So the Netherlands works with the The Programme for the International Assessment of Adult Competencies (PIAAC) to organise these literacy tests with universities and government institutions. In the Dutch PIAAC report it states that CBS (Central Statistical Office) does the sampling (
source). How does CBS organise these samplings? CBS simply make a mathematical number and then decides the sub-groups and then decides how many people are needed for each sub-group for an accurate rapport, they then times it by that entire sub-group population (1 test taker represents 1000 per example).
The juicy part, CBS does the sampling by simply seeking
volunteers online who then do tests at a location with a tablet. If they don't have enough of a certain sub-group they simply
compensate it by increasing how much 1 person represents (instead of 1/1000 it becomes 1/2000). For 2022-2023 they only tested
4257 people (1 person represents 4200 people) between the ages of 16-75 and as a results 16% of those test takers were considered illiterate. The results which they consider has a 95% confidence interval (
source).