Citation

Paris H, Van den Broucke S. Transp. Res. F Traffic Psychol. Behav. 2008; 11(3): 168-180.

Affiliation

Department of Mobility and Safety, Ministry of the Flemish Community, Belgium; Research Group for Health Psychology, Katholieke Universiteit Leuven, Tiensestraat 102, 3000 Leuven, Belgium (stephan.vandenbroucke@psy.kuleuven.be)

Copyright

(Copyright © 2008, Elsevier Publishing)

DOI

10.1016/j.trf.2007.09.002

PMID

unavailable

Abstract

This article describes the development and validation of a self-report questionnaire to measure the determinants of speeding behaviour in road traffic, based on the theory of planned behaviour. A provisional questionnaire measuring self-reported speeding behaviour as well as its determinants as predicted by the TPB model (attitudes towards speeding and towards respecting speed limits, social norms, perceived behavioural control, and intentions) was completed by 116 drivers. Separate principal component analyses on the items measuring attitudes, social norms and perceived control resulted in two component solutions, explaining between 46.5% and 57.5% of the variance, for each determinant. After Promax rotation, these solutions were used to construct eight scales comprising 2-7 items each, measuring negative and positive attitude towards speeding, negative and positive attitude towards respecting speed limits, explicit and implicit social norm, and perceived internal and external control, respectively. Internal consistencies ranged from .51 to .82. Scale scores accounted for a significant proportion of the variance in intention and self-reported behaviour. Intentions were most strongly predicted by explicit social norms and negative attitude towards respecting speed limits. Self-reported speeding was predicted by intention and perceived internal control. In contrast, actual speeding behaviour was not significantly predicted by intentions and perceived control. The study demonstrates the validity of the theory of planned behaviour to predict self-reported speeding behaviour and provides a valid and reliable measure of the cognitive concepts featured in this model, but suggests that actual speeding behaviour can only partially be predicted from these concepts.