05 Aug When Maths and Reading Difficulties Appear Together
Parents are usually told about one problem at a time. The school raises reading. Or the school raises maths. Rarely does anyone say that maths and reading difficulties frequently travel together, and that when they do, the picture underneath is different from either difficulty on its own.
A systematic review with meta-analysis published in Educational Psychology Review on 6 July 2026 by Waaland, Rønneberg and ten Braak has now examined this question directly. It asked a simple thing that had never been properly answered: when a child has difficulties in both maths and reading, does maths intervention actually help? The answer is encouraging, and the detail matters a great deal if you are trying to work out what your child needs.
How common are co-occurring maths and reading difficulties?
Far more common than the way schools discuss them would suggest. Estimates vary across studies, but the research points to somewhere between 30 and 50 percent of children with reading difficulties also having difficulties in maths.
Think about what that means in practice. If your child has been flagged for reading, there is close to a coin-flip chance that maths is also affected. If the school only screened reading, the maths side of the picture may simply never have been looked at.
This happens for a structural reason. Most school screening is subject-specific. A literacy screen finds literacy problems. A numeracy screen finds numeracy problems. Neither is designed to tell you whether both are present in the same child, and almost no screening tool looks underneath both at the cognitive processes they share.
Key Takeaway
Maths and reading difficulties overlap far more often than most parents are told. Between 30 and 50 percent of children who struggle with reading also struggle with maths. A major new meta-analysis published in July 2026 found that well-designed maths teaching does work for these children, producing gains of roughly half a standard deviation. The interventions that worked shared four features: explicit instruction, clear structure, deliberate repetition, and taught strategies. The same research also found that the threshold a school uses to decide a child “qualifies” for support varies enormously, which means a child identified in one school may be missed entirely in another.
Why two difficulties are not simply double the trouble
Here is the finding that changes how parents should think about this. Children with co-occurring maths and reading difficulties do not just have two separate problems stacked on top of each other. They tend to show greater challenges in three underlying cognitive areas than children with either difficulty alone:
Executive function. This is the set of processes that let a child hold information in mind while working with it, resist the pull of distraction, and switch attention when the task changes. A 2023 meta-analysis by Viesel-Nordmeyer and colleagues found something striking here. For maths and reading skills themselves, the co-occurring group looked roughly like the two isolated groups added together. For executive function, the pattern was different, which suggests executive function is a shared root rather than a separate problem in each subject.
Language. Children with co-occurring difficulties perform lower on language measures than children with isolated difficulties. Willcutt and colleagues found greater challenges in verbal comprehension. Snowling and colleagues found more severe language difficulties in the co-occurring group.
Rapid naming. This is the speed at which a child retrieves and says familiar items, such as letters, digits or colours. It is linked to fluency in both reading and arithmetic, and it accounts for part of the shared variance between the two.
The practical consequence is significant. These are precisely the abilities a child needs to take in an instruction, hold it, and act on it. When they are weaker, classroom support that works for other children may not land the same way. This is why a child can sit through the same intervention as a classmate and get much less from it.
There is a further detail worth knowing. Children with co-occurring difficulties perform worse on word problems than children with maths difficulties alone, and this remains true even when the word problems are read aloud to them. Removing the reading demand does not remove the gap. Whatever is driving the difficulty sits deeper than decoding the words on the page.
What the 2026 meta-analysis actually found
The researchers searched the literature through January 2026 and applied strict criteria. Studies had to involve children in kindergarten through third grade, had to formally identify children with difficulties in both domains, and had to compare them against a control group of children with the same profile receiving ordinary classroom teaching.
Only nine studies met the bar. That is a small evidence base, and the authors are appropriately cautious about it.
The result was positive. The overall mean effect was g = 0.47 including outliers, and g = 0.52 with outliers removed. In plain terms, an effect of around half a standard deviation would take a child sitting near the 25th percentile to somewhere around the 40th. That is not a cure, and it is not trivial either. It is meaningful movement in a group that is often quietly written off.
Why this matters: earlier work had painted a gloomier picture. A 2013 meta-analysis by Zheng and colleagues reported a negative effect on word problem accuracy for this group. A 2022 meta-analysis by Myers and colleagues concluded that word problem interventions were less effective for children with co-occurring difficulties by around 0.60 of a standard deviation compared to children with maths difficulties alone.
The distinction is in the question being asked. Those earlier studies largely asked whether children with co-occurring difficulties respond as well as children with maths difficulties alone. The answer to that is no. The 2026 study asked something more useful to a parent: does intervention help this child compared to a similar child who receives only standard classroom teaching? The answer to that is yes.
What the effective interventions had in common
The researchers did not stop at the headline number. They analysed what the working interventions shared, and grouped those features by the challenge they addressed.
Explicit instruction. The teacher shows exactly how, step by step, rather than expecting the child to infer the method from examples or discovery activities. This was the feature most consistently present, and it is classified as particularly relevant for lower-performing learners.
Overall structure. Predictable session shape, clear sequence, consistent routine. This reduces the load on executive function, because the child is not spending capacity working out what is happening next.
Repetition. Deliberate, planned revisiting rather than single exposure. This supports retention when working memory and retrieval speed are weaker.
Strategy learning. The child is taught named, reusable methods rather than left to invent their own. Children with co-occurring difficulties are known to use counting strategies less efficiently, so handing them an explicit strategy replaces a weak one rather than adding to it.
Notice that three of these four target executive function directly. That is not a coincidence. It reflects what the cognitive research says about where the shared difficulty sits.
If you are evaluating a support programme your school proposes, these four features are a reasonable checklist. A programme built on open-ended exploration, minimal teacher modelling, and single-pass coverage is unlikely to serve a child with this profile well.
Where the line gets drawn changes who gets help
The third finding of the review is the one with the most immediate consequence for internationally mobile families.
The researchers found considerable variation in the cut-off criteria studies used to identify children with co-occurring maths and reading difficulties. Some used the 25th percentile. Others used far stricter thresholds. There is no settled standard.
This is not merely an academic problem. Cut-points impose an artificial line on abilities that are genuinely continuous. Shift the line slightly and a different set of children falls inside it. A child at the 22nd percentile is identified under one threshold and invisible under another, with no change whatsoever in the child.
For a family moving between school systems, this is a live issue. A child receiving structured support in a British curriculum school in Dubai may arrive at an IB school in Zurich and be told they do not meet criteria. Nothing about the child has changed. The threshold has.
The protection against this is a properly documented assessment with full standardised scores, referenced to internationally recognised diagnostic frameworks rather than to one school’s internal threshold. Scores travel. Local eligibility decisions do not.
What this means if your child is struggling in one subject
Three practical conclusions follow from this research.
One difficulty warrants looking at both. If reading has been raised, ask what the maths data shows, and vice versa. Given the overlap rates, assuming the other domain is fine is not a safe assumption.
The subjects are not the whole picture. Executive function, language and rapid naming sit underneath both. An assessment that measures only attainment in reading and maths tells you where your child is, not why. Without the why, the support plan is guesswork.
Early matters. The reviewed studies focused on kindergarten to third grade for a specific reason. At that stage mathematical skills are still closely interrelated and highly responsive. Support delivered at seven does more than the same support delivered at eleven.
What a comprehensive assessment examines
A psychoeducational assessment that addresses this profile properly needs to look across all of it in one process, rather than producing two partial pictures at different times.
At Global Education Testing, assessments are conducted by HCPC-registered educational psychologists and cover cognitive ability, attainment across both literacy and numeracy, the underlying processing skills implicated in co-occurring difficulties, and attention and executive function. The battery includes the WISC-V for children under 18, the WIAT-3 UK for attainment across both domains, the CTOPP-2 for phonological processing and rapid naming, and the MOXO continuous performance test for attention. Reports are written to DSM-5-TR and ICD-11 standards, which is what makes them portable between school systems and examination boards.
Assessments are delivered by secure video link, which means a family in Bangkok, Riyadh or São Paulo works with the same psychologist and the same standardised instruments as a family in London.
Questions worth asking your child’s school
- Which screening tools have been used, and did any of them cover both maths and reading?
- What percentile is my child at in each domain, not just which band they are in?
- What threshold does this school use to decide a child receives support?
- Does the intervention being proposed include explicit instruction, taught strategies and planned repetition?
- Has anyone assessed working memory, processing speed or rapid naming?
Alexander Bentley-Sutherland is the CEO of Global Education Testing, the leading provider of Learning Development Testing tailored specifically for the International and Private School community worldwide.