Examination outcomes: what the published pass rates leave out

A state's overall pass rate for a major examination is one of the most widely quoted education statistics, and one of the least examined. It compresses tens of thousands of individual results into a single percentage, which makes it easy to cite and correspondingly easy to misread.

Working through published examination reports over the last several terms, we found that the compilation method itself deserves more attention than it usually gets. Some releases calculate pass rate as a share of candidates who sat the exam; others calculate it as a share of candidates who registered, which includes those who did not sit at all. The two produce meaningfully different figures from the same underlying results.

A gap that isn't always stated

When a published bulletin doesn't specify which denominator was used, we note that ambiguity directly in our report rather than picking one interpretation silently. It matters: in examinations with high absenteeism on the day, the difference between the two methods can shift a headline pass rate by several percentage points, enough to change how a year is characterised.

There's a second layer worth naming too. A state-level pass rate says nothing about the spread beneath it — whether results cluster near a pass threshold or are more evenly distributed, and whether particular subjects are dragging the average down while others hold it up.

What we do instead

Where the underlying subject-level breakdown is published, we include it in our change-over-time strips so a single headline figure can be checked against its components. Where a school or examination body responds to a report with additional context, that response is recorded alongside our findings rather than folded into the text unattributed.

None of this changes what the exam results actually were. It simply gives readers a fuller basis for judging what a published pass rate is, and isn't, telling them.