← All proof trees

1. 1
   How it was computedhistogram of school-level below-90 shares

   universe

   Same as citywide_trend

   packet block

   distribution

   Uses: Chronic absenteeism as 100 minus the 90%-attendance share

   Evidence: generator — search "// 7 — distribution"

2. 2
   What went inInputs

   • sdp_attendance_90_school — read directly from the publisher’s file (Student attendance — % with 90%+ attendance, by school and year). findings
3. 3
   The generated fileThe committed data file

   Generated 2026-09-08 by one script from the inputs above; the page renders only from this file.

   This block’s raw values (JSON)

   {
    "bins_pp": 5,
    "histograms": {
     "2021-22": {
      "counts": [
       3,
       3,
       10,
       9,
       17,
       10,
       18,
       22,
       20,
       19,
       15,
       26,
       13,
       15,
       9,
       3,
       2,
       0,
       0,
       0
      ],
      "n_schools": 214,
      "bin_edges": [
       0,
       5,
       10,
       15,
       20,
       25,
       30,
       35,
       40,
       45,
       50,
       55,
       60,
       65,
       70,
       75,
       80,
       85,
       90,
       95,
       100
      ]
     },
     "2024-25": {
      "counts": [
       2,
       4,
       9,
       11,
       17,
       25,
       23,
       19,
       22,
       22,
       18,
       17,
       14,
       9,
       2,
       2,
       0,
       0,
       0,
       0
      ],
      "n_schools": 216,
      "bin_edges": [
       0,
       5,
       10,
       15,
       20,
       25,
       30,
       35,
       40,
       45,
       50,
       55,
       60,
       65,
       70,
       75,
       80,
       85,
       90,
       95,
       100
      ]
     }
    },
    "n_above_50_by_year": {
     "2020-21": {
      "n_above_50": 43,
      "n_schools": 213,
      "pct": 20.1878
     },
     "2021-22": {
      "n_above_50": 83,
      "n_schools": 214,
      "pct": 38.785
     },
     "2022-23": {
      "n_above_50": 71,
      "n_schools": 215,
      "pct": 33.0233
     },
     "2023-24": {
      "n_above_50": 64,
      "n_schools": 216,
      "pct": 29.6296
     },
     "2024-25": {
      "n_above_50": 62,
      "n_schools": 216,
      "pct": 28.7037
     }
    }
   }

   Evidence: packet JSON · generator script

4. 4
   How it was checkedChecked: passed

   • ✓ passedHistogram counts and the above-50% counts recomputed independently from the district CSV and matched (full QA §D).

   The QA report behind this status · the report also lists what it does not check.

   Each input metric’s own checks are on its metric tree (links above). Only the district-wide all-students series is covered by the packet’s 15 source-vs-database checks; that coverage is not claimed for any other block.

5. 5
   Assumptions & caveatsAssumptions and caveats

   None beyond the input metrics’ own assumptions.