In the last year, a national debate over whether there’s a better way to judge teachers has exploded. The Obama administration has pushed schools to use what’s known as “value-added data” to evaluate teachers and decide their pay.

A Los Angeles Times investigation last year pushed the issue out of academia and into the mainstream, inspiring strong backlash and fevered discussion.

The idea is simple: Measure how much each child improves over time, instead of simply how high they score.

It is a powerful way to examine the impact of educators and schools, proponents argue, because it strips away kids’ inherent advantages and disadvantages.

The idea is also deeply controversial. Statisticians debate whether it is technically possible to tease out how a teacher impacts a student from a multitude of other factors. Teacher unions say tests are a shoddy way to measure teaching in the first place.

San Diego Unified has turned away from Obama’s plan and panned the idea of using the data to rate teachers. It has embraced a softer touch: using similar information to study what gets good results.

Last week we surveyed our members (those who have donated to support to ask them:

How should schools use “value-added data” to measure teacher performance?

We sent our emails to the 818 members for which we have email addresses. Ninety-nine responded. (If you’d like to participate in our surveys, donate to become a member and enjoy this and other benefits.)

The question is a complicated one, and it generated a lot of thoughtful and complex answers, many sharing similar opinions:

  • Teachers should be measured, one way or another.
  • Unions are a problem.
  • It’s difficult to rate teachers based upon their students’ performance alone.
  • Why should teachers bear most of the credit or blame, when they have little or no control over the students’ lives and educations outside of school or in previous years?

Below are just a few of the member comments, sorted by broad themes.

Is Using Value-Added Data a Good Idea?

Bill Hampton and Sylvia Hampton:

All teachers should not be subject to the same evaluation model. A different method should be done for master teachers, experienced teachers, less experienced and probationary teachers. Whatever kind of testing data is used, it should not substitute for direct and on-going classroom observations by qualified administrators and master teachers. … You could have two teachers with the same raw data, but one could be a bad teacher in how they relate to the kids.

Tyler Cramer:

It’s no secret I believe value-added measures (i.e. correlations between specific educational inputs, including teacher efforts, and student achievement over time) can and eventually will be an essential tool for improving student learning rates. I am also aware of the technical and political challenges which currently inhibit the effective use of VAMs (Value-Added Metrics). Nonetheless, I think two points will become clear: (1) all education inputs (not just teachers) need to be evaluated at least in part by VAMs to determine their effects on student learning rates; and (2) financial resources must be more efficiently allocated to those specific educational inputs which have been proven by VAMs to accelerate student learning rates.

Paisley Cato:

It should not be used as the sole measure for evaluation, but it does have solid merit as a tool and should be used in a rational, regular method.

Monika Gardner:

There are so many flaws in using value-added data for teacher performance that we should not waste money and resources on this as a meaningful tool of evaluation.

What’s a Good or Bad Teacher?

Jeff Toister:

We need to have a working definition of “good teacher” before we can start evaluating whether individual teachers fit that description.

Allen Hemphill:

EVERYONE knows who the poor teachers are — the administration knows, their fellow teachers know, the students know, the classified staff know and the parents know. The only ones who do not know are the taxpayers who pay the salaries!

The Value of Tests

Kay McElrath:

NCLB’s obsession with standardized tests was and is a mistake. Standardized tests do little to measure the kinds of skills that are needed for current generation technology and careers. They’re a vestige of the 20th century industrial approach to K-12. Great teachers hate them. Poor teachers know nothing but to teach to them. Schools are damned if they dare ignore them.

Brian Adams:

I could support “value-added data” if they actually had quality tests that measured the skills that we care about most (critical thinking and analysis). But the tests, and testing procedures, are so flawed that I don’t think they can really be used as an accurate measure of “value-added.” As it currently stands, the tests push teachers to focus on test-taking skills rather than substance.

Richard Childers:

What gets measured gets managed.

James Long:

Different social and economic factors make it difficult to evaluate a teacher’s performance. I would bet you take the entire faculty from La Jolla High and transplant them at Lincoln High and the academic results would be the same or worse. They are entirely different worlds.

What Should Be Measured?

Joan Larsen:

The best way to evaluate the value of a teacher is to see the improvement in their students’ knowledge. I refer you to the dictionary for the definition of the word “teacher.” All the other stuff is BS.

Linda Canada:

It seems reasonable to include as one of many standards by which to measure teacher effectiveness, the new technique of “Value Added Data.” After all, isn’t teaching children how to learn one of the primary goals of teaching? Don’t we measure how well they have mastered this task by assessing what and how much the children have learned? It doesn’t seem such a big stretch to me to infer that there is a correlation between what children have learned and how effective their teacher is.

Controlling for Factors Other Than Teachers’ Skill

Oscar Ramos:

It is too difficult to measure the impact of each individual teacher, isolated from all the other factors in a student’s life. If my student attends an after-school tutoring program, the added value would be measured in my classroom, even though I may not the reason for it.

Nathaniel Cohen:

The only sensible way must use a method which eliminates the variability of each student’s learning ability. This is impacted by IQ, home environment, mental distractions of every kind.

Lee Terry:

How do you account for the mobility of today’s students, moving from school to school?

Charlie Powell:

Until we have methods or laws that require parents to do their job of parenting, including ensuring students take school work seriously and complete their work, you can’t hold teachers completely responsible for the students’ success.

Jeanne Brown:

It has always been difficult to teach students who have a harder time learning. Many schools know that it can be very depressing for a teacher to be assigned all lower-level classes. So the teacher who has all advanced classes has an exciting, invigorating day while the teacher with a lot of lower-level classes can have a very frustrating day. So is there any wonder why it is easier to find teachers to teach the advanced classes? And now they want to reward those teachers because their students do better? How will you ever find any teachers willing to take on the slower students? Not only is it frustrating and exhausting, now they will pay you less or threaten to take your job.

Maria Penny:

I taught high school physics for ten years in an excellent suburban high school in Maryland. In my opinion, the parents’ socioeconomic level and their investment in the child’s education since birth account for 75% of the child’s incentive to learn; the teacher and the classroom environment are the rest. I wouldn’t have minded being evaluated by the change in my students’ achievement because they came to me ready to learn and interested in studying physics. If there was a problem, I could call the parents and get support to resolve the learning issue. I don’t think the average teacher has a situation as good as mine was; it would be unfair to expect the average teacher to overcome parental indifference or inability to be involved in the classroom process.

Union Involvement

Gary Cristofani:

There can be NO proper performance measurements in a government-run, monopolistic, non-competitive conglomerate run by the very group (teacher labor unions) whose members are to be the evaluation subjects. So I personally would not waste too much time trying to implement this. They have a long history of co-opting and corrupting every education reform devised or implemented.

Mark Lewis:

This is the “toilet to tap” issue in teaching. No one understands how it’s really going to work, so the union scare tactics rule the day.

Bill Bradshaw:

As a former human resources director, I’ve often wondered how the teacher unions get away with preventing any effective performance-based evaluations of their members, as though the profession was uniquely hard to evaluate compared to, e.g., engineers and scientists. Here’s an inside secret: All performance evaluations are substantially subjective, but in the real world they are done nevertheless, because it’s necessary to cull out the weaker performers in any occupation.

Comments quoted here may have been edited for style, spelling and clarity.

I’m Grant Barrett, engagement editor for Drop me a line at, call me at (619) 550-5666, and follow me on Twitter @grantbarrett.

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.