Teachers unions mean better teachers, new study says

[applegrove]

Teachers unions mean better teachers, new study says

by Laura Clawson at the Daily Kos

http://www.dailykos.com/stories/2016/7/22/1550632/-Teachers-unions-mean-better-teachers-new-study-says

"SNIP..............


All that stuff you hear about how teachers unions protect bad teachers through the evils of due process and their general uniony badness? A new study for the National Bureau of Economic Research says nuh-uh. EduShyster interviews the study’s author, Eunice Han:

By demanding higher salaries for teachers, unions give school districts a strong incentive to dismiss ineffective teachers before they get tenure. Highly unionized districts dismiss more bad teachers because it costs more to keep them. Using three different kinds of survey data from the National Center for Education Statistics, I confirmed that unionized districts dismiss more low-quality teachers than those with weak unions or no unions. Unionized districts also retain more high-quality teachers relative to district with weak unionism. No matter how and when I measured unionism I found that unions lowered teacher attrition.

This isn’t all theoretical. Thanks to Republican state governments in Indiana, Idaho, Tennessee, and Wisconsin, Han had the chance to see how this played out in recent years:

If you believe the argument that teachers unions protect bad teachers, we should have seen teacher quality rise in those states after the laws changed. Instead I found that the opposite happened.


................SNIP"

[Gabi Hayes]

1. Trump's reaction:

''If you have a good teacher, and he/she's in a union, KILL 'em! You can hire a new one at a third the price, then, when she gets too old, you can kill HER, too. Just Like I'm going to do with my third wife, Milena....zat her name? I forget.''

[malaise]

2. ROFL

[Gabi Hayes]

3. what's so funny? I were a nun-oniun skewl studnet... thimk I are kidning?

?

ahh kepe inphaurmmd, cee?

[malaise]

4. LOL

[MichiganVote]

5. When he loses, that marriage is going belly up.

[Igel]

6. I have problems with the study.

1. It assumes that the sampling is random, and it probably is. The census can do random.

But it then assumes that the distribution of high quality teachers and collective bargaining is random. That is, you have the same chance of a school having strong unions in the middle of nowhere in the South as you do in, say, urban California.

While I don't know how to fix this, it strikes me as unlikely.

2. It assumes that the distribution of salaries is dependent on unionization when there's a correlation, but, in fact, those rural districts in the middle of nowhere are going to have lower salaries. This is a confound. It also assumes teachers are fine living anywhere.

3. It gauges high-quality by two standards. The first is meeting NCLB requirements, typically state-certification. The second is student achievement for the school, not by classroom. But it evaluates elementary schools and high schools by high-school graduation rate. That not only isn't a great standard, it can be easily fudged. In Texas, you start "home schooling" (which may mean de facto drop out) but it's a transfer.

On the one hand, many high school teachers' success depends not on their abilities but on preparation in earlier grades. On the other, a lot of sucky districts have a lot of transient students. Who gets credit for their achievement? And, since we're talking Shiva and there are more than 2 hands, on the third hand we have charter schools, barely included. A lot of dismal students drop out of high school and go to charter schools, but those drop outs do not count against the district. They're "transfers". These are more likely in high-population density areas than in rural areas. Ah, again, the distribution has a geographical bias.

4. It distinguishes between those teachers dismissed for being low-quality and those who leave for other reasons. Now, this is very pertinent to collective bargaining. Why? Because in my experience in two districts and four schools (albeit in Texas), firing a teacher for being low-quality requires paperwork and appeals. But if you threaten a teacher and get her to quit and move on--"Resign and you still get a good reference"--then that resignation is not because she's low qual. In other words, there's a principled bleeding of the teachers the study wants to include from the data categories that must contain all teachers fired for issues of student achievement.

So while this is CB-relevant, it also undermines the validity of any conclusion. I don't know how you can compensate for these known unknowns (at least to teachers). Rule 1 for any research, understand your data.

I'm not sure what to make of the timing of the data sets--just before and after the 2008 crash. The first set featured a lot of large cohorts of teachers but not many new hires--tough job market. The second was a tough job market because RIFs were on-going, and if you RIF a teacher it's more likely to be a low-qual or low-seniority teacher. Overall, not sure what I'd predict biases in that data set would be.

I could go on, and I'm not even done with just the data portion. I can't comment on the model or the analysis, but I simply don't trust the data.

http://edushyster.com/wp-content/uploads/2016/07/Han_Teacher_dismissal_Feb_16.pdf

Even the title is misleading. Most teachers notorious for not being fired aren't under the gun for being low-qual, but for other kinds of incompetence that her evaluation metrics ignore. Many low-qual teachers don't make it through probation--they know they're incompetent, and that stress can be overwhelming. Student teaching often strips out the incompetent before they're hired, probationary years do the same.

CV for author here. Visiting lecturer 2 years out of grad school when appointed visiting lecturer, one publication, lots of working papers. Second NBER working paper, and I suspect that her research is interesting for some who are supporting her appointments but she's not getting things through peer review. Advocacy research can be like that. Don't know--maybe she'll get tenure-track job this fall, maybe she'll bounce around as post-doc or lecturer for a while.