The short answer
The PISA reading gap between boys and girls remains an educational imperative. In 2022, girls outperformed boys in reading in almost every participating education system, while 31% of boys and 22% of girls across the OECD did not reach baseline reading proficiency.`
The OECD-wide gap did not widen between PISA 2018 and PISA 2022. It narrowed from around 30 points to 24. But that does not make the issue less serious. Reading performance fell overall, and a large proportion of boys remain below the baseline level needed to understand, use, evaluate and reflect on written texts.
The appropriate response is not a nostalgic return to “back to basics”, nor technology for its own sake. It is expert teaching, knowledge-rich curricula, sustained attention, early support for struggling readers and deliberate decisions about when technology helps learning.
What does PISA 2022 tell us about boys’ and girls’ reading?
PISA assesses the reading, mathematics and science of 15-year-olds. Its reading assessment is not simply a test of whether students can decode words. It looks at whether they can understand, use, evaluate and reflect on texts.
In 2022, girls outperformed boys in reading by 24 score points on average across OECD countries. Girls performed better in 79 of the 81 participating countries and economies where comparisons were available.
The difference also appears among students who are struggling most. Across the OECD:
- 31% of boys did not reach Level 2, PISA’s baseline level of reading proficiency.
- 22% of girls did not reach Level 2.
- Girls outperformed boys by 34 points among the lowest-performing students.
- Girls also outperformed boys among the highest-performing students.
These figures should not be used to make crude claims about what boys or girls are naturally capable of. The gap varies considerably between countries and over time. That alone tells us it is shaped by culture, experience, teaching and wider social conditions, rather than being something schools should simply accept.
Source: OECD, PISA 2022 Results, Volume I.
Is the PISA reading gap widening?
Not across the OECD in the latest published PISA comparison.
In PISA 2018, girls outperformed boys in reading by almost 30 points across OECD countries. In 2022, the gap was 24 points. On that measure, the gap narrowed.
That is an important correction to make, particularly when public discussion is quick to turn a worrying trend into a more dramatic one. But it would be a mistake to confuse a narrower gap with a satisfactory outcome.
Reading performance fell by around 10 points across the OECD between 2018 and 2022. A gap can narrow because one group’s performance falls more quickly. The more important issue is that too many young people, and disproportionately boys, are leaving compulsory education without secure reading proficiency.
Source: OECD, PISA 2018 Results, Volume II and OECD, PISA 2022 Results, Volume I.
Why does the Nordic context matter?
The Nordic countries are often treated as a single education story. They are not. Their systems, outcomes and trends differ.
But the PISA 2022 results give us good reason to pay attention to reading and to the gender gap within the region.
Norway is particularly striking. Girls outperformed boys in reading by 42 score points. Thirty-four per cent of Norwegian boys were below Level 2 in reading, compared with 20% of girls. Finland was also among the countries with the largest reading gaps in favour of girls.
Sweden and Denmark performed above the OECD average in reading in 2022, but the broader pattern remains. Girls outperformed boys in reading across the Nordic countries, and overall performance cannot be the only measure of whether a system is serving all its young people well.
Source: OECD PISA 2022 country note for Norway.
Why is reading proficiency an educational imperative?
Reading is not simply one subject among others.
It is one of the main ways young people encounter ideas beyond their immediate experience. It helps them follow an argument, distinguish evidence from assertion, understand complex instructions, build knowledge and participate in public life.
The arrival of generative AI gives this added urgency. Students will increasingly encounter fluent, plausible text that may be incomplete, misleading or simply wrong. They need the reading knowledge and habits to assess it. They need enough background knowledge to recognise when something does not make sense. They need the confidence to pause rather than accept the first apparently convincing answer.
That is why I find the discussion about boys’ reading more important than a league table issue. A persistent gap matters because it can become a gap in later study, work, confidence and participation.
Does this mean a return to “back to basics” teaching?
No. I do not think “back to basics” is a useful answer to a complex educational problem.
Deep learning requires rigour, but rigour is not nostalgia. It is not simply more worksheets, more silent compliance or a simplified story about how children used to learn.
Getting a young person to read something difficult, stay with it when they do not understand it immediately, connect it with previous learning, discuss it, question it, retrieve it later and use it in a new setting is demanding work. It requires teachers with secure subject knowledge and strong pedagogical judgement.
Teachers need to decide when to model, explain, practise, question, revisit, extend and give students room to think. They need to know their students well enough to recognise the difference between productive struggle and a child quietly falling behind.
That is highly skilled professional work.
Are laptops and AI the solution to weak reading?
No. A laptop is not a pedagogy. AI is not a pedagogy. Access to information is not the same thing as learning.
Technology can support excellent teaching. It can provide access to rich texts, useful feedback, carefully designed practice and new ways to explore ideas. It can also make distraction easier, reward superficial completion and allow students to produce work they have not understood.
The question is not whether technology belongs in education. It plainly does. The question is what students are being asked to do with it, what they need to know before using it and whether it strengthens or weakens their attention to the task.
In an AI-rich world, schools should be even more deliberate about creating opportunities for students to read carefully, write independently, discuss ideas and stay with material that does not offer an immediate answer.
What should schools and systems do?
There is no single intervention that will close a reading gap. The work sits across curriculum, teaching, assessment, school culture, family life and technology use.
A serious response would include:
- identifying reading difficulty early and responding before students decide that reading is simply not for them;
- protecting regular time for sustained reading of worthwhile texts;
- building students’ background knowledge, because comprehension depends on more than generic reading strategies;
- supporting teachers to teach reading, language and disciplinary literacy with confidence;
- maintaining high expectations for boys without assuming that boys are one kind of learner;
- making deliberate decisions about devices, platforms and AI, rather than treating digital access as an educational strategy in itself;
- looking closely at local data, since averages can hide the students and schools who need the most support.
The point is not to create a separate educational model for boys. It is to refuse the idea that weak reading is an inevitable part of being one.
A better response than panic or complacency
The PISA reading gap between boys and girls is longstanding. It has not recently widened across the OECD, but it remains large, and in parts of the Nordic region it is particularly pronounced.
We should be careful with the data. We should resist easy stories about boys, girls, screens or supposedly simpler times. But we should not learn to live with a situation in which so many young people, disproportionately boys, leave school without secure reading proficiency.
The work is more demanding than buying another platform or reviving an old slogan. It means investing in expert teachers, serious reading, rich knowledge, sustained attention and technology used with clear educational purpose.
That seems a more credible place to begin.