An experimental approach towards the evaluation of a seat belt campaign with an inside view on the psychology behind seat belt use

A Belgian national safety belt campaign was evaluated by means of a questionnaire survey in an adolescent sample. The evaluation was done through a three group after-only design with the use of one control group and two experimental groups. The first experimental group, the attentive group, was exposed to the campaign material in a very direct, attentive way, whereas the second experimental group, the pre-attentive group, was exposed rather inattentively. The framework of the Theory of Planned Behaviour (TPB) was applied and extended with a habit and a past behaviour variable in order to verify whether seat belt usage is to be understood as an automaticity mechanism (i.e., habitual or repeated past behaviour) or as planned behaviour. In terms of campaign effect, the comparison of the pre-attentive group and the control group revealed no significant differences. However, the attentive group and the control group differed significantly regarding two basic dimensions of perceived behavioural control (i.e., confidence and motivation), habit, past behaviour, behavioural intention and behaviour. In terms of explaining seat belt usage, linear regression models were fitted and gave most support for the repeated past behaviour approach. According to the latter, using seat belts is recycling an originally reasoned behaviour, yet without systematically going through the whole underlying reasoning every time a situation in which the decision to wear a seat belt (or not) presents itself. The practical implications of these findings are discussed more in detail.

► We evaluate the effectiveness of a seat belt promoting campaign.
► We find small positive effects on several socio-cognitive determinants of behaviour.
► We explain seat belt (non)usage as the recycling of an originally reasoned action.
► Campaign effectiveness however is dependent upon exposure conditions.
► We recommend explicit rather than subliminal exposure to campaign stimuli.