Dear NEPS-team,
In my research on NEPS data, I exploit differences in the timing of the surveys. Using tx8620m and tx8620y I explored the dates, or rather the months, in which the parental surveys were conducted in SC2 in the different waves. Here I found that the moment of the parental interview in some waves differs depending to which of the three subsamples the student belongs (original sample - in a neps school, original sample - not in a neps school, or refreshment sample). This seems to be the case for wave 3, 4, and 6.
The field reports provided me with an understanding for differences in some waves.
For wave 3, it is clear that the refreshment sample was split into two distinct tranches, explaining why plenty of interviews in the refreshment sample are conducted after surveys of the original sample are already finished.
For wave 6, it became clear that one group of parents were only approached later in the fieldwork period, because they just participated in the individual follow up as their child did not attend a NEPS school.
However, for wave 4 I cannot find any suggestion in the field report on why the interviews with parents of children in the the original sample (whether in a NEPS school or not) often took place in February while for the parental interviews for parents of the refreshment sample there are hardly interviews in february but mostly in March and April. Do you have any idea what could be the reason for this?
Kind regards,
Kars van Oosterhout
Dear Kars,
thank you for your contribution.
Unfortunately, we at the RDC LIfBi don’t have any further information on this beyond what you’ll find in the field reports. However, I can well imagine that the assignment of surveys to the interviewers regarding the three subgroups at the field institute isn’t random. In my opinion, it’s conceivable that all interviewers started with cases from the existing subgroups in February and only processed the cases from the refresher sample afterwards. This could explain why some interviewers had already finished with the first two subgroups in February and then started with cases from the refresher sample in February, while others didn’t do so until March or April. Does that sound like a plausible explanation to you?
Best regards and good luck with your analyses,
Bennno Schönberger
Dear Benno,
Thank you for your reply.
Okay, then I think we indeed won’t have a definitive answer.
The differences in wave 4 stand out compared to other waves, were wither there were no differences between the subgroups or, in case there were differences, they mention different approaches between subgroups were discussed in the field report. So I was curious whether there would be a specific reason for this non-random pattern.
But if there is no known specific reason, then I think the most plausible and simple explanation is indeed that intentionally or unintentionally the original sample was (generally) assigned to interviewers before the refreshment sample.
Best regards,
Kars