← All proof trees

1. 1
   How it was computedvalue = mean of the subtopic scores within (school, year, topic, respondent)

   1. Unzip the survey archive and read the school-level subtopic-scores workbook.
   2. Keep rows for this respondent type; roll subtopics up to the topic by averaging their scores.
   3. 'n.a.' and 'i.s.' are stored as suppressed.

   Algorithms applied downstream: Survey topic scores per respondent type

2. 2
   The load that wrote it1 data load wrote this metric

   Loaded	Source file	Years	Cells	Suppressed	File hash	Commit
   2026-06-04	pses_topic_subtopic_scores_all_years	2024-25	245	0	142800f6ae81…	f75e3313

   The hash identifies the exact file version read; the commit identifies the exact code. Both are recorded at load time.

   Evidence: pses.ts

3. 3
   The publisher's filePhilly School Experience Survey — topic and subtopic scores, 2019-20 to 2024-25

   Philly School Experience Survey — topic and subtopic scores, 2019-20 to 2024-25 — School District of Philadelphia (Research, Evaluation & Accountability). Survey scores by school, year, topic, subtopic, and respondent type (students, teachers, parents, support staff).

   format

   ZIP of XLSX

   join key

   ULCS Code

   landing page

   https://opendataphilly.org/datasets/sdp-philly-school-experience-survey/

   • 'n.a.' (no data) and 'i.s.' (insufficient sample) are both stored as suppressed.

   Evidence: source registry (sources.yaml) · pses_topic_subtopic_scores_all_years: first-fetch findings

4. 4
   How it was checkedChecks: 1 passed

   • ✓ passedComputed-value check: our rollup reproduces the district's published values

   Evidence: Computed-value check

5. 5
   Assumptions & caveatsAssumptions and caveats

   Assumptions

   • A topic score is the unweighted mean of its subtopics (the district publishes topic scores aggregated across respondents; per-respondent topic scores are our rollup).

   Caveats

   • Year coverage varies by topic and respondent — some topics were not asked every year.

To see the tree for one school’s number, click any dotted-underlined figure on a school, Outliers, or Movers page.