Toward Comprehensive Assessment of Beliefs and Attitudes Related to Physical Activity in Young Adults: Pilot Study

Studies show that despite the positive effects of physical activity (PA), most university students are not active enough. For interventions, it is necessary to understand the determinants of PA behavior. Theory of planned behavior (TPB) is one of the most widely used frameworks to describe the psychological determinants of health behavior. Research has shown that in addition to the determinants included in TPB (attitudes, subjective norms, perceived control, and intention), fear of negative outcomes (eg, discomfort or pain) is a major barrier to increasing one's PA. It has been shown that accepting the possibility of unpleasant outcomes may help in reaching one's PA goals.

The purpose of this study was to create a questionnaire of PA determinants based on TPB and complemented by the topic of acceptance of unpleasant outcomes. The questionnaire is thus meant for evaluating the effectiveness of health psychological PA interventions in university students and young adults.

This study was carried out using qualitative and quantitative methods and consisted of three phases: (1) an elicitation study for item generation, (2) pretesting for clarity and understanding, and (3) item selection using conceptual and psychometric criteria (based on a pretest with N=447) to maximize domain coverage and avoid redundancy.

A questionnaire covering the core topics of TPB plus acceptance of negative outcomes was constructed, using a combination of qualitative and quantitative methods. The final shortened questionnaire consists of the following question blocks: positive and negative behavioral beliefs, acceptance of negative outcomes, subjective norms (injunctive and descriptive norms, and motivation to comply), and control beliefs. In terms of predicting self-reported PA, the shortened questionnaire was equal to the unabridged version. Notably, negative behavioral beliefs and acceptance of negative outcomes had opposite-signed correlations with self-reported PA (-0.22 and 0.32, P<.001). Despite the aim of avoiding redundancy, several item bundles (eg, positive and negative behavioral beliefs, and acceptance of negative outcomes) were highly homogenous in the final version, and are thus usable as psychometric scales.

This questionnaire can assess a range of PA determinants and has good psychometric properties. The questionnaire can be used to assess the beliefs and attitudes (behavioral beliefs, perceived norms, control beliefs, and acceptance of negative consequences) related to PA in young adults when planning interventions, as well as evaluating the effects of health psychological interventions aiming to increase PA.