Harno alarmed by growing maths gap between Russian and Estonian pupils
Estonia's Education and Youth Board (Harno) has expressed concern over a widening gap in mathematics exam results between Russian-speaking and Estonian-speaking students. Overall state exam scores remained stable or improved compared to last year, though the narrow-course mathematics average dropped by 3.7 points. A total of 12,391 students sat state exams in 2026.
EstoniaEstonia's Education and Youth Board (Harno) has sounded the alarm over an increasingly pronounced divide in mathematics performance between Russian-speaking and Estonian-speaking schoolchildren, following the release of this year's state examination results.
Overall Results Stable
A total of 12,391 students sat state exams across Estonia in 2026, around 700 more than the previous year. Of those, 9,790 were upper secondary students, 1,953 attended vocational schools, and 716 had previously completed secondary education. Average scores across all exams either rose or held steady compared to 2025, with one notable exception: the narrow-course mathematics exam average fell by 3.7 points to 35.4. The wide-course mathematics average edged up by 0.9 points to 59.2.
Alge Ilosaar, head of the Harno assessment centre, noted that the addition of a formula sheet to this year's narrow-course mathematics exam did not produce better results. «This confirms that the decisive factor is still the ability to understand and correctly apply formulas,» she said.
The Russian-Estonian Maths Divide
The most pressing concern flagged by Harno is the growing performance gap in mathematics between students who sit exams in Russian and those who do so in Estonian. According to Ilosaar, that difference has now grown to 12.49 percentage points. «As in previous years, the ever-widening gap between maths exam results for Russian-speaking and Estonian-speaking children remains a source of concern,» she said.
Results for the Estonian language as a second language exam also gave cause for worry: fewer than half of students, just 46.8%, reached the B1 language proficiency level in the basic school exam.
Ilosaar acknowledged that a thorough study would be needed to understand the root causes of weaker mathematics results. «We need to find out what the profile of teachers is and what methods they use,» she said.
Teacher Training as a Key Factor
Hele Liiv-Tellmann, chief curriculum expert at the Ministry of Education, emphasised that teacher competence and methodology are the most critical elements in mathematics education. «We can improve exam results by training teachers. At the same time, we know it is very difficult for teachers to adopt new methods once old ones have become ingrained,» she said. Schools themselves can also act, for example, by analysing whether teachers need upskilling, collaborating with parents to boost pupil motivation, or experimenting with different teaching approaches.
Polarisation in Maths Performance
Beyond the language divide, Ilosaar also highlighted growing polarisation in mathematics outcomes overall. While nearly a third of basic school students scored in the 90-100 percentile range, more than in any other subject, almost a quarter scored below 50%. «This polarisation is a cause for concern,» she said.
Top Scores and Language Exams
Eighty-one students achieved a perfect score of 100 points across all state exams this year, down slightly from 85 last year. Of these, 65 achieved full marks in mathematics. Eight students scored 100 in Estonian language, seven in Estonian as a second language, and just one in English.
In the English language state exam, 84.2% of the 5,384 participants reached the B1 or B2 level required by the national curriculum. Meanwhile, 92.5% of the 4,920 students who sat the Cambridge C1 Advanced international exam reached C1 or above, a strong result that reflects steady year-on-year improvement.
Gustav Adolf Grammar School led the table for perfect mathematics scores with 10 students, followed by Hugo Treffner Gymnasium with 7 and Tallinn Secondary School of Science with 6.
Open in app →