Miami-Dade · metric methodologyClimate and attendanceSafetyderived

SESIR: weapons incidents per 100 students

Reported weapons incidents in the School Environmental Safety Incident Report divided by October enrollment, per 100 students.

key sesir_weapons_per_100 · unit rate per 100 students · lower is better · rated (peer and district percentiles)

This metric is not re-read cell by cell by the validation harness yet. It is written by the same loader as a checked metric from the same file family (see below), and its coverage is listed here; a component for it can be added in components.yaml.

Definition

Reported weapons incidents in the School Environmental Safety Incident Report divided by October enrollment, per 100 students.

Incident counts from the School Environmental Safety Incident Report (SESIR) divided by October membership and multiplied by 100. Categories are groupings of the state's incident types: violence (battery, fighting, homicide, kidnapping, physical attack, robbery, sexual battery, sexual offenses, threat/intimidation), weapons, drugs and alcohol (including tobacco and vaping), bullying and harassment; 'all incidents' is every type.

Derived metric: computed by sesir.ts from the SESIR incident counts by type and the same year's October membership (enrollment_oct); the inputs have their own pages.

Source files

Florida Department of Education, Office of Safe Schools (source_id: fldoe_sesir_school_2024-25)

Year: 2024-25
File: SESIR2425a-h.xlsx (xlsx)
Fetched: 2026-09-15 · sha256 268ad6792244 · 6,414,462 bytes · fetched via origin host
First-fetch findings: fldoe_sesir_school_2024-25.md
Registry note: No suppression; Dade 14,850 rows (~550 schools x 27 types). Series 2005-06 -> (three URL forms). Definitions change over time — check Appendix P per year.

Loader: scripts/loaders/miami/sesir.ts · registry: docs/cities/miami/sources.yaml

What is served

Years: 2024-25
519 school-year cells across 519 schools; 0 stored as suppressed (publisher withheld)
438 cells carry a peer-group percentile (all-students rows of schools with a peer group)

What it cannot show

SESIR counts what schools report. Reporting practice differs across schools and principals, so a low rate can mean a safe school or an under-reporting one; the survey items on safety are the independent signal.
The category regexes are this project's grouping, not the state's; the state's own groupings (violence, weapons, drugs, other) are similar but not identical.
Counts are unsuppressed, so small schools show large swings per 100 students.

Comparability across years

Only 2024-25 is loaded so far. The total is the sum of every incident type the file lists for the school (its own 'SCHOOL TOTAL' line is used as a cross-check, not added in); the four category rates sum to less than the total because some types belong to no category.

How it was checked

Every served number runs through the four-check taxonomy described on the methodology page: (i) completeness, (ii) an exact re-read of the file by independent Python, (iii-b) a second publisher where one exists, (iv) recomputation of derived values.

Harness: methodology and known gaps · components · closure-track verdict (93ea735c8bb9) · outcomes verdict (REVIEW)

In the ratings

Schools are ranked on this metric within their peer group (40 nearest schools on demographics and size, same level, sector and magnet status) and district-wide, with the rank inverted where lower is better so a high percentile always means a better outcome. See outliers and movers for this metric, and the ratings overview for how peer groups are built.

Related metrics

Corrections log

None yet. Corrections to published figures are recorded here with the date and the affected numbers.