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

Saturday, June 6, 2015

"Partially Ineffective" and "Ineffective" Are NOT the Same Teacher Rating

So the first "report" came out about New Jersey's new teacher evaluation system, aka Operation Hindenburg. And guess who was on the TV talking about it?

Thanks to Briana Vannozzi for including me in her piece. There's much more to say, but I want to highlight one thing Vannozzi pointed to that several reporters have glossed over; here's a graph from the report:

The report itself (and most of the media reporting that has followed) talks about the approximately 3 percent of teachers who are Ineffective or Partially Ineffective. It then compares this number to the 0.8 percent of teachers found "not acceptable" under the previous system.

First of all: I have no idea where the state got this 0.8 percent figure. There is no documentation or data source cited in the report for this figure. The state does collect this information through its data systems, but I am unaware of any auditing that occurred to determine the validity and reliability of this figure.

Let's, however, assume it's correct; there is still no evidence that "not acceptable" in the old system is equivalent to "Ineffective or Partially Effective" in the new system.

And that's a problem.

Think about the term "Partially Effective"; yes, you are only doing things right some of the time, but not all of the time. If you were a total screw up, you'd be "Ineffective," and your administrators would either be applying a very intensive intervention or, more likely, showing you the door.

One of the myths that has been perpetuated by people who don't know how schools work is that teacher evaluations before "reform" were simple fine/not fine ratings with no differentiation. But that's simply not true: good principals in well-run districts always gave their teachers more feedback than a simple binary rating. Evaluation systems always included conversations with administrators about how a teacher could improve his or her practice. Teachers rated "acceptable" always had to have professional development plans designed to help with their continuing growth.

Undoubtedly, there were teachers deemed "acceptable" who were told that they were barely acceptable, and that they would have improve in order to avoid dismissal or tenure charges. As I wrote back in 2011:
Is there a way to get this benefit without eliminating tenure? You bet: streamline the dismissal process. Cap the time for a dismissal and appeals at 90 days. Send the cases to arbitrators who specialize in teacher dismissals. You'll cap costs and make it much easier to dismiss these hypothetical "bad" teachers.

Still, most agreed that the small numbers of tenure charges filed with the state are really only a fraction of the cases of low-performing teachers for whom the formal filing is a last resort, a vast majority of them eased out of the classroom as the complaints mount.
'You don’t see these statistics, but I would say that hundreds of teachers who receive the first tenure charges resign,” said Eugene Liss, general counsel to the Newark Teachers Union. "Maybe the case didn’t go all the way to Trenton, but many who sit with us, they end up leaving the profession."
Newark has a system in which teachers receiving unsatisfactory ratings are required to undergo additional training through Seton Hall University. Last year, it was 90 teachers, all but 12 of whom returned to the classroom, he said. Those 12 all resigned, none by tenure charges.

The plain fact is that any of the advantages of eliminating tenure could be gained by making a few simple changes to the system. And all of the advantages to keeping tenure cannot be replicated through other policies.
Again, the idea that all "acceptable" teachers were treated the same is a myth. So we really don't know if AchieveNJ is identifying more "bad" teachers than the old system.

Let's be blunt: this report is really a commercial the NJDOE made for itself, not a serious program evaluation. There are many things we still don't know about AchieveNJ, and I believe the only way to really evaluate the program is to turn the data over to an outside source and let them conduct a proper assessment.

Until then, let's all agree on one thing: the number of truly "bad" teachers in New Jersey is quite small. Most are probably working outside of "tested" areas -- math and language arts. Many may not even be teachers, but instead student support personnel. I'm all for requiring remediation for these folks; if that doesn't take, they should be dismissed.

But let's finally put to rest the notion that hordes of "bad" teachers roaming the halls of New Jersey's schools should be our primary concern in state education policy. We've got much larger issues to address.

It always seems to come down to this, doesn't it?

1 comment:

edlharris said...

In Prince George's County, Maryland, teachers are being assigned a rating that appears to be sliding.
As it was explained:
The rating scale used to determine ineffective/effective/highly effective uses a Z-score, with cuts at -1.5 and +1.5.

So, a teacher who got an overall score last year of 84 was rated "Highly Effective." This year, when the teacher's indicator score increase such that overall they are 84.9, they have been ranked "Effective."

The County Executive and superintendent have said the county needs to attract more teachers and hold onto the ones they have. The county has turned over 7,000 teachers in the 8 years. There are usually about 10,000 teachers