Methodology · Decision Rule v1
The MOI turns 12 public data points into one score from 0 to 100 for each neighborhood, then into one recommended type of action. The full rule is on this page, so a researcher, an agency analyst or a resident with a calculator can check any score by hand.
Rule locked April 11, 2026 · Developed by Maillard Howell · New York City
Read this first. Phase 1 scores are research-informed estimates. They come from published public data and peer-reviewed research, and they show how the framework works. They have not yet been confirmed against agency data. That check happens in Phase 2, described in section 04.
01 · The rule
Each step uses a fixed formula. The same inputs always give the same result.
Every indicator has a floor (scores 0) and a ceiling (scores 100), set from public benchmarks. A neighborhood's value is placed on that range.
Each dimension has 3 indicators. Its score is their plain average, rounded to a whole number.
The MOI score is the plain average of the 4 dimensions. Each one counts for 25%.
The score sets urgency. The weakest dimension sets the type of fix. Together they pick 1 of 12 action classes.
02 · Worked example
The same rule gives two very different answers. A neighborhood like Brownsville has the lowest score in Phase 1. Washington Heights has the highest.
Chronic disease rates are high enough to limit how active people can be. Clinical care comes first: cardiometabolic screening, diabetes prevention and metabolic health programs with DOHMH and NYC Health + Hospitals. Physical activity programs follow.
The spaces already exist. The fix is more program hours, longer hours of operation, and community groups funded to run activities in parks, schools and rec centers that are already there.
Same 4 steps, same thresholds. One neighborhood needs clinical care first. The other needs more programs in buildings it already has.
03 · Indicators
Three indicators per dimension. Each one is something a city agency already counts or surveys.
| Indicator | Direction | Source |
|---|---|---|
| D1 · Supply · Can people get to a place to be active? | ||
| Parkland acres per 1,000 residents | higher is better | NYC Parks Open Data · NY4P Open Space Index |
| Recreation centers within 0.5 miles, open and operating | higher is better | NYC Parks Open Data · NYC Parks Capital Tracker |
| Schoolyard open space open to the community | higher is better | DOE Schools · NYC Parks |
| D2 · Delivery · Is anything running in those places? | ||
| Free program hours per 1,000 residents per week | higher is better | NYC Parks Recreation Programs · DYCD After School |
| DYCD after-school physical activity programs active | higher is better | DYCD After School |
| Senior center physical activity hours per week | higher is better | DFTA Quarterly Reporting |
| D3 · Coverage · Are people taking part? | ||
| Adults meeting weekly physical activity guidelines (%) | higher is better | DOHMH Community Health Survey |
| Youth enrolled in after-school physical activity programs (%) | higher is better | DYCD After School |
| Seniors enrolled in DFTA mobility and wellness programs (%) | higher is better | DFTA Quarterly Reporting |
| D4 · Burden · What is the health cost? | ||
| Adult diabetes (%) | lower is better | DOHMH Community Health Profiles 2021 · EpiQuery |
| Adult obesity (%) | lower is better | DOHMH Community Health Survey · EpiQuery |
| Adult hypertension (%) | lower is better | DOHMH Community Health Profiles 2021 |
The floor and ceiling for each indicator are available on request (see section 04). Phase 2 may add youth fitness data such as DOE FitnessGram. The model stays at 4 dimensions.
04 · Confidence and access
Phase 2 rebuilds each dimension from finer census tract data and checks the results against these Phase 1 estimates. Some dimensions are easier to measure than others, so each one carries a confidence rating. Every change is logged here with a date.
Park access and facility locations are well documented in public data.
No public dataset counts free programs cleanly. It is the weakest of the 4, and the scores should be read with that in mind.
Survey and visit data exist, but they are thinner in lower-income neighborhoods.
Chronic disease rates are published for every census tract in the city.
Public
All 10 Phase 1 scores are in the Neighborhood Dashboard.
On request
MNHowell00@gmail.com