Tuesday, September 30, 2014

VAMs: Stupid Economist’s Tricks



The formula to the left will allow you to calculate your Value Added Score. Please don't forget to show your work.

I am not an economist. I had one intro to economics course in college. Everything I have learned about economics since then has come from reading Paul Krugman in the New York Times. I consider it a good month economically if my checkbook balances and my 401k doesn't tank.

But I am thinking of economists today because I just read a review by Krugman of a new book called Seven Bad Ideas: How Mainstream Economists Have Damaged America and the World, by Jeff Madrick. The book is a chronicle of all the wrong headed advice mainstream economists have been dishing out since before the debacle of 2008 and up until now. 

Here is a brief restatement of what the economists got wrong.
1.      Failed, despite all the warning signs, to predict the 2008 recession, and, in fact, argued that it could not happen.
2.      Failed to agree on a response to the recession once it did happen
3.      Opposed the stimulus package and raising the minimum wage as government interference in the free-market which would, according to these economists, right itself. How is that going?

How could economists get things so wrong? Well mainly, according to Madrick and Krugman, because of a slavish belief in the free market’s ability to manage itself, along with a faith based love of mathematical models that are elegant on paper, but have no connection to the realities of the real world.

Interestingly, while economists were not busy destroying the economy for all of us except the 1%, they have busied themselves with attempts to apply their cockamamie “models” to teacher evaluation schemes. The movement toward value-added measurements has been led by, you guessed it, economists. People such as Stanford’s Erik Hanushek, Harvard’s and Bill and Melinda Gates Foundation’s Thomas Kane, Harvard’s Raj Chetty and Columbia’s Jonah Rockoff.

Madrick apparently does not mention Value-Added Measures (VAMs) of teacher accountability as one of the “seven bad ideas”, but it is at the top of my list. Talk about a faith in an unrealistic mathematical model, VAMs take the cake. What really cheeses me off about this rat pack of economists is that many of them sit in tenured university positions, while they attempt to deprive teachers of their job protections with junk science.

If you have not already read Audrey Amrein Beardsley’s book, Rethinking Value-Added Models in Education Critical Perspectives on Tests and Assessment-Based Accountability, I recommend it to you. In it you will find a compendium of problems with VAMs. Another good resource is Linda Darling-Hammond's commentary here

In order for the “sophisticated statistical formula” of VAMs to work you need to make several assumptions. You must assume that
·         student learning is measured well by a given test.
·        student learning is influenced by the teacher alone.
·         students are randomly assigned to classes
·         student learning is independent from the growth of classmates and other aspects of the classroom context.

That’s a lot of assuming to do. Oh and by the way, none of these assumptions is well supported by current evidence.

Here are some other real world issues that are unaccounted for in VAMs.
·         School factors such as class sizes, curriculum materials, instructional time, availability of specialists and tutors, and resources for learning
·         Individual student needs and abilities, health, and attendance
·         Peer culture and achievement
·         Prior teachers and schooling, as well as other current teachers
·         Special needs of second language learners
·         Differential summer learning loss, which especially affects low-income children
·         The specific tests used, which emphasize some kinds of learning and not others and which rarely measure achievement that is well above or below grade level.

But the number one reason to reject VAMs as a measure of teacher accountability? They don’t provide any useful feedback to the teacher. Right now if you are a teacher with a low VAM score, your best advice would be to wait a year, because the scores are so volatile, you could get a high score next year. As a profession, we would be wise to stop chasing our VAM tail and move on to some real reforms in teacher evaluation that would provide formative feedback that could help us refine practice.

When it comes to economists I would like to propose a simpler model we could use to deny them their job protections. I take my model, not from economics, but from baseball: three strikes and you’re out.
·         Failed to predict the recession – Strike One
·         Failed to agree on how to fix it – Strike Two
·         Failed to support measures to ameliorate the impact – Strike Three
·         You’re Out, Mr. Economist! – No more tenure for you! Welcome to the “free market.”