Measuring What Matters: How We Track Whether This Actually Works
· Sanctuary Community Initiative · 2 min read
Every nonprofit says its work matters. Fewer publish the scoreboard. We think donors, partners, and especially the people we serve deserve to see whether this actually works, so here's how we measure, what the numbers say, and where the honest limits are.
Our headline outcomes
Among participants who have received Sanctuary Community Initiative's wraparound support:
- 94% have not been rearrested: across all participants and time windows we track
- 100% non-rearrest among participants housed at 60–89 days: the group whose housing held through the critical window
- 52% employment engagement: participants working or actively engaged in the employment pipeline
Why the baselines make these numbers remarkable
Numbers only mean something against a backdrop. Nationally, Bureau of Justice Statistics data shows about 44% of people released from state prison are rearrested within their first year, and roughly two-thirds within three years. Prison Policy Initiative research puts unemployment among formerly incarcerated people at 27%, while our participants' employment engagement runs roughly double the national employment picture for this population.
The 100% figure for housed participants deserves its own sentence, because it aligns perfectly with what the research predicts: federal reentry studies identify stable housing in the early months as one of the strongest protective factors against reincarceration. Our data doesn't just report success, it confirms the mechanism.
How we measure (and the honest caveats)
We track participant outcomes through program records and follow-up, and we report them with approximations clearly marked. And here's what transparency requires us to say plainly:
- Our sample is small. We're a young, focused organization. Small samples produce impressive percentages more easily than large ones. As we grow, we expect these numbers to face harder tests, and we'll publish them either way.
- Selection effects are real. People who engage with support may differ from those who don't. Our comparisons to national baselines are context, not a controlled trial.
- Recidivism isn't the whole story. Researchers rightly note that rearrest data is shaped by policing patterns, not just behavior, and that outcomes like housing stability, employment, and family reconnection tell a fuller story. We track those too.
Why we publish anyway
Because the alternative (asking for trust without evidence) isn't good enough for the people whose recoveries are on the line or the donors funding them. Our commitment is simple: real numbers, clear baselines, honest caveats, updated as we grow.
The model already works. The measurements are how we prove it, and how we get better.
Explore the full data, sources, and the donation-impact calculator on The Numbers.