Abstract
Abstract Proxy responses in travel surveys can produce several biases in the results. Thus, examining survey results with actual survey completer information is important; however, this information is not always available. To address this problem, a method has been proposed to infer the proxy-response probability for survey data with an unknown completer. This study validated the method using the reported time of trips. Respondents in paper-based travel surveys often report the departure/arrival time of their trips in rounded time such as 00, 15, 30, and 45 min. Proxy respondents will report these rounded times more often than self-respondents. The results, using the 2012 Kumamoto Person Trip survey data in Japan, showed that inferred proxy respondents tended to report departure/arrival time more often in 00 and 30 min, and less often in nonmultiples of 5 min than inferred self-respondents. The results indicated that the inference provided by the method was reasonable.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
