Pay for performance in education

Today the Wall Street Journal reports that School ‘Bonus’ Plan Comes Up Short.

In short, $57 million has been spent in a school bonus plan in New York City Public Schools since 2007 to see whether it could improve test scores.  It didn’t.

As pointed out in the article, part of the problem might be with the bonus plan design.  Generally, the bonuses were not paid to top-performing teachers.  Rather the bonuses were pooled at the school level based on test scores and distributed evenly.  That might be part of the problem.

Personally, I don’t expect much test score improvement from such pay-for-performance attempts for two reasons.

First, I don’t think test scores are a good measure on which to base performance.  There are too many variables that impact test scores when averaged at a teacher or school level.

I don’t choose new restaurants based on test scores, nor do I choose which school to send my kid to based on test scores.   I generally choose both based on reputation and recommendations of my friends and family.

The better measure of performance is the percentage of parents who would recommend the teacher to other parents and/or the number of parents who request a specific teacher.

In a world with a competitive education market, the teachers who garner high recommendations/requests from parents are rainmakers for the school.  Administrators would generally hire, retain and pay teachers who bring in students.  This would act as a buffer against the politics and arbitrary decision-making they currently fear from their administrators.  There would be no need for the faux and limited-dimensional accountability system that branches up to the highest levels of government.

Second, I don’t think pay-for-performance operates in the manner many people think it should.

Generally, when I hear pay-for-performance discussed, it seems people believe that it is supposed to improve the performance of the existing teacher population.  They seem to think it will make bad teachers mediocre and mediocre teachers better.

But, that’s not how pay-for-performance can improve education quality.  It improves education quality by attracting better and more talented teachers.

If you put me on an NBA team, do you think paying me more to produce better results will cause me to produce better results?  Probably not much.  The NBA pay-for-performance works because it attracts people with immensely greater basketball skills than most.