France’s Classroom Warning: PISA Keeps Exposing the Same Failure

France has spent 25 years measuring weaknesses in its schools – without finding a convincing way to reverse them. Institut Montaigne argues that PISA 2025 should not become another burst of national alarm followed by another round of reforms and another three years of drift.

France remains around the OECD average, with too many struggling pupils, too few top performers and stubbornly severe educational inequalities. The deeper failure is strategic: governments keep changing the machinery before they have properly tested whether the previous changes worked.

France knows the problem already

PISA has been examining French 15-year-olds since 2000, and the diagnosis has become painfully familiar.

France combines a substantial group of pupils lacking basic skills with a relatively small elite reaching the highest performance levels. Social background also continues to weigh heavily on educational outcomes.

Institut Montaigne’s point is blunt: another ranking matters less than whether France finally learns to act on what decades of evidence have already shown.

Reform overload is wrecking evaluation

The pupils tested in PISA 2025 entered primary school in 2015 and have already lived through eight education ministers and successive policy changes.

Some reforms directly affected them, others arrived too late, while Covid-19 disrupted their schooling between primary and secondary education.

This constant churn creates a basic governance problem. Policies need years to reach classrooms, be absorbed by teachers and produce measurable effects. France repeatedly changes direction before that cycle is complete.

The extremes expose the weakness

Average scores can hide the scale of France’s problem.

In PISA 2022 science, only 8 per cent of French pupils reached the highest proficiency levels, compared with 24.4 per cent in Singapore. At the other end, 24 per cent of French pupils fell below the basic proficiency threshold, against 7.8 per cent in Singapore.

Applied to France’s roughly 781,000 15-year-olds, a Singapore-like distribution would mean around 128,000 more high performers and nearly 127,000 fewer pupils below the minimum threshold.

That gap matters far beyond school rankings.

Education is becoming a sovereignty problem

Institut Montaigne wants France to treat PISA with the seriousness it eventually gave international university rankings.

Weak basic skills increase the risk of fragile educational and employment trajectories. Too few exceptional pupils, meanwhile, means a smaller future pool for science, technology, research and innovation.

France cannot demand technological sovereignty and stronger competitiveness while ignoring the school system producing its future engineers, researchers and skilled workers.

Paris keeps reaching for another reform

The analysis warns against blaming PISA 2025 on policies that could not possibly have affected the pupils being tested.

The reduction of class sizes in disadvantaged primary schools began when this cohort was already too old to benefit. More recent reforms of the first years of secondary education arrived later still.

There are tentative positive signals – PIRLS 2021 showed French reading performance stabilising after 15 years of decline – but Institut Montaigne stresses that policies need time before governments can credibly judge them.

AI raises the stakes again

Artificial intelligence makes France’s strategic confusion even more dangerous.

Schools must now decide which skills pupils must master independently, what can sensibly be delegated to AI and how knowledge should be assessed when technology can assist with schoolwork.

By 2029, PISA itself will directly examine media and AI literacy. France therefore faces another transformation before it has solved many of its older weaknesses.

The stark truth: France has diagnosed enough – now it needs direction

Institut Montaigne calls for fewer priorities, greater stability and serious evaluation instead of endless policy churn. Governments should keep measures that work even when they were introduced by political predecessors.

Another disappointing PISA result would not be the real embarrassment. Repeating the same debate without building a durable strategy would be.