The effects of teacher effectiveness on student achievement

Using data from Tennessee, North Carolina, and Washington State, a 2017 study published in the Statistics and Public Policy journal examines how teachers of various levels of effectiveness impact student achievement. Researchers analyze multiple years of educator data, using as many data points as were available for each teacher.

Not surprisingly, the authors find that the distribution of teacher effectiveness resembles a bell curve. As Figure 1 demonstrates, educators who are nearer to the middle of the curve have similar effects on student achievement, regardless of which percentile the teachers fall into. The impacts of those in the 45th and 55th percentiles are quite close, for example. This is not, however, true of those at the tail ends of the bell curve—teachers who are very bad or very good. Compared to more average educators, the impacts of those in the 2nd and 12th percentiles, for instance, are markedly different, with the bad-but-not-as-bad teachers producing much better student results. And the same goes for those at the high end. Students educated by teachers in the 98th percentile are much better off than their peers whose teachers fall into the 88th percentile.

Figure 1. Gains in student achievement associated with gains in teacher effectiveness

Source: Dan Goldhaber and Richard Startz, “On the Distribution of Worker Productivity: The Case of Teacher Effectiveness and Student Achievement.” Statistics and Public Policy (2017).

These findings have important implications for teacher policy, argue the study’s authors. From a student outcome standpoint, it might be much more cost effective to invest in efforts that concern teachers at the tail ends of the performance spectrum—the bottom and top 10 percent, perhaps—than implement policies meant to improve every teacher.

What this means for the real world is that districts might consider moving away from expensive teacher-workforce-wide strategies like professional development—which cost on average $18,000 for each educator per year yet does not lead to systemic improvement in teacher effectiveness—and moving toward policies that dismiss, reject, or improve the very worst educators, and retain the very best. States and districts are often complaining about their levels of funding and searching for more. Being smarter and more discerning in which policies they choose to implement is an important counterpart to that.

SOURCE: Dan Goldhaber and Richard Startz, “On the Distribution of Worker Productivity: The Case of Teacher Effectiveness and Student Achievement.” Statistics and Public Policy (2017).