However, without additional resources, raising the top tier of teachers to $100,000 or more would necessitate adjustments on the bottom end to compensate for that. But just how much would that be, and would that mean the same would be spent on teacher salaries? I decided to take a look, using the much-talked about model from Harrison, Colorado, which recently adopted a compensation system that was approved by its teachers and required the hiring of several new assistant principals to implement.
Note: The infrastructure necessary to implement a really thorough and effective evaluation system will be a topic of a future post.
In it, teachers are compensated within bands of pay based on evaluations. There are five categories:
Novice, Progressing, Proficient, Exemplary, and Master. Novice teachers make $35,000 a year. Progressing teachers receive $38,00-$44,000. Proficient teachers make between $48,000 and $60,000. Exemplary teachers receive $70,000 to $80,000 and the Master teachers make $90,000. These aren't perfect bands of compensation; there are fixed amounts depending on how you rank in each category, but for the sake of argument, let's assume they represent true bands and the evaluations will allow folks to determine precisely where the teacher should be paid along that band's compensation continuum.
Ohio's new teacher evaluation system envisions a four-tier system of Accomplished, Effective, Developing and Ineffective. So, I decided to take a run at translating the Harrison system into Ohio's, using the same pay scale (adjusted for cost of living, which is a bit more than 3% higher in the Harrison system than Cleveland).
Since the names and categories are different between the two states' systems, I took some license to say that ineffective teachers in Ohio would represent the Novice and bottom categories of Harrison's Progressing designation. Ohio's developing teachers would be the top Progressing and bottom two proficiency categories in Harrison. Ohio's Effective teachers would encompass the top two categories of proficient teachers and the bottom category of exemplary teachers in Harrison. And Ohio's Accomplished teachers would equal Harrison's top Exemplary category and its Master category.
I then assumed that 15% of teachers would rate Accomplished or Ineffective, with 45% being effective (so 60% are effective or better) and 25% being developing (meaning 40% aren't effective). In that case (assuming even distribution of salaries within the bands), Cleveland would spend about 6% less overall on its teachers (using the 2011 Cupp Report FTE figure). If there's a perfectly even distribution between the categories (25% of teachers rank in each), there would be a 7% reduction in the overall amount Cleveland pays teachers. If the 45% of teachers are developing rather than effective (meaning 60% of Cleveland teachers aren't effective), then it would be an 11% reduction in salaries. And if half the teachers in the ineffective category are let go, for the evaluation system contemplates that teachers who are ineffective after a year of peer review would be let go, then it would be a 15% reduction in salaries.
Again, all this assumes the same revenue stream and the same number of FTE teachers in Cleveland as there are on the 2011 Cupp Report. And this is all just crunching publicly available data; it's not looking at what the overall costs actually are and what they'll actually be. I just wanted to see if, hypothetically, rewarding the top teachers could actually cut overall teacher costs. And it appears you could do that.
What does this mean? It means, if you raise salaries for the top rated teachers -- and keep those top ratings a fairly exclusive club -- without an accompanying increase in revenue, you may drive down your overall teacher cost. It's counter intuitive, but that's what it appears to mean. In Cleveland, it would be an overall reduction of between 6% and 15%, depending on the distribution of the ratings. The only way they would be more expensive is if roughly 70% of teachers rate effective or higher, with about an equal 35-35 split between the top two categories. Keeping the top rating at 15% would mean 65% of teachers would need to be in the effective category to keep the overall costs about the same.
Note: While teacher evaluations have historically been very generous (with some districts putting upwards of 99% of their teachers in the top category under the old "fails to meet" "meets" or "exceeds" expectations system), the whole point of the new, four-tiered system is constructing a more realistic distribution of teacher performance. So it's unlikely that 99% of teachers in a district will rate Accomplished, for example. It will likely be a much more limited number.
This brings up all kinds of questions. Such as, what happens if a district is facing a deficit either due to an economic downturn or continued state divestment from education? This system would allow a district to cut their overall compensation figures pretty flexibly without losing teachers. They would simply need to put more teachers in the bottom two categories. While I'm all for keeping teachers in the classroom during tough budget times, without checks, this system could incentivize artificially bad evaluations to do it. And that can lead to problems.
What if 70% of the teachers should rate effective or higher under the new evaluation system? Would the temptation to limit cost be too great for administrators to evaluate the teachers as they should and put 70% in the lower categories instead?
And if you artificially de-flate their ratings, would teachers then be in the position of facing potential disciplinary action because they dropped a category, even though the drop was budget, not performance driven? And doesn't it undermine the credibility of the evaluation tool if the administration games it to save money?
Again, I'm all for good, objective evaluations of teachers and administrators. I think teachers and administrators are as well. My concern is that any system that's developed in Ohio or Cleveland should be of the highest integrity and not able to be gamed for budgetary purposes. Evaluation must produce true results in order to be most effective at improving teaching.
If significant numbers of teachers under the new system end up being highly rated, they should be rewarded, not graded on a curve based on a district or state's budgetary constraints.
