The Six-Dollar Fix Nobody Funds: Where the First Dollar Against Chronic Absenteeism Should Go

Sep 2, 2026 · 8:42 PM · ~9 min read

Abstract

The 2024-25 state attendance data landed this summer, and it says the recovery has stalled. Across 44 states and DC, 22.6% of students were chronically absent in 2025, against 15% in 2019: a rate still half again above the pre-pandemic baseline, and falling by less than a point a year. Meanwhile the 2026 legislative session produced 70 absenteeism bills in 24 states, and the ones that carry money buy attendance officers, dashboards, court referrals, and parent fines. None of those has a randomized trial showing it moves attendance. The lever that does, personalized absence notices mailed or texted to parents, has been tested in at least five randomized trials covering well over 100,000 students, costs roughly $4 to $7 per student per year, and cuts chronic absenteeism by about 10-15% in its best versions. It is at least two orders of magnitude cheaper per student than home visits or mentoring, which also work. This post ranks the levers by evidence per dollar, asks why the cheapest proven one has not scaled, and proposes the one condition a state should write into its FY27 attendance grant.

Finding 1: The recovery has flattened well short of baseline

Nat Malkus's Return to Learn tracker at AEI, updated June 30, 2026, is the most complete count. Chronic absenteeism (missing 10% or more of enrolled days) peaked at 28.3% in 2022, then fell to 25.7% in 2023, 23.5% in 2024, and 22.6% in 2025. The declines shrank each year: nearly 3 points, then just over 2, then under 1. Only 29% of students attend districts on pace to hit the "50% Challenge" of halving 2022 rates by 2027, down from 34% a year earlier. Malkus's conclusion is that at the current pace "the country is not on track to return to pre-pandemic attendance in the near term—or likely at all."

The distribution is worse than the mean. Attendance Works' analysis of 31 states' 2024-25 data reports 46% of schools still at high (20-29%) or extreme (30%+) chronic absence, and 58% of the highest-poverty schools at the extreme level. NWEA's July 2026 analysis of 2,000 districts finds those with persistently high absenteeism about 50% further from math recovery. The absenteeism problem and the learning-loss problem are the same problem in the same schools.

Finding 2: Ranked by evidence per dollar, the cheap lever wins by a hundredfold

The table pulls the cost and effect figures from each intervention's own evaluation. Costs are per student per year unless noted; "evidence" is the strongest design available.

Lever	Cost per student	Effect	Evidence
Personalized absence mailers (Rogers & Feller, Philadelphia)	$6.60 per household for up to 5 mailings	Chronic absenteeism down >10%; ~$6 per added day	RCT, 28,080 K-12 students
Mailers, early grades (Robinson et al., 10 California districts)	~$5.68 per family	Chronic absenteeism down 15%; $10.69 per added day	RCT, 10,504 K-5 students
Personalized messages via SIS (NCRERN, 47 rural districts, 16 states)	$4.07	Absences down 1.7% (intent-to-treat) to 4.5% (among families actually reached)	RCT, 2022-24
Six-district pooled trials (JREE, Nov 2025)	Not restated; Evidence for ESSA lists the program at $3-10 per enrolled student	Absences down 1.9% (95% CI 0.6-3.1%); best versions cut chronic absenteeism 10%+	6 RCTs, 78,000+ students, preK-12
Weekly text alerts on absences and missing work (Bergman & Chan, 22 West Virginia schools)	Variable cost $63 for 32,000+ texts	Class attendance up 12%; course failures down 27%	RCT
Home visits (Connecticut LEAP)	$10.7M for ~8,700 students, roughly $1,200 each (my arithmetic)	Attendance rate up ~8 points (PK-5) and ~16 points (6-12) nine months after first visit	Matched comparison, not randomized
Check & Connect mentoring	~$1,955 (2017 dollars)	Absences down 4.2 days (22.9%) in grades 5-7; no detectable effect in grades 1-4	RCT, Chicago
Truancy court, diversion, fines	Not costed; court and officer time	Minnesota: no detectable effect over 10 years; South Carolina: youth missed 5 more days after justice involvement; Texas: 90% drop in filings, attendance unchanged (+0.13%)	Difference-in-differences; administrative cohort; before-after

Three things stand out. First, the nudges are real. Five randomized trials by different teams in urban, suburban, and rural districts land in the same range, and Johns Hopkins' Evidence for ESSA rates the mailer program "Strong." Robinson and colleagues put it bluntly: their mailers cost $10.69 per incremental school day, while "interventions that employ mentors and social workers can cost over $120 per incremental school day." The 2025 pooled paper calls the approach "more than one order of magnitude more cost effective than current absence-reduction best practices"; on the per-student figures above the gap is closer to two orders.

Second, the expensive levers are not wrong, just expensive. Connecticut's home visits produced the largest effects in the table, the state now funds them with $7 million of its own money, and its chronic absence rate fell from 23.7% in 2021-22 to 16.4% in 2025-26. But at $1,200 to $2,000 a student these are Tier 3 tools for the students furthest gone, not something a state can buy for the one child in four who is chronically absent.

Third, enforcement has the weakest record, and states keep funding it. Lee and colleagues, using ten years of Minnesota data and a difference-in-differences design, "failed to reject the null hypothesis of no program effect" for court diversion, and showed that pre-post comparisons overstate effects because chronically absent students regress toward the mean on their own. The CSG Justice Center's South Carolina study found youth missed five more days after juvenile justice involvement, and that court oversight designed specifically to improve attendance did not. When Texas decriminalized truancy in 2015, court filings fell roughly 90% and statewide attendance moved by 0.13%. Yet in 2026 Vermont added a $1,000 parent fine and a court-petition path, Utah made it a Class B misdemeanor for parents to skip a meeting with school officials, and Tennessee expanded juvenile court reporting.

Finding 3: Why the six-dollar lever has not scaled

If mailers are this cheap and this well-tested, why does the marginal state dollar keep going elsewhere? The rural trial is the most honest account, because its authors wrote up the failures. The intervention was built into a student information system the districts already owned, so it should have been turnkey. It was not. Message templates had to be copied by hand into the right staff group; when the absence filter was not re-applied each month, families got the wrong message, hitting up to 17% of the treatment group in some districts. The messaging platform defaulted to opt-in, so districts that never flipped it could reach only families who had signed up. In the 2023-24 cohort, fewer than half of treatment-group caregivers received any message in 9 of 38 districts. The authors' summary: "Regardless of how straightforward and low effort the intervention is designed to be, there are significant barriers to start-up that need to be proactively mitigated."

That is the scaling problem in one paragraph. The letter costs $6. A district reliably knowing which families to send it to, holding a current address or number, computing the peer comparison, and repeating it every month for eight months is what nobody has funded. It is dosage, not design.

The second reason is legibility. Jeremy Singer's 2025 study of 47 high-absenteeism Michigan districts found principals' improvement plans favored "new organizational infrastructure and hire new personnel, with less emphasis on specific practices." Legislatures do the same. Mississippi's SB 2103, enacted this year, funds one attendance officer per 4,000 compulsory-age students at a starting salary of $29,528 plus a $5,000 raise: about $8.60 per student in salary alone, more than a full year of mailers for every child, spent on a position with no randomized evidence behind it. An officer is a line item a legislator can point to; half a day per student per year is invisible on a dashboard even though it is the best return in the table.

Recommendation: one condition on the FY27 attendance grant

States should not create a new program. They should add one condition to the attendance money they already send districts. Something like:

Any district receiving funds under this grant shall, from October through May, deliver each month to the guardian of every student who missed two or more days in the prior month a personalized notice, by mail or text, stating the student's cumulative absences for the year alongside the average for the student's grade. The district shall report monthly the share of such guardians actually reached. Funds for attendance staff, software, or enforcement under this grant are released only after reported reach exceeds 80%.

The wording does three things the evidence says matter. It fixes the content (cumulative absences plus a peer comparison, the two elements the trials found corrected parents' misbeliefs). It fixes the dosage (monthly, eight months). And it makes reach, not adoption, the gate for everything else, which is exactly what the rural trial showed goes wrong. A state with a million students could cover every family for $4-7 million, less if it targets the students who have already missed days. Mississippi's pending SB 2506, which would require the state board to adopt rules mandating evidence-based interventions, is the right instinct; it needs the dosage line to have teeth.

Then, and only then, spend the second dollar on home visits for the students the letters did not reach.

Limitations

The nudge trials report small absolute effects, roughly half a day per student per year; the relative effect on chronic absenteeism depends on how many students sit near the 10% threshold. The rural trial found heterogeneous effects, including a statistically significant increase in absences among Black students, which deserves replication before any universal mandate. The Connecticut LEAP evaluation is a matched comparison, not a randomized trial, and its per-student cost is my division of funding by students visited, not a published figure. EveryDay Labs' $19-per-day and "20 times more cost-effective" figures are vendor claims and are not used here. The Texas figure is a before-after comparison from a think-tank analysis of state data. Costs span 2014-2024 and are not inflation-adjusted. The 2024-25 national figures cover 44-45 jurisdictions, not all 50 states.