<$BlogRSDUrl$>

My personal musings about anything that gets on my radar screen--heavily dominated by politics.

2004-12-27

Teacher Turnover, Test Scores, and Bad Reporting

The Friday Rocky Mountain News ran this story about teacher turnover. It starts out well enough:

About one in five Colorado teachers changes schools in a typical year, a rate of flux that is above the national average . . .

but within the first paragraph begins to unravel:

. . .and one that appears to be directly linked to student achievement on state exams.

• Teacher turnover or "churn" is higher where test scores are lower, with a 17 percent churn rate in schools rated excellent by the state and spiking to 23 percent in schools rated low and unsatisfactory.

What the data here is reflecting is a COROLLARY relationship between teacher turnover and student achievement; that is, these two occurrences are happening simultaneously. What the story is trying to suggest is a CAUSAL relationship, saying that teacher turnover is the cause of low student achievement. Any person who has ever taken a statistics class, the difference is pretty significant. It's akin to saying that being married is the cause of divorce; of course the one exists in relation to the other, but it's much harder to say that one is the cause of the other.

In this case, the causes of low achievement and teacher turnover could easily--and likely--be factors completely outside of the school. Issues of student mobility, discipline, and support from home have enormous effects on both student achievement AND teacher job satisfaction (and thus, retention). There is also the external factor of difficulty attracting teachers to "difficult" schools, and thus these schools needing to hire from the ranks of the young and inexperienced. Such teachers leave the field at a far greater rate than older teachers, and for many different reasons, not the least of which is discovering that they don't like teaching.

It isn't, to me, that the story is dealing in with turnover; it's that it's doing it in a way that is misleading from a strictly statistical standpoint. I should think that this case could be made using better numbers or more accurate reporting of numbers.

All that said, this IS an interesting story. That there exists a correlational link between turnover and achievement is only half of the story; the next obvious question is "what are school districts doing to attract and retain teachers in low achieving schools?". Fortunately, Denver has at least begun to address the problem:


He said he expects to propose early next year the creation of school "zones" free from some contract restrictions. He cited initiatives elsewhere, including New York and San Francisco, where school leaders have identified groups of underperforming schools for special interventions.

In San Francisco, for example, some low-achieving schools are being converted to "Dream Schools" with tougher curriculums, longer school days and student uniforms. Teachers in those schools are being asked to reapply for their jobs, which has touched off union protests.

I would have never guessed that the Denver School system would become the incubator of serious education reform in the state, but between this idea and the performance-pay initiative last year, this district is quickly leaping to the head of the class for education reform. I wonder if other school districts--and, more importantly--teacher unions are watching (ahem. . .calling Jefferson County Education Association).

In one other little note, the article points out:

• Teachers in charter schools posted higher turnover rates than teachers in traditional schools in each of the four years studied.

I saw numbers last year that showed the same sort of phenomenon is even more pronounced for charter school administrators. Surely, this data tracks correlationally to the data that seems to be coming out about charter school test scores. I've also seen data that shows charter school employees citing "intrusive" and "interfering" parents as the biggest source of job dissatisfaction, leading to this kind of turnover. Not coincidentally, I think it's also that sort of thing which leads "traditional" schools and educators to turn their noses up at charter schools. Perhaps part of the long-term solution for charter schools should include efforts to retain teachers by buffering the staff from parents.

I know, that fundamentally alters what charter schools are supposed to be about. On the other hand, failure will also fundamentally alter charter schools, as will continued reliance on new teachers who are not invested in a philosophy and self-serving parents who either want what's best for only their child or who only want to get their hands on a little fiefdom.

I like charter schools, and would like nothing better than for them to prove that different models can be remarkably effective; but I think the movement needs to address a few things to make the whole thing work just a bit better.
Posted by: Michael / 12/27/2004 01:14:00 AM

Weblog Commenting by HaloScan.com

This page is powered by Blogger. Isn't yours?