Seven Questions About DeKalb County's Attendance Boundary Process

Board Policy AD — the policy governing how DCSD draws school attendance boundaries — was last revised in 2013. The field has since moved to equity-informed, multi-criteria approaches that include socioeconomic data, academic performance, school choice modeling, and feeder pattern analysis as standard inputs — criteria that peer districts treat as foundational, not optional. DCSD's process includes none of them.

This analysis asks seven questions about that gap.

1. 1. Why is Policy AD from 2013 and not updated to current redistricting standards?
2. 2. How does a majority-minority district that talks about equity decide not to include any equity factors in its boundary process?
3. 3. Is HPM performing the redistricting analysis — and does reducing their visibility address the trust concerns that paused the previous SAP?
4. 4. HPM's own planning lead personally used equity and SES methods for redistricting at his previous firm. Why aren't those methods being applied here?
5. 5. Policy AD requires "condition of facility" data. Not all schools have been visually inspected — so why was $300,000 spent on building-size assessments instead?
6. 6. Why are choice school communities excluded from the engagement process while the district says magnet placement is on the table?
7. 7. Why is the redistricting process being used to address capital budget shortfalls — and why were Kittredge and Allgood targeted before any analysis began?

Board Policy AD was last revised on August 5, 2013. Here is everything it includes:

None of the criteria are weighted or prioritized. The final secondary criterion allows a criterion to be introduced the night of the vote. And the following criteria — standard in districts that have updated their policies in the last decade — are entirely absent:

These are not aspirational additions. A 2021 Urban Institute report analyzing every attendance boundary in U.S. metropolitan areas — including Atlanta — found that even small boundary changes can significantly improve integration. A 2026 New America report argues boundary placement should be "systematic and intentional." The National School Public Relations Association identifies establishing clear guiding principles and criteria before engaging the community as a core best practice. Districts that have modernized their redistricting — including Howard County Public School System, MD (HCPSS), Winston-Salem/Forsyth County Schools, NC (WS/FCS), Fayette County (KY), Dublin City Schools (OH), and Champaign Unit 4 (IL) — treat socioeconomic composition, academic performance, and demographic impact assessment as foundational inputs. And while race-based student assignment is constitutionally constrained, Georgia law already treats socioeconomic status as a permissible lever: state charter statutes (O.C.G.A. § 20-2-2066, amended in 2015) expressly allow weighting admission lotteries toward educationally disadvantaged — including FARM-eligible (Free and Reduced-Price Meals) — students.

The SAP website frames the project around "strengthening opportunity for every DeKalb student." The district's own planning process — displayed on its SAP page — commits to a "Study" step in which "district staff review enrollment trends, demographics, transportation, school capacity, and educational opportunities." But "demographics" appears nowhere in Policy AD's criteria — not as a primary criterion, not as a secondary one. The process page promises demographic analysis. The policy that governs the process contains no demographic criterion. The process includes no mechanism to define, measure, or evaluate equity.

Source: DCSD SAP Website, "How We're Planning" (accessed September 3, 2026). The "Study" step promises review of "demographics" — a term that appears nowhere in Policy AD's primary or secondary criteria.

The engagement video and presentation materials describe the data inputs driving this process. The confirmed inputs are: enrollment, instructional capacity, geographic proximity, facility condition, building size, feeder patterns, and utilization trends. These are facilities and logistics metrics. They answer: how do we balance building capacity? They do not answer: how do we balance educational opportunity?

DCSD is 89% students of color (56.4% Black, 22.4% Hispanic/Latino, 6.5% Asian, 11.1% White per U.S. News, 2024–25) and 82% FARM-eligible (DCSD School Meals), with 105 of 131 schools on the Community Eligibility Provision (CEP). But that 82% average masks enormous variation across schools. The six clusters being redistricted include schools with FARM rates well below the district average alongside schools at or near 100%. Boundary changes that ignore this variation can concentrate poverty into some schools and disperse it from others — with no policy requirement for anyone to measure the effect.

The Engagement Structure Inverts the Standard Model

In districts that have modernized their redistricting practices, the standard engagement model has clear roles. The community provides input on values and priorities — what criteria matter most, how competing objectives should be weighed. The district or its consultant develops boundary scenarios using board-approved criteria informed by those community-weighted priorities. In Howard County, the Office of School Planning develops the boundary review report and scenarios; a third-party consultant facilitates; community members provide feedback at input workshops and public hearings. In Winston-Salem/Forsyth County, over 8,000 survey responses determined the relative weights of competing objectives — SES (socioeconomic status) integration, transportation efficiency, feeder pattern stability — that fed into the optimization model. The community shaped what the model valued. The district ran the model.

DCSD's engagement structure inverts this. The Engagement Series 1 survey asks parents to rank their top three from the existing Policy AD secondary criteria — and only those criteria. There is no option to suggest or rank criteria the policy does not already include. No socioeconomic composition. No academic performance. No English learner data. No demographic impact assessment. The only mechanism for a parent to suggest a criterion outside the existing framework is Question 9: a 300-character text box, framed "Keeping Board Policy AD in mind." The engagement video states explicitly: "this is the framework we will be using to consider any feedback provided during this process" (Engagement Video, 5:40).

The community is not being asked what criteria should exist. It is being asked to rank criteria the district chose in 2013. In best-practice districts, community input shapes the framework. Here, community input operates within a framework the community had no role in shaping — and the framework itself is missing the criteria that peer districts treat as foundational.

The previous SAP phase was paused after significant community backlash. Interim Superintendent Sauce acknowledged in May 2026 that the process had been "too top-down" and said going forward, conversations would involve "independent facilitators" alongside district representatives. However, HPM personnel — including Scott Leopold, HPM's Director of Planning Services — are leading focus groups in the current engagement series. The community was told district officials would drive this process. HPM is at focus groups. At a recent online focus group, Leopold confirmed that HPM is conducting the redistricting analysis.

The trust problem that paused the previous SAP was about what data was used, what criteria were applied, and how decisions were made. The response was not to change the methodology, update the policy criteria, or add equity data. The response was to reduce HPM's visibility. If the same team is doing the same work with the same data inputs under the same policy — just with less public acknowledgment — that does not address the substance of what broke trust.

Scott Leopold, HPM's Director of Planning Services, is leading focus groups in DCSD's current engagement process. Before joining HPM, Leopold was a Partner at Cooperative Strategies — the firm districts hire specifically for equity-informed redistricting.

In June 2022, Leopold personally presented redistricting scenarios to the Howard County (MD) school board as the Cooperative Strategies lead. The scenarios, he said, were created with a focus on promoting "a diverse and inclusive student body" at each of the schools by examining socioeconomic compositions of school populations. HCPSS's process uses 14 criteria including FARM ratios, academic performance, and English learner levels — confirmed in the HCPSS redistricting overview video narrated by Leopold himself.

Leopold also led the Champaign Unit 4 redistricting in Illinois, where Cooperative Strategies was hired to balance schools by SES percentages.

Leopold described his own methodology as examining socioeconomic compositions to promote diverse student bodies. He used academic performance data. He balanced schools by SES percentages. He did this work for other districts as his professional service. The question the community should be asking: is the absence of these methods in DCSD's process a contract scope limitation, a policy limitation, or a choice — and who made that decision?

Policy AD's secondary criterion is "condition of facility." In facility management, that means a Facility Condition Assessment (FCA): an evaluation of a building's physical systems — roof, HVAC, structural integrity, plumbing, electrical. FCA data tells you whether a building is safe, functional, and worth investing in. It is the standard basis for capital planning decisions in school districts nationwide.

DCSD has FCA scores for all schools — but for many, those scores are modeled estimates based on building age and size, not visual inspections. At the February 26, 2025 board retreat, COO Erick Hofstetter explained that HPM would conduct FCAs on a rolling basis — approximately 40 to 50 schools at a time — using what he called an "integrated approach," rather than assessing all schools comprehensively. He described the previous model — hiring a firm to deliver a single report — as producing information that "becomes obsolete." Under the new approach, HPM would assess batches of schools, build the district a database, and update it annually. Board Member Eyasu raised concerns about the pace: "My concern is we have over 100 schools, which are not getting younger. Every year we wait, the needs are increasing." At the July 13, 2026 board meeting, Board Member Whitney McGinniss asked for "FCA [Facility Condition Assessment] scores for all those schools" before the board could evaluate the capital deferral plan — a plan that included deferring Kittredge ($21.4M) and Allgood ($34M) to an unfunded ESPLOST VII. The current, inspection-based FCA data she was asking for does not exist for all schools in scope.

What the District Paid For Instead

HPM's contract with DCSD includes FCA work — the rolling inspections Hofstetter described. But HPM also offered — and DCSD purchased — a separate product: its Facility Educational Adequacy Assessment (FEAA), priced at $300,000 as an Add Alternate to the contract. That $300,000 went to a building-size measurement — while the visual facility-condition inspections Policy AD actually requires remain incomplete for schools in scope.

HPM's FEAA is not educational adequacy as the term is understood in facility planning — an assessment of whether a building supports its educational program. It functions as a construction planning checklist: seven sub-categories that measure room counts and square footage against a standardized template (78,000 sq ft for a 624-student elementary school). The sub-categories — academic core, administration, food service, media center, PE/athletics, special education, visual arts — are all physical space measurements. None measure student outcomes. None measure educational quality.

A small school that serves its students well — but lacks the square footage for a regulation gymnasium, full-size media center, or dedicated art room to HPM's specifications — scores poorly not because it fails students but because it doesn't fit the template. Kittredge Magnet — 57,163 sq ft, adequacy score 70.2 — is the #1 elementary school in Georgia with 99% proficiency. An independent analysis of the FEAA methodology found that for 4-section elementary schools (n=41), building size alone explains 95.7% of the variance in HPM's adequacy scores (R² = 0.957). The assessment measures building size. It does not measure educational adequacy.

The decisions that flow from this data — which schools receive capital investment, which are deferred, which are considered for program relocation — require accurate facility condition data. The district spent $300,000 on a building-size proxy instead.

DCSD's engagement instructions tell specialty school families to "select your home attendance area (zoned) school as the school you are representing." The Engagement Series 1 survey lists only zoned schools in its dropdown — Kittredge, Wadsworth, Robert Shaw, DeKalb Arts Academy, DeKalb School of the Arts, Warren Technical, and Coralwood are not options. Families at these schools are routed to their home-zoned cluster, not the school their child attends.

Yet the district's own presentation acknowledges that magnet placement is part of this process:

Source: DCSD SAP Engagement Series 1 Presentation, "Timeline and What to Expect" slide (Engagement Video, 15:06). Note the final bullet: "Any changes will need to be approved by the Board of Education." Under Board Policy JBCC, program moves require board approval upon superintendent recommendation.

The district acknowledges it may relocate programs and sever feeder patterns — while the engagement structure excludes the families most affected from representing their schools. No Community Advisory Team (CAT) member represents these schools. No public comment slot addresses their specific enrollment or overcrowding issues. The engagement video confirms CATs are "made up of 10 representatives from each school in the six overcrowded clusters" (17:16) — specialty schools are not in those clusters.

At a recent online focus group, HPM's Director of Planning Scott Leopold confirmed this directly. According to attendee notes, he said that programs are definitely part of the discussion — but that Kittredge was not included in the survey options because, in his framing, this is a boundary discussion and Kittredge does not have a boundary. However, Board Policy JBCC (Student Assignment), revised July 13, 2026, states: "The board, upon recommendation by the superintendent, may move a school choice to a different facility or cease the operation of a school choice program." The board — not the superintendent alone — has authority over program placement.

Choice Schools Need Choice School Tools

Attendance zone schools fill seats through geography. Choice schools fill seats through applications and qualifications. These are different enrollment systems. Moving a magnet to a different building doesn't reduce demand — it changes who can access it. The right tools for choice school overcrowding are enrollment management, capacity expansion, and transportation access — not boundary changes.

Three Choice Schools This Process Cannot See

The district's own overcrowding map includes these schools. The survey and engagement structure exclude them. Here is what the process is missing:

Source: DCSD SAP Engagement Series 1 Presentation, Slide 18 (Engagement Video, 10:41). Kittredge appears on the district's own overcrowding map at 99 students over capacity (red circle). Yet the Engagement Series 1 survey does not list Kittredge as a selectable school.

Kittredge was named twice — once before any analysis existed, and once alongside a budget error the administration's own numbers showed was already covered.

February 2025: Kittredge Named Before Any Assessment

At the Board of Education retreat on February 26, 2025 — 33 days after the board approved HPM's contract and before HPM had conducted any facility assessments — COO Erick Hofstetter laid out the capital budget pressures facing the district. Cross Keys High School (~$100M) and Sequoyah (~$130–135M) were consuming the ESPLOST VI budget simultaneously. The program has no bond financing — it is cash-flow at its limit. Hofstetter named Kittredge specifically and linked its capital deferral to anticipated SAP outcomes:

A board member responded: "Okay so that's good... Now that's a logical hold." Hofstetter was not saying "we will defer until the budget allows." He was saying "we will defer because we might move the program" — framing a capital decision as contingent on a program relocation before the analysis to justify either decision existed.

July 2026: Kittredge Named Again

At the July 13, 2026 board meeting, the district presented its plan for addressing an estimated $12 million ESPLOST VI distribution error, disclosed the previous month. After voters approved ESPLOST VI in November 2021, the district failed to submit the required certificate of distribution to the Georgia Department of Revenue — resulting in DCSD collecting more than its share from Atlanta Public Schools ($2.9M) and City Schools of Decatur ($9.1M) for four years.

The administration's own presentation showed that this error was already covered. COO Hofstetter explained that approximately $12 million in interest earnings on ESPLOST VI collections would replace the overpayment, leaving a positive projected variance of $10 million. The deferrals were not necessary to absorb the error.

Despite this, the administration presented a recommended action plan that deferred Kittredge ($21.4M) and Allgood Elementary ($34M) to the unfunded ESPLOST VII — while recommending that Stoneview ($35M) proceed.

Source: DCSD Board Presentation, July 13, 2026 — "Impact of Sales Tax Distribution Adjustment on E-SPLOST VI Revenue." This slide appeared at the same board meeting where the $12M distribution error was disclosed.

Two board members challenged the framing directly.

Board Vice Chair Eyasu questioned why the deferrals were being presented at all:

When asked to explain the rationale for recommending Stoneview over Kittredge, Chief of Capital Improvements Darrell Stallings acknowledged that Kittredge had been overlooked:

The Board Has Authority — But Not the Information

Under Policy AD, the board has authority to accept, modify, or reject the superintendent's redistricting recommendation. Under Policy JBCC, the board votes on moving or ceasing school choice programs upon superintendent recommendation. The board has clear authority over both boundary changes and program moves. But the engagement process is not giving the board the information it needs to exercise that authority — no choice school input, no demand data, no academic performance analysis, no feeder pattern impact assessment.

The COO is an operations official. He does not set academic program policy. But Hofstetter singled out Kittredge by name twice — once in February 2025, framing its deferral as a possible program relocation ("is the student assignment process going to look at that and say hey, maybe that gets moved to a different building?"), and once in July 2026, when the school lost the $21.4M in capital work it needs. An operations official is driving what is effectively an academic program decision through capital recommendations. Under JBCC, a program move requires a board vote — but if the board is never presented with the choice school data, community input, and impact analysis that an informed vote requires, the authority exists on paper without the substance to exercise it. Kittredge was named for deferral in February 2025 — before any facility assessment. It was named again in July 2026 — alongside a budget error that was already covered. Policy JBCC gives the board authority over program moves. But authority without information is not oversight. When the #1 elementary school in Georgia, with 742 applications for 146 seats and 100% proficiency, can be targeted for deferral and potential program relocation without the board receiving choice school input, demand data, or academic impact analysis, the process has a gap the board should close before December.

What the Board Can Do