This report compares the value of a diary, and telephone interviews covering either the previous 24 hours or the previous week as methods of collecting data from young families on perceived health problems and the measures taken to deal with them. The population of 310 families was randomly divided into three groups. The study design allowed the evaluation of both qualitative and quantitative aspects of the different methods. For each individual a health problem was perceived every four days and care for health problems was reported every six days. The frequency of participation was highest in the 24-hour recall group. In the diary group a weekly telephone call elicited a higher compliance than a weekly letter of reminder. In spite of a higher reporting of symptoms in the diaries the relative number of measures taken was greater in the two recall groups. When comparing the information obtained by one-week recalls with that of 24-hour recalls it was found that the risk of forgetting a perceived symptom or measure taken was low but the risk of overreporting symptom days may increase with time. The distribution of symptoms in eight main categories was equal as reported by the different methods. The cost for the investigator per individual day covered was more expensive when using the 24-hour recall method compared to the diary method.