Stories · Impact
What we can prove, and what we cannot.
Attendance, return rates, a 12-question survey, and a section about what we do not measure. Published every year, including the flat numbers.
- Attendance
- 94%
- Return rate
- 78%
- Survey response
- 91%
- Waitlist
- 611
The headline numbers
Attendance is measured as kids present divided by kids enrolled, averaged across the year. It is high because kids choose to come, not because we chase them; we do not require attendance and we do not penalize absence. The return rate is the percentage of fall-term kids who enrolled again in winter or spring.
The kid survey
Twelve questions, paper, twice a year, administered by a staff member who is not the kid's program lead. 91% response rate. Kids answer on a four-point scale from 'not true for me' to 'very true for me.' The questions are not about enjoyment. They are about the things we think the program actually builds:
| Statement | 2024 | 2025 | Change |
|---|---|---|---|
| I can keep going when something is hard. | 3.1 | 3.2 | +0.1 |
| I can explain what I made to someone else. | 3.3 | 3.4 | +0.1 |
| I know how to ask for help. | 3.2 | 3.2 | flat |
| I can fix something that is broken. | 2.9 | 3.2 | +0.3 |
| I feel like an adult here listens to me. | 3.4 | 3.4 | flat |
| I would come here even if my friend did not. | 3.0 | 3.2 | +0.2 |
Two flat numbers, published every year: asking for help and feeling listened to. We have tried three things to move them and none of them have worked. The most likely explanation is that both scores are near the ceiling and the scale is not sensitive enough; the other explanation is that we are not as good at listening as we think. We are testing the second one with a kid-led interview project in 2026.
What we do not know
- Whether kids do better in school because of us. We do not have access to grades, and a correlation would not tell us much.
- What happens to the 39% of alumni we cannot reach. Our tracking is opt-in and incomplete.
- Whether the Nothing program has a measurable effect. We have anecdotes and a waitlist, not a study.
- Whether a kid who attends for one term gets the same thing as a kid who attends for three years. We suspect not, but we have not measured it.
The waitlist
611 kids. It is the most honest number on this website and the one we like least. At our current cost per seat ($2,730 per kid per year), clearing the waitlist would require about $1.67 million in additional annual funding and a seventh site. We are working on the site. We are not going to pretend the money is easy.
Every nonprofit says 'we measure what matters.' We are trying to say what we do not measure, because that is the part a funder should be allowed to ask about.
Read the annual report
The full budget, the audit, the survey instrument, and the things that did not work.