New data: reading scores plunge from 59% to 29%

New Ministry of Education data suggests the sharp early gains reported in the Government's phonics checks are not holding up over the course of a school term, reigniting debate over how far the structured literacy reforms have actually moved the dial on reading.

According to figures obtained under the Official Information Act by the Aotearoa Educators Collective and released by the Te Tai Tokerau Principals Association, 59 percent of students were at or above expectations at the 20 week Phonics Check in Term One this year. By the 40 week check, that figure had fallen to 29 percent. Term Two followed a similar pattern, with 54 percent at 20 weeks dropping to 27 percent at 40 weeks.

The association says the wider data set shows achievement patterns were broadly unchanged from the previous year and that progress between checkpoints had slowed. It also says the gaps affecting Māori and Pacific learners, and students in communities facing greater barriers, show no statistically significant improvement. That matters because closing those gaps has been one of the stated goals of the literacy overhaul.

The record

The Government's curriculum reset made structured literacy the mandated teaching approach in primary schools, built around explicit phonics instruction. The Phonics Check, sat at set points in a child's first year of school, was introduced as one of the key tools to track whether that approach was working.

Early results from those checks have previously been cited publicly as evidence the reforms were lifting achievement quickly. The newly surfaced data compares the 20 week snapshot, taken relatively early in a child's phonics instruction, against the 40 week checkpoint later in the same term. The gap between those two numbers, rather than either figure alone, is what the principals' association says raises questions about how sustained the improvement really is.

The association is careful to frame this as a data problem, not a verdict on the teaching method. It says the figures do not show structured literacy is ineffective. What they challenge, it argues, is the certainty with which earlier claims of broad, sustained success across all schools and communities have been made.

What doesn't add up

The most obvious gap is the size of the drop itself. A fall from 59 percent to 29 percent, and from 54 percent to 27 percent, is not a marginal wobble. The article does not explain why the proportion of children meeting expectations would roughly halve within the same term, and that is the central question hanging over this data.

There is also a definitional issue the association itself flags. The Phonics Check measures phonics achievement specifically, not overall reading ability, which includes comprehension, fluency and vocabulary. That means neither the 59 percent nor the 29 percent figure tells us how well these children can actually read and understand a text. What the checks can compare is consistency over time within the same measure, and on that narrower question the numbers do not look settled.

The equity question is harder to explain away with a definitional caveat. If the reforms were meant to lift outcomes for Māori and Pacific learners and for children in higher needs communities in particular, a data set showing no statistically significant reduction in those gaps is a direct test of that promise. The article does not record any explanation offered for why that gap has persisted.

Finally, there is a timeline gap. The article does not say whether the Ministry of Education itself has published or endorsed this comparison between the 20 week and 40 week checks, or whether it was assembled independently by the two education groups from raw OIA data. That distinction matters for how much weight the comparison should carry.

The other side

No response from the Ministry of Education or the responsible Minister is included in the article, so it is not possible to report their explanation for the gap between the two checkpoints.

The strongest case for the reforms, as it appears in the article, comes from the principals' association's own caution. It explicitly says this data does not demonstrate structured literacy is ineffective, and that phonics checks capture only one slice of reading ability. Early gains at 20 weeks may reflect real early skill acquisition even if they are not yet fully embedded by the 40 week mark, and a term is a short window in which to judge a multi year curriculum change.

What happens next

The Ministry of Education holds the full data set and is the body best placed to explain the pattern between the 20 and 40 week checks. Whether it publishes a fuller breakdown, or responds directly to the equity findings, will shape how this is understood.

The figures are likely to feature in the wider election year debate over the curriculum reset, particularly around outcomes for ākonga Māori. With the general election set for 7 November 2026, expect further OIA requests, select committee questions and competing interpretations of what these checkpoints actually prove.