Development and Validation of the Cognitive Behavioral Physical Activity Questionnaire

Abstract

Purpose: Develop and demonstrate preliminary validation of a brief questionnaire aimed at assessing social cognitive determinants of physical activity (PA) in a college population.

Design: Quantitative and observational.

Setting: A midsized northeastern university.

Subjects: Convenience sample of 827 male and female college students age 18 to 24 years.

Measures: International Physical Activity Questionnaire and a PA stage-of-change algorithm.

Analysis: A sequential process of survey development, including item generation and data reduction analyses by factor analysis, was followed with the goal of creating a parsimonious questionnaire. Structural equation modeling was used for confirmatory factor analysis and construct validation was confirmed against self-reported PA and stage of change. Validation analyses were replicated in a second, independent sample of 1032 college students.

Results: Fifteen items reflecting PA self-regulation, outcome expectations, and personal barriers explained 65% of the questionnaire data and explained 28.6% and 39.5% of the variance in total PA and moderate-to-vigorous-intensity PA, respectively. Scale scores were distinguishable across the stages of change. Findings were similar when the Cognitive Behavioral Physical Activity Questionnaire (CBPAQ) was tested in a similar and independent sample of college students (40%; R (2) moderate-to-vigorous-intensity PA = .40; p < .001).

Conclusion: The CBPAQ successfully explains and predicts PA behavior in a college population, warranting its incorporation into future studies aiming at understanding and improving on PA behavior in college students.

Keywords: College Students; Health focus: fitness/physical activity; Manuscript format: research; Outcome measure: behavioral; Physical Activity; Prevention Research; Questionnaire; Research purpose: instrument development; Setting: school; Social Cognitive Theories; Strategy: skill building/behavior change; Structural Equation Modeling; Study design: nonexperimental; Target population age: young adults; Target population circumstances: education/income level.

Figure. Cognitive Behavioral Physical Activity Questionnaire* and Questionnaire Instructions†

*On the left is the hierarchical model of the Cognitive Behavioral Physical Activity Questionnaire (CBPAQ), depicting standardized factor loadings between questionnaire items and their constructs and path coefficients between the constructs and the general higher-order factor. Measurement error variances and disturbances were estimated in the model; however, they are not depicted. Model fit statistics: χ²:df = 3.0 (χ² = 258.58; df = 86), comparative fit index = 0.947, and root mean square error of approximation = 0.064 (0.055–0.073). †On the right are the questionnaire instructions and items along with the scoring protocol.