This chapter investigates the ecological validity of the data type known as sociolinguistic interviews. The data type is widespread and perhaps unavoidable for studies of variation and change in entire speech communities. The first question to be asked is whether there is any speech event like the sociolinguistic interview as such. This is answered in the negative. However, passages in sociolinguistic interviews resemble what you find in one or several other situations. This is detailed on the basis of data from the Danish LANCHART study, and John R. Rickford’s work is consistently used as a reference point. Consequences for future work are spelled out.