The No-Stats All Star Retires

June 18, 2014

(Guest Post by Matthew Ladner)

Shane Battier, the man dubbed by Michael Lewis as the No-Stats All Star– has announced his retirement from the NBA at age 35 to take a college basketball analyst position with ESPN.  Battier never looked like much on the stat-sheet but when the statisticians got around to crunching the NBA they discovered that all he does is little things-like help his team win basketball games.  Battier-type “White Space” employees raise important questions about how to approach employee evaluation including teachers.

John White and I spoke on a panel together a few years ago and the topic of evaluation came up.  I sounded a note of caution but Superintendent White saw my bet and raised me by opining that we were in danger of making a fetish out of value added scores and that ultimately we should rely upon the professional judgement of administrators informed by data rather than merely the data itself. At least that is how I interpreted what White said, and if so, I agree with him.

Greg has been saying all along that ultimately this system requires choice.  Give parents meaningful choice, let Principals hire their own teams, have Superintendents evaluate Principals on the basis of the health of their school.  This strikes me as not only as the best way to do teacher eval, but also the only way to create a system to recognize the value of woefully under-appreciated highly effective instructors.  To choose another sports analogy developed by Michael Lewis, the pay of Left-Tackles took off after the advent of free-agency in the NFL.  Once a true market for players had been established, guys who had the skills to block a Lawrence Taylor found themselves in high demand, whereas the old system kept their compensation under wraps.

There are only a few states where we might be inching towards meaningful levels of parental choice, probably fewer still if any where the school leader has anything approaching a free hand to choose their own team. Mechanistic programs that attempt to identify and reward and remove instructors will be better than a unconditional tenure and dance of the lemons system but will never match a system in which trained professionals with healthy incentives exercise professional discretion. The Heat for instance hired Battier because they understood that there is a great deal more going on than the stat sheet, and won a couple of championships.

The primordial soup is slowwwwly starting to bubble…

Now imagine a burnt out and disgruntled Charles Barkley riding the bench of the Heat as a player in 2014 drawing a bigger salary than LeBron because the coaches can’t make best use of their salary cap…


Room for Debate on Teacher Assessment at the NYT

September 6, 2010

(Guest Post by Matthew Ladner)

Lance and Marcus enter a bar brawl over at the NYT on value added assessment. Watch out for the guy holding the pool stick upside down!


NYT on LA Times Value-Added Bombshell

September 2, 2010

(Guest Post by Matthew Ladner)

Check it out. To do this right (aka as best we can) schools need to have multiple tests to get much more data and thus much less error. The state of the art with this involves teachers drawing up their own common assessment items based on state academic standards, giving monthly assessments, and tracking student learning gains together as departments. Teachers can own this process, and either remediate or weed out ineffective instructors themselves.

Fantasy? Nope- it is already happening, and it is not rocket science.

Even improved scores should also be only a (big) part of an assessment, and the goals should be communal as well as individual.

All this reactionary hand-wringing about the measures not being perfect is a waste of time. We need to get these measures as close to perfect as we can and then run with them. Stringing together three crappy state tests in a row is NOT as close to perfect as we can get, but it is much better than nothing.

I’m not willing to settle for better than nothing. Rock star pay for rock star teachers or bust baby!


Rock Star Pay for Rock Star Teachers Part Trois

May 7, 2009

(Guest Post by Matthew Ladner)

A few months ago I got an angry email from an Arizona teacher claiming that her school had been terribly underfunded, and that she had 32 students in her classroom. I wrote to her:

If you have 32 children in your classroom, my first question is what is your school district doing with all of that revenue?

The JLBC put the statewide spending per pupil in Arizona at $9,399. A classroom of 32 at the statewide average would mean $300,768 in revenue from the students in your class.

Her response:

1-teacher, 1ELL teacher, 1 Special ED teacher, reading specialist, principal, janitor, secretaries, music, art, PE, computer teacher, Cafeteria workers, Para-educators, paper, textbooks, hands on science materials, Computers (this is the 21st century learning) building up keep, electricity, water, tables, chairs , etc…..

She forgot to mention administrative salaries from central command. There is one tiny little problem with all of this. According to the 2007 NAEP, 44 percent of Arizona 4th Graders scored BELOW BASIC in reading.

In other words, as Dr. Phil likes to say, how’s that hiring your average teacher from the bottom third of university students and supplementing them with crowds of others working out for you?

Shape up people!

The sad reality of American public education is that our schools have become revenue and employment maximizers that all too often are profoundly unfocused on the bottom line: student learning.  Public schools ought not to be jobs programs, but focused on their mission of equipping students with the academic skills necessary for success in life.

So, if you’ve got $300,000 in revenue from a classroom (many states have more) call me crazy, but I think you’ve got $100,000 for what research shows to be going away the most important factor for student learning gains: a high quality teacher. When I say a high quality teacher, I mean a verified high quality teacher whose student learning gains are being tracked over time by both administrators and parents on a continuous basis.

The best platforms for ongoing value added assessment are web-based data products that allow teachers to develop common assessment items based on state standards. If there are state standards for a subject, you can do value added analysis on it. When schools really get going on this, they give monthly assessments. This gives ongoing assessment data that greatly drops the amount of error (using only state tests, some of the pioneering value added models require 3 years worth of data).

Overall, it isn’t very hard to imagine a system that would improve upon the status-quo in these practices. We can no longer in good conscience socially organize our efforts to teach children to read along the lines of: let’s hire an army of people who want job security and summers off , do absolutely nothing to reward merit, and hope for the best.

This must change, and it will change.