March 24, 2010

Teachers and student test scores

Colossus R&D man Gooch sends me this link about a Florida state senator who wants to base teacher firing/promotion primarily on student test scores:

The bill, sponsored by state Sen. John Thrasher, the new head of Florida’s Republican Party, would require that school systems evaluate and pay teachers primarily on the basis of student test scores. What would not factor into teacher pay would be advanced degrees and professional credentials, including National Board Certification, which requires teachers to pass a competitive series of tests that is considered the gold standard for educators.

It gets worse: Experience in the classroom wouldn’t matter either. And student test results would determine which teachers get targeted when layoffs are necessary.

Would it be fair, say, to fire doctors because his/her patients failed to follow his/her directions in taking their medication (meaning their health got worse)? If not, how is this radically different from Thrasher's proposal?

I once went back and forth via e-mail with DE state Senator Dave Sokola about this very issue (among others). I never did get a worthy response. Sokola's reply consisted of him explaining that he "knows what he's talking about" because he did a year-long stint as a substitute teacher back in the day, and that teachers are really not different from a company. Regarding the latter, I had asked Sokola how teachers aren't different from private companies as they do not control the factors or production. They have to "produce" a product solely based on what they're given -- whether they want the "factors of production" or not. Can companies do this? Sokola retorted that a person at a manufacturing co. could get canned if one of the co.'s suppliers didn't come through with some materials needed to build a [finished] product (on time). Well, perhaps. But that co. could easily ditch that supplier for another, not to mention make amends (monetarily or materially) to whomever for the lateness of the company's finished product.

In addition, Sokola could offer no explanation as to how it was fair that a then-proposed state teacher evaluation system that based 20% of every school's teacher's evaluation on test scores -- tests that were, like the current DSTP, reading, writing, and math. So how is that fair to teachers who don't teach those subjects? 20% of an art teacher's evaluation ... based on students' math scores?? Say whaaaa ...?

Hey look, those who know me know that I am hardly a hardcore union type who is against any sort of education reform. Indeed, one proposal from the article I don't have much of a hassle with (at a cursory glance, however, to be sure) is "Newly hired teachers would be on probation for five years and then work on annual contracts for the rest of their careers." (Right now, in Delaware, teachers usually are on a probationary period for three years and then get tenure that next year. And contrary to popular belief, tenure does not mean a teacher cannot be fired; it ensures certain steps must be followed and can, admittedly, prolong the process and make it difficult to ax a lemon teacher.) Annual contracts are what many charter schools utilize, so I don't see many reasons why they couldn't be used in public schools. (As long as the primary basis for a firing isn't student test scores, that is!)

Also, I am not one who seeks to blame everyone but teachers and schools for lousy performance. Clearly, teachers (and schools) play a big role in shaping students' lives. Lemon teachers can clearly exacerbate even more the problems of students who already have myriad issues, and great teachers can assist in alleviating [some] of these. But ultimately, a teacher would have to be "in control," for lack of a better term, of a student's life for much more than the 45-to-60 minutes per day that he/she sees him/her (and that's all individual teachers see students per day, especially 6th grade and up -- not the full 7 and a half hours that critics claim) to make "student test score teacher evaluations" fair.

What would be a "fair" evaluation system for teachers? The way the system is currently set up across the country, [school] administration does teacher evaluations based on a few classroom visits. (Again, keep in mind that the frequency of these visits vary from not only state to state, but district to district and not all states/districts may use such methods.) The inherent problem with this is, while administrators may be well versed in general pedagogy, they may not know a whit about the actual subject matter. I'd recommend assembling a small cadre of veteran (good) teachers, one for each subject area in both elementary and secondary levels, that would periodically visit teachers' classrooms for evaluations. This would ensure that the evaluators would actually know something about the course being taught. (This idea doesn't address the issue of cost; however, various districts might be able to pay EPER -- Extra Pay for Extra Responsibility -- or offer "clock hours" towards recertification, which are required by the state.)

As for counting students' test results towards evaluations, I've little problem with it as long as it makes logical sense. Sokola's and the former DE legislature's inane 20% for all teachers regardless of subject matter certainly doesn't qualify. The DSTP goes a step towards the right direction in that since children are tested every year, one can see progress up or down, and somewhat correlate it to the teacher. By "somewhat," I mean it is not clear cut. For example, a student may have never been taught (or taught properly) his times tables in elementary school, so why should that reflect [moreso] on a subsequent teacher whose primary job is to teach him middle school algebra? An 8th grade algebra teacher's evaluation who had such a child may be more negative than that of another algebra teacher who had children who benefitted from an excellent past teacher of multiplication ... even though the former may be a superb algebra teacher. One of Thrasher's ideas in Florida is to test students in every subject every year not already done so by state or other assessments. That's a good idea, if you can come up with such [good] assessments. (Band? Chorus? Art? Shop?) And for introductory classes, what would be the baseline assessment?

There are many, many questions involved in utilizing student test scores for teacher evaluations, pay, and hire/fire decisions.

Posted by Hube at March 24, 2010 06:36 PM | TrackBack

Comments  (We reserve the right to edit and/or delete any comments. If your comment is blocked or won't post, e-mail us and we'll post it for you.)

Saw an interesting proposal to use the 4th grade and 8th grade tests to establish a plus or minus factor in math and reading to grade schooling by state.
Delaware's line looked like this:
Math 05-09 Reading 05-09
4th 8th diff 4th 8th diff

DE 3 14 -11 11 17 -6

So Delaware's total rank would be minus 17 indicating a good deal of lost ground. Rankings are against other states plus DOD schools and DC.

Posted by: kidney at March 25, 2010 12:49 PM