Kirabo Jackson
by James A. Bacon
There is near-universal agreement that teachers are a key component of the schooling environment. And few observers would dispute that some teachers are better at their jobs than others. But there is little but discord and controversy over how to evaluate teachers. Teacher-advocate groups like the Virginia Education Association, for example, reject the use of standardized testing data to judge teacher performance. (See the previous post, “What the Virginia ‘Education’ Association Is Trying to Hide.“)
One body of research has shown that teachers who improve test scores (high value-added teachers) also improve students’ longer run outcomes such as high school graduation, college attendance and earnings. But other research shows that noncognitive skills not captured by standardized tests, such as adaptability, self-restraint and motivation, also influence adult outcomes. Thus, judging teachers on the basis of test scores only would overlook important contributions they may make in building students’ character.
Now comes C. Kirabo Jackson with Northwestern University, whose paper, “What Do Test Scores Miss? The Importance of Teacher Effects on Non-Test Score Outcomes,” has been published by the National Bureau of Economic Research.
Drawing upon rich administrative data for all public school 9th graders in North Carolina from 2005 to 2012, Jackson captured proxies for non-cognitive skills such as suspensions, attendance, course grades, and on-time grade progression, each of which is strongly correlated with certain non-cognitive skills.
“The results,” writes Jackson, “support an idea that many believe to be true but had not previously been shown — that teacher effects on test scores capture only a fraction of their effect on human capital. This underscores the need for holistic evaluation approaches that account for effects on cognitive and non-cognitive skill.”
Bacon’s bottom line: I have no ability to critique Jackson’s social scientific methodology, but I find his conclusions intuitively plausible. Common sense tells us that it takes more than intelligence to succeed in life. It takes grit, drive, determination, focus and self-discipline. Insofar as teachers may influence those behavioral traits, standardized test scores may not capture their full contribution.
Some might throw up their hands and say, “Well, there you go, there’s no point in evaluating teachers with test scores.” I would argue, to the contrary, if Jackson’s analysis survives scholarly scrutiny, his paper suggests that schools use a broader set of metrics to evaluate teachers than standardized test scores alone.
The larger truth still holds: Teachers vary in quality and effectiveness. Some teachers just aren’t cut out for teaching, and they should be culled from the profession. Because the profession is disinclined to judge anyone harshly, subjective evaluations are largely worthless. School systems need objective metrics by which to evaluate teacher performance — and ideally those metrics should made available to parents. Here in Virginia, if Standards of Learning (SOL) test scores or Student Growth Percentiles (SGPs) provide an incomplete portrait, then by all means let’s gather the data that can.
