How robust will the data be?

In technical terms, the data will be very robust.  The individual assessments of children have been shown to have high reliability, with inter-rater correlations of 0.98 and high predictive validity, with correlations up to 0.68, with later assessments of children at age 11.  It has been translated into 10 languages and has been proven to work in countries where traditions of pre-schooling and modes of learning are very different UK.
The sampling processes and other procedural aspects of the assessments will conform to internationally recognised, technical standards.
The assessments themselves, will be thoroughly trialled in each administration where they are taking place for the first time, to ensure that they are culturally and linguistically appropriate, while retaining a high level of international comparability.