School Discipline

School Discipline

School discipline indicators measure the percentage of students subjected to different types of disciplinary actions, such as suspensions or expulsions.

Research shows that exclusionary discipline practices place students at greater risk for numerous academic and personal consequences, including lower achievement, disengagement from school, and increased risk of dropping out.

How is it measured?

Discipline rates are usually measured in one of two ways:

  1. The percentage of students experiencing one or more disciplinary actions (e.g., number of students suspended or expelled divided by the total number of students); or
  2. The number of disciplinary incidents (e.g., the total number of suspensions or expulsions in the school). This measure is also often reported as the number of suspensions or expulsions per 100 students.

All school districts are required to report certain discipline data to the U.S. Department of Education’s Civil Rights Data Collection (CRDC) for each school. The CRDC collects data on the number of students (by race and disability status, as well as for English learners) receiving in-school suspensions, out-of-school suspensions, expulsions with and without educational services, corporal punishment, and referrals to law enforcement, respectively. For out-of-school suspension only, districts are required to report counts of students who were disciplined only once, and the number of students disciplined more than once, separately. Under ESSA, this information must also appear on school report cards.

What do the research/data tell us?

  1. Research shows that some groups of students — especially African American students and students with disabilities — are more likely to be subjected to disciplinary practices than their peers.1
  2. On average, schools with higher disciplinary rates have lower proficiency rates, and vice versa. This is especially true in high schools, and holds for students overall and for each group of students. But there are exceptions to this pattern — there are schools that have high achievement and high discipline rates, and others that have low achievement and low discipline rates.2 In other words, knowing a school’s proficiency rate does not necessarily tell us whether the school is suspending lots of students.
  3. Suspension and expulsion rates are much lower in elementary school than in secondary school. At the elementary level, discipline rates may look very similar in most schools, so — at least at the elementary level — the measure may not meet the “meaningful differentiation” requirements in ESSA (see “Indicators: What to Include in School Ratings”).3

What are the benefits and risks of including this measure in a school rating system?

  • Including discipline measures in school ratings could draw attention to exclusionary discipline practices that research shows a) negatively impact students and b) are not used equitably.
  • High discipline rates identify an actionable problem: Research has identified appropriate interventions that improve school climate and reduce suspensions and expulsions.
  • While schools with high proficiency and graduation rates generally have lower discipline rates, there are schools that do well on these measures, but suspend or expel high percentages of students. These schools could slip under the radar if school ratings don’t take discipline rates into account.
  • Discipline data may be easy to game. If schools know that their ratings depend in part on their discipline rates, they may stop reporting accurate suspension and expulsion rates.
  • If schools (and districts) know that reducing discipline rates will improve their school ratings, they may respond in unhelpful ways – e.g., by disallowing the use of suspension, without introducing practices such as restorative approaches or providing other teacher training.

If your state is considering including school discipline measures in school ratings, what questions should you ask? What should you watch out for?

What types of discipline actions will the state include?
Watch out for the inclusion of some disciplinary actions, but not others (e.g., including suspensions, but not expulsions). Doing so may incentivize schools to use the action that they aren’t held accountable for.

How will the state measure discipline rates? As a percentage of students disciplined? Or as the number of disciplinary actions?
Each of these measures shows something important, but misses something important, too. The first shows how many students are subjected to disciplinary action, but makes no distinction between students who experience that action just once, and those who experience it multiple times. For example, a student who is suspended once counts the same as a student who is suspended five times. The second method captures the total number of discipline actions, but does not show how many students are affected. This means that if a school has five suspensions, we don’t know whether one student has been suspended five times, or five students were suspended once. States will have to decide whether to use one of these approaches, or combine the two.

How will the state ensure that the data are accurate? Is there a process for auditing districts to verify that data are correct? What does that process involve?
If schools know that their ratings depend in part on their discipline rates, they may try to “game” their data — for example, by reporting inaccurate numbers of disciplinary incidents. If states are planning to hold schools accountable for discipline rates, they need to have quality controls in place to make sure the data are accurate, as well as a process for reviewing the data once it becomes part of the accountability system. For example, if discipline rates drop dramatically — rather than steadily — once schools are held accountable for reducing them, that may be an indication of gaming.

1 The Council of State Governments Justice Center and Public Policy Research Institute, “Breaking Schools’ Rules: A Statewide Study of How Discipline Relates to Students’ Success and Juvenile Justice Involvement,” July 2011.
2 Based on an Education Trust analysis of data from the Kentucky Department of Education and Massachusetts Department of Education.
3 Ibid.