Citation

Sakashita C, Senserrick TM, Boufous S, de Rome L, Elkington J, Ivers RQ. Traffic Injury Prev. 2014; 15(5): 491-500.

Affiliation

George Institute for Global Health, School of Public Health , University of Sydney , Sydney , New South Wales , Australia.

Copyright

(Copyright © 2014, Informa - Taylor and Francis Group)

DOI

10.1080/15389588.2013.837576

PMID

24678572

Abstract

OBJECTIVES: While self-report methods to collect exposure information have large practical advantages in many research contexts, little research has specifically investigated the reliability and validity of motorcyclists' self-reported exposure. The present study aimed to examine the reliability and validity of different self-report exposure measures and to provide recommendations on best practice self-report riding exposure questions.

METHODS: The reliability and validity of different self-report exposure measures were examined amongst novice motorcyclists through t-tests, Bland Altman plots, coefficients of variation, and correlations.

RESULTS: The most valid and reliable data was provided when riding exposure was asked for the current average week rather than earlier and longer periods, and in units of time rather than distance or number of trips. The greater reliability of riding exposure found amongst commuting and rural riders compared to recreational and metropolitan riders respectively and at the second interview compared to the first suggests that factors such as riding purposes, geographical locations, and riding experience can contribute to measurement error.

CONCLUSIONS: If self-reported odometer readings are used, questions on whether the respondents share their own bike or ride more than one bike, and a built-in process to ensure respondents report the exact odometer reading on their bike are recommended. It is recommended that self-report riding exposure questions ask about the hours of riding for the current average week, and data on riding purposes, locations, and experience are also collected.

Language: en