Where language gaps begin — development data and the early-years evidence. 12 verified findings — quoted verbatim, fully cited, independently checked (register v1.3.0, verified 2026-08-24).
← All domains, the protocol, and the register’s through-lines
Argo teaches from around Year 3 onward, so early childhood is not a service we offer. But it is the register's first chapter for a reason: the gaps we work with in primary and secondary students rarely start in primary or secondary school. The Australian Early Development Census (AEDC) — a national, teacher-assessed check-in on roughly 288,000 children in their first year of full-time school — gives the clearest picture of where things stand before formal schooling has had a chance to act. In 2024, the picture worsened. 23.5% of children starting school were developmentally vulnerable on at least one of the AEDC's five developmental domains, up from 22.0% in 2021 [ECD-01]. More seriously, 12.5% were vulnerable on two or more domains, up from 11.4% — the highest rate recorded across all six AEDC collections since the census began in 2009 [ECD-02]. The share of children on track across all five domains fell from 54.8% to 52.9%, though it remains above the 2009 baseline of 50.7% [ECD-03].
The decline was not confined to one area of development. Every one of the five AEDC domains got worse between 2021 and 2024, and three — social competence, emotional maturity, and physical health and wellbeing — hit their highest vulnerability levels since records began [ECD-04]. The two domains most directly tied to literacy readiness moved the same way after a period of improvement: 'language and cognitive skills (school-based)' rose to 7.7% vulnerable, and 'communication skills and general knowledge' rose to 8.9%, both having fallen between 2009 and 2018 before turning upward again [ECD-05] [VOC-11]. Within that communication domain, First Nations children faced a developmental-vulnerability rate of 17.3% in 2024 — roughly double the national rate, though this is down substantially from 21.3% in 2009, indicating a real but still wide and only slowly narrowing gap [VOC-12].
These averages sit on top of steep gradients. Sixty-one percent of children in the least-disadvantaged areas were developmentally on track across all five domains in 2024, compared with 41% in the most-disadvantaged areas — a 20-point gap tied to socioeconomic status alone [ECD-06]. Geography compounds it: 54% of children in major cities were on track, against 51% in regional areas and 43% in remote and very remote areas [ECD-06]. Royal Far West's analysis of the same data, independently corroborated on the on-track figure, found that more than one in five children in remote and very remote areas were vulnerable on two or more domains — worth noting that this specific figure carries a lower-tier source rating and an editorial correction history, so it should be treated as indicative rather than a settled headline statistic [ECD-07].
Home environment shows up early and specifically around language. Longitudinal data from the Longitudinal Study of Australian Children (LSAC), following 2,369 children, found that low levels of parent-child book reading in infancy (10 minutes or less per day) significantly raised the risk of poor vocabulary at school entry: nearly a quarter of children with low reading exposure across early childhood had poor vocabulary near school entry, compared with fewer than 8% of children with high exposure [ECD-08]. The same LSAC-based work links this further downstream — children read to daily at age 2–3 had higher average Year 3 NAPLAN reading scores than children read to less often, and having more than 30 children's books at home at that age was positively associated with higher Year 3 reading and numeracy results [ECD-09]. As a national baseline, most Australian families are already doing this: in 2017, 79% of children aged 0–2 were read to or told stories at least three days a week, with 60% read to almost daily — but 16% were not read to at all, which is the group these figures suggest is most exposed to the downstream vocabulary and reading risk [ECD-10]. All of this is correlational, drawn from observational cohort data rather than controlled trials, and family reading habits are entangled with broader socioeconomic factors that independently predict outcomes — the register does not claim a clean causal reading-to-outcome pathway here, only a consistent and fairly large association.
Preschool attendance shows a similar-shaped association. Matched AEDC and enrolment data found that children who received some early childhood education and care were far less likely to be developmentally vulnerable on one or more domains (19.9%) than children who received none (38.5%) [ECD-11]. Again, this is not evidence that preschool caused the difference on its own — families who do not access early childhood education differ systematically in ways (income, remoteness, parental work status) that also predict developmental outcomes. A methodologically stronger, covariate-adjusted analysis using LSAC data did find a significant positive association between preschool attendance and Year 3 NAPLAN results even after controlling for socio-demographic factors, which is a somewhat firmer basis for the association without being a randomised causal claim [ECD-11].
Where the register does have a genuine randomised trial in this domain, the honest result is a mixed one, and it is reported here without spin. An independent, Evidence for Learning-funded RCT of MiniLit — a small-group Tier 2 literacy intervention for struggling Year 1/2 readers, tested with 237 students across nine NSW schools — found no statistically reliable impact on the primary outcomes of reading accuracy, rate, or comprehension at 12 months, and the evaluators themselves rated those primary-outcome findings as low security [ECD-12]. It did find a significant, sustained improvement in letter-sound knowledge at six and twelve months, and implementation was uneven — only 55% of students received the recommended dose, with stronger outcomes concentrated among those who attended more consistently [ECD-12]. This is not a case against early intervention; it is a reminder that even well-designed, well-funded programs can fail to move headline reading outcomes in a randomised trial, and that dosage and fidelity matter as much as the program design itself. The register does not cheer for interventions, including ones aligned with literacy instruction generally — it reports what the trial found.
For a parent whose child is already at school, the honest takeaway is not that the early years are destiny. It is that the gaps visible at five are real, they are unevenly distributed by postcode and income, and language exposure in the home in the first few years leaves a measurable trace in vocabulary and NAPLAN scores years later. Those early gaps do not stay still — they tend to compound through the same mechanism that drives reading development at every later age: children who read more, in or out of school, encounter enormously more words than children who read less, at ratios research has put as high as 200 to one between high- and low-volume child readers, meaning a full year of one child's reading can equal roughly two days of another's [RDG-09]. That is the same Matthew effect — the rich get richer, the poor get poorer — playing out on a bigger stage than infancy, which is exactly why Argo's own work from Year 3 onward treats reading volume and vocabulary exposure as central levers rather than side details. Early is easier than late, because gaps have had less time to compound. But nothing in this domain's evidence says the door closes at five. It says the door was easier to walk through earlier, and that the same mechanisms — exposure, volume, consistency — are still the ones doing the work whenever a family or a school decides to act on them.
Share of Australian children starting school developmentally vulnerable on two or more AEDC domains — the 2024 figure is the highest of the six collections since 2009. Only the two most recent published collection values are drawn. Figures from #ECD-02.
In 2024, 23.5% of Australian children starting school were developmentally vulnerable on one or more of the five AEDC domains, up from 22.0% in 2021.
Developmentally vulnerable on one or more domains (DV1): The percentage of children developmentally vulnerable on one or more domains increased from 22.0 per cent in 2021 to 23.5 per cent in 2024.
In 2024, 12.5% of Australian children starting school were developmentally vulnerable on two or more of the five AEDC domains, up from 11.4% in 2021 — the highest level recorded across all six AEDC collections since 2009.
Developmentally vulnerable on two or more domains (DV2): The percentage of children developmentally vulnerable on two or more domains increased from 11.4 per cent in 2021 to 12.5 per cent in 2024. This is the highest level of developmental vulnerability in two or more domains across all six AEDC collections.
The proportion of children developmentally on track across all five AEDC domains fell from 54.8% in 2021 to 52.9% in 2024, though this remains above the 2009 baseline of 50.7%.
At a national level, the AEDC data shows the percentage of children who were on track on all 5 domains decreased from 54.8% in 2021 to 52.9% in 2024. However, results remain above the 2009 baseline (50.7%).
In 2024, 7.7% of Australian children were developmentally vulnerable on the language and cognitive skills (school-based) domain, an increase of 0.4 percentage points since 2021; emotional maturity had the largest single-domain increase (10.0% vulnerable, up 1.5 percentage points since 2021).
Emotional maturity: 10.0 per cent of children are developmentally vulnerable on the emotional maturity domain, an increase of 1.5 percentage points since 2021. This represents the largest increase in developmental vulnerability among all domains. Language and cognitive skills (school-based): 7.7 per cent of children are developmentally vulnerable on the language and cognitive skills (school-based) domain, an increase of 0.4 percentage points since 2021.
Developmental outcomes on the AEDC follow a clear socioeconomic gradient: 61% of children in the least-disadvantaged areas (SEIFA quintile 5) were developmentally on track on all five domains in 2024, compared with 41% of children in the most-disadvantaged areas (quintile 1); a geographic gradient is also present, with 54% on track in major cities versus 43% in remote/very remote areas.
61% of children living in the least disadvantaged areas (quintile 5 of the SEIFA Index of Relative Socio-economic Disadvantage) were on track in all five domains, compared with 41% of children living in the most disadvantaged areas (quintile 1). 54% of children in major cities were developmentally on track in all five domains, compared with 51% in regional areas and 43% in remote and very remote areas.
In 2017, almost four in five (79%) Australian children aged 0-2 were regularly read to or told stories by a parent at least 3 days a week; approximately 60% were read to 6-7 days per week, while 16% were not read to at all.
In 2017, almost 4 in 5 children aged 0–2 (738,000 or 79%) were regularly read to or told stories by a parent
In the 2024 AEDC collection, developmental vulnerability increased across all five domains compared to 2021, with three domains (social competence, emotional maturity, physical health and wellbeing) reaching their highest recorded levels since the census began in 2009; the language and cognitive skills (school-based) domain and the communication skills and general knowledge domain had fallen from 2009 to 2018 before trending upward again through 2024.
When we look across all domains, we can see that a greater proportion of Australian children were found to be developmentally vulnerable across all five domains compared to the last census in 2021. The 2024 AEDC saw record levels of developmental vulnerability in three domains: 'social competence', 'emotional maturity' and 'physical health and wellbeing' - the highest levels since the census began. The proportion of children developmentally vulnerable in the other two domains, 'communication skills and general knowledge' and 'language and cognitive skills', decreased from 2009 to 2018, but has since trended upwards.
Using LSAC data (2,369 children), the risk of poor vocabulary at school entry was significantly increased for children who experienced low levels of parent-child book reading in infancy (10 or fewer minutes per day): nearly a quarter of children with low levels of parent-child book reading across early childhood had poor vocabulary near school entry, compared with fewer than 8% of children with high levels of parent-child book reading.
the risk of poor vocabulary development was significantly increased for children who had experienced low levels of parent-child book reading (10 or fewer minutes/day). Nearly a quarter of the children who had low levels of parent-child book reading across early childhood had poor vocabulary around the time of school entry compared to fewer than 8% of the children who had high levels of parent-child book reading across early childhood.
AIHW reporting of LSAC data finds children whose parents read to them every day at age 2-3 had higher average Year 3 NAPLAN reading scores than children read to less frequently, and that having more than 30 children's books at home at age 2-3 was positively related to higher Year 3 NAPLAN reading and numeracy scores.
children whose parents read to them every day when they were 2–3 year olds had higher Year 3 NAPLAN reading scores on average, than children whose parents read to them less frequently... Having more than 30 children's books at home when children were aged 2–3 was positively related to higher NAPLAN scores in reading and numeracy in Year 3
Analysis of matched AEDC/preschool-attendance data (SCRGSP 2019, drawing on 2015 collection) found children who received some Early Childhood Education and Care (ECEC) were substantially less likely to be developmentally vulnerable on one or more AEDC domains (19.9%) than children who received no ECEC (38.5%).
In 2015, children who received some Early Child Education and Care (ECEC) were less likely to be developmentally vulnerable on 1 or more domains (19.9 %), compared to children who did not receive any ECEC (38.5%) (SCRGSP 2019).
An independent randomised controlled trial of MiniLit (a Tier 2 small-group literacy intervention for struggling Year 1/2 readers), commissioned and funded by Evidence for Learning through its Learning Impact Fund and conducted with 237 students across 9 NSW schools, found no statistically reliable impact on the primary outcomes of reading accuracy, rate, or comprehension at 12 months, but found significantly improved letter-sound knowledge at 6 months that was sustained at 12 months; the primary-outcome results were assigned a low security rating by the independent evaluators.
This randomised controlled trial found that MiniLit, as experienced in this trial, did not have an additional impact on reading, as measured by the YARC – Passage Reading (YARC-PR) tests on Accuracy, Rate and Comprehension but there was evidence of significant improvements in foundational reading skills at six months, particularly Letter Sound Knowledge which was also sustained at 12 months.
Royal Far West's analysis of the 2024 AEDC finds that only 43% of children in rural and very remote areas were developmentally on track across all five domains, and more than 20% of children in remote/very remote locations were developmentally vulnerable on two or more domains.
The latest AEDC figures show that only 43% of rural and very remote kids are developmentally on track in five key developmental areas before they start school, and over 20% of children in remote or very remote locations are developmentally vulnerable in two or more domains.
Cite any finding by its anchor — for example #ECD-01 — and the link resolves to the claim, quote and source above. The verification protocol lives on the register’s main page.