Countries with the Smallest Class Sizes Worldwide 2026

Countries with the smallest class sizes in the latest comparable international data are mostly education systems where primary classrooms average between 14 and 18 students. The latest OECD comparison uses reference year 2024 (school year 2023/24) and is based on the UNESCO-UIS/OECD/Eurostat education statistics collection administered in 2025/26. Across OECD countries, average class size is 21 students in primary education and 23 students in lower secondary education. [OECD]

Smallest Primary Class Sizes by Country

The smallest reported primary class size in the 2024 dataset is Peru, with an average of 14 students per primary class across all institutions. Costa Rica follows at 15 students. Croatia and Luxembourg report 16, while Greece and Latvia report 17. Estonia, Italy, Poland, and Slovenia average 18 students.

The ranking above uses the all institutions column because public and private school coverage differs across countries. Data source: OECD Table D2.3 [OECD].

Full Available Primary Class Size Ranking

The broader ranking shows that most reported primary class sizes fall inside a fairly narrow band. Many systems sit between 18 and 23 students. The OECD average is 21 students per primary class, while the EU25 average is 19 students. These averages are useful reference points, but they do not replace the country-level values.

Table values are average number of students per class in all institutions. Missing and not applicable observations are excluded from the ranking instead of being estimated.

Smallest Lower Secondary Class Sizes

Lower secondary classes are often larger than primary classes because subject-based teaching can change how students are grouped. In the 2024 table, the smallest all-institution lower secondary averages are 18 students per class in Croatia, Latvia, and Poland. Denmark, Finland, and Luxembourg report 19 students; Belgium also reports 19 through the TALIS-based value included in OECD Table D2.3.

Primary and Lower Secondary Differences

A small primary average does not always carry into lower secondary education. Costa Rica, for example, reports 15 students per primary class and 28 students per lower secondary class. Peru reports 14 at primary level and 23 at lower secondary level. In contrast, Croatia, Luxembourg, Latvia, Estonia, Finland, and Poland remain at or below 20 students at both levels.

What “Class Size” Means in International Data

Average class size is not the same as the number of students assigned to one teacher. OECD defines class size as the number of students following a common course of study, based on the highest number of common courses (usually compulsory studies), while teaching in subgroups is excluded. The comparable class-size table covers primary and lower secondary education. [OECD]

The student-teacher ratio uses a different calculation: full-time equivalent students divided by full-time equivalent teachers at the same education level. In 2024, the OECD average was 14 students per teacher in primary education and 13 in lower secondary education, while average class sizes were 21 and 23 respectively. A country can therefore have a low student-teacher ratio and still have larger classes because staffing and classroom grouping measure different parts of school organisation. [OECD] UNESCO’s glossary also notes that pupil-teacher ratios are not equivalent to average class size. [UNESCO UIS]

Why Some Countries Are Missing From the Ranking

Some large education systems do not appear in the ranking because the relevant average class size value is missing, not applicable, or not reported in the same table column. That does not mean they have large classes. It means the table does not provide a comparable value for the selected indicator. This distinction prevents missing observations from being mistaken for large class sizes.

Countries Not Ranked Because the Average Class Size Value Is Missing

Data Patterns Behind Smaller Class Sizes

Smaller average class sizes usually reflect a mixture of school organization, student population size, teacher supply, school network density, and public-private sector structure. The data do not prove one single cause. A country may have smaller classes because rural schools remain open, because enrolment has declined, because staffing policies preserve class sections, or because schools divide students differently across grades and subjects.

Primary Schooling Shows the Smallest Averages

Primary education often has smaller or more stable grouping because one teacher may stay with the same class for several subjects. This does not happen everywhere. Still, the lowest primary values in the dataset sit well below the OECD average of 21 students. The cluster of 14 to 18 students is therefore a clear low-class-size group among countries with reported data.

Lower Secondary Can Change the Picture

Lower secondary education tends to introduce more subject specialization. Schools may reorganize students for mathematics, languages, science, arts, and optional courses. In many countries this level has larger classes than primary education. Yet several systems keep both levels compact: Croatia, Latvia, Luxembourg, Finland, and Poland report lower secondary averages below 20 students, while Estonia is at 20.

Public and Private Values Should Be Read Carefully

Private school class-size values can be much smaller in some systems, but the share of students in private institutions differs by country. The all institutions measure gives a more consistent basis for cross-country comparison because it does not rank systems solely from one school sector. Public-private values remain useful for examining differences within individual countries.

Public, Private, and All-Institution Values

The all-institution value combines public and private institutions where both are reported. Sector differences remain visible in the 2024 data. Türkiye reports 23 students per public primary class, 19 in private institutions, and 23 across all institutions. Latvia reports 18 in public primary classes, 12 in private institutions, and an all-institution average of 17.

Reference Year and Coverage

The ranking uses OECD Table D2.3 for reference year 2024, corresponding to school year 2023/24. The data were collected through the UNESCO-UIS/OECD/Eurostat education statistics process administered by the OECD in 2025/26. Values marked missing or not applicable are not estimated for the ranking. [OECD]

Smaller Class Size Is an Input Indicator

Class size is a learning environment indicator. It is not a direct achievement score, graduation rate, PISA result, or teacher-quality measure. Smaller classes may allow more individual feedback and easier classroom monitoring. Yet education outcomes also depend on curriculum, teacher preparation, school leadership, family background, assessment design, language of instruction, attendance, and many other factors. A low class-size rank should not be read as a full education-system rank.

Country Averages Hide Local Variation

National averages can hide differences between regions, municipalities, school types, urban schools, rural schools, grade levels, and subject areas. A country average of 18 students does not mean every classroom has 18 students. Some classes will be smaller. Some will be larger. The value is still useful because it gives a comparable national-level measure.

Comparable Class-Size Coverage Is Limited

Comparable average class-size data cover fewer countries than broader education indicators. UNESCO and World Bank datasets widely report pupil-teacher ratios, but those values measure staffing rather than the number of students grouped in a class. They can support staffing analysis, but they should not be substituted for average class size when countries are ranked by classroom size.

Most Compact Class-Size Group in the Latest Data

The countries that form the lowest primary class-size group are Peru, Costa Rica, Croatia, Luxembourg, Greece, Latvia, Estonia, Italy, Poland, and Slovenia. Each reports a primary all-institution average of 18 students or fewer. At lower secondary level, Croatia, Latvia, and Poland have the lowest all-institution average at 18 students; Denmark, Finland, and Luxembourg follow at 19.

The smallest comparable primary class sizes in the latest international data range from 14 to 18 students per class. Countries without comparable average class-size observations remain outside the ranking even when pupil-teacher ratios or other staffing indicators are available from separate datasets.