Application of Theory of Planned Behavior to Transit Use in the Houston Train Study

Abstract

Transit-related physical activity (PA) can contribute to total daily PA. Public transit use is related to increases in PA. Identifying the determinants of public transit use could be important in understanding ways to promote PA. The Theory of Planned Behavior (TPB) has been used to predict many health behaviors, but not transit use. TheTransit-Related Activity in Neighborhoods (TRAIN) Study, a prospective natural experiment of transit use following expansion of the light-rail infrastructure in Houston, TX, may be used to address this question. PURPOSE:To examine the utility of theTPB model for predicting transit use in a low-income, minority population. METHODS:This study used baseline TRAIN Study data (2013-2015) which included reported transit use items and key TPB constructs. Structural equation modeling tested the fit of the TPB model for explaining the number of trips taken on buses or light rail in a typical week. The latent variables were: attitudes, subjective norms, perceived behavioral control, and intentions related to transit use. RESULTS: Participants included 695 (63% women) with a mean age of 52 (SD = 14) years and the majority (63%) were Black or Hispanic. The average number of reported transit trips per week was 8.7 (min = 0; max = 102) with 28% reporting zero trips. Standardized item-factor loadings for TPB constructs were all statistically significant (p<0.001), ranging from 0.69 to 0.87. The model demonstrated adequate fit with the data: χ2 (100, N=593) = 375.3, p < 0.001, CFI = 0.93, RMSEA = 0.067. Attitudes (β = 0.97, SE = 0.08) and subjective norms (β = 0.36, SE = 0.08) were significantly related to intentions (p < 0.001). Intentions (β = 3.32, SE = 0.59), age (β = -0.1, SE = 0.041), education (β = -1.1, SE = 0.34) and reported concern about becoming the victim of crimes in the neighborhood (β = 1.25, SE = 0.49) were significantly associated with reported transit use (p < .05). The variance in intentions and the outcome accounted by the model were 68.5% and 13.5%, respectively. CONCLUSION: Results suggest that TPB is a promising theoretical model to explain transit use. Although more data are needed to confirm these results, this model may be useful to guide future interventions to promote transit-related PA among adults in low-resource communities. Funded by NIH R01 DK101593