Teacher evaluation systems and their reforms were built, in part, on the idea that rubric-based classroom observations, paired with feedback, would improve teacher performance and raise student achievement. Not every reform delivered on this promise, but there were bright spots. Tennessee’s evaluation system—the Tennessee Educator Acceleration Model (TEAM)—was among a small group of system reforms that were associated with gains in math and reading scores.[1] Whether TEAM is connected to other important outcomes, like student discipline, is a question worth asking.
In a new study, Seth Hunter of George Mason University, Adam Kho of the University of Kentucky, and I asked whether the teacher observations Tennessee assigns to teachers in TEAM affect student discipline outcomes (SDOs). We focused specifically on exclusionary SDOs (in-school suspensions, out-of-school suspensions, and expulsions).
Why discipline?
Teachers and administrators have considerable discretion over how students are disciplined, and those decisions can bring longer lasting effects. Students who receive exclusionary discipline are more likely to drop out of school, more likely to be incarcerated, and less likely to attend college.[2] These negative consequences are particularly alarming given clear evidence of racial bias in student discipline decisions.[3] If classroom management can be improved, and improving it can curb exclusionary discipline and the longer- term consequences, then it is worth investigating already-existing levers that may do so.
See NCTQ’s Classroom Management Framework for more on why classroom management matters and what teacher preparation programs can do to foster classroom management skills before teachers are on their own in the classroom.
What did we find?
Each year, Tennessee policy requires administrators to conduct a specific number of TEAM observations for each teacher, with more frequent observations of teachers who scored lower the prior year or who are earlier in their careers. A teacher rated 5 on the state’s 1-to-5 effectiveness scale receives a single observation, while a teacher rated 1 receives four. Teachers in the middle range of effectiveness receive two or four observations, depending on years of experience.[4] We draw on six years of Tennessee student and teacher data. The student data include every in-school suspension, out-of-school suspension, and expulsion. The teacher data include each observation a teacher received and its score, along with their overall effectiveness ratings and years of experience.[5]
Our study examined whether the total number of observations assigned to a school’s teachers was related to the share of its students who received at least one exclusionary SDO. We found that as state policy assigned more observations within a school, a smaller share of students received an exclusionary SDO. In the average Tennessee school during the study period, 8.45 percent of students experienced a suspension or expulsion. Within a school, an additional 25 policy-assigned observations was associated with a statistically significant reduction of about 0.3 percentage points in the share of students receiving an exclusionary SDO. For the average school, that is roughly a 3.5 percent decrease.
This overall reduction was driven primarily by in-school suspensions, which was expected. The behaviors that lead to an out-of-school suspension or expulsion are more severe, and therefore harder to disrupt, than behaviors that ultimately lead to in-school suspensions.
How do more observations decrease exclusionary discipline outcomes?
We tested several hypotheses and found that improved teachers’ classroom management skills account for part of the decline in in-school suspensions. In practical terms, teachers got better at managing and de-escalating student misbehavior, which reduced the need to issue an office referral that can ultimately lead to suspension. More policy-assigned observations presumably mean that teachers receive more feedback and principals gain a clearer view of what is happening across classrooms, which may help them identify school-wide areas to improve, like classroom management, and address them through professional development or other initiatives.
Importantly, these findings apply specifically to policy-assigned observations. When principals exercise discretion and observe teachers more often than policy requires, we do not see the same benefits. Rather, results suggested the opposite—more discretionary observations are associated with higher rates of exclusionary discipline. The likeliest explanation is reverse causation. Principals decide to increase observations beyond what policy determines in response to discipline problems that already exist. Because of that, we cannot separate any effect of these additional discretionary observations from the conditions that prompted principals to conduct them.
What is the policy implication?
Results suggest that policymakers or district leaders who want to reduce exclusionary discipline and, in turn, improve related life outcomes for students, should consider evaluation systems that determine the number of observations each teacher receives using two criteria: a teacher’s prior year performance and years of experience. Assigning observations using a set rule can plausibly strengthen classroom management school-wide and, in turn, reduce the share of students who are suspended.
Editor's note: This was first published by NCTQ.
___
Endnotes
1. Bleiberg, J., Brunner, E., Harbatkin, E., Kraft, M. A., & Springer, M. G. (2025). Taking teacher evaluation to scale: The effect of state reforms on achievement and attainment. Journal of Political Economy Microeconomics, 3(3), 568–610.
2. Bacher-Hicks, A., Billings, S. B., & Deming, D. J. (2024). The school-to-prison pipeline: Long-run impacts of school suspensions on adult crime. American Economic Journal: Economic Policy, 16(4), 165–193. https://doi.org/10.1257/pol.20230052; Sorensen, L. C., Bushway, S. D., & Gifford, E. J. (2022). Getting tough? The effects of discretionary principal discipline on student outcomes. Education Finance and Policy, 17(2), 255–284. https://doi.org/10.1162/edfp_a_00341
3. Hayes, M. S., Liu, J., & Gershenson, S. (2023). Who refers whom? The effects of teacher characteristics on disciplinary office referrals. Economics of Education Review, 93, 102376. https://doi.org/10.1016/j.econedurev.2023.102376
4. Certification status effectively distinguishes early career, apprentice, teachers with fewer than four years of experience from professional teachers, with four or more years of experience. Teachers rated 2 through 4 in the prior year receive four observations if apprentice and two if professional. This description reflects the policy in place during the study period (2012 through 2018). The underlying logic of assigning more frequent observations to lower-rated and less-experienced teachers remains in place.
5. The data also include student, teacher, and administrator demographics, as well as responses from the Tennessee Educator Survey.
