Assessing attitude: The case of health-enhancing physical activity

Objectives: The basic phases of constructing compact questionnaire scales for health-promotion interventions are demonstrated in the case of attitudes towards two modes of physical activity, outdoor aerobic exercise (OAE) and everyday commuting activity (ECA). The scales were to show good psychometric properties and to provide adequate indicators of attitude. Methods: Behaviours and criteria were specified for both OAE and ECA. The items describing the emotional (affect) and cognitive (outcome-expectations) aspects of attitude towards OAE and ECA behaviours were selected on the basis of a pilot study. The scales under development were used in a Finnish population survey on health, personal resources and physical activity (n=1,516). Item analyses were completed for purposes of refining the scale; principal component analyses and factor analyses were used to establish the dimensional structure of the scales. Evidence of discriminatory validity was provided by comparing the scale scores of the different groups of respondents with regular, with subcriterion, and without any OAE and ECA behaviours, respectively. Results: The purported psychometric properties were met in the case of the affective and, less obviously, the positive outcome-expectation scales. The negative expectation scales need to be reconstructed with additional items. Conclusions: The psychometric properties of the affective scales and the positive expectation scales allow conclusions to be drawn regarding group differences in attitudes towards OAE or ECA behaviours. These scales are ready to be cross-validated in a new sample. Test-retest reliability should also be established.