Closer to the total? Long-distance travel of French mobile phone users

Abstract

Analyzing long-distance travel demand has become increasingly relevant because the share of traffic induced by journeys related to remote activities which are not part of daily life is growing. In today’s mobile world, such journeys are responsible for almost 50 percent of all traffic. Traditionally, surveys have been used to gather data needed to analyze travel demand. Due to the high response burden and memory issues, respondents are known to underreport their number of long-distance journeys. The question of the actual number of long-distance journeys therefore remains unanswered without additional data sources. This paper is the first to quantify the underreporting of long-distance tour frequencies in travel diaries. We took a sample of mobile phone billing data covering five months and compared the observed long-distance travel with the results of a national travel survey covering the same period and the same country. The comparison shows that most of the estimates of the number of missing tours by researchers have thus been too low. Our work suggests that the actual number of long-distance journeys is twice as high as that reported in surveys. Two different causes of underreporting were identified. Firstly, soft refusers travelled long distances but reported no long-distance tours. Secondly, respondents underestimated their number of long-distance tours. Consequently, there is a need to use alternative data sources in order to gain better estimates of long-distance travel demand.