K-12 Student Infractions and Discipline

Highlights

Record Level: student x incident x year 
Record Count: 111,000 records/year 
Years Covered: 2011-present 
Population Coverage: public school students in Michigan

The State of Michigan provides discipline data in a single dataset that includes all reported disciplinary incidents in Michigan schools from 2011 to present. Suspensions are required to be reported in all years for students with IEPs, and expulsions are required to be reported in all years for all students.

There is a large increase in reporting starting in the 2017-2018 school year, when reporting suspensions for all students became mandatory for all Michigan schools. However, this policy ended after the 2017-2018 school year, so beginning in 2018-2019 schools were no longer required (but were still encouraged) to report suspensions for students without an IEP. In all years, schools were able to voluntarily report suspensions for students without an IEP. A suspension is an event resulting in a student’s removal from their usual education setting for one half of the school day or longer.

The raw data included identifiers such as student Id (RIC), school year, school and district code; incident date, and type of incident; type and length of disciplinary action; and any administrative follow up to address the incident (e.g. referred to mental health agency). These data are intended to be unique at the student x year x incident level, but additional de-duplication will be required to get the data to this level of uniqueness as an explicit incident identifier is not provided. Evidence suggests deduplication can be accomplished by grouping at the student x year x school x district x incident date level.

Incident reporting changes after the 2010-11 school year, with 34 incident types used in 2010-11, but only 12 aggregated types used in the subsequent years. Definitions of each can be seen in the tables below. The majority of incidents in each year are coded under the “Other” designation. In addition, each incident is assigned to 1 of 5 disciplinary actions (in-school suspension, out-of-school suspension, removal by hearing officer, unilateral removal, and expulsion). Definitions of each can be seen in the tables below. While there are some guidelines and requirements around the severity of the disciplinary action, in many cases this is left to school/teacher discretion. Researchers should note that there are no disciplinary codes in the 2010-11 school year.

Out-of-school suspensions are the most common type of disciplinary action, encompassing approximately 75% of disciplinary actions across years. In-school suspensions account for an additional roughly 22-25% of disciplinary actions, and expulsions account for 1-2% of disciplinary actions. There are very few instances of unilateral removal or removal by a hearing officer.

Data Quality & Summary

There are a number of unique features of the discipline data that researchers should note. First, as mentioned above, there are no disciplinary codes for the 2010-11 school year, so those values will show up as missing for those years. There is no missingness for disciplinary codes in other years. There is no missingness on the incident codes across years (though most incidents are coded in the “Other” category). The length of the disciplinary action (ranging from 0.5 days to 888 days) is missing for 2010-11, but there is no missingness in other years.

Second, reporting of suspensions for students without IEPs was only mandatory for schools in the 2017-18 school year, but schools could report this voluntarily before and after, which led to a large increase in suspensions in 2017-18 and a decrease thereafter. There is also a large decrease in incidents reported in the 2020-2021 school year, which is likely related to the COVID-19 pandemic and virtual schooling.

Finally, the US Department of Education collects school-level aggregate data on discipline through the Civil Rights Data Collection (CRDC). These data are collected every other year and are collected for students both with and without disabilities. School districts report these data directly to the Office for Civil Rights (OCR), so there is no guaranteed relationship between the data reported to CEPI for inclusion in the K-12 discipline dataset and the data found in the CRDC. These data are the most correlated in 2017-18, when disciplinary incidences were required for all students (r=0.84 for out-of-school suspensions (OSS) and r=0.75 for in-school suspensions (ISS)). In general, however, correlations are high across years for students with IEPS (r~0.85 for OSS and r~0.75 for ISS), but low for students without IEPs in years where reporting for students without IEPs was not required (r~0.15).

Current reporting guidelines for incident types (post 2010-11):

Disciplinary Action Descriptions