By Clay Shumate
Affective statements are the smallest move in restorative practices: one sentence naming how a specific behavior affected you, with no judgment of the student attached. They sit at the informal end of the restorative continuum, below questions, conversations, circles and conferences. No study has ever tested them on their own, so treat the evidence as bundled.
Key Takeaways
- An affective statement names a behavior and its impact, not a character trait. The example the practice’s own originators use is “I’m uncomfortable when I hear you tease Sandy” — a feeling, a specific action, nothing about the student as a person.
- No study has measured affective statements by themselves. The study teachers cite for them combined four practices into a single score, and the affective-statement scale had a reliability of α = .59, which the authors themselves call only fair.
- In 29 high school classrooms and 412 students, student-reported use of restorative practices predicted teacher respect (β = .12, p < .01) while teacher-reported use predicted nothing (β = .05, not significant) — and the two ratings were unrelated to each other. Teachers’ sense of how often they do this does not match what students notice.
- In the same study, the racial referral gap was 1.69 versus 9.13 referrals for teachers using fewer restorative practices and 0.77 versus 2.92 for those using more. The gap narrowed. It did not close, and the design cannot show the practices caused the narrowing.
- The one randomized district trial found suspensions fell 36 percent in restorative-practice schools against 18 percent in control schools — but in grades 6–8, suspension rates did not change and academic outcomes worsened. That result belongs in any honest middle-school conversation about this.
Most restorative training starts at the circle. Twenty chairs, a talking piece, a ninety-minute block nobody has. Teachers leave the professional development impressed and implement none of it, because the smallest thing they were shown costs more than a class period.
Affective statements are the opposite problem. They cost one sentence, they are the bottom rung of the whole framework, and almost nobody coaches them. Below: what they are, where they sit, what they sound like, what the research actually supports, and the middle-school finding that cuts against the sales pitch.
What are affective statements in restorative practices?
An affective statement is a short sentence in which an adult expresses their own emotional reaction to something a student did, naming the specific behavior and its impact rather than labeling the student. The International Institute for Restorative Practices describes the term as “another way of saying ‘expressing your feelings’ or ‘sharing impact’”.
They work in both directions, which is the part people miss. The research definition covers reactions to “both positive and negative events”. A sentence naming what a student’s work did for the room is an affective statement too, and it is not the same thing as praise, for reasons worth being precise about further down.
Where do affective statements sit on the restorative continuum?
Affective statements are the first and most informal element of the restorative practices continuum, which runs in five steps from a single sentence to a formal conference. Every step above them costs more time, more people, or both.
| Element | What it is | Time it costs | Who is involved |
|---|---|---|---|
| Affective statements | One sentence naming a behavior and its impact | Seconds | Teacher and one student |
| Affective questions | Questions that surface impact, such as “What impact has this incident had on you and others?” | A minute or two | Teacher and one or two students |
| Small impromptu conversations | A short exchange in the hallway or after the bell | Several minutes | Teacher and those involved |
| Circles | Structured group discussion with a protocol | A class period | The whole class |
| Formal conferences | Facilitated meeting after serious harm | An hour or more, plus scheduling | Students, families, staff, a facilitator |
The structure explains the implementation failure. A school that trains only circles and conferences has trained the two most expensive items on the list, and the PACE implementation guidance names that exact pattern as a sticking point: some schools end up “implementing conflict-resolution practices only”. The cheap rung gets skipped because it does not look like a program.

What does an affective statement actually sound like?
An affective statement has two parts: the specific behavior, and what it did to you or to the room. It has no third part about the student’s character, motives or future. The IIRP’s own worked example — “I’m uncomfortable when I hear you tease Sandy” — contains exactly those two parts and nothing else.
Here is the same pattern applied to five situations that actually happen in grades 6–12. The left column is the version that comes out when a teacher is tired. The right column keeps the standard identical and moves the subject of the sentence.
| What usually gets said | The same moment as an affective statement |
|---|---|
| “You are being disrespectful.” | “It bothers me when someone talks over an answer while a classmate is still giving it.” |
| “Why can’t you ever be ready?” | “I get frustrated starting the same lesson twice because half the room is still finding a pencil.” |
| “You clearly do not care about this class.” | “I am worried, because three weeks ago you were arguing with me about sources and this week you have not turned anything in.” |
| “Great job, buddy.” | “I appreciated that you asked the question nobody else wanted to ask. It changed what the rest of the period was about.” |
| “That is not how we treat people here.” | “I felt uncomfortable hearing that comment, and I do not think the person it was about laughed.” |
Two rules keep these from going wrong. First, name the behavior, not the person. “I am frustrated that you are so lazy” is an insult with the word “I” stapled to the front. Second, stop talking. The sentence is the whole intervention. A teenager given three seconds of silence usually fills it, and what they fill it with is more useful than anything else you had planned to say.
Affective questions are the natural next rung when the statement is not enough. The IIRP treats any question that elicits feelings and addresses impact as a usable affective question, and offers forms like “What impact has this incident had on you and others?” That is the move when two students are involved and you do not yet know what happened.

What does the research on affective statements actually show?
There is no study of affective statements on their own. The study everyone cites bundled them with three other practices into one score, and it was correlational. Gregory, Clawson, Davis and Gerewitz surveyed 412 students in 29 high school classrooms across two large, diverse East Coast high schools in their first year of restorative implementation.
Students and teachers both rated how often the teacher used four things: affective statements, restorative questions, proactive circles and fair process. Those four were combined into a single factor score per rater. So every number below describes the bundle, not the sentence.
What it found. Student-reported use of the practices correlated with student-reported teacher respect at r = .58, p < .01, and held in a multilevel model at β = .12, p < .01, accounting for 17 percent of the between-teacher variance in respect. Teachers’ use of the practices also tracked the racial referral gap: for Latino and African American students, β = .44, p < .05. Across all 29 teachers, Asian and white students received an average of 1.28 referrals against 6.34 for Latino and African American students, t(29) = 3.63, p = .001. Split at the mean, teachers using fewer restorative practices issued 1.69 against 9.13; teachers using more issued 0.77 against 2.92.
Now the three findings that should change how you read all of that.
- Teacher self-report predicted nothing. Teacher-reported use of the practices came in at β = .05, not significant, and teacher ratings were not significantly related to student ratings at all. Only what students noticed mattered.
- The affective-statement scale was the weakest one in the study. α = .59 for the student version, which the authors describe as fair; the teacher version reached .80. The scales were newly developed with no prior validation.
- It cannot show causation and says so. The authors state the design “cannot support causal conclusions,” and flag shared method variance, because the same students rated both the practices and the respect. Students who like a teacher may simply report both more generously.
The wider literature is no more settled. The broadest review of quantitative research on restorative justice in US K–12 schools concludes that “available evidence suggests that RJ programs can improve school climates and reduce student misbehavior and school discipline” while results for bullying, absenteeism and academic performance are “more mixed,” and that there is “still relatively little quantitative research regarding its effectiveness.” I read that review’s abstract and metadata; the full text sits behind a paywall, so I am citing it at that level and saying so.
And the measurement guidance written for districts is blunt about the gap that matters most here. The PACE brief lists affective statements as a measurable teacher behavior — “Teachers using/modeling affective statements (‘I feel …’)” — and then declines to claim any practice-level result. It raises the question of which practices carry the effect and does not answer it, and notes the survey scales it recommends still “require future psychometric testing.”
Does any of this improve achievement in middle school?
In the one randomized district-wide trial, no — and in grades 6–8 the academic results went the wrong way. The RAND evaluation of restorative practices in Pittsburgh Public Schools randomized 44 of the district’s schools and ran for two years.
Days lost to suspension fell 36 percent in the program schools against 18 percent in control schools, and the suspension gap between African American and white students narrowed. Teachers reported a safer environment and better conduct management. Then the part that gets left out of the slide deck: “Restorative practices did not improve academic outcomes, nor did they reduce suspensions for middle school students”, and in grades 6–8 academic outcomes “actually worsened in the treatment schools.” There was also no reduction in suspensions for violent offenses, for male students, or for students with IEPs.
Read that carefully before you discount it or over-read it. The trial tested a two-year district rollout of a whole program, not a teacher using one sentence well. It does not show that affective statements lower test scores. What it does show is that the grade band this site writes for is the band where the program results are weakest, and that anyone promising a middle school both fewer suspensions and better scores is promising something this trial did not deliver.

What are affective statements not?
An affective statement is not a consequence, not a substitute for one, and not praise. Confusing it with any of the three is how the practice gets a bad reputation in a building within one semester.
It is not a consequence because nothing happens to the student as a result of it. If a situation calls for a consequence that actually follows from the behavior, the sentence goes before the consequence and does not replace it. A school that tells teachers to use affective statements instead of their discipline ladder has not adopted restorative practices; it has removed a ladder. That distinction is also what you tell a parent who has heard that the teacher “just talked about their feelings” instead of handling something: the sentence was the opening, the consequence still applied, and here is what it was.
And it is never the response to conduct that triggers a reporting duty or a safety protocol. Threats, harassment, weapons, anything that reaches a mandatory-reporting standard, anything involving suspected harm to a student — those follow your district’s procedure immediately and completely. An affective statement is a tool for the ordinary friction of a school day. Using one where a protocol applies is not restorative practice; it is a failure to report, and no framework on the continuum changes that.
It is not praise because praise evaluates the student and an affective statement reports an effect. “You are such a strong writer” is a verdict. “Your second paragraph made me change my mind about the question” is information the student can do something with. The distinction matters more in grades 6–12 than it does with younger students, who have not yet learned to discount an adult’s verdict.
It is also not a conversation-ender for serious harm. The continuum has four rungs above it for a reason, and the research on where restorative practice does and does not hold up is clear that a half-implemented version is barely different from none at all.
How do you use affective statements without sounding fake?
Say true things in your own words and keep them short. Teenagers detect scripted emotion instantly, and a sentence that sounds workshopped reads as manipulation rather than honesty. If you are not actually frustrated, do not say you are frustrated.
Four things keep it credible. Use a feeling word you would use off the clock. Use them for the good events too, or the practice becomes a tell that someone is in trouble. Say it quietly and near the student rather than across the room, which is the same logic behind close-range private redirection with teenagers. And do not run three in a row on the same student in one period, because at that point you are not sharing impact, you are campaigning.
There is one more thing the Gregory study implies and nobody says out loud: your own count does not matter. Teacher self-report predicted nothing in that data. The only version of this practice that showed up in the results was the version students noticed. If you want to know whether you are doing it, the people to ask are sitting in the room.
Two honest objections, because they come up every time this gets presented to a staff. The first is that some students will mock it, and in grades 6–12 a few of them will. The answer is volume and setting: said quietly, to one student, about one specific thing, there is nothing to perform against. Said to a whole room as an announcement, it invites exactly the reaction teachers fear. The second is that not every adult wants to disclose feelings at work, and that is a legitimate position rather than a deficiency. The practice tolerates a narrow emotional range — “that bothered me,” “I appreciated that” — and does not require anyone to open a vein in third period.
The line that matters most is the one between sharing impact and applying pressure. Telling a student once that something bothered you is information. Returning to your own disappointment repeatedly, with the same student, until they manage your feelings to get out of the conversation, is leverage. Teenagers are extremely good at spotting the difference, and the second version costs you the first one permanently.
What should a school decide before putting this on a walkthrough form?
Decide whether you are measuring a frequency or a relationship, because the evidence only supports the second. Counting affective statements during a twenty-minute observation measures the one informant whose rating predicted nothing in the research.
Four practical cautions for anyone leading this. First, a scale built on a reliability of .59 is not an evaluation instrument; the PACE brief recommends its survey items for implementation support and says they still need psychometric testing. Second, if you are rolling this out in a middle school, go in holding the RAND result rather than discovering it later from a board member. Third, train the cheap rung first. A building where every adult can say one honest sentence about impact has more restorative capacity than a building with a circle protocol nobody uses, and the teachers who most need it are the ones with the hardest rosters. Fourth, be careful what you require. A district can require that staff name behavior rather than character; requiring an employee to disclose a feeling on demand is a different thing, and it is the version that produces the scripted sentences students see straight through.
For a coach, the usable implication of the Gregory finding is a change of instrument rather than a change of effort. Tallying affective statements during an observation measures the informant whose ratings predicted nothing. Asking four students in that class whether their teacher talks to them about how things affect people, and whether they feel respected, measures the one that did. It is a harder data collection and it is the only one with a result attached.
The cost argument is worth stating plainly to a board, because it is unusual. This is the only rung on the continuum with no budget line: no curriculum, no substitute coverage, no facilitator, no scheduling. That is a genuine advantage and it is also why it gets skipped, since a practice with no invoice attached is hard to point at in a board report.
Clay’s take
I think this is the most undersold item in the whole restorative toolkit, and the reason is unflattering to all of us. It cannot be turned into a program. There is nothing to purchase, no protocol to laminate, no two-day training to send four people to. It is one sentence, said honestly, by an adult who is willing to be a person in front of a sixteen-year-old.
What I will not tell you is that it fixes a hard class. The evidence does not support that and the middle-school finding above argues against it. What it does is change what a correction is about. “You are disrespectful” is a claim about who a student is, and a teenager has to defend themselves against it. “That bothered me” is a fact about me, and there is nothing to defend against.
And the finding I keep coming back to is that teacher self-report predicted nothing. That is a useful piece of humility to carry into any practice we are told to adopt.
What to do next
Pick one sentence you say often that starts with the word “you” and is really about how a behavior affects the room. Rewrite it once, on paper, into the two-part shape above. That is the entire first step, and it takes less time than reading this paragraph.
Then use it for a week on the good events as well as the bad ones, and do not count how many times you did it, because that number does not predict anything. From here, the full picture of what restorative practice does and does not deliver is the next thing worth reading, and the consequence side is what this practice sits beside rather than replaces. Everything on this site is free and nothing asks for your email address.
An affective statement is the cheapest rung on the continuum; the circle is several rungs up and costs real time. Most teachers should get fluent at the first before convening the second. When the thing you need is the group rather than one student, the seven rounds of a circle and the 80/20 split are the practical version, including when a circle is the wrong tool.
Frequently Asked Questions
What is an affective statement in restorative practices?
An affective statement is one short sentence in which an adult names a specific behavior and the effect it had on them, without judging the student’s character. The International Institute for Restorative Practices describes it as expressing your feelings or sharing impact, and offers the example “I’m uncomfortable when I hear you tease Sandy.” It is the most informal element of the restorative continuum.
What are some examples of affective statements for middle and high school?
Keep two parts and drop the third. Instead of “you are being disrespectful,” try “it bothers me when someone talks over an answer while a classmate is still giving it.” Instead of “you clearly do not care,” try “I am worried, because three weeks ago you were arguing with me about sources and this week nothing has come in.” Use them for good events too.
Is there research proving affective statements work?
Not on their own. The study most often cited combined affective statements with restorative questions, proactive circles and fair process into one score across 412 students and 29 high school classrooms, and the affective-statement scale had a reliability of only .59. The design was correlational and the authors state it cannot support causal conclusions.
Do restorative practices improve academic achievement?
The randomized evaluation of a two-year district rollout in Pittsburgh found they did not. Suspensions fell 36 percent in program schools against 18 percent in control schools, but academic outcomes did not improve overall and in grades 6 through 8 they worsened. The broadest review of the research calls academic results mixed and the quantitative evidence base small.
What is the difference between an affective statement and an I-statement?
They share a grammar but not a purpose. An I-statement is a general communication technique for owning your own perspective in any conflict. An affective statement is specifically about sharing the impact of a behavior as the lowest rung of the restorative continuum, which runs upward through affective questions, impromptu conversations, circles and formal conferences.
Do affective statements replace classroom consequences?
No. Nothing happens to a student as a result of an affective statement, so it is not a consequence and cannot stand in for one. The sentence goes before a consequence that follows from the behavior, not instead of it. A school that tells teachers to use these in place of a discipline ladder has removed the ladder rather than adopted restorative practice.
How do you know if you are actually using affective statements well?
Ask students, not yourself. In the Gregory study, teacher-reported use of restorative practices predicted nothing, student-reported use predicted teacher respect, and the two ratings were not significantly related to each other. Your own count of how often you do this has no demonstrated relationship to whether students experience it.
Sources
- Gregory, A., Clawson, K., Davis, A., & Gerewitz, J. (2015). The promise of restorative practices to transform teacher-student relationships and achieve equity in school discipline. Journal of Educational and Psychological Consultation, 25, 1–29. 412 students, 29 teachers with one randomly selected focal classroom each, two large diverse East Coast high schools in their first implementation year. Affective statements measured as one of four combined scales (α = .59 student, .80 teacher). Student-reported practices and teacher respect r = .58, β = .12 (p < .01), 17 percent of between-teacher variance; teacher-reported practices β = .05, not significant, and unrelated to student ratings. Referral gap 1.28 vs 6.34 overall; 1.69 vs 9.13 (low use) and 0.77 vs 2.92 (high use). Correlational; the authors state it cannot support causal conclusions, and flag shared method variance, unvalidated scales and a small teacher sample.
- Augustine, C. H., Engberg, J., Grimm, G. E., Lee, E., Wang, E. L., Christianson, K., & Joseph, A. A. (2018). Restorative practices help reduce student suspensions (RB-10051-DOJ; full report RR-2840-DOJ). RAND Corporation. Randomized controlled trial, 44 of 54 Pittsburgh Public Schools, K–12, June 2015 to June 2017. Days lost to suspension fell 36 percent in treatment schools against 18 percent in control schools; disparities narrowed; gains concentrated in elementary grades. Middle-grade suspension rates did not change and grades 6–8 academic outcomes worsened in treatment schools. No reduction for violent offenses, male students or students with IEPs. The brief reports no effect sizes.
- Darling-Hammond, S., Fronius, T. A., Sutherland, H., Guckenburg, S., Petrosino, A., & Hurley, N. (2020). Effectiveness of restorative justice in US K-12 schools: a review of quantitative research. Contemporary School Psychology, 24(3), 295–308. Finds evidence that restorative programs can improve school climate and reduce misbehavior and discipline, with results on bullying, absenteeism and academic performance “more mixed,” and notes there is “still relatively little quantitative research regarding its effectiveness.” Read at abstract and metadata level only; the full text is paywalled, so study counts, designs and grade bands are not reported here.
- Restorative practices in the classroom (Chapter 1 excerpt). International Institute for Restorative Practices. Sets out the five-element continuum from informal to formal: affective statements, affective questions, small impromptu conversations, circles, formal conferences. Source of the definition (“expressing your feelings” / “sharing impact”), the worked example “I’m uncomfortable when I hear you tease Sandy,” and the affective-question form “What impact has this incident had on you and others?” The excerpt does not state a publication year.
- Darling-Hammond, S., & Gregory, A. (2023). Measuring restorative practices to support implementation in K–12 schools. Policy Analysis for California Education. Lists “teachers using/modeling affective statements (‘I feel …’)” as a measurable practice and supplies teacher and student survey items. Names the pattern of schools “implementing conflict-resolution practices only.” Reports no practice-level outcome findings and states its adapted scales “require future psychometric testing.”
- Center for Dispute Resolution, University of Maryland Francis King Carey School of Law (2022). Restorative approaches lesson 9: affective statements and questions. Supports the point that these tools are for everyday interactions and that not every conflict requires a circle. Cited as practitioner training material; the page’s sample statements live in a slide deck that is not included, so no examples are quoted from it.
- The before-and-after rewrites in the table are my own worked examples built on the two-part shape the IIRP excerpt demonstrates. They are not quotations from any source and are not transcripts of real classroom exchanges.



















