Five households. Five alarm clocks. The same city.
What determines whether a child makes it to school today?
In Madison, Wisconsin, nearly one in three public school students misses more than 10 percent of the school year.1 The district's chronic absence (CA) rate of 29.6 percent is classified as "High" severity by state standards. But that number is an average. Averages hide the geography beneath them and mask the household circumstances that produce them.
The Madison Equity Atlas maps 22 data layers across 125 census tracts and 54 schools to show where structural conditions converge to keep children out of school. This document tells the story of five households — composites drawn from real tract-level data — to show what those conditions look like at 6:45 on a Tuesday morning.
Every statistic in this document is drawn from public data, processed in the Atlas. Every household represents a real neighborhood archetype. The names are invented. The numbers are not.
Each household is located using the Neighborhood Attendance Risk Index (NARI) — a composite score that synthesizes eight structural indicators (poverty, racial composition, youth population, transit access, housing cost burden, uninsured rate, eviction filings, and health care shortage) into a single 0–100 percentile ranking for each of Dane County's 125 census tracts. Tracts scoring in the 80th–100th percentile are classified as "Critical"; those below the 20th percentile are "On-Track." The NARI is the Atlas's primary lens for identifying where structural risk concentrates.
Maria Garcia wakes at 5:40. Her shift at the food processing plant starts at 7:00. Her two children, ages 6 and 9, attend an elementary school 2.3 miles away. There is no before-school childcare in her census tract — it is designated a childcare desert, with zero licensed providers within walking distance.2 The nearest bus stop is a 12-minute walk, and during the AM peak, buses arrive fewer than 2 times per hour.3
On mornings when the bus is late or the walk takes too long, Maria faces a choice: leave the children home with a neighbor's teenager, or arrive late to work. A third late arrival this month means a written warning. She chooses work. The 6-year-old misses first period. By December, she has missed 14 school days.
Maria earns $38,000. Her neighborhood's median household income is $42,000, placing it in the lowest income bin in the Atlas. Schools serving neighborhoods below $45,000 median income show an average chronic absence rate of 27.8 percent — compared to 17.6 percent in neighborhoods above $100,000.4
MMSD already provides free Madison Metro bus passes for high schoolers who live more than 1.5 miles from school and qualify for Free and Reduced Meals, plus universal free summer passes for grades 6–12. But Maria's children are in elementary school, and the school-year passes are need-based and grade-limited. The issue is not fare — it is frequency. If her tract had 15-minute bus headways during AM peak and an on-site childcare provider at her children's school, the Atlas estimates her neighborhood's chronic absence rate could decline by 5 to 7 percentage points.3,5
Three interventions have demonstrated measurable effects on attendance in conditions like Maria's:
Madison Metro is in an active service planning cycle, and MMSD already invests in student transit (need-based school-year passes, universal summer passes, yellow bus for K–8 beyond 1.5 miles). What remains unaddressed is AM-peak frequency in the 31 tracts with zero bus stops and before-school childcare at elementary sites. These are the gaps the Atlas measures — and the gaps where precedented, evaluated interventions exist.
Terrence Johnson received an eviction filing in October. The filing did not result in removal — most don't — but it triggered a forced move. He and his daughter, a seventh-grader at a nearby middle school, relocated to a cousin's apartment on the east side, doubling up in a two-bedroom. His daughter transferred schools in November.
In their old tract, the eviction filing rate is 5.12 per 100 renter households — more than double the Dane County average of 2.06.6 The cost burden in that tract is 17.7 percent, meaning nearly one in five households spends more than 30 percent of their income on housing.7 In Dane County, 51 percent of eviction defendants are Black, despite Black residents representing 5.8 percent of the county population.8
His daughter missed 18 school days during the transition. By the time she enrolled at her new school, she had fallen behind in math and stopped attending advisory period.
If Terrence had been in a Common Wealth Development (CWD) affordable housing unit, the eviction filing would not have occurred. CWD manages 165 affordable housing units in and adjacent to priority tracts.5 Stable housing is attendance infrastructure. The Atlas estimates that reducing eviction rates to the county median in affected tracts would remove the primary source of forced school transfers for hundreds of families. Dane County's eviction diversion program targets these geographies, but coverage remains incomplete.8
The research base on housing stability and attendance is among the strongest in the field:
If eviction filing rates in the 14 priority tracts fell to the county median of 2.06 per 100 renter households, the Atlas model projects hundreds fewer forced school transfers annually — each one representing 10 to 20 school days a child does not lose to relocation.6,9
Sarah Larsen is a project manager. Her husband teaches at the university. Their combined household income is $165,000. Their 10-year-old walks seven minutes to an elementary school where the chronic absence rate is under 13 percent. There are two licensed childcare centers within a half-mile. The nearest bus stop, three minutes away, sees a bus every 8 minutes during AM peak.
When their son is sick, Sarah works from home. There is no forced choice between employment and caregiving. Their neighborhood's poverty rate is 1.1 percent. Their eviction filing rate is 0.09 per 100 renter households.6 They have not thought about housing instability because it has never happened to them.
The Larsens live 6.2 miles from the Johnsons. They live in the same school district.
The Larsens do not need an intervention. They need to see the Atlas. Their neighborhood's structural advantages — transit, childcare, housing stability, income — produce attendance rates that look like personal virtue but are, in fact, infrastructure. The gap between the Larsens and the Garcias is not one of effort. It is one of bus schedules, eviction filings, and licensed childcare slots.10
The Larsens receive every intervention simultaneously — by default. The evidence says that is exactly why it works:
The Larsens' neighborhood delivers the full stack. Policy should aim to deliver it everywhere.
Amina Abdi's three children attend schools at three different levels — elementary, middle, and high. Her eldest, a sophomore, has missed 34 days this year. In MMSD, Black high school students have a chronic absence rate of 59.9 percent.11 Her younger children's attendance is better, but declining: the elementary child's rate is 42.8 percent, already above the district average for all students.11
Amina's tract carries federal Health Professional Shortage Area designations in all three categories: primary care, mental health, and dental care.12 Her eldest has not seen a therapist in eight months despite an anxiety diagnosis, because the nearest provider accepting her insurance has a four-month waitlist. When the anxiety peaks, he stays home. His school marks it as an unexcused absence.
Amina earns $62,000. Her neighborhood's median income is $58,000. She is not poor by federal definitions. Her family's attendance outcomes look like those of families earning half as much.
Amina's son needs a therapist, not a truancy notice. MMSD opened behavioral health clinics in all four high schools during the 2023–24 school year, providing in-school therapy for students with significant needs. This is a real investment. But the clinics serve students already enrolled and attending — they do not reach the student who has stopped showing up. And no full-service primary care, dental, or vision clinics exist in MMSD schools. All 14 priority tracts carry triple federal HPSA designations — primary care, mental health, and dental care.12 Expanding community-based behavioral health access beyond the school building is Demand 4 of the Atlas's six institutional demands. UW Health and the City Health Department are named. The timeline is 18 months. The metric is trackable year over year in the Atlas.14
Race-explicit and health-access interventions with measured outcomes:
The 25.8 percentage-point racial gap that persists after controlling for income is not explained by poverty. It is explained by race-specific stressors: housing discrimination, concentrated eviction, health access deserts, and institutional distrust. Income-targeted programs alone will not close it. The interventions above address the transmission mechanisms.13
Linh Nguyen's daughter attends Lincoln Elementary. The neighborhood is classified as Severe-tier on the NARI — high poverty, moderate transit, elevated cost burden. The school's predicted chronic absence rate, based on its structural neighborhood context, is 30.8 percent. The school's actual chronic absence rate is 13.3 percent.15
Lincoln Elementary beats its prediction by 17.5 percentage points. It is one of four positive-outlier schools identified by the Atlas's regression model — schools whose attendance outcomes are significantly better than their neighborhoods would predict.15
Linh does not know why. She knows the principal calls when her daughter is absent. She knows the school provides breakfast. She knows there is a family engagement coordinator who speaks Vietnamese. She does not know that the school's practices are, according to the data, the most effective attendance intervention in the district.
Lincoln Elementary demonstrates what is possible. Its practices — not its demographics — explain its outcomes. MMSD already operates eight Community Schools (Hawthorne, Kennedy, Lake View, Leopold, Lori Mann Carey, Mendota, Orchard Ridge, and Sandburg) that co-locate family services and community partners inside school buildings. But 8 of 30-plus elementary schools is not district-wide, and the community school model has not been paired with a positive-deviance analysis to identify which school-level practices actually drive attendance gains. The Atlas identifies positive-deviance schools as the natural sites for learning what works. UNICEF's 2024 Data Must Speak framework validates this methodology across 15 countries. No Wisconsin district has published a positive-deviance analysis of school attendance. CWD's application to a U.S. metropolitan area is novel.16
Positive-deviance analysis reverses the deficit narrative. Instead of asking what is wrong with failing schools, it asks what is right with succeeding ones:
CWD is positioned to lead this work. The Atlas's regression-residual methodology, validated by UNICEF's Innocenti research office across 15 countries, identifies both where to invest and where to learn.16
The five households in this document live in the same school district. Their children attend schools governed by the same board. The gap between them is not one of effort, values, or culture. It is a gap of bus schedules, eviction filing rates, childcare slots, therapist waitlists, and housing costs.
These are structural conditions. They are measurable. They are mapped. And they are changeable — by the institutions named in the Atlas's six demands, on the timelines specified, using the metrics tracked year over year.
The Madison Equity Atlas does not recommend. It measures. Measurement is the beginning of accountability.