Citation

Thigpen C. Transp. Res. F Traffic Psychol. Behav. 2019; 60: 453-461.

Copyright

(Copyright © 2019, Elsevier Publishing)

DOI

10.1016/j.trf.2018.11.002

PMID

unavailable

Abstract

Though panel data have substantial advantages over cross-sectional data in the study of travel behavior change, it also introduces logistical and methodological challenges for researchers. Retrospective surveys have the advantage of addressing many of these challenges, though the measurement validity of recalled answers remains an important concern. In this study, I assess the measurement validity of recalled characteristics related to bicycling: usual use of a bicycle to commute, bicycling attitude, and bicycling skill. I statistically compare data collected prospectively in the 2012-13, 2013-14, 2014-15, and 2015-16 UC Davis Campus Travel Surveys (CTS) with corresponding data collected retrospectively in the 2016-17 CTS. I find that prospective answers are strongly associated with recalled responses, that the passage of time has a weak, uncertain influence on recollection, and that present-day attributes attenuate the relationship between prospective and retrospective answers. I conclude that the recalled answers display a sufficient degree of correspondence to prospectively-collected response to have faith in analyses using retrospective survey data regarding traits within the past five years or less. Further research should be done on the measurement validity of retrospective travel survey data to contribute to the knowledge base on this important methodological topic.

Language: en

Keywords

Attitudes; Behavior; Retrospective; Skills; Survey; Validity