Central Data Catalog

The sample is drawn using a two-stage stratified cluster sampling design. The country is divided in 16 strata.
In the first stage, a random sample of schools is selected in each stratum, using probability proportional to size.
In the second stage, a random sample of 20 students ( for grade 3 and grade 6) and 30 students (for grade 9) is selected at each targeted grade among those who are present at school on the day of the assessment.


Oversampling is used in districts (administrative units) that have less students than expected in the sample design. Survey weights are computed to make the following adjustments for the computation of estimates: probability of being sampled, non-response and post-stratification.