Test Score Incentives Don’t Work

By Michael Diedrich, Policy Associate, February 9, 2012 at 11:00 am

Having talked about why linking test scores to retention or pay misunderstands the theory of motivation, it's time now to look at what's happened when schools have tried linking teacher pay, promotions, or retention to test scores.

Minnesota's moving down this path; we're developing a statewide evaluation system that mandates at least 35% of a teacher's evaluation be linked to test scores while we watch the legislature argue about dropping seniority as the determinant of teacher retention. What would replace seniority? No one's sure, but there are certainly plenty of people who'd like it to be test scores.

We aren't the first place in the country to try this, and the data we've gleaned from other places where it's been tried don't support pursuing this course. As one very concrete example, New York City adopted a “performance pay” plan linked to test scores for three years. They abandoned it a few months ago when the RAND Corporation found zero impact of the program.

Nor is New York alone in this. More comprehensive studies of value-added models (VAM) – a “fairer” version of pay-for-scores that judges teachers based on the growth of their students rather than absolute mastery – find that they're too unreliable to work.

One study [PDF], sponsored by the American Educational Research Association and the National Academy of Education and focused on Houston, found “most researchers have concluded that value-added modeling (VAM) is not appropriate as a primary measure for evaluating individual teachers.”

This is reinforced by a primer [PDF] from Henry Braun, writing on behalf of the Policy Information Center of the Educational Testing Service (the people who create the GRE and administer the SAT). He concludes that “VAM results should not be used as the sole or principal basis for making consequential decisions about teachers (concerning salaries, promotions and sanctions, for example).” [Emphasis his.]

These are just a few examples of research finding test scores an unsuitable basis for improving teacher effectiveness. While there isn't much formal research on this, we also know that most charter schools in Minnesota don't use seniority, and for the most part their results aren't any better than nearby public schools.

We must demand clarity about what would replace seniority as the standard for teacher retention. Too many people want it to be test scores, but we now know that just doesn't work.

Posted in Education | Related Topics: K-12 education  Student Assessment  Teachers  Teacher Assessment