Everything on the main page comes from documents the district has already published. Nothing here uses student records, and nothing here could — student-level data is protected and was never public. This page states exactly how each figure was produced and where it is soft.
Capacity and projection figures are parsed directly from the projection tables in RSP's March 2026 Enrollment Analysis (slides 49–52). As a check, the parsed East feeder rows reproduce the table in the 22 June BOE deck exactly — Corinth 609/638/643/655/674 and Highlands 255/245/244/244/242. These numbers are the district's, unaltered. The same table supplies the spare-seat counts for every East feeder school used in the island analysis.
RSP renders every map in the report from one GIS base template. We tested this by isolating the
purple district-boundary outline on nine separate map pages and cross-correlating them against a
reference page across a range of vertical and horizontal shifts. Every page returned a best
alignment at dy=0, dx=0.
Because the pages are pixel-registered, the elementary boundary map (slide 9) and the median home value map (slide 31) can be overlaid directly. No projection, warping, or control-point fitting is involved, so no georeferencing error is introduced.
Legend swatch colours were sampled for the five assessor value bands and all 34 school fill colours. Every pixel in the map panel was assigned to its nearest legend colour within a fixed tolerance; unmatched pixels (roads, labels, parks, water) were discarded. School masks were cleaned by morphological closing, hole filling, clipping to the eastern strip, and keeping the dominant connected component. Results were validated by checking that each school's centroid falls in its correct real-world position relative to the others — all eight do.
Limitation. These figures are the share of land area in each value band. They are not household counts and not student counts. They describe housing stock composition, which is what RSP's map depicts. A household-weighted version would require the district's parcel data.
RSP publishes a student density heat map (slide 85) with a continuous colour ramp. We sampled that ramp and inverted it to recover relative density per pixel, then integrated density within each attendance area.
Across schools, the ratio of known enrollment to integrated density varies (roughly 0.04–0.19), because the heat surface is kernel-smoothed. So we do not use it to predict enrollment. We use it only to apportion students within a single school, calibrated so that the school's total equals RSP's own published figure. Under that calibration the unknown scaling factor cancels out.
Limitation. Kernel smoothing spreads density across boundaries, so apportionment near an edge is less reliable than in the interior. Figures are rounded and presented with a tilde throughout.
Four streets anchor the analysis, each located a different way and each stated here so it can be checked.
83rd Street. Corinth Elementary sits at 83rd and Mission, so the school's own map marker locates it. We isolated the dark maroon school markers on slide 9 and took the one inside the Corinth zone: it sits 32% of the way down from the zone's northern edge.
Somerset Drive. Labelled on slide 9. We digitised control points along the visible centreline, fitted a smooth curve, then drew that curve back onto the source map and checked it by eye against the road. It follows Somerset, and the cut region uses the curve rather than a straight line.
89th Street. Located by running OCR over the southern part of the Corinth zone; the '89TH' label resolves at high confidence.
Lee Boulevard and Mission Road. Read from their rotated labels on the same map.
Interactive tile boundaries. The tappable map on the main page splits Corinth into twelve tiles bounded by streets located from RSP's own map. Somerset Drive is traced from the map; 83rd Street from the school marker; 89th Street from OCR; Mission and Lee from their labels. The remaining north–south edges — 85th, 91st, 93rd and 95th — are interpolated at 21.83 pixels per street number between the 83rd and 89th anchors. Those interpolated edges are the softer ones, and 95th is the furthest from an anchor. Each tile carries its own child count and its own housing mass, so any combination of tiles scores exactly rather than approximately: children are summed, and mixed housing is recomputed as the $400–600k housing mass remaining divided by the total banded mass remaining. Percentages are never averaged. The per-tile figures are published as 10_corinth_tiles.csv.
Approach 1 removes two areas from Corinth, both named on slide 33: north of 83rd Street to Belinder, and west of Mission Road to Briarwood. Together they move about 210 children — the western piece is small, roughly four.
Approach 2 is drawn on slide 34 as two islands: north of 83rd and south of 89th. RSP published no student counts, no demographic figures and no receiving schools for it, so we measured the two areas off their own map against the same density surface. Both together move 443 children and leave Corinth at 231, which is 69 below the 300-student floor the district set in its own three-step enrollment procedure.
Whether those areas are islands depends on which school receives them, which RSP never states. Absorbing them into the neighbours would be contiguous — but the seats aren't there. Island 1 would put Belinder at 114% of capacity and island 2 would put Briarwood at 133%; Briarwood and Trailwood together hold 191 seats against island 2's 237. The only building with real room is Highlands at 358 spare seats, and it borders neither area. Those capacity figures are transcribed from the same RSP projection table as everything in section 1.
RSP's building-level projections stop at 2030/31, so the five-year comparison on the main page extends one year beyond published data under three stated assumptions: Corinth continues at +16 children a year (the mean of RSP's own 2026/27–2030/31 year-over-year changes); Corinth tracks the district's projected elementary decline of −2.8% over five years (RSP's figure); or Corinth is flat. We show all three rather than picking one, and on the first of them the district's plan performs better than ours — which is stated on the page.
Limitation. This is arithmetic on RSP's numbers, not a demographic model. The figures that would settle it exist in RSP's model and were not published.
If any figure here is wrong we want to know, and we will publish the correction with the same prominence as the original claim. The district can settle most of these questions by releasing data it already holds.