Recent years have seen an increasing interest in probability sampling, and some proponents of the method have urged its use in public opinion surveys. This article reports on a comparative study of area-probability and quota sampling, undertaken to determine the adaptability of the former method to opinion research. The authors' general conclusion is that while probability sampling is sound in theory, its practicability for opinion measurement is open to question. The call-back problem and the dangers of repeated interviewing of the same individuals are cited as particularly difficult aspects of the method.