Abstract
This meta-analysis synthesised quantitative research into potentially modifiable psychological correlates of car use and intentions to drive. Online psychology and transportation databases were searched, and inclusion criteria applied to potentially relevant records. An ancestry approach was also employed to search selected publications. Meta-analyses of effect size rs were performed on 23 unique study datasets. Results generally supported the predictive utility of variables derived from the Theory of Planned Behaviour, though cognitions towards not driving displayed uniformly larger effects than were observed for car use cognitions. There was also a strong effect of habit on behaviour. Support for effects of pro-environment cognitions on driving was weak. However, a dearth of available evidence limited our findings and precludes development of clear evidence-based recommendations for intervention design. Directions for future research are discussed. (C) 2008 Elsevier Ltd. All rights reserved.