Discussion of “Probability vs. Nonprobability Sampling: From the Birth of Survey Sampling to the Present Day” by Graham Kalton

Let me first thank Dr. Kalton for his amazing historical review of the development of survey sampling from its origin, contrasting purposive sampling, until now, where some elements of purposive sampling in terms of web or big data seem to supersede the well-elaborated theory of survey statistics. Shall the message be that we do not need any sampling courses at universities anymore, that official statistics should turn to modelling using data with unknown data generating processes, or actually even be substituted by (commercial) data krakens? Hardly so! Graham Kalton emphasises a modern thinking about the use of these new data sources which may also have some advantages and he urges future research on data integration methods using (very) different kinds of data while strongly taking quality aspects into account.