I am working with the SC2. In the data manual, when it comes to the ID_i it says “identifies the respective educational institutions such as kindergartens or daycare centers, schools, universities, etc. The variable ID_i is unique across waves and starting cohorts” I understood by this that there would also be ID_i for secundary education institutions (i.e.: gymansiums and so on). However, the ID_i seems to stop at wave 6 in this dataset. Are there identifiers for the institutions that the children transition to? Or is it impossible to cluster them afterwards?
thank you for your question. Indeed institutional identifier variable ID_i is only available for the first 6 waves. This is simply because the children were only tested together in the context of kindergarten during waves 1 and 2, and in the context of primary school during waves 3-6 only. Contextual information about the school administrators, caregivers, and teachers is only available for these waves. For the vast majority of the children, a change to a different school was imminent after the 4th grade. Thus, the children who were originally in the NEPS schools were distributed among a large number of secondary schools, which is why the survey and testing of the children from that point onward could only take place at the individual level, and no contextual or group effects could be calculated.