Edit: Also where do you get "require" from. Not all teachers belong to a union? Did you read the article at all?
Many states require non-union employees to pay the union. I believe DC does as well.
Your only choices are to be taxed by the union with voting rights or be taxed by the union without voting rights.
Besides the end game of unions are what happened in Detroit.
But thats a Danish dispute which I doubt many here is familiar with, hence the Detroit example.
And it is hardly "one specific" example: its multiple agreements with multiple companies.
I'd rather work in a system where the quality of my work determines my wages and not my seniority or how many masters degrees I've obtained.
I'd rather work in a system that doesn't strong arm my employer into paying benefits they can't afford. I don't know about you, but I'm not worried about my 401k being there in 40 years, but I'd be really scared if I had a municipal pension.
Let's assume we are in a performance based pay system. How do we measure performance? Is it average student grades? If so, then do we pay teachers who work in affluent areas more because every effort is afforded the students (tutoring etc)? Do we pay teachers who choose to work in more challenging schools less (and more likely have lower grades but possibly greater impact on students' lives with their successes)?
What do we do with teachers hand picking elite classes to improve their perceived performance? It already happens even without performance pay? Or it could just be the luck of the draw. Sometimes I end up with difficult student groups simply because of a timetabling issue. Less pay?
There are other metrics for teacher performance that have been proposed, and all the ones I have heard have been very flawed.
Just to touch on the abuse bit at the start, if there was one employer in your town (as there effectively is in many rural areas when it comes to schools), should you relocate because of abuse you have received? (or indeed at all? Who doesn't deserve to feel safe from abuse in their workplace?)
You build a statistical predictor of student performance, then you pay teachers proportional to Actual Performance - Predicted Performance.
If teachers pick elite classes to improve their performance, the bar is raised. If their class has a predicted score of 80%, then the teacher needs an actual score of 90% to get a bonus. Conversely, if you get a class of low quality students, the predicted score might be 20%. To get the bonus, you need to achieve 30%.
The only incentives you have are hand pick classes of students that you can improve. If you are great at moving students from the 50'th percentile to the 60'th percentile, but suck with other student groups, your incentive is to get a class of students in the 50'th percentile.
I suppose you could factor all that into the model, though I'd question the feasibility of doing so.
I'd also argue that grades aren't the only important factor in schooling. Building social skills, discipline, time management, helping to realize goals... All ways in which good teachers can influence students, and only some of them reflect in neat packages like test scores.
Edit: I'd like to clarify that in my previous post I mentioned getting more difficult classes sometimes as the luck of the draw. I didn't mean low ability classes. I meant classes with students with serious behavioral issues, students with learning disabilities and so on. These will very often negatively impact the learning outcomes of the entire class, and is something you very rarely if ever have to deal with in top end classes.
This would only matter in the event it did not average out over a teacher's entire cohort of students. I.e., you would need to get a situation where all the students who wanted to move into the workforce were taught by teacher A, and all the students who wanted to complete school were taught by teacher B.
The law of large numbers helps here.
Also, this would only matter insofar as this factor is uncorrelated with the things you do use as predictors - parental income, race, prior grades, etc. If all the low income students want a job, and all the high income students want to complete school, the predictor already incorporates this factor.
I meant classes with students with serious behavioral issues, students with learning disabilities and so on. These will very often negatively impact the learning outcomes of the entire class, and is something you very rarely if ever have to deal with in top end classes.
If this factor is strongly correlated with income (as you suggest), the predictor will already incorporate it.
"if there was one employer in your town (as there effectively is in many rural areas when it comes to schools), should you relocate because of abuse you have received?"
Yes. That, or start a competing company. If the abuses are really bad, you'd no doubt attract a lot of other disgruntled employees who'd prefer to work for a better company. Jobs are not a right, no matter how badly people want them to be.
So uh, that'd pretty much be like a union, right? A group of people effectively saying that they won't work under certain conditions? ;)
I suspect we'll fundamentally disagree on the first point about parents voting with their feet/wallets, since that kinda describes the private school system, and I'm a firm believer in the same education being available to everyone (freely, if possible, and cheaply if not).
A popular conservative position is to give all students the same educational opportunity by allowing parents to take their children's education dollars to private schools.
(Of course I suppose the conservative counter to that is to get a better job, and that they'd have enough money if taxes weren't contributing to education dollars :)