From the Interested Participant comes a fascinating tale of teachers complaining about high test scores, and superintendents claiming that it's all about the money:
Thousands of Arizona students struggling to learn English are about to lose extra help because a new state test shows they can read and write in English, though educators fear many of them really can't. In some districts, students are passing a new state test that says they are proficient in English at nearly double the rate of last year. But educators say the test is just easier and these kids aren't likely to pass regular classes without help.Arizona Superintendent of Public Instruction Tom Horne said the criticism comes because some school districts don't want to lose the extra money, about $350 per English-language learner. He said some schools have kept students designated as English-language learners for years because it means more money.
What a mess. Certainly, though, if the teachers are correct in their suspicions about the test, the results should be pretty much immediately apparent; a doubling of the overall pass rate is certainly cause for concern. However, I'm not sure if they're comparing apples to apples here:
Arizona school districts used to have the choice of one of four language tests, but the federal government now requires states to select one test for consistency. Arizona chose a Harcourt Assessment test, along with several other states..."We're very concerned," said Cindy Segotta-Jones, director of language acquisition for Cartwright Elementary School District in Phoenix. She said about 2,200 Cartwright students passed the language test, nearly double the usual rate.
But on what test was the baseline set? If 1100 students passed last year, was that over all four tests? If so, is it possible that the other tests were too hard, and this one is closer to being just right? There's no way to tell from this information, of course, but it would help if I knew the pass rate on this exam from last year vs. this year. Now that everyone is using the same, it will be possible to compare results between schools in way that wasn't feasible before.
Sal Gabaldon, a language acquisition specialist for the Tucson Unified School District, said the new test is far different from previous ones used in Tucson schools.The previous test used in Tucson required students to pass all three parts - reading, writing and oral - to be declared proficient in English, he said. The new test uses a composite score so it's possible that a somewhat lower score in writing could be offset by a higher oral score.
Becuase the previous test was triply conjunctive, one could argue that it was more difficult to pass, even if the individual items were the same difficulty as on this exam. However, is it a bad thing if the test isn't conjunctive? I'm not a content specialist in the area of English proficiency, so I don't know (a) how possible it is to get wildly different scores in the areas of reading, writing, and speaking, or (b) whether having wildly differing scores is predictive of disaster. Is it highly unlikely that a student will be a disaster in one area and great in another? If a student is good enough to pass in any one of the three areas, will they have a decent chance at succeeding later on without additional remediative help? I'd have to know all of those things before I could judge whether the test is useful for this purpose.
Tucson's teachers apparently think these results presage disaster, but Interested Participant wonders if some of that is hurt feelings from having control over exam choice taken away.
Posted by kswygert at July 12, 2005 12:11 PM