Thea's post made me realize that I have probably concentrated too much on the positive sides of CASAS. Partly, it was my intention, as I did not want to transfer my own judgment to the innocent people. The truth is, that in the year I worked in CASAS-served program, I was its most ardent opposer. There are several things about CASAS that seem weird and/or unfair to me, so now I will try to present a more balanced account.
- As all standardized tests CASAS fails to capture a lot of things (writing and speaking, to start with). And, of course, there are people, who just don't do well on the tests, no matter what kind of songs and dances you do as a teacher to relieve their anxiety. On the other hand, CASAS is mostly used in state-funded programs and the state wants to know if the money was put to a good use. The teacher's "Oh, now they make less mistakes and actually understand instructions" is no less subjective than the bland score on the test sheet and I am not even entering the field of statistical implications of comparing "They are doing so well" with "They are making adequate progress", and "Gosh, they finally learned something". Finally, if the rest of US education is dominated by SATs, ACTs, GREs, TOEFLs, GMATs and other devilish contraptions, why should ESL be different? Standardized tests are evil, but they are the best we've come up with so far.
- I have not really discussed it in class, but the way things work between CASAS, programs, and states is that the state usually gives money to the programs based on the number of gains shown on tests. In Kansas, the program is required to have 65% of gains on participants to receive the same funding it did the year before. First of all, let's talk about the term "participant". If the person has 12 hours with the program - they are it. What are 12 hours? 2 days of testing/orientation and 2 class days. If the person went to the orientation, visited your class for a week, and then disappeared (a VERY frequent situation) you are screwed. Or, to put it in different words, you now need to have a 70% gains rate to cover for that idiot, who could not decide that they didn't need ESL before they were on your balance sheet.
So, we need 65% of gains on participants. Great! What is a gain? CASAS has a system of levels, based on the person's test scores. So levels are, for example, 160-185, 185-200, 200-210, 210-225, 225-250 etc (I don't remember the exact numbers, so :)). A gain is jumping from one level to another, and here several problems come up.
1. Students need gains in their lower score, so if someone has 180 in reading and 187 in listening they need a gain in reading. Even if their next listening test is 211, they still need the gain in reading, no matter what.
2. A gain is only jumping from one level to another. So, if student A had a score of 186 and now has a score of 199 it is not a gain, although t is an amazing progress. If student B had a score of 184 and got a score of 185 - GAIN. Although, there is probably zero progress involved.
3. Theoretically, all test scores are commensurable. Meaning that a 210 on an A-level test is equal to a 210 on a D-level test. It is not so. Yes, you need to answer 90% of A-level test questions to get the 210 score, and about 10% of D-level questions, but they require totally different levels of vocabulary, reading and analytical skills. So, moving from 220 on a B-level test to 220 on a C-level test IS a big deal.
The issues with teaching to the test, irrelevant vocabulary (aerogram) are also a big part of the problem.
Subscribe to:
Post Comments (Atom)
3 comments:
The section where you talked about trying to have 65% in gains. This is something that educators in all areas are fighting in the current time of No Child Left Behind. But to focus on the ESL students, so many of these students are members of migrant families. Their attendance is just like the student who left after a couple lessons and no follow up testing. They aren’t receiving consistent education. These significant gaps in education make learning English hard as well as learning content. This is compounded by their having ___ years in the country and then by testing standards not qualifying as ESL students.
Then the “standards” for gain are so arbitrary as you wrote. Blech to standardized testing.
Thank you for responding to my thoughts. It was interesting to see your actual opinions
I think that 65% of gains sound a bit worse than they actually are, at least with the lower level students. The thing is that it's a LOT easier to get gains on beginners than on advanced students. Finally, even if you're teaching without regard to the competencies, you are still covering some of them, whether you realize it or not. I practically skipped the whole picking competencies-teaching only test competencies thing my first semester, and still had an average of 75% gains. The number of tests, which can get stressful for students, also helps the teachers, as it raises our chances of actually catching a person and testing them.
"Standardized tests are evil, but they are the best we've come up with so far." I agree with his because it reminds me TOEFL exam I took it was really insane but thanks to its result I am here now!
Post a Comment