Challenge Periods in School Discipline: Identifying Consistent Patterns in out-of-School Suspension Usage throughout the School Year

How does the use of exclusionary discipline vary over the course of a school year? Studies of out-of-school suspension rates typically target annual usage and changes over multi-year periods. Empirical efforts have not yet examined the within-year patterns sometimes professed by educators and school social workers, whereby after a “honeymoon period” of low misbehavior early in the year, there is said to be an increase in referrals and suspensions as the year progresses. Because to date, no systematic analyses have estimated these patterns, schools and districts do not have an evidence-base from which they can anticipate challenge periods and strategically deploy targeted resources at key points in the year.

In response, the current study leveraged a multi-year dataset from a large, diverse, and urban middle school in the northeast United States. Model estimation analyses examined the number of suspensions administered each week of the school year over a five-year period, and then fit linear and non-linear models to the pattern to estimate the best fit over time. Results suggest that suspensions follow a bimodal cubic sequence: administrators issue out-of-school suspensions more frequently during two “challenge periods”, which run from mid-October until the Thanksgiving break (roughly weeks 9 through 14), and then again between early May through the end of the school year in early June (weeks 35 to 41). Over five school years the cubic function representing this pattern were consistently significant (p<0.001), and explained between 19% and 58% of variation in suspension rates annually. An analysis of referrals confirmed this cubic pattern over the five-year period, although there was more variation year-to-year.

This is the first study we are aware of that has produced precise estimates of suspension and referral patterns across and within the school year. The results here have important implications for researchers, as additional studies across multiple schools and geographic contexts are needed to verify the patterns observed here. Qualitative studies are also needed to identify the contextual factors that contribute to rising suspension rates in challenge periods. Of particular import is determining whether these patterns are attributable to either perceived or actual student and teacher behavioral changes, and which environmental factors contribute to these changes in behaviors or perceptions of behaviors.

Although results here are preliminary, early feedback from practitioners suggests that they hold face validity with teachers and social workers in schools. If validated by future research, these challenge periods can be anticipated. As a result, supports around positive classroom management training, restorative practices, mental health counseling, and relevant program referrals can be strategically allocated by school leaders and social workers to maximize the effectiveness of what are often limited resources in urban school contexts. This work is especially important considering the overwhelming evidence that exclusionary discipline is harmful to youth. As such, a nuanced range of resources to support non-punitive alternatives are desperately and strategically needed in these challenge periods.