Affecting real change at BVSD

We had a very interesting conversation in the comments of the Camera article Boulder Valley teachers turn down 1 percent cost-of-living raise. (This by the way is the Camera commenting at its best – an thoughtful discussion between people who disagree listing out the reasoning behind their points, and coming to agreement on some of the issues.)

Let's look at the facts. First BVSD has only 70% of its students at proficient or better. That means it is failing to do its job for 30% of the children entrusted to its care. Second, over the last 10 years of CSAP testing (CSAP pre-dates NCLB) there has been no improvement in BVSD's effort. Nothing, nada, zilch. Third, in the world today not having a college degree is a sentence of a lifetime of poverty for most. As most of the 30% BVSD is failing to educate are poor, this perpetuates a cycle of poverty.

At the same time, we have two major items that will let us effect significant change that will lead to a more effective system. First the teacher contract is expired. BVSD can come back with anything they wish, including the changes needed to improve the system. Second, in the present economy there are a lot of people well qualified to teach who are desperate for a job. So if many of the existing teachers turn down the offered contract, it is easy enough it find qualified replacements.

Out of this, based on the ideas of myself and others commenting in the article, I think BVSD should consider implementing the following:

1. The total budget allocated for teacher pay be increased by 1%.
2. Principals can select the teachers they want in their schools irrespective of tenure. All teachers not selected regardless of tenure are let go.
3. The school district sets individual pay anywhere within 20% of the presently set union rate for each teacher. But the total matches the present total plus 1%. In other words, say the union rate for a given teacher is 82K. Then the district can pay that teacher between 65.6K and 98.4K. But when adding up all salaries for all teachers, it needs to equal the present total plus 1%. So in total the teachers got a 1% raise, but some received a cut of up to 20% while other received a raise of up to 20%.
4. 20% of each teacher's salary is a bonus based on merit (merit definition below).
5. Half of the salary for principals & administrators is also a bonus based on merit. For merit pay to work, it needs to start at the top and the higher up you are, the higher a percentage of your pay is tied to performance. And it has to be tied to the same metrics. Otherwise it doesn't work.
6. Lousy principals & administrators are fired immediately. (And if administration bonuses are dependent on performance, they will be.)
7. The school board members promise to begin each re-election speech listing the percentage of students proficient & above and the percentage improving 1.2 years/year both when they first took office, and today. In other words they will run for re-election based on what they have, or have not, accomplished in improving the system.

So how do we pay out these bonuses? It is paid proportionally with the full amount for a 2%/year improvement in the number of students advancing 1.2 years each year (1.2 because we have a lot of catching up to do). This is measured on the advance of the student's (which we are now tracking). So a teacher is measured by the students in their class, the principal by the students in their school, and the administrator by the students in the district.

As to the claim that it's impossible to measure teachers accurately – bullshit. Teachers manage quite well distilling their full evaluation of a student in their class to a single letter grade. We've got very good tools in place to measure the improvement in students. Not perfect, but they are quite good. And life always requires us to make decisions, even critical decisions, with incomplete & imperfect information.

And as to the worry that teachers will teach to the test – yep. The trick is to continue to refine the tests so that teaching to the test is exactly the instruction we want. The tests at present range from excellent (generally the math & science ones) to ok. The big problem is that testing for the ability to think, to write, to persuade (all key today) requires essays. And those are expensive to grade. But it can be done.

My question to BVSD is why would you not implement that above? Because the present approach is clearly a failure and small tweaks are not going to affect any real change.