TIMSS 2023 Data
All statistics were computed using sampling weights.
All achievement data are presented as five plausible values. All required statistics were calculated separately for each of the plausible values and the resulting statistics averaged to get the result.
Standard errors for significance testing of achievement results were computed using:
- The jackknife repeated replication (JRR) method to estimate sampling variance
- Differences between the five plausible values to estimate measurement variance
General note: TIMSS data are designed for group-level reporting; students’ scale scores in the database are not suitable to measure performance of individual students or very small groups.
TIMSS data and the International Database (IDB) are freely available.
For secondary analyses, data can be analyzed with IEA IDB Analyzer and the TIMSS 2023 International Database. These tools take into account the characteristics of the sampling and assessment design and implement the jackknife replication method with the appropriate plausible values.
- Achievement items, including item information, IRT item parameters, and item percent correct statistics (see below under item release policy)
- Background questionnaires (school, teacher, student, and home/early learning survey)
- Curriculum questionnaires
- Test-curriculum matching analysis (TCMA) data
For data protection reasons, not all variables are included in the public-use files (PUF). While the PUF provide broad access for secondary research, some variables are only available under restricted-use conditions and require researchers to apply and receive approval from IEA through its study data repository. A third group of variables is kept fully secure and not released at all, in order to ensure the possibility of measuring trends over time. This approach allows for meaningful secondary research while safeguarding respondent privacy and maintaining long-term data consistency.