Best Victorian government secondary schools, measured against similar schools

How to read this

Why the default ranking is not the study score

Sorting Victorian schools by raw results mostly sorts their intakes: the academically selective schools come first, then the wealthiest suburbs. That tells a parent which school admits the strongest students, not which school teaches well. The Department of Education already solves this by grouping each school with similar schools, so the default ranking here is the average gap between a school and its own Similar Schools group across the comparable measures. Every raw measure is still available in the Rank by menu.

What “above similar schools” measures

For almost every measure, the Department of Education publishes three numbers: this school, the average for its Similar Schools group, and the state. Subtracting the second from the first gives the school's gap on that measure, in percentage points. This column averages those gaps across the 11 measures that carry both figures, so it answers one question: does this school do better than schools with a similar intake?

0 is level with its Similar Schools group. A school at +10 is ten percentage points better than comparable schools, averaged across those measures. Negative is below. For scale, across the 315 schools ranked here: 159 sit at or above 0 (50%), and 10 reach +10 — roughly one school in 32. The range runs from -26.8 to +21.2.

Why small schools are scaled back

A gap measured on thirty students swings a long way on chance alone. Across this dataset the spread of gaps is about twice as wide for schools under 60 students as for schools of 350 to 1,200, which is what sampling noise looks like rather than small schools genuinely differing more. So each school's gap is scaled by the share of it that is signal, estimated from its enrolment: a 25-student school keeps about a fifth of its measured gap, a 400-student school about four fifths, and a large school almost all of it. Without this the top of the table was mostly schools of nine to thirty students. The unscaled figure is still available as Unadjusted gap in the Rank by menu, and as similar_schools_advantage in the JSON alongside each school's similar_schools_reliability.

One caution. The measures do not vary by the same amount between schools: student survey endorsement spreads across tens of percentage points, while VCE completion is squeezed into the few points below 100. So a school well clear of the pack here is often well clear on the survey measures in particular. A standardised version that corrects for this, similar_schools_index, is published in schools.json for anyone who wants it.

Disadvantage (the SFOE band)

The Student Family Occupation and Education index is the Department's measure of a school community's socio-educational disadvantage, reported as a band:

• Low — least socio-educational disadvantage
• Low-Medium — below average socio-educational disadvantage
• Medium — above average socio-educational disadvantage
• High — most socio-educational disadvantage

Suppressed figures

The Department withholds a figure when a cohort is too small: NDA where there is no data available and NDP where publishing it would risk identifying students. Those become null in the JSON, with the reason kept on the measure, and the school is simply not ranked on that measure.

Source and licence

School register: School Locations, published by the Department of Education on DataVic under Creative Commons Attribution 4.0. Performance measures: each school's Annual Report to the School Community, addressed by Department of Education school number, with the document's URL and checksum recorded on every school record. The Department's own guide to these measures is here.

JSON

rankings.jsonschools.jsonindex.jsonopenapi.json