Key findings
Of Sweden's 4,684 active compulsory schools, 912 have access to a school library and 3,616 lack one. An additional 156 schools have no data on this.
Merit value grade 9
+1.5
points higher with library
228.2 vs 226.7
Certified teachers
+2.1
percentage points higher
72.9% vs 70.8%
SALSA deviation
+2.4
points better with library
1.7 vs -0.7
The SALSA deviation is the most important measure here: it controls for students' socioeconomics and shows the school's own contribution. A difference of +2.4 points suggests that the effect partly persists even when accounting for background conditions.
Detailed comparison
| Measure | With library | Without library | Difference |
|---|---|---|---|
| Number of schools | 912 | 3,616 | — |
| Merit value grade 9 | 228.2 (494 schools) | 226.7 (1081) | +1.5 |
| NP Swedish grade 9 | 13.0 (488) | 13.0 (1039) | 0.0 |
| NP Mathematics grade 9 | 11.3 (494) | 11.3 (1058) | 0.0 |
| NP English grade 9 | 15.8 (494) | 15.8 (1060) | 0.0 |
| Certified teachers | 72.9% (855) | 70.8% (3511) | +2.1 pp |
| Students/teacher | 12.1 (912) | 12.1 (3590) | 0.0 |
| Eligible for vocational programmes grade 9 | 85.0% (494) | 86.6% (1055) | -1.6 pp |
| SALSA deviation (merit value) | 1.7 (487) | -0.7 (1013) | +2.4 |
pp = percentage points. The number of schools in parentheses refers to those with data for the respective measure. SALSA data from 2024/25.
Visualisation
SALSA deviation — the most important measure
Raw merit values are difficult to compare directly, since schools' results are strongly influenced by students' socioeconomic background. Therefore we use the National Agency for Education's SALSA model (Skolverkets Arbetsverktyg for Lokala Sambandsanalyser), which adjusts for:
- Parents' education level
- Proportion of recently immigrated students
- Gender distribution
The SALSA score shows how much better or worse a school performs compared to what can be expected based on these factors. A positive value means the school outperforms its predicted level.
With library
+1.7
SALSA deviation (average)
487 schools with SALSA data
Without library
-0.7
SALSA deviation (average)
1013 schools with SALSA data
Schools with a library have an average SALSA deviation of +1.7 merit value points, while schools without a library average -0.7. The difference of +2.4 points thus persists even after socioeconomic adjustment.
Methodology
The analysis is based on data from the National Agency for Education's open API (school year 2023/24) and the SALSA model (2024/25).
We filtered to active compulsory schools (school type GR) and grouped them by the hasLibrary field.
For each group, arithmetic means are calculated — each school weighs equally regardless of student count.
The SALSA score (mr) is sourced from the National Agency for Education's SALSA database and linked via school unit code.
Only schools appearing in both the school database and the SALSA database are included in the SALSA comparison.
Important caveats
Correlation does not equal causation. The fact that schools with a library perform better on SALSA deviation does not necessarily mean the library causes higher results. Other factors may co-vary:
- School size: Larger schools more often have a library and may have other structural advantages.
- Organiser type: Municipal schools may have different patterns than independent schools.
- Data quality: The school library field (
hasLibrary) is self-reported and may be outdated for some schools. - Definition: What counts as a "school library" varies — some schools count a book room, others a staffed library function.
- SALSA limitations: The model controls for three variables but misses other factors that may affect results.
Despite these caveats, it is noteworthy that the difference in SALSA deviation (+2.4 points) persists after socioeconomic adjustment. Research generally supports that school libraries with trained staff have a positive effect on reading comprehension and academic results.
Statistics: school year 2023/24. Source: National Agency for Education open data and SALSA 2024/25. Processed by Skolkoll. Glossary · About the data.