ABSTRACT
ON VIRTUALLY every subject ever tackled by a survey one could make a list of relevant questions far longer than any reasonable questionnaire could include-not merely because
wording variations enable one to ring numerous changes on a single question, but because there are generally so many aspects of the subject that could be explored. Suppose one wished to know the attitudes of a group of workers towards their management. At once a number of possible questions suggest themselves: 'Do you think the management is making a good or a bad job of things?'; 'How well qualified do you think the managers are for their task?'; 'In what ways do you think the relationship between management and workers could be improved?'; 'What do you think of the management's part in improving working conditions?'
