The theory of planned behaviour predicts self-reports of walking, but does not predict step count

Objectives This paper compares multiple measures of walking in two studies, and the second study compares how well Theory of Planned Behaviour (TPB) constructs perform in predicting these different measures.

Methods In Study 1, 41 participants wore a New Lifestyles NL-2000 pedometer for 1 week. Subsequently, participants completed a questionnaire containing measures of the TPB constructs and two self-report measures of walking, followed by two interview measures of walking. For Study 2, 200 RAF trainee aircraftsmen wore pedometers for 2 weeks. At the end of each week, participants completed the questionnaire and interview measures of walking.

Results Both studies found no significant association between questionnaire measures of walking and pedometer measures. In Study 1, the interview measures produced significant, large correlations with the pedometer measure, but these relationships were markedly weaker in the second study. TPB variables were found to explain 22% of variance in intention to walk in Study 1 and 45% of the variance in Study 2. In Study 2, prediction of subsequent measures of behaviour was found to be weak, except when using a single-item measure of walking.

Conclusions Recall of walking is poor, and accurate measurement by self-report is problematic. Although the TPB predicts intentions to walk well, it does not predict actual amount of walking, as assessed by pedometer. Possible reasons for these findings include the unique nature of walking as an activity primarily used to facilitate higher order goals. The use of single-item measures may exaggerate the effectiveness of the TPB model for walking, and possibly other forms of physical activity.