Research · Regression analysis

ICSEA vs NAPLAN — schools that beat their demographics

What looks like a high-performing school often just looks like a high-ICSEA cohort. This page strips out the demographic effect to show which schools are genuinely outperforming what their profile would predict.

Schoogaroo Research·Method: linear fit per state · 4 states · Source: ACARA / MySchool

What "residual" means

ACARA's ICSEA score captures parental occupation, parental education, school remoteness and the proportion of Indigenous students. Higher ICSEA usually means higher NAPLAN; the line on each chart shows the best-fit relationship for that state's schools.

A school's residual is the gap between its actual mean NAPLAN score and what the line would predict from its ICSEA percentile alone. A positive residual means the school is doing better than its demographics predict — that is the signal worth paying attention to.

Two schools with very different ICSEA profiles can both be over-performers. A high-ICSEA school that out-paces its line is adding value beyond what its cohort gives it. A low-ICSEA school that out-paces its line is doing the same — sometimes more impressively.

NSW — 2,047 schools, 2025

3284044805556310255075100NAPLAN meanICSEA percentile →GovernmentCatholicIndependent

Top over-performers (NSW)

SchoolSuburbSectorICSEAMean scoreResidual
Sydney Grammar SchoolDarlinghurstIndependent1274625.4+91.8
St Joseph's Primary SchoolJerilderieCatholic1011567.0+81.1
AbbotsleighWahroongaIndependent1217608.8+75.2
Al-Faisal College - CampbelltownMintoIndependent1100593.2+74.8
St Hurmizd Assyrian Primary SchoolGreenfield ParkIndependent954532.4+72.9
St Joseph's Catholic Primary School PortlandPortlandCatholic924520.8+71.4
Sutherland Public SchoolSutherlandGovernment1101588.2+69.8
Ermington Public SchoolWest RydeGovernment1117586.8+64.4
Biraban Public SchoolTorontoGovernment871502.6+63.4
Hurstville Public SchoolHurstvilleGovernment1102581.6+62.2
Beecroft Public SchoolBeecroftGovernment1187595.4+61.8
A.G.B.U. Alexander Primary SchoolDuffys ForestIndependent962523.6+60.0
Northcross Christian SchoolRydeIndependent1194593.6+60.0
Matthew Pearce Public SchoolBaulkham HillsGovernment1177592.6+60.0
Wyong Public SchoolWyongGovernment931511.0+59.6

VIC — 1,602 schools, 2025

3514124745355960255075100NAPLAN meanICSEA percentile →GovernmentCatholicIndependent

Top over-performers (VIC)

SchoolSuburbSectorICSEAMean scoreResidual
Heyfield Primary SchoolHeyfieldGovernment956546.0+85.1
Lighthouse Christian College CranbourneCranbourneIndependent1093588.4+75.9
Koorlong Primary SchoolKoorlongGovernment983545.2+73.1
Coleraine Primary SchoolColeraineGovernment975531.8+63.4
Ballarat Clarendon CollegeBallaratIndependent1172590.2+62.7
Wheelers Hill Primary SchoolWheelers HillGovernment1151585.4+60.8
St Andrews Christian CollegeWantirna SouthIndependent1166587.0+60.5
Presbyterian Ladies' CollegeBurwoodIndependent1207588.8+60.4
Haileybury College - Keysborough BoysKeysboroughIndependent1191588.4+60.0
Oakleigh South Primary SchoolOakleigh SouthGovernment1163585.4+58.9
Wendouree Primary SchoolWendoureeGovernment917505.8+57.1
Willow Grove Primary SchoolWillow GroveGovernment1002535.2+55.6
Huntingtower SchoolMount WaverleyIndependent1201584.0+55.6
Gormandale And District Primary SchoolGormandaleGovernment940509.0+53.8
Beverley Hills Primary SchoolDoncaster EastGovernment1153579.2+53.6

QLD — 1,160 schools, 2025

2693494285085870255075100NAPLAN meanICSEA percentile →GovernmentCatholicIndependent

Top over-performers (QLD)

SchoolSuburbSectorICSEAMean scoreResidual
Mount Tyson State SchoolMount TysonGovernment941555.4+116.4
Mount Marrow State SchoolMount MarrowGovernment955529.6+83.7
Banana State SchoolBananaGovernment941521.8+81.7
Yarwun State SchoolYarwunGovernment999539.0+71.2
Lake Clarendon State SchoolLake ClarendonGovernment961513.8+65.6
Crawford State SchoolCrawfordGovernment913496.4+65.5
Sunnybank Hills State SchoolSunnybank HillsGovernment1119581.6+64.1
Charters Towers School of Distance EducationMillchesterGovernment948506.2+63.8
Townsville Central State SchoolNorth WardGovernment930496.6+61.1
Ipswich Grammar SchoolIpswichIndependent1114574.2+57.8
Jondaryan State SchoolJondaryanGovernment948500.2+57.8
Babinda State SchoolBabindaGovernment931492.8+57.3
Dimbulah State SchoolDimbulahGovernment869477.6+56.0
Cardwell State SchoolCardwellGovernment932491.6+55.0
Federal State SchoolFederalGovernment981511.6+53.0

WA — 783 schools, 2025

2343214084945810255075100NAPLAN meanICSEA percentile →GovernmentCatholicIndependent

Top over-performers (WA)

SchoolSuburbSectorICSEAMean scoreResidual
Australian Islamic College (Dianella)DianellaIndependent1018557.2+79.1
Morawa District High SchoolMorawaGovernment902505.2+78.2
John Calvin School (Albany)AlbanyIndependent1000540.2+71.4
Greenfields Primary SchoolGreenfieldsGovernment928501.4+67.4
Narembeen District High SchoolNarembeenGovernment995533.0+66.5
Australian Islamic College (Kewdale)KewdaleIndependent1037553.2+65.8
Ravensthorpe District High SchoolRavensthorpeGovernment978522.8+65.6
Darkan Primary SchoolDarkanGovernment963514.6+65.5
Leinster Community SchoolLeinsterGovernment926494.6+60.6
OneSchool Global WAWillettonIndependent986518.0+57.3
Wongan Hills District High SchoolWongan HillsGovernment920487.8+56.2
Cervantes Primary SchoolCervantesGovernment955501.4+55.8
Shark Bay SchoolDenhamGovernment901482.6+55.6
Meekatharra School Of The AirBluff PointGovernment975510.8+54.8
Condingup Primary SchoolCondingupGovernment970505.2+51.5

Methodology

For each state we computed each school's mean Year 5 NAPLAN score in the most recent year, then fit a simple linear regression of mean score against ICSEA percentile. Schools with fewer than 3 reported subjects in the year are excluded. Residuals are the difference between the actual score and the line, in NAPLAN-points.

A residual of +50 means the school scored 50 NAPLAN points higher than the average school at the same ICSEA percentile in that state. Residuals are not directly comparable across states because each state has its own line.

?

Frequently asked questions

Should I just pick a high-residual school?

Probably not on that fact alone. A high residual is useful because it suggests the school is doing better than its demographic profile would predict, but it does not tell you whether the school is the right environment, whether the result is stable over time, or whether the cohort is large enough for you to trust the size of the gap.

Why use ICSEA percentile rather than raw ICSEA?

Mostly for modelling convenience and readability. Percentile keeps the x-axis on a familiar 0–100 scale. The broad pattern should be similar with raw ICSEA, so the main analytical choice here is not the exact x-scale but the decision to compare schools relative to their demographic context at all.

Why are there small schools with very large positive or negative residuals?

The most likely explanation is volatility. Smaller cohorts can move around more sharply from year to year, so a large residual at a very small school should be read as a lead worth checking rather than as a settled conclusion about school quality.