Evidence on the Comparison of Telephone and Internet Surveys for Respondent Recruitment

Internet surveys have a potential use for survey research when compared against costs and declining response rates of traditional modes as they form a powerful tool for reducing respondents' burden in complex questionnaires. On the other hand, there exists scepticism about the reliability and robustness of the collected data. Arenze et al . (2005) argue that case studies involving Internet surveys cannot be generalised to other countries and have recommended systematic collection and reporting of experiences worldwide. Such studies have had limited exposure in the transport literature. This paper provides empirical evidence on the comparison between telephone and Internet surveys in the context of a car ownership study. The comparison between telephone and Internet modes focuses on performance measures such as response speed, response rates, survey costs, demographic profiles and geographical representation of the sample. The results indicate the cost effectiveness of Internet surveys. Moreover, they show that the time and cost for data collection significantly vary by sampling and recruitment method. Finally, Internet survey response rates are lower than those in the telephone interview, which implies that Internet surveys can only be used to complement traditional data collection methods.