Problems of obtaining a sample in a study of deviancy: A case of failure