Writing is the one that spreads students apart.
In every other NAPLAN domain, students look more alike as they get older. In writing they look less alike, year after year, in every state that reports it. That is the most useful thing in the file for a parent deciding when to pay attention.
Writing is the onlyNAPLAN domain in which the gap between students widens as they get older — and it does so in eight of the eight jurisdictions that report the data, growing in each of the last four years.
Standard deviation of Victorian scores, 2026: writing rises from 65.2 at Year 3 to 97.6 at Year 9. Reading falls, from 85.9 to 75.1. Spelling falls further, 84.7 to 66.8.
Every other subject converges. Writing diverges.
Standard deviation is a plain measure of how spread out a set of scores is: a small one means students clustered together, a large one means they are strung out. Comparing Year 3 with Year 9 in the same year of testing shows whether a domain pulls students together or apart as they progress through school.
Reading and spelling both converge: by Year 9 students are measurably more alike in those domains than they were at Year 3. Grammar and punctuation is flat. Numeracy spreads somewhat. Writing spreads roughly twice as much as numeracy and in the opposite direction to both reading and spelling — and reading and writing are supposedly neighbours.
Not a Victorian quirk, and not one bad year
A single cut of a single year would not be worth publishing, so it is worth saying how the pattern holds up. Three ways, independently.
Writing is the largest riser in eight of the eight jurisdictions with reportable data. Reading spread rises in none.
Tasmania is excluded because its 2026 figures are suppressed in the published file. The other eight — the seven remaining states and territories plus the national total — all show writing as the domain that spreads most.
Second, it is not a one-year artefact. Victoria’s writing spread has widened in each successive year of testing since the current scale began:
Third, the same thing shows up in the proficiency levels, which are built quite differently from the score scale. The Strongband — the large middle where most young students sit — collapses in writing and holds everywhere else.
Victorian writing, Year 3 to Year 9: Strongfalls 70.5% → 41.0%, while Exceeding rises 9.3% → 25.4% and Needs additional supportrises 3.6% → 8.8%.
Both ends grow at once. That is the signature of a distribution pulling apart rather than sliding up or down, and writing is the only domain that shows it. The middle-band collapse holds in eight of the nine jurisdictions.
What the numbers do and do not license
The comparison above is between four separate groups of students sitting in the same year — this year’s Year 3s against this year’s Year 9s — each assessed against what is expected of their own year level. It is not one cohort followed through school. So it describes how the skill is distributed at each age, not the path any individual child takes.
That distinction matters for what can honestly be concluded. The data shows a distribution that spreads with age, consistently and in every jurisdiction. It does not show why, and it cannot show that any particular intervention closes the spread. A reasonable explanation is that writing is cumulative in a way the other domains are not: it asks a student to make something rather than recognise something, and it draws on sentence control, structure, vocabulary and judgment simultaneously, so small early differences have more room to compound. That is an interpretation, offered as one, and this practice has an obvious interest in it being true — which is exactly why it is labelled rather than asserted.
What this means for your child
This is the finding in the whole dataset with the clearest practical bearing, and the useful part is about timing rather than alarm.
- A middling writing result at Year 3 deserves more attention than a middling reading result.Reading converges — students tend to look more alike by Year 9. Writing does the opposite. The same starting position implies a different trajectory in the two domains.
- Early is cheaper than late.Because later writing tasks assume control built in earlier ones, a student who reaches Year 7 without a sentence under control does not simply score lower — they lose access to the analytical paragraph, the argument, and eventually the VCE text response. Regular writing at Year 3, read back closely, is worth more than a term of coaching at Year 9.
- Do not read a strong Year 3 writing result as settled. The middle band is where most young students sit and it is the band that empties. A comfortable position at Year 3 is the least informative result in the set, because it is the one most likely to move in either direction.
- What actually helps is unglamorous.Writing often, reading widely, and having someone competent read the writing back and name what is slipping. Not test coaching — the writing test rewards the same craft good writing rewards, which is why formula coaching can lift a score and still cost a student later.
How this analysis was done
Standard deviations and proficiency percentages are read directly from ACARA’s published NAPLAN national results dataset, sheet “NAPLAN Results”, all-students rows. The arithmetic is subtraction: Year 9 value minus Year 3 value, within one year of testing and one jurisdiction. Nothing is modelled or re-estimated.
Three limits. Comparisons run across year levels within a single year of testing, so they describe four separate cohorts rather than one followed through school. The window starts at 2023, when proficiency levels replaced the ten-band scale, and cannot be extended back across that break. And Tasmania is excluded wherever jurisdictions are counted, because its 2026 figures are suppressed in the published file — the participation column still carries computed values for those rows, which would produce a badly wrong chart if read without checking the reporting flag.
Sources
- Australian Curriculum, Assessment and Reporting Authority (ACARA), NAPLAN national results dataset, 2026.Every score, percentage, effect size and classification quoted across these pages.
- ACARA, NAPLAN proficiency level descriptions.Definitions of the four levels: Needs additional support, Developing, Strong, Exceeding.
- ACARA, NAPLAN 2025 Technical Report, pp. 86–87.The effect-size and nature-of-difference methodology: the 0.2 and 0.5 thresholds separating “close to”, “above” and “substantially above”.
- ACARA, NAPLAN (National Assessment Program), and the Victorian Curriculum and Assessment Authority, NAPLAN assessments.What the assessment is, who sits it, and how it is administered in Victoria.
Is your child’s writing on the widening side?
A twenty-minute discovery call about the writing itself — what it does well, what is quietly slipping, and what would move it. Free, and without obligation.
New to all this? Start with what NAPLAN actually is →