This paper describes a new tool for data collection — electronic questionnaire system — and suggests that under some condition it may be a very useful tool at a low cost for social research. 969 users were selected at random from the directory of a nationwide computer network system in Japan on which over 48,000 users were registered on April 1989. A letter was sent to respondent by both ordinary and electronic mail to ask for responding to the survey by electronic questionnaire. Respondents must access to the network by their own computer in order to answer the electronic questionnaire and send a command "GO XXX". Then they can get into the Electronic Questionnaire System. At first the system sent a message how to reply to the questionnaire, and started to send a first question automatically. After they answer it correctly, the system goes to next one and so on. Their answers were all stored in the host computer of the network and the researcher can collect them at the end of research. Data on 213 users were obtained from 969 users (response rate was 22.0%). 60% of 213 has an experiences of having answered some electronic questionnaires before. The result shows that they feel less psycological barrier to answer electronic questionnaire than interview method. Most of them (over 90%) answered that electronic questionnaire method is easier, more frankly and more honestly to respond than interview method. However the result shows three problems of using the electronic questionnaire, (a) possible violations of privacy or lack of anonymity, (b) a small number of network users, and (c) difficulty of using computer or word processor.