How will you take into consideration the different cultures with differing behaviours and understanding?
This issue relates specifically to the assessment of children’s personal, social and emotional development, which is carried out by means of teacher ratings rather than an objective assessment. There may be considerable cultural differences in expected norms of behaviour between one country and another. For example, what is perceived to be hyperactive behaviour in Japan, may be seen differently in the US. In order to be able to carry out valid international comparisons, we need to ensure that we can benchmark teachers’ judgements across cultures whilst establishing the reliability and the relative severity/leniency of each rater. This will be done using video snapshots of children in different countries. All teachers involved in the project will be shown around six short clips from different cultural settings and their own, and asked to rate them on the iPIPS scales. They will also rate their sample of students. Using the Facets model in Rasch, we will be able to establish the leniency and severity of the raters as well as their consistency (reliability). The exercise will also establish the cultural norms (leniency/severity) and whether indeed the construct can be reliably assessed within that cultural context. This will allow us to estimate the degree of error and bias within a system and establish reliable and valid comparisons.