Every school I have worked with has a climate survey, and almost every one of them has the same problem. The numbers come back fine. Not glowing, but fine. Then something happens in March that the numbers did not predict, and in the inquiry afterward it turns out that half the building knew.
The instinct is to blame the instrument, or the response rate, or the week it was issued. I think the real problem is more interesting, and a researcher at Yale has spent her career on it.
Jessica Hoffmann is an assistant professor in the Yale Child Study Center and the director of research at the Yale Center for Emotional Intelligence. She directs the Center's two high school initiatives, RULER for high schools and inspirED, and the reason her name should matter to anyone running a high school is simple arithmetic about where the field puts its effort. Social and emotional learning is well developed for young children. It thins out dramatically by ninth grade, which is precisely when the self-editing this series is about begins in earnest. She works the end of the age range that has the least built for it.
Her most practically useful contribution is not a program. It is a method for getting an honest reading out of adolescents.
Consider what you are actually asking when you put "I feel I belong at this school" on a survey in front of a sixteen year old. You are asking a young person who has spent four years learning what adults in that building want to hear, to rate, on a scale, in a form with her name attached or not attached in a way she does not entirely trust, a feeling she may not have words for. The output of that request is not a measurement. It is a performance of the measurement, which is why it keeps coming back fine.
Hoffmann's alternative, the School Climate Walkthrough, has students assess their school across a set of named domains rather than report a feeling. It is a different cognitive task. You are no longer asking how she feels about the building. You are asking her to appraise specific, observable features of it, which is a thing teenagers are extremely good at and rarely invited to do.
Then inspirED takes the step most climate work never takes. The students who generate the picture are the ones who act on it. They are not respondents. They are the intervention.
This is the fourth book in The Unmistakably You Series, Stop Suppressing, Start Unleashing, and its argument is that what looks like a teenager's emotional control is often emotional suppression wearing better clothes, and that the two have opposite long-run prices. Hoffmann's work holds the same line at the level of a school. She co-developed the Student Emotion Regulation Assessment, published in 2025 with Zi Jia Ng, Cynthia Willner and colleagues, which measures regulation across grades one to twelve. Measuring regulation as a developing skill, rather than scoring students on how calm they appear, is the whole distinction the book is making, built into an instrument.
Her 2025 paper with Zorana Ivcevic and a colleague in Thinking Skills and Creativity makes the same point from another angle, linking school climate to students' attitudes toward creativity. A student's willingness to offer an unfinished idea is not a personality trait she brings to your building. It is partly a read on your building.
Three things follow for a school leader, and none of them need a budget.
Stop asking for feelings and start asking for appraisals. Replace at least one feeling item with a question about something students can actually observe, and the quality of what comes back will change immediately.
Give the data back to the students who produced it. The fastest way to teach a cohort that honest answers are pointless is to collect them, take them away, and return with a plan. The fastest way to teach the opposite is to hand the findings to a group of students and ask what they want to do about the worst one.
And separate climate from satisfaction. A school can be well liked and still be a place where students put most of themselves away before first period. Those are different measurements, and only one of them predicts the thing you are worried about.
Here is the uncomfortable part. If you run this properly, the numbers will get worse before they get better, because you will have stopped collecting the performance and started collecting the thing. That dip is not a failure of the new approach. It is the first evidence it is working, and the first honest answer your building has given you in years.
Want to explore these ideas further? Learn more about my work on leadership, identity and conviction at www.theglennllopis.com.
This essay was first published on Conviction Chronicles

