School groups vs SALSA

Summary

Groups with revenue data

421

With SALSA data (≥ 2 schools)

20

Performing above expectation

14

Performing below expectation

6

The SALSA model calculates an expected merit value per school based on parents' education level, share of recently immigrated students, and gender distribution. The average SALSA score per school group shows whether their schools as a group perform better (+) or worse (−) than what the student composition predicts — regardless of whether they have favourable students or not.

Revenue is sourced from BV's annual report data and normalised against the total pupil count in the group's active compulsory schools. Note that revenue is not the same as profit (operating margin) — it includes all income.

Revenue per pupil and SALSA score

Sorted by revenue per pupil (highest revenue at top). Schools with fewer than 2 SALSA matches are excluded.

School groups without sufficient SALSA data

These school groups have revenue data but fewer than 2 SALSA-matched schools, which makes the SALSA average unreliable. They are reported separately for completeness.

Method and limitations

Revenue data comes from the Swedish Companies Registration Office's (BV) annual report database. Not all independent school groups have reported data or are represented in Skolkoll's BV integration — coverage is therefore partial.

The SALSA score per school group is calculated as the average of all constituent schools' residuals (mr). School groups with fewer than 2 matched schools are excluded because an average based on 1 school is sensitive to random variations.

Revenue ≠ profit: High revenue per pupil may be due to high rents, large staffing costs, or the group's business model. A fairer profitability measure would be operating margin, which is not available in this dataset.

The analysis is descriptive — it shows correlations, not causation. The fact that a school group has high revenue and a positive SALSA score does not prove that the money creates the results. It could equally be a selection effect (students with favourable backgrounds choose these schools regardless of demographics adjustment) or third factors.