For more than two decades, South Korea has shown up near the top of one of the most closely watched measures of literacy on the planet.
Every three years, the OECD tests hundreds of thousands of 15-year-olds worldwide on how well they can read, interpret, and reason through text, and every three years, Korean students land in the group of countries doing it best. By most conventional measures, this counts as an enormous success story.
What happens to those same readers a decade or two later tells a much less tidy story.
What the test scores show
In the most recent PISA 2022 results, Korean students scored 515 points in reading, placing the country between second and 12th among the 81 participating countries and economies, and between first and seventh among the 37 OECD member nations. That score sits roughly 39 points above the OECD average, at a moment when reading scores fell across most of the developed world following pandemic-era school disruptions. Korea’s own score barely moved during that period, which is itself notable. Whatever the country’s schools are doing to teach reading comprehension, it has proven remarkably resistant to the kind of disruption that dragged other systems down.
What the adult surveys show
Ask the same population, ten or twenty years further into adulthood, how much they actually read, and the picture changes dramatically. Korea’s Ministry of Culture, Sports and Tourism has tracked adult reading rates since 1994, when 86.8 percent of adults reported reading at least one book that year. By the survey covering September 2022 to August 2023, that figure had fallen to 43 percent, the lowest reading rate recorded since the survey began. Roughly six in ten adults surveyed had read nothing at all that year. Among the adults who did read, the average came to 3.9 books annually, a figure that dropped to 1.7 once webtoons were excluded and only printed books were counted. The same survey found a wide gap by income, with a reading rate of just 9.8 percent among adults earning under 2 million won a month, compared with 54.7 percent among those earning more than 5 million won.
The decline didn’t stop there. By the most recent tally, the adult reading rate had slipped further to 38.5 percent in 2025, down from that 43 percent two years earlier. Over three decades, the share of Korean adults who read even a single book in a given year fell by more than 48 percentage points, from 86.8 percent to 38.5 percent, even as the country’s teenagers kept scoring near the top of the world on reading tests.
Where the two numbers actually diverge
Put side by side, the two results don’t contradict each other so much as measure two different things that get talked about as if they were one. PISA is testing a skill: can a 15-year-old decode a passage, follow an argument, and answer questions about what it means. That skill is teachable in a fairly direct way, through curriculum, practice, and the kind of sustained, structured instruction Korean schools are well known for. Whether a person chooses to spend a free evening with a book once no one is testing them on it anymore is a different question entirely, shaped less by instruction and more by habit, competing demands on time, and whatever reading has come to feel like by the time school pressure lifts.
What seems to hold up across both data sets is a fairly simple distinction. Being taught to read well is something a school system can engineer and measure with real precision. Wanting to keep reading once nobody is grading it is a separate outcome, and it doesn’t automatically follow from the first one just because both involve the same activity on paper.
Policymakers are already treating these as two different problems
South Korea’s own government appears to have reached the same conclusion. In April 2026, the Ministry of Culture, Sports and Tourism launched a nationwide campaign called 2026 Book-Reading Korea, running customized reading programs at roughly 200 bookstores and late-night reading events at 140 more, aimed squarely at adults rather than students already inside the school system that produces those strong PISA scores.
Announcing the campaign, Culture Minister Choi Hwi-young said, “In the age of artificial intelligence, we will work to strengthen the foundation of a humanities powerhouse and spread reading culture so that the joy of reading can lead to the power of thought.” Notice the shape of the campaign. There’s no curriculum reform in it, no push to teach comprehension differently, because that part of the system already scores near the top of the world. What’s being funded instead is bookstore events, reading challenges, and celebrity endorsements, the kind of effort aimed at rebuilding a habit that classroom instruction was never actually designed to instill.
Why this gap matters beyond South Korea
South Korea’s numbers are unusually stark, but the underlying pattern isn’t unique to one country. Plenty of education systems chase strong test scores as the goal, treating a rising PISA rank as proof that the reading problem has been solved. The adult readership figures suggest a rising score can coexist with a shrinking reading culture, and that the second problem doesn’t fix itself just because the first one looks good on paper. A population can be highly literate in the technical sense and still be reading less, for shorter stretches, and with less enjoyment than it did a generation ago.
None of this means Korea’s education system has failed at its stated job. It has succeeded, repeatedly and by a wide margin, at teaching reading comprehension to teenagers. The reading survey data simply points to a second, separate job that test scores were never designed to measure: whether people go on choosing to read once the tests stop counting.
