Are beliefs elicited biased by question order? A theory of planned behaviour belief elicitation study about walking in the UK general population


Correspondence should be addressed to Catherine D. Darker, School of Sports and Exercise Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK (e-mail:


Objectives. To elicit salient beliefs about walking for an average of 30 minutes per day, with the aims of investigating whether the order of Theory of Planned Behaviour (TPB) belief elicitation questions affects the number and types of beliefs elicited and whether affective and instrumental questions elicit different beliefs.

Design. A 3 × 2×2 × 2 mixed factorial design was employed, with order of behavioural, normative and control questions, and affective and instrumental questions as between-subjects variables, and affective/instrumental and positive/negative questions as within-subjects variables.

Method. Quota sampling with regards to age and gender (N = 180) was employed to obtain a sample that was broadly representative of the adult general population.

Results. The order in which behavioural, normative and control beliefs were asked had little impact on the number or type of beliefs elicited. The affective/instrumental attitude distinction was supported. Few differences were apparent between older and younger respondents and between men and women.

Conclusion. TPB belief elicitation studies are not biased by order effects. Interventions to promote walking should consider targeting affective beliefs, e.g. stress relief, in addition to beliefs about health, which is the traditional focus of health campaigns. Given the similarities in beliefs across demographic groups, ‘one size fits all’ interventions to promote walking are appropriate.