A statewide, statistically tested analysis of what Michigan's 2016 Rethink Discipline law (Public Acts 360–366) actually changed, using state enrollment and discipline data from 2013 to 2025.
Comparing the pre-reform period (2013–2017) to the post-reform period (2017–2025, excluding the 2020–2021 pandemic outlier), the statewide student expulsion rate dropped 20.8% — from 0.1038% of enrolled students to 0.0822%, a reduction that's statistically significant at the individual-student level (Z = 13.87, p < 0.001). That translates to an estimated 2,137 fewer students expelled between 2017 and 2025 than pre-reform rates would predict — roughly 305 fewer expulsions per year.
The law also produced a modest but statistically significant shift toward in-school discipline: the ratio of out-of-school to in-school suspensions moved from 3.27-to-1 before the reform to 2.99-to-1 after (Z = 16.48, p < 0.001) — an estimated 22,320 incidents shifted from out-of-school to in-school suspension over the period, or about 3,190 incidents per year.
The analysis relies on Center for Educational Performance and Information (CEPI) and Michigan Department of Education (MDE) discipline and enrollment data, and accounts for a 2017 state reporting-compliance change that expanded suspension logging (explaining the volume jump in 2017–18 data) as well as the COVID-19 pandemic's disruption to normal discipline patterns in 2020–21.