than tugging at our heartstrings with anecdotes about poor test-takers, the reporter should be asking why someone who's made straight A's in four years of science is struggling with this material.
Thanks to Daryl for the link.
Posted by kswygert at May 24, 2004 12:44 PM