On the relationship between travel time and travel distance of commuters. Reported versus network travel data in the Netherlands
This paper gives a detailed empirical analysis of the relationships between different indicators of costs of commuting trips by car: difference as the crow flies, shortest travel time according to route planner, corresponding travel distance, and reported travel time. Reported travel times are usually rounded in multiples of five minutes. This calls for special statistical techniques. Ignoring the phenomenon of rounding leads to biased estimation results for shorter distances. Rather surprisingly, the distance as the crow flies and the network distance appear to be slightly better proxies of t... Mehr ...
Verfasser: | |
---|---|
Dokumenttyp: | Artikel |
Erscheinungsdatum: | 1999 |
Reihe/Periodikum: | Rietveld , P , Zwart , B , van Wee , B & van Hoorn , T 1999 , ' On the relationship between travel time and travel distance of commuters. Reported versus network travel data in the Netherlands ' , The Annals of Regional Science , vol. 33 , no. 3 , pp. 269-288 . https://doi.org/10.1007/s001680050105 |
Sprache: | Englisch |
Permalink: | https://search.fid-benelux.de/Record/base-27619806 |
Datenquelle: | BASE; Originalkatalog |
Powered By: | BASE |
Link(s) : | https://research.vu.nl/en/publications/3540e7c1-fe20-492e-bc11-b4ed9e691bd0 |
This paper gives a detailed empirical analysis of the relationships between different indicators of costs of commuting trips by car: difference as the crow flies, shortest travel time according to route planner, corresponding travel distance, and reported travel time. Reported travel times are usually rounded in multiples of five minutes. This calls for special statistical techniques. Ignoring the phenomenon of rounding leads to biased estimation results for shorter distances. Rather surprisingly, the distance as the crow flies and the network distance appear to be slightly better proxies of the reported travel time compared with the shortest network travel time as indicated by the route planner. We conclude that where actual driving times are missing in commuting research the other three indicators mentioned may be used as proxies, but that the following problems may emerge: actual travel times may be considerably higher than network times generated by route planners, and the average speed of trips increases considerably with distance, implying an over-estimate of travel time for long distance commuters. The only personal feature that contributes significantly to variations in reported travel times is gender: women appear to drive at lower average speeds according to our data. As indicated in the paper this may be explained by the differences in the car types of male and female drivers (females drive older and smaller cars) as well as higher numbers of stops/trip chaining among women. A concise analysis is carried out for carpoolers. Car-pooling leads to an increase in travel time of some 17% compared with solo drivers covering the same distance. In the case of car poolers, the above mentioned measures appear to be very poor proxies for the actual commuting times.