A subjective measure of car dependence was developed based on people’s own assessment of their reliance on car use. The measure supplements the commonly used objective measure based on actual car use. Structural equation models (SEM) were estimated to quantify the subjective dependence and to examine its determinants: demographics, socioeconomics, and land use and transit access. The comparison between subjective dependence and actual car use discloses significant differences between both measures despite their statistical linkage. They also differ significantly in terms of how they are influenced by the determinants. Segmenting the population by both measures reveals 20% of the sample with contrasting subjective and objective measures. The SEMs examine relations between subjective car dependence (attitude), actual car use (behavior), and the intent to reduce car use (intention), after controlling for the determinants. Given the cross-sectional nature of the data, causality cannot be proven. Two plausible structural relationships were tested: (a) actual car use determines subjective car dependence, and (b) no direction of causality is assumed and the model only examines their association. Subjective car dependence mediates the impact of car use on the intent to reduce it: The direct effect of car use on the intent to reduce it is 0.2, the indirect effect via stated car dependence is –0.6; the total effect is –0.4. Actual car use explains approximately 50% of the variation in subjective car dependence, which, together with the car use, explains approximately 60% of the variation in people’s intent to reduce car use.
Full paper