The Effect of Evaluation on Performance: Evidence from Longitudinal Student Achievement Data of Mid-career Teachers

As states and districts seek to overhaul
teacher-evaluation systems, this NBER paper answers a salient question: Do
evaluations actually improve a
teacher’s performance? That’s one hope of reformers and unions alike—that clear
and regular feedback will help instructors improve their craft. Based on eight
years of data from Cincinnati’s Teacher Evaluation System (TES), the answer is
yes—in math, anyway. TES is an evaluation system that uses periodic,
unannounced classroom observations coupled with student-work portfolios. For
this report, researchers examined data from 2003-04 to 2009-10 to ascertain the
impact of TES on mid-career teachers (those in the system for five to nineteen
years). Building on performance-evaluation research, these analyses looked not
just at any immediate improvements incurred during a teacher’s evaluation
period, but at the long-term impacts resulting from participation in TES itself.
They do this by comparing achievement of students taught before teacher participation in TES with student achievement during or after
TES participation, while
also controlling for students’ prior achievement, teacher experience,
and relevant demographic variables. Though there were no significant
differences found in reading, teacher performance
in math improved both during the evaluation period and afterwards. For
example, a teacher whose pupils had typically scored in the
50th percentile on math tests before being evaluated begins to see
results in
the 55th percentile range in the years after evaluation. Teachers who
scored in
the lowest quartile on their evaluations showed the greatest
improvements. As
we rethink teacher evaluation, these are promising findings indeed. But
forewarned: A system like TES comes with a lofty price tag—roughly
$7,500 for
each teacher evaluated (over the course of the six studied years). If
or states plan on taking it to scale, some financial juggling will be in

Click to play

Click to listen to commentary on the NBER paper from the Education Gadfly Show podcast


Eric S. Taylor and John H. Tyler, “The Effect of Evaluation on
Performance: Evidence from Longitudinal Student Achievement Data of Mid-career
,” (Cambridge, MA: National
Bureau of Economic Research, March 2011).