Behavior: procedures, rules, and protected time · Science of Learning Institute

This page covers one thing: the behavior that gets in the way of teaching. It is a first page, not a full library, and it is graded like everything else on the site.

The short version. Behavior is mostly a leadership issue. Teachers need procedures, not more rules; three rules or fewer, and procedures practiced until they run without thinking. Only the adults who run the building can uphold the rules every time and protect instructional time, and that consistency is what separates schools where teachers can teach from schools where they can't.

What the evidence supports

Practice	Grade	What it rests on
Teach classroom routines explicitly, like content, and practice them until automatic	Evidence grade: Strong Strong	Federal practice guide rates "modify the classroom environment" and "teach and reinforce new skills" as strong evidence (IES 2008(opens in a new tab)); classroom management training shows consistently positive, small to medium effects across 31 studies, mostly elementary (EEF 2019(opens in a new tab)); a meta-analysis of 76 controlled studies in primary grades found g = 0.23 (Korpershoek et al. 2026(opens in a new tab))
A few clear rules, posted and taught; frequent chances to respond; specific praise; brief correction of errors	Evidence grade: Strong Strong	Twenty practices with at least three studies each, reviewed by Simonsen and colleagues (Simonsen et al. 2008(opens in a new tab)); large classroom-level effects on disruption in a Campbell review of 12 studies (Oliver, Wehby & Reschly 2011(opens in a new tab))
Leaders set the culture: rules upheld every time, by every adult, with certain rather than severe consequences	Evidence grade: Promising Promising	The strongest statement is Tom Bennett's 2017 review for England's Department for Education, built on school visits and practitioner evidence, not trials: "Teachers alone, no matter how skilled, cannot intervene with the same impact as a school leader can" (Bennett 2017(opens in a new tab)); the EEF guidance makes consistency its sixth recommendation
Protected instructional time: no interruptions, pull-outs or announcements during core blocks	Evidence grade: Promising Promising	Sensible and widely recommended; not tested as its own intervention. The schedule section on the building leaders page covers how
Removing a disruptive student during instruction, then returning them for independent work and catching them up one to one	Evidence grade: Promising Promising, untested	No trial has tested this procedure. Bennett's review supports removal rooms as "a positive solution," and the EEF says isolation rooms have not been robustly evaluated. Suspension, the blunt version, lowers the suspended student's own scores (Lacoe & Steinberg 2018(opens in a new tab)). The design below keeps the student in school and brings them back the same lesson
Whole-class reward games (the Good Behavior Game)	Evidence grade: Promising Contested	A Baltimore first-grade trial found lasting effects into adulthood, concentrated in aggressive boys (Blueprints summary(opens in a new tab)); a 77-school trial in England found no effect on reading, concentration or disruption, with a quarter of schools dropping out (EEF 2018(opens in a new tab))

The procedure, in plain words

Three rules or fewer. Short enough that every student and every adult can say them. Everything else is a procedure, not a rule.

Procedures practiced until they run themselves. Entering the room, starting the warm-up, getting whiteboards up, moving to groups, packing up. Each one is taught the way content is taught: explained, modeled, practiced, and re-practiced after every break. A procedure that has to be announced each time was never taught.

Removal during instruction, return during practice. When a student disrupts while the teacher is teaching new material, they leave the room to a supervised space, calm down, and come back once the class is working independently. The teacher then works with them one to one to catch up on what they missed. The lesson was never stopped for the other 28 students, and the student who left is back and caught up the same period. This needs an adult and a place to send the student, which is the leader's job to provide.

Leaders uphold the rules. The same consequence, every time, from every adult. Bennett's finding that certainty matters more than severity is the whole point: a mild consequence that always happens beats a harsh one that sometimes does.

Protected time. The reading and math blocks are not interrupted by announcements, pull-outs, assemblies or visitors. If the schedule doesn't protect them, nothing else on this page can.

What a leader looks for in 5 minutes

Can students say the rules? Are there more than three?
Did a procedure happen without being announced?
When a student was corrected, was it brief, and did teaching resume?
Is there a place a removed student goes, and an adult there?

Where the evidence runs out

Most of the trial evidence is from elementary classrooms in the United States; secondary evidence is thin. The removal-and-return procedure above is a design built from the better-evidenced pieces, not a tested program. Whole-school systems have small average effects in trials, and the two best-known programs disagree with each other. We say this plainly because behavior is where the most confident claims and the least evidence meet.

Go deeper

Researcher: Brandi Simonsen and George Sugai (classroom management practices); Hill Walker and the IES practice guide panel. Translator: Tom Bennett, Running the Room (John Catt, 2020) and the 2017 DfE review; Doug Lemov, Teach Like a Champion 3.0, on routines; Anita Archer and Charles Hughes, Explicit Instruction, on teaching procedures. Practitioner: to be added.

People named here are described by what they have published. Being named does not mean they endorse this site.

What the evidence supports

The procedure, in plain words

What a leader looks for in 5 minutes

Where the evidence runs out

Go deeper

Sources