A Statistically Valid Way

I’ve been arguing in my writing–both candidacy paper and literature review–that we (researchers, administrators, policy makers, etc.) need to adopt a new view of teachers as professionals, people who actually know what they are doing, people who work in incredibly complex environments.  This is not really th kind of job that would seem to be best described by a numerical rating.  But, for the second time just today, I read an article about school divisions who are trying to do just that…base teacher salaries and raises on a number.

Here are the first few paragraphs from the article in today’s Dallas Morning News about the Dallas school district’s plan to assign a Classroom Effectiveness Rating (CEI) to teachers:

“The Dallas school district wants to base a $22 million employee bonus project on an arcane, decade-old teacher rating system that, until now, has been largely ignored.Through a complex statistical analysis, the evaluators say, they can convert student test scores into a number, 1 through 100, that measures a teacher’s effectiveness. The district’s plan would give bonuses of up to $10,000 to teachers with the best “Classroom Effectiveness Index” ratings, or CEI for short.

DISD administrators say the CEI is a statistically valid way of identifying good teachers based on their students’ scores on standardized tests.”

These CEIs have been around for a decade, but no one really understood them or what they were good for.  I’m so glad that DISD has figured it out!  Not surprisingly, teachers are a bit mistrustful.  And, ultimately, the debate gets at the larger philosophical debate about teaching:

This sort of information, if reliable, is gold to education reformers, some of whom believe the biggest factor holding back public education is the quality of its teaching force. Evaluation systems that claim to quantify the quality of individual teachers are despised by many educators and teacher union officials who say teaching is more an art than a measurable science.”

Art? Science? A mix of both?  I’m probably somewhere in the middle, but the idea that somehow what I do every day in the classroom can get translated into statistics that yield one score for my effectiveness just seems ridiculous to me.  Am I that effective for every student?  Or is it just overall?  In this age of NCLB, shouldn’t I get an effectiveness score for each subgroup?

I think teachers should get paid more…not a bonus for some bogus statistic.

Leave a Reply