Poverty-targeted staffing reaches MCPS's poorer elementary schools most clearly, its middle schools less so, and its high schools least

What Title I and Focus designations provide

MCPS's most explicit equity policy names schools: 38 Title I elementaries (federal program) and 38 Focus schools (MCPS's own extension of Title-I-style staffing to near-miss schools) — together 69 of the 137 elementary schools, averaging 67.7% FARMS vs 25.7% elsewhere — plus 5 Title I middle schools. There are no Title I high schools. The designations are not symbolic — they carry staff, dollars, and smaller classes.

Staffing pooled per 100 enrolled students (2025); per-pupil medians (2025, MSDE); class sizes from the 2024-2025 class-size analysis — elementary uses the published Title I/Focus category averages, middle school averages the per-school English/other-subject figures across the five Title I schools vs the rest. Smaller classes are the designation's core mechanism. Federal per pupil rises steeply with poverty even within elementaries: median $424 in the lowest-poverty quartile → $2,357 in the highest. There are no Title I high schools — high-school gradients appear in the FARMS-based sections below.

Source: MCPS School Profiles staffing (2025); MSDE per-pupil expenditure file (2025); this site's class-size analysis.

The dollars: spending rises with poverty

MSDE has published school-level per-pupil spending since 2019. In every year, the poorest quartile of elementaries out-spends the richest — the gap was +$4,602 in 2019 and is +$4,824 in 2025 — including the state and local dollars that pre-2020 studies found flowing the other way. The premise OLO carried forward from its 2014–2019 predecessors seems to be outof date.

Median per-pupil spending by FARMS quartile, 2025

Quartiles are by FARMS rate — the share of students eligible for free or reduced-price meals (MSDE Report Card), the standard school-poverty measure; not household income or Census poverty. Each level is quartiled separately. Federal aid concentrates the same way: elementary Q4 median $2,357 vs Q1 $424. For the trend chart, FARMS quartiles are recomputed within each year; 2020 unpublished (pandemic).

Per-pupil spending vs. school poverty, every school

Each dot is a school (2025). Correlations with FARMS: state/local dollars ES +0.33, MS +0.58, HS +0.45; total dollars ES +0.49, MS +0.65, HS +0.48. The gradient is not perfectly monotonic — state/local spending peaks around the third quartile — but the direction is unambiguous. The Blueprint for Maryland's Future (concentration-of-poverty grants and compensatory per-pupil funding phasing in since 2020) reinforces a gradient that school-level data shows predating it.

Source: MSDE per-pupil expenditure file (2019–2025); MSDE Report Card FARMS rates

Extra staffing

Beyond dollars, position counts: at 2025 staffing rates, MCPS's lower-income schools operate with about +1,174 more professional staff and +241 more paraeducators than they would hold at other schools' per-student rates — concentrated on roughly 81k students.

Extra FTE = positions at target schools beyond what baseline-group per-100-student rates would provide (elementary: 69 Title I/Focus schools vs 68 others; middle/high: the higher-FARMS half vs the lower; "(n.s.)" = school-level correlation with FARMS not significant). All teacher types together: +679 elementary, +206 middle, +253 high (the three teacher rows shown don't fully sum to these — the total also carries smaller teacher categories). Special education is approximately zero everywhere — SPED staffing follows disability counts, not school income. The extra elementary staffing is mostly categorical: ELD and intervention roles exceed pure classroom additions; middle and high schools get almost no extra classroom teachers. The need-gradient view of the same data is on the staffing page.

What those extra positions cost each year (estimated)

Priced at all-in compensation ($125k per professional FTE, $60k per paraeducator; assumption range in the comparison below). Per target student that is $2,529 at elementary, $1,797 at middle, $1,340 at high school.

How the pricing assumptions move the total

The junior-pay bar reprices every extra professional at $95k — the concern that high-poverty schools get cheaper staff (see the experience section below). Even the conservative pricing leaves a nine-figure annual total, most of it riding inside general-education staffing formulas rather than a named budget line.

Source: MCPS School Profiles staffing and enrollment (2025); MSDE Report Card FARMS rates (2025); cost model in the dataset metadata

Is the extra staffing growing? 2023–2025

School-level staffing data exists for three years. Over them, the extra teaching staff at higher-poverty schools has been essentially stable at elementary and high school and slightly wider at middle school — no sign of erosion, and none of widening either. OLO's Finding 12 predicts the FY27 budget will widen service-group inequities; these are the numbers to watch.

The bold line marks a gap of one extra teacher per 100 students: elementary schools clear it comfortably, high schools fall just short. Pooled positions per 100 students at schools at-or-above the level's median FARMS rate, minus the below-median pool — a consistent split across years (the Focus-school flag isn't published for 2023, so this trend deliberately uses the FARMS split everywhere; the staffing chart above uses the Title I/Focus baseline at elementary, so levels differ slightly). Three data points; the 2023 profiles also use an older label taxonomy — only label-stable categories are compared.

Source: MCPS School Profiles staffing, 2023–2025; MSDE FARMS rates, matching years

Does the extra staffing reach Black and Latinx students?

OLO's central concern is Black and Latinx students, not poverty per se. In MCPS the two are nearly the same map: school FARMS rates and Black+Latinx enrollment shares correlate at ρ = 0.96. The extra staffing tracks race nearly as strongly as it tracks income — ES ρ = +0.76, MS ρ = +0.86, HS ρ = +0.83 for teachers per 100, and per-pupil dollars at +0.48, +0.58, +0.48 — so income-targeted mechanisms are, in practice, reaching the schools Black and Latinx students attend. What they demonstrably do not equalize is the experience mix below.

Each dot is a school (2025); Black+Latinx share is the sum of MCPS's Black/African American and Hispanic/Latino enrollment percentages. School-level resources cannot show within-school allocation — OLO's interview findings about how students experience resources (Findings 13–15) are outside what this data can test.

Source: MCPS enrollment by race/ethnicity (2025); staffing and per-pupil sources as above

Who teaches where: poorer schools get junior teachers

The clearest counter-pattern in the source data runs the other way. Split MCPS schools into poverty quintiles and the median share of teachers with under three years' experience roughly doubles across the gradient — 6.4% at the least-poor fifth vs 14.1% at the poorest in 2025 — and Title I/Focus elementaries run 14.1% vs 9.3% elsewhere. The quintile gap has narrowed somewhat since 2019 (12 → 8 points) but persists. There is no clear evidence here that it changes student outcomes — inexperience measures tenure, not teaching quality, and the outcome sections below do not tie it to lower results — but it is worth watching: poorer schools get more positions, and more junior ones, which is why the dollar model above carries a junior-cost variant.

Each row is the distribution across the ~41 schools in that poverty band: the box is the middle half (25th–75th percentile), the tick the median, the whisker the 10th–90th. The boxes overlap — a few low-poverty schools run high and some high-poverty schools low — but the whole distribution slides toward more inexperience as poverty rises, not just the median. FARMS quintiles are computed across all MCPS schools each year (a MoCo-specific split; the state's top-vs-bottom quartile tiers are a narrower, poorer comparison group, also in the dataset). "Inexperienced" is MSDE's under-3-years definition; it measures tenure, not quality, but experience is the one staffing attribute that consistently runs the other way, toward wealthier schools.

Source: MSDE Report Card educator qualifications, 2019–2025; MSDE FARMS rates, matching years

The gap the money has not closed

The extra staffing and dollars have not translated into equal outcomes — and where the gap lives depends on the subject and the level. In reading, the median school's gap between not-economically-disadvantaged and economically disadvantaged students is essentially the same at elementary and middle school (31.1 and 26.8 points on MCAP ELA in 2026) and has not narrowed since 2022. It looks smaller at high school (21.8 points), but only on English 10 — the one full-cohort high-school test — so that reflects the measure more than a closing gap. In math the story is different in kind: by middle school the gap stops being a score gap and becomes a placement gap.

Gap = median school's not-disadvantaged minus disadvantaged proficiency on MCAP ELA only (127 elementary, 40 middle, 24 high schools with both groups published). ELA is the honest cross-level comparison: it is a full-cohort test at every level, while high-school MCAP math covers only the students who reach Algebra 1 in high school — a group most stronger students have already left (see below) — which floors every group's score and fakes gap compression. The narrower HS reading gap comes with its own caveat: by grade 10 the lowest-performing students are likelier to have disengaged or left entirely.

Matching for poverty shrinks the race gap but does not close it

To separate race from poverty, compare only low-income (FARMS) students to each other — district-wide, so no school-mix bias — matching for poverty does shrink the race gap, but does not erase it, and the residual differs sharply by group. Among low-income students, White out-read their Black peers by a modest margin (-2.6, 11.4, and 8.4 points at elementary, middle, and high). The gap to low-income Hispanic students is two to three times wider (14.4 / 19.4 / 16.7) — many are multilingual learners still building English, and they are the furthest behind at every level.

District-wide MCAP reading proficiency among FARMS (free-and-reduced-meals) students, 2026 — the composed race×FARMS value MSDE reports for each grade (PercProfLEA), averaged over each level's tested grades (elementary 3–5, middle 6–8, high = English 10). Because these are district figures, not a median of the few schools with reportable low-income White groups, they carry no school-composition bias. In math the same ordering holds at elementary (FARMS White 21% vs Black 18.9% vs Hispanic 8.8%); middle-school math is omitted because most students take Algebra 1 there, leaving the grade-8 test a non-representative remainder.

Low-income students' reading scores are higher at lower-poverty schools

The school a low-income student attends is associated with how they do. Taking Black and Hispanic students on FARMS and sorting by their school's own poverty level, both groups read measurably better at the district's lower-poverty elementaries than at its poorest: low-income Black students 40% proficient at the lower-poverty third of schools versus 33.7% at the poorest (r = -0.33); low-income Hispanic students 25% versus 21.8% (r = -0.31). School context tracks outcomes even holding the student's own poverty status fixed — and more clearly than the staffing inside those schools does.

Median MCAP reading proficiency of FARMS Black (108 schools) and FARMS Hispanic (107 schools) students, 2025, elementary, grouped into thirds by the school's FARMS rate (averaging 24%, 49%, and79% FARMS). Observational, not causal: "FARMS" is a yes/no line, and a FARMS family at a lower-poverty school is likely nearer the eligibility threshold — less deeply poor — than one at a high-poverty school, and families who reach those schools differ in ways this can't see. So it points to a school-context association, not proof that changing schools would raise scores.

Inside schools, by school poverty quartile (2026)

Medians across ~193 schools with both subgroups published (Economically Disadvantaged vs Non-economically Disadvantaged (MSDE); MCAP ELA and math pooled; 2021's anomalous administration excluded). Read the compressed gap at poorer schools carefully: it is not evidence the gap is solved there — both groups score lower, and the "non-disadvantaged" population at a high-FARMS school is itself close to the FARMS line. Within Title I/Focus elementaries the median internal gap is 14.1 points vs 32.5 at other elementaries, for the same reason. None of this is causal — families are not randomly assigned to schools — but it bounds what school-level resources alone can be expected to do. And school poverty is far from the whole story: among schools at similar poverty (40–65% FARMS, 54 schools), disadvantaged students' results still range nearly two-to-one (20.8% to 46% proficient), and none of the inputs this page can measure — spending, staffing rates, teacher experience — predicts who beats expectations.

Math becomes a placement gap in middle school

Reading proficiency holds roughly flat from grade 3 through grade 8. Math does not: the full cohort slides from 50.5% proficient in grade 3 to 32% in grade 6 — the last grade where nearly everyone takes the same test. After that, MCAP math stops measuring one thing: students split between grade-level courses (Math 7, Math 8) and accelerated ones (Algebra 1, Geometry), and the students still in grade-level math are, by construction, those not moved ahead — only 2.2% of Math 8 takers score proficient. Where a student lands on that ladder is the real middle-school math outcome.

District-wide MCAP proficiency by grade, 2025, all students, count-weighted across schools. Grades 3–6 cover essentially the full cohort in both subjects. In grades 7–8 the math bar covers only students still in grade-level math (70% and 40% of the grade) — the rest test in Algebra 1 or Geometry instead (28.6% proficient pooled across middle- and high-school takers), so the grade-7/8 collapse reflects the acceleration split itself, not the performance of all students.

Who gets past Math 8 — and who did in 2019

In 2019, taking Algebra 1 (or higher) by 8th grade was the MCPS norm everywhere: 88% of 8th graders district-wide, and 82% even at the poorest quartile of middle schools. The pandemic-era placement change ended that. By 2025 only 60% of 8th graders are past Math 8 — and the retreat was far larger at poor schools: the poorest quartile fell to 36% while the least-poor quartile kept 79%. An acceleration gradient that barely existed in 2019 is now one of the sharpest poverty splits in the district's data.

Median share of 8th graders testing beyond Math 8 (Algebra 1 or Geometry), by FARMS quartile of the district's ~40 middle schools: 2019 88% · 2022 65% · 2023 61% · 2024 62% · 2025 60% district-wide. Computed as 1 − Math-8 testers ÷ grade-8 enrollment per school (a school reporting no Math-8 testers counts as fully beyond; the estimate assumes Math-8 testers are 8th graders, so accelerated 7th graders sitting Math 8 would push true shares slightly higher). The correlation with school FARMS went from r = -0.28 in 2019 to r = -0.69 in 2025; grade-7 acceleration is unchanged, so the entire shift is the 8th-grade Algebra decision. Placement is not the whole story either: even among middle-school Algebra takers the median school posts 38% proficiency, and school poverty still predicts passing (r = -0.73). None of this shows why placement changed — course-placement criteria are not in the public data — but 2019 demonstrates that near-universal early Algebra was recently the district's own practice at every poverty level.

The gap widens at each high-school gate

Early Algebra is one gate in a sequence, and the gap widens at each. Looking only within the same high school — so this is tracking inside a building, not poor students attending different schools — the White−Hispanic gap grows from 16 points at grade-9 on-track status, to 27 at Algebra 2 completion, to 29 at college-and-career-ready. White−Black widens6 → 15 → 28. The White−low-income gap (13 → 26 → 33) tracks White−Hispanic almost exactly — most of the "race" gap here is a poverty gap.

Median within-school White−minority gap (percentage points) at three successive high-school gates, 2024 (31 high schools reporting both groups). Each panel is the same schools — the fan-out is tracking inside buildings, not between them. Gates are cumulative filters, so later gates measure a more selected group; and race and poverty are statistically inseparable here (the low-income line ≈ the Hispanic line), so no bar is a pure race effect.

The gap is wider at the Advanced level

The gap in reaching the top is larger than the gap in clearing the proficiency bar. Counting the share of elementary schools where a group actually posts an Advanced (Level 4) reading rate of at least 5% — a measure that needs no estimating around suppressed cells — White students reach Advanced in 76.0% of schools, but Black in 32.0%, Hispanic in 24.0%, and economically disadvantaged students in only 17.0%.

Share of schools (by level, 2025) where the group's Advanced MCAP-reading rate is reported and at least 5%; a suppressed or absent cell counts as not clearing, so no imputation is involved. This "presence" measure re-expresses the ordinary proficiency gap somewhat (schools that clear the Advanced bar average higher proficiency overall) — but it shows the disparity is, if anything, larger at the top of the scale than at the proficiency line.

Race and income together

The bulk files publish race and income separately, but the report-card explorer's district-level filters compose — so the intersections OLO's central findings turn on can be tested. They cut both ways. Among economically disadvantaged students, race gaps shrink to a fraction of their headline size (White−Black 11.5 points, White−Hispanic 21.2). But amongnon-disadvantaged students they persist at roughly triple that (White−Black 30.0, White−Hispanic 40.9) — a binary FARMS line cannot equalize the family resources above it, and race predicts outcomes beyond what the income categories capture. That is OLO's Finding 8, confirmed with public data.

District-wide MCAP proficiency, 2026, ELA and math pooled across all tested grades. Each figure is the district value the report-card explorer returns for that composed race×income filter — a value the downloadable files never publish (they carry marginal groups only) — read as the consensus across the schools that enrol the group. The pattern is stable across 2022–2026: within the disadvantaged group the White−Hispanic gap has run 15.2–21.2 points by year, versus 38.9–41.2 among the non-disadvantaged. The income split is the binary MSDE category; "not disadvantaged" spans everything above the FARMS line, and those families differ by race in ways this measure cannot see — one reason gaps persist in the top row.

Source: MSDE MCAP proficiency by student group (school level), 2022–2026; district race × income cells via the report-card explorer (2026); MSDE assessment files (per-assessment tested counts) and MCPS enrollment by grade for the placement ladder; MSDE FARMS rates; inputs as elsewhere on this page

Suspensions: a gap that lives in secondary school

The starkest race and poverty gap on this page isn't a test score — it's who gets suspended, and it barely exists until middle school. In elementary school out-of-school suspension is vanishingly rare for every group (Black students %, White %). At middle school it detonates: the Black rate reaches % and the low-income rate % versus % for non-low-income students — a × Black-to-White ratio. High school sits just below middle. The entire disparity is a secondary-school phenomenon.

Out-of-school suspension rate — unique students suspended as a share of subgroup enrollment,2024 (0 elementary, 0 middle, 0 high schools). Low-income (FARMS) vs not: ES % / %, MS % / %, HS % / %. Two caveats: the near-zero elementary rates are partly a Maryland policy floor (out-of-school suspension of young children is legally restricted), not purely behavioral; and White rates are census-suppressed below 5%, so rates are reconstructed from counts — the direction is robust, the exact magnitudes approximate.

Source: MCPS suspension counts and enrollment by subgroup (2024); reconstructed per method above

Every school

Staffing rates, poverty, designation, and per-pupil spending for all 202 schools. Click a school for its full profile.

Source: MCPS School Profiles staffing and enrollment (2025); MSDE Report Card FARMS rates (2025); MSDE per-pupil expenditure file (2025) — all as published in this site's open data. Read alongside OLO Report 2026-9, Budget Equity in Montgomery County Public Schools (July 2026). Method: pooled per-100-student rates; "extra FTE" relative to baseline-group rates; Spearman correlations. Generated 2026-07-31.