Memo: Proposed Census Tract Qualification Methodology
Margin of Error (MOE) Provision -- 2026 Application Cycle

1. Background

The ENOUGH Grant Program determines census tract eligibility based on child poverty rates derived from the American Community Survey (ACS) 5-Year Estimates published by the U.S. Census Bureau. These estimates use the Official Poverty Measure (OPM) and represent the only methodology available for neighborhood-level poverty measurement. The statutory threshold requires that an eligible community include at least one census tract where more than 30% of children live in poverty.

ACS estimates for small geographies (census tracts) carry significant margins of error due to limited sample sizes. The Census Bureau publishes a margin of error (MOE) alongside each estimate, representing a 90% confidence interval. For many Maryland census tracts, the MOE is substantial:

• Average MOE: 12.4 percentage points
• Median MOE: 9.7 percentage points
• ~49% of tracts have MOE exceeding 10 percentage points
• 189 tracts shifted more than 10pp between the 2023 and 2024 ACS releases

2. The Proposed Formula Change

For the 2026 application cycle, we propose expanding the Census Tract Poverty Rate Qualification with a two-path approach:

The dual requirement (upper bound exceeds threshold AND previous-year confirmation) ensures that the provision captures tracts with genuine sustained poverty that appears to have improved due to sampling variability, rather than tracts that were never truly above the threshold.

3. Quantitative Impact

3.1 Scale

Category	Count	% of All Tracts
Criteria 1: Point estimate >30%	180	12.3%
Criteria 2: Upper bound >30% + previous year confirmation	159	10.9%
Total Census Tract Poverty Rate Qualified (either criteria)	216	14.8%
Net tracts added by MOE provision	36	2.5%

The MOE provision adds 36 net tracts to the qualified pool -- a 20% increase over the point-estimate-only approach. This is a conservative expansion. Many of the 159 tracts meeting Criteria 2 also meet Criteria 1; the overlap accounts for the modest net gain.

3.2 Impact on Full Eligibility

Of the 216 Census Tract Poverty Rate Qualified tracts, 181 also pass the school test (School Poverty Rate Qualified), achieving full ENOUGH eligibility. The school test -- not the tract test -- is the binding constraint for 35 tracts that are tract-qualified but cannot achieve full eligibility.

4. The Case for Approval

4.1 Addresses a Known Data Limitation

ACS tract-level estimates have documented high uncertainty. The Census Bureau publishes MOE specifically so that data users can assess reliability. Using the upper bound for threshold decisions is consistent with Census Bureau guidance for applications involving small-area estimates.

4.2 Prevents Harm from Statistical Noise

189 tracts shifted more than 10 percentage points between the 2023 and 2024 ACS releases. These shifts do not reflect genuine one-year changes in community economic conditions -- they are artifacts of the 5-year rolling sample methodology. Without the MOE provision, communities lose access to anti-poverty programming based on sampling artifacts rather than genuine improvement.

4.3 Includes a Safeguard Against Over-Expansion

The dual-year requirement (previous year must also exceed 30%) ensures that:

• Only tracts with demonstrated sustained poverty qualify via Path 2
• Tracts with low poverty but high MOE (due to very small populations) do not qualify spuriously
• The provision acts as a stability mechanism, not an open-ended expansion

4.4 Aligns with Federal Statistical Practice

Using confidence interval bounds for threshold decisions is standard practice. HHS, HUD, and the Census Bureau itself recommend considering MOE when making program eligibility determinations based on ACS estimates. The ENOUGH program's adoption of this approach is consistent with how other federal and state programs handle small-area estimate uncertainty.

4.5 Conservative in Application

The net effect is 36 additional tracts (20% expansion) -- not a dramatic change. The provision primarily stabilizes eligibility for communities already at the margin, rather than bringing in fundamentally new geographies. Most of its benefit accrues to tracts that were above 30% in one ACS release but slipped marginally below in the next.

4.6 Promotes Program Continuity

Communities already receiving ENOUGH funding, or in the process of building partnerships for application, should not have their eligibility stripped due to year-over-year sampling variation when poverty conditions have not genuinely improved. The MOE provision supports the program's place-based, long-term anti-poverty strategy.

5. Distinction from School Threshold

It is important to note that the 75% school CPG threshold is statutory for Year 3 and does not require the Secretary's approval. The school threshold change (from 80% in Years 1-2 to 75% in Year 3) is settled law. Only the census tract methodology change (MOE provision) requires administrative approval.

6. Recommendation

We recommend the Secretary approve the proposed upper-bound methodology. It is statistically sound, conservative in application (36 net tracts, 20% expansion), includes appropriate safeguards against over-expansion, and prevents communities from losing eligibility due to measurement artifacts rather than genuine improvement.