I will protect your pensions. Nothing about your pension is going to change when I am governor. - Chris Christie, "An Open Letter to the Teachers of NJ" October, 2009

Wednesday, September 4, 2013

What the Newark Merit Pay Experiment Is Telling Us

Right now, in the Newark, NJ school district, a grand merit pay experiment is underway. Understand that merit pay has never worked in schools and most likely won't work again in the Brick City. Still, the contract was negotiated and everyone signed off on it; merit pay is a reality in Newark for two more years, like it or not.

We may as well use this time to thoroughly vet the Newark merit pay program to see if it works. Which means that someone with the chops to handle a serious policy analysis should be releasing regular, yearly reports to the public about how this program is faring.

Until that day comes, however, we can look at the data we do have to draw some conclusions. Case in point:

Lisa Fleisher of the Wall Street Journal has followed the story closer than just about any other reporter. And she recently released some very interesting information about Newark's distribution of merit pay: 

Last week, 190 Newark public-school teachers learned they’d be getting bonuses in a controversial merit-pay program funded by Facebookcreator Mark Zuckerberg‘s foundation. Not all of Newark’s 3,200 teachers were eligible to begin with, and even fewer – only 11 teachers – qualified for the full bonus amount of $12,500.
Here’s a breakdown of how Newark’s bonuses were handed out, and more about the program.
Opt-out
The new contract approved in November by the Newark Teachers Union was a big leap of faith for some teachers. The district and union agreed to allow some teachers with advanced degrees to opt-out of the new pay scale, and for some it made more financial sense to stick with the old pay system. Only about 20% of the 1,300 teachers with masters and doctorates chose to try out the new system. Newark superintendent Cami Anderson told NJSpotlight she was happy with the figure.
[...]
Performance
Teachers had to score highly effective on the district’s four-tiered evaluation system to get a bonus. But regardless of which pay scale teachers were on, they were only guaranteed an annual experienced-based pay bump if they scored effective or highly effective. Here’s a breakdown of how all teachers rated, both those on the new pay scale and those who opted out:
  • Highly effective –11%
  • Effective – 69%
  • Partially effective – 16%
  • Ineffective – 4%
All 190 teachers who received bonuses were rated highly effective.
Let's play around a bit with the numbers. There are five data points I take away from this report:

  1. Total teachers: 3200 
  2. Teacher getting bonuses: 190 
  3. Teachers w/advanced degrees: 1300 
  4. % of teachers w/adv degrees who opted into merit pay: 20% 
  5. % of teachers rated Highly Effective (HE): 11% 
That is enough information for us to be able to draw some conclusions. You can check out my calculations at the end of the post, but here -- assuming Fleisher's numbers are correct -- are the big takeaways:
  • The pool of "merit pay eligible" teachers has 2160 members; 190 of those got merit pay because they were rated "Highly Effective."
  • The pool of "opted out of merit pay" teachers has 1040 members; 162 of those were rated "Highly Effective."
In other words, about 9% of the teachers who were eligible for merit pay were HE; however, 16% of the teachers who opted out of the merit pay system were HE.


Remember: according to the Memorandum of Agreement between the NTU and the state-run district,  only teachers who were more senior and had earned advanced degrees were eligible to opt out of the merit pay system; teachers without masters/doctorates and who were new to the district had to opt in. But one group was clearly more likely to earn a rating of HE than the other.

Newark teachers who are more senior, have advanced degrees, and who opted not to compete for merit pay were more likely to get a rating of "Highly Effective" than newer teachers without advanced degrees who were competing for a merit pay bonus.

Here are two different ways to look at the extrapolated data:


The numbers of HE teachers in and out of the merit pay pool are similar, but there are far more teachers not HE in the merit pay pool than out of it. Therefore...


The percentage of teachers out of the pool who are HE is higher than the percentage of teachers in the pool who are HE. Were I the state superintendent, Cami Anderson, I'd be wondering what I needed to do to keep my senior teachers with advanced degrees, as they are more likely to be highly effective.

Is it possible I'm missing some variable in this analysis? Absolutely -- which is why we need to release as much data as possible about the merit pay program in Newark. Only then can we determine whether merit pay is working and why.

No teacher's union in New Jersey, or anywhere else in the country, should accept any agreement containing merit pay unless and until the Newark district releases as much data as possible about the state-run program.

Isn't that fair? Isn't that prudent? Isn't that reasonable? I mean, we're making decisions based on data, aren't we?

Aren't we?

Er...

Um...


Well...

I'm sorry, what was the question?


Calculations:

Total teachers 3200
Teachers getting bonuses 190
Teachers w/advanced degrees 1300
%Teachers w/adv degrees who opted into merit pay 20%
%Teachers rated Highly Effective (HE) 11%

Teachers w/o adv degrees 1900
Teachers w/advanced degrees who opted IN 260
Teachers w/advanced degrees who opted OUT 1040

Sumcheck: Total Teachers 3200

Teachers in merit pay pool 2160
Teachers out of merit pay pool 1040

%ALL teachers who got HE 11%
Teachers who got HE (in & out of merit pay pool) 352
Teachers who got HE IN merit pay pool 190
Teachers who got HE OUT of merit pay pool 162

%Teachers who are HE IN merit pay pool 9%
%Teachers who are HE OUT of merit pay pool 16%


ADDING: I'd be very curious to see the percentage of HE teachers in of the pool of teachers with advanced degrees who opted into the merit pay system. How much do you want to bet it's higher even than 16%? That there is some self-selction going on?

NPS and NJDOE: data wants to be free. Release the numbers!

4 comments:

Anonymous said...

It would be interesting to know how many of those that were rated highly effective were actually evaluated by the standardized tests and out of the HE + subject (23) how many were Math teachers? It is interesting to note that English teachers cannot get the HE + subject bonus but Spanish teachers can!

20% were rated below effective, which means (.2*3200) 640 teachers did not increase in step. Using the average step increment on the Universal guide 13-14, ($2652) this means they saved approximately $1,697,280. If the bonuses cost them 1.4 million then they are still $300,000 ahead of the game without spending any of the FB money!

giuseppe said...

From bluejersey.com: What does the Newark community think today after Christie's insult? On NJToday Governor Christie was asked yesterday whether he would renew Cami Anderson's contract as Superintendent of Newark schools. He said he and Commissioner Cerf would. He added, "We don't care about the community criticism. We run the school system in Newark, not them." Callous and insensitive maybe? How long can he get away with statements like this?
Wow, did Christie actually say that? Incredibly arrogant and nasty! Double wow!

Duke said...

Anon: Great point which I will now steal!

g: See my next post.

LI Teacher said...

Your point that teachers with advanced degrees and more experience are much more likely to be rated Highly Effective is well taken.

However, I think the more interesting point that can be taken from this data is that teachers who opt out of the merit pay pool are almost twice as likely to be rated Highly Effective as are teachers who opt in. Supervisors have a financial incentive to rate teachers lower on the HEDI scale, and are more likely to give an HE rating if it doesn't cost the district any extra cash.