This paper analyses the quality of information collected by a self-administered survey responded to by a general population, who were offered the possibility of answering using the post or Internet. The analysis will be focused on the use of three grid questions with using rating scales with 6, 6 and 8 items, respectively. There was a polar point labelled scale (0-10) with verbal labels only at the endpoints. The postal survey showed greater acquiescence, a greater choice of extreme response categories and 'easy' answers, and a greater number of incomplete questions.