Gates Foundation Researchers Say They Know The Best Way To Evaluate Teachers

Candian Pacific / Flickr

Gates Foundation researchers say they believe schools can accurately assess teacher performance using a statistical formula.

The Gates Foundation says teacher performance can be accurately evaluated using data-based statistical formulas, but the best teacher evaluations also include student ratings and classroom observation.

That’s the conclusions from a three-year, $45 million study of a number of big school districts across the country including Hillsborough County, Charlotte, Dallas, Denver, Memphis, New York City and Pittsburgh.

The most definitive conclusion is likely to be the most controversial. Gates researchers say that a teacher’s so-called value-added scores accurately predict a student’s future performance.

Value-added uses a complex statistical formula which includes a number of factors to predict how a teacher will affect a student’s performance. The scores have been criticized for their large margins of error, year-to-year swings and their heavy reliance on standardized test scores.

But Gates researchers say value-added is essential to any teacher evaluation.

“The research confirmed that, as a group, teachers previously identified as more effective caused students to learn more,” the report concludes. “Groups of teachers who had been identified as less effective caused students to learn less.”

Critics are still picking apart the research, but one early view is that the Gates report is a self-fulfilling prophecy.

“The researchers find – to no freakin’ surprise – that prior year value added is, among all measures, the best predictor of itself a year later,” Rutgers University professor Bruce D. Baker wrote on his blog, School Finance 101. “Wow – that’s a revelation!”

The Gates report also tries to evaluate how much weight to put on their three recommended evaluation components.

The model that boosts state test scores the most would use state testing gains as 81 percent of the total evaluation score (Most Florida districts use 50 percent). Student surveys would be 17 percent of the total score and observations 2 percent of the total score.

The most reliable model weights the three components equally, the researchers said.

One other interesting note: Nate Silver, the most well-known practitioner of statistical analysis, had some interesting comments on data-based teacher evaluations.

Silver took to Internet site Reddit yesterday for one of their “ask me anything” sessions.

The first question was about value-added. Here’s the question, and Silver’s response (h/t to Larry Ferlazzo):

GrEvTh 971 points 20 hoursago

What are your thoughts on data-driven metrics for teacher evaluation? Do you think a system that accurately reflects teacher value could ever be created, or will it always be plagued by perverse incentives (teaching to the test, neglecting certain types of students, etc)?

• permalink

 

[–]NateSilver538[S] 771 points 18 hours ago

There are certainly cases where applying objective measures badly is worse than not applying them at all, and education may well be one of those.

In my job out of college as a consultant, one of my projects involved visiting public school classrooms in Ohio and talking to teachers, and their view was very much that teaching-to-the-test was constraining them in some unhelpful ways.

But this is another topic that requires a book- or thesis-length treatment to really evaluate properly. Maybe I’ll write a book on it someday.