Exploring the psychometric properties of a tripartite model of risk perception (TRIRISK) in a general U.S. population sample.

Objective: Risk perceptions are key constructs in some theories of health behavior. A tripartite model of risk perception, the TRIRISK model, was developed to assess deliberative, affective, and experiential components of risk perception. The current paper attempts to replicate the factor structure of the TRIRISK measure for cancer and extend the structure to respiratory illness. Methods and Measures: Participants 18 or older were recruited using an address-based sample in New York State to participate in a Web-based survey. We employed the TRIRISK questionnaire with respect to cancer and respiratory illness. Confirmatory Factor Analyses were conducted in Mplus to validate the TRIRISK model in our sample. TRIRISK model fit across demographic and behavioral groups was tested using multiple-group models. Results: Of the 704 people included in the analysis, the mean age of participants was 46.9, the majority reported being female (58.5%), and most were White (81.7%). For cancer and respiratory illness, items loaded on the respective constructs as expected. Overall, the TRIRISK model framework fits well across differing subgroups, suggesting that this is a valid model of risk perception to use in a general population sample. Conclusion: These results provide further evidence that the TRIRISK model is a good model to use for risk perceptions in tobacco control research. The TRIRISK model can be used to communicate risk to encourage positive health behaviors among most sociodemographic groups.