Which is the hardest place in Australia to pass a driving test?

That question cannot be answered nationally, because six of the eight jurisdictions publish nothing at centre level that has been read here. Within the Western Australian release held here, Midland is the lowest named centre at 28.3 percent and Cannington the highest at 56.4 percent, a gap of 28.1 percentage points inside one state.

Why not show a table of every test centre?

Because the numbers for most of them are not held. The Western Australian release covers 22 centres and two of them are named in the material here. Filling in the other twenty, or inventing rows for other states, would turn a real finding into a fake table. The two named centres are shown, the rest are marked as not held, and that is the whole story.

Does a low centre pass rate mean the testing officers are harsher?

It does not follow. A centre sits in one place, and its candidates learn to drive on the roads around it. Traffic, road types, the mix of first attempts against repeat attempts and the local supply of lessons all differ. Outcome data records what happened, not why, and any release that reports outcomes cannot on its own separate a hard route from a hard cohort.

Study

The hardest and easiest places to pass

Inside one Western Australian release, two assessment centres sit 28.1 percentage points apart. That gap is real and it is published here. A national table of test centres is not, because those numbers are not held.

Reviewed on 2026-08-28

2 of 22 centres named

Lowest named centre: 28.3%Midland, Western Australia
Highest named centre: 56.4%Cannington, Western Australia
Gap inside one state: 28.1 pointsAgainst a statewide figure of 42.3% across 244,557 assessments

The centres that are actually known

Two rows, because two centres are named in the material held here. The release covers 22. The other 20 are not reproduced, and no row for them is estimated.

Named assessment centres against the statewide figure

Three bars and one gap. The two centres sit either side of the statewide figure, inside one state, under one set of rules.

Cannington56.4%The highest centre named in the release held here
Western Australia statewide42.3%Across 244,557 assessments in the same release, covering 22 centres.
Midland28.3%The lowest centre named in the release held here
The other 20 centresNot reproduced hereThey are in the release and they are not in the material held on this site, so no bar is drawn for any of them.

Bars run from zero to 100 percent of assessments passed. 3 of 4 rows have a figure. The other 1 row is drawn as a gap rather than filled in.

The gap of 28.1 points is the higher named centre minus the lower one. Both inputs are on the chart, and the statewide bar between them is there so neither centre is mistaken for the state.

Freedom of information release WA Department of Transport, freedom of information release 2026-0140. Read on 27 August 2026.

Named assessment centres from the Western Australian release, with their pass rates
Centre	Jurisdiction	Pass rate	Against the statewide figure	Standing in the release
Midland	Western Australia	28.3%	14 points below 42.3%	The lowest centre named in the release held here
Cannington	Western Australia	56.4%	14.1 points above 42.3%	The highest centre named in the release held here

A candidate at the lower centre and a candidate at the higher one sat the same assessment under the same rules, set by the same agency, in the same state. Whatever explains the gap, it is not the written rules.

A low centre figure is not proof of a harsh testing officer

A centre sits in one place, and its candidates learn to drive on the roads around it. Traffic, road types, the share of first attempts against repeat attempts, and how easy it is to get lessons nearby all differ from suburb to suburb. Outcome data records what happened. It cannot separate a hard route from a hard group of candidates, and any site telling you which centre to book at because it is easier is guessing on your behalf.

The gap is still worth publishing. It tells a learner that where they sit the test is not a neutral detail, and it tells anyone reading a single statewide average that the average is hiding a wide spread underneath it.

Who holds what, jurisdiction by jurisdiction

Before anyone builds a national centre table, this is the state of the evidence. Two columns: what exists in public, and what is held here.

Centre level outcomes, jurisdiction by jurisdiction

A filled cell means something at centre level is held on this site. An empty one means nothing is, and the table under it says whether that is because nothing exists or because what exists is already published somewhere better.

NSWNew South WalesStatewide figure only. The per centre table is public and is already republished elsewhere, so there is nothing to add by copying it.
WAWestern AustraliaStatewide figure, the number of centres, the total assessments, and two named centres.
QLDQueenslandStatewide figure. The area breakdown is in the open dataset and is the right place to read it.
VICVictoriaStatewide figure and the attempt number breakdown. No Victorian centre figure will be published here while the only available numbers come from a single school's own candidates.
SASouth AustraliaNothing.
TASTasmaniaAn unattributed statewide figure, flagged as such.
NTNorthern TerritoryNothing.
ACTAustralian Capital TerritoryNothing.

5 of 8 jurisdictions have something here. The rest are marked empty rather than filled in from a source that has not been read.

Availability of test centre level driving test outcome data by Australian jurisdiction
Jurisdiction	What exists in public	What is held here
New South Wales	Transport for NSW publishes outcomes for roughly 150 individual test centres through its Registration and Licensing Statistics portal.	Statewide figure only. The per centre table is public and is already republished elsewhere, so there is nothing to add by copying it.
Western Australia	Not published as a matter of course. Centre level outcomes for 22 assessment centres were released under freedom of information application 2026-0140.	Statewide figure, the number of centres, the total assessments, and two named centres.
Queensland	Practical driving test results are published as open data, broken down by local government area rather than by individual centre.	Statewide figure. The area breakdown is in the open dataset and is the right place to read it.
Victoria	Nothing at centre level. A freedom of information request for per centre outcomes was refused, and the figures that circulate come from one driving school reporting on its own students.	Statewide figure and the attempt number breakdown. No Victorian centre figure will be published here while the only available numbers come from a single school's own candidates.
South Australia	Nothing read.	Nothing.
Tasmania	Nothing read.	An unattributed statewide figure, flagged as such.
Northern Territory	Nothing read.	Nothing.
Australian Capital Territory	Nothing read.	Nothing.

New South Wales publishes its per centre outcomes openly and a commercial site, audrive.net, has already republished 191 centres nationally from that data and its own request to the Western Australian department. Copying a table that already exists twice would add nothing, so this page does not try.

How this page was made

What the release containsPractical Driving Assessment outcomes across 22 Western Australian assessment centres and 244,557 assessments, obtained under freedom of information rather than published as a matter of course.
What is reproduced hereThe statewide figure, the number of centres, the total assessments, and 2 named centres. Everything on this page is one of those five things.
The one calculationThe gap of 28.1 points is a subtraction between the two named centres. Nothing is averaged, ranked or weighted, and no centre outside this release appears anywhere on the page.
Why there is no national tableSix of the eight jurisdictions have nothing at centre level that has been read here, and the two that do publish already have their tables in public. A national table would be two real states and six invented ones.

What has been drafted to close the gap

Two freedom of information letters sit in this repository. Their status is stated plainly, because a page that implies a request is already in flight when it is not is doing the same thing as a made up number.

Western Australia: Department of Transport, freedom of information office; A repeat of release 2026-0140: Practical Driving Assessment counts, passes and fails by centre and by month, for the period after the one already covered. Letter drafted and not yet sent. Nothing on this site will move until a reply arrives, and a reply is not guaranteed.
Victoria: Department of Transport and Planning, freedom of information unit; A refresh of the statewide drive test dataset, with the attempt number and early termination detail kept in. Letter drafted and not yet sent. A per centre request has already been refused once, so this one asks for the refresh rather than repeating a request that failed.

When a reply arrives, the figures go on this page with the release number and the date, the same way the current two did. Until then there is nothing more to show, and nothing will be filled in to look busier.

Questions people actually ask

Question: Which is the hardest place in Australia to pass a driving test?; That question cannot be answered nationally, because six of the eight jurisdictions publish nothing at centre level that has been read here. Within the Western Australian release held here, Midland is the lowest named centre at 28.3 percent and Cannington the highest at 56.4 percent, a gap of 28.1 percentage points inside one state.
Question: Why not show a table of every test centre?; Because the numbers for most of them are not held. The Western Australian release covers 22 centres and two of them are named in the material here. Filling in the other twenty, or inventing rows for other states, would turn a real finding into a fake table. The two named centres are shown, the rest are marked as not held, and that is the whole story.
Question: Does a low centre pass rate mean the testing officers are harsher?; It does not follow. A centre sits in one place, and its candidates learn to drive on the roads around it. Traffic, road types, the mix of first attempts against repeat attempts and the local supply of lessons all differ. Outcome data records what happened, not why, and any release that reports outcomes cannot on its own separate a hard route from a hard cohort.

Last checked 28 August 2026.

The jurisdictions behind these figures

Every figure above is read off one of these pages, where it sits next to the agency that published it and the date it was checked.