Loading school data…

Teacher certification and results — what remains after SALSA adjustment?

What remains when we partly account for pupil background?

Use this for

Get the main pattern quickly, find a local angle and jump straight to the charts that carry the story.

Data & format

Statistics story · 3 visualisations · copy reviewed 18 October 2026 · Skolverket, Kolada and SCB.

Everyone says the teacher matters. The difficult question is how much of school results can be linked to teacher certification and how much reflects pupils' circumstances. By comparing raw results with SALSA-adjusted ones we can see what remains of the correlation. At municipal level no correlation is visible. A weak correlation cannot be ruled out, but nothing close to the school-level one.

The classic correlation: certification and merit score

At school level there is a moderate positive correlation between the share of certified teachers and the merit score in year 9 (r = 0.32 for 1,479 schools). The spread is large: schools with high certification can have mediocre results, and schools with lower certification can reach higher than expected.

Certification co-varies with pupil composition, the housing market and recruitment capacity. The chart therefore does not show how much of the difference is due to the teachers.

Each bubble a school. The correlation is positive but moderate.

Open Certification vs Merit value fullscreen

Certification against SALSA

When we instead use the SALSA deviation as the outcome measure, we account for the three background factors the model includes: parents' education, the share of newly arrived pupils and the share of boys. The chart compares municipalities' share of certified teachers with their average SALSA deviation.

No correlation is visible here (r ≈ 0 for 289 municipalities). Municipalities with a high share of certified teachers are no more likely than others to land above the model's expectation. This also holds when weighting by pupil numbers or looking only at municipalities with several SALSA schools, and a school-by-school comparison gives the same result.

No clear difference: municipalities land above and below the zero line at every certification level.

Open Teacher certification vs SALSA fullscreen

Geography — where the certified share is lowest

The certified share also varies geographically. County medians differ by about 13 percentage points, from Jämtland to Uppsala, but the spread between schools within a county is larger. Uppsala and Östergötland rank highest, the metropolitan counties in the upper half and Jämtland and several central Swedish counties lowest.

The teacher question is therefore not just about education policy but also about regional attractiveness, working conditions and where teachers want and are able to work.

Share of certified teachers per school unit and county — median and spread.

Open Teacher certification by county fullscreen

At school level, higher certification goes together with higher merit scores, but when pupil background is partly adjusted for with SALSA, no correlation is visible between municipalities. That does not mean teachers do not matter: the measures are coarse, they refer to different years, and SALSA accounts for only three background factors.

Anyone who wants to apply the question locally needs to look at their own municipality's schools, pupil intake and development over time. A comparison between municipalities at one point in time is not enough to say what a single investment would achieve.

What the data does not show

The copy was reviewed on against the certified share for the 2025/26 school year (Kolada and school level), merit scores for 2025/26 and SALSA for 2024/25. The charts load the latest published data and may show a newer period than the copy.