Rebecca A Ferrer1, William M P Klein2, Alexander Persoskie3, Aya Avishai-Yitshak4, Paschal Sheeran4. 1. Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Rockville, MD, USA. ferrerra@mail.nih.gov. 2. Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Rockville, MD, USA. 3. Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Silver Spring, MD, USA. 4. University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
Abstract
BACKGROUND: Although risk perception is a key predictor in health behavior theories, current conceptions of risk comprise only one (deliberative) or two (deliberative vs. affective/experiential) dimensions. PURPOSE: This research tested a tripartite model that distinguishes among deliberative, affective, and experiential components of risk perception. METHOD: In two studies, and in relation to three common diseases (cancer, heart disease, diabetes), we used confirmatory factor analyses to examine the factor structure of the tripartite risk perception (TRIRISK) model and compared the fit of the TRIRISK model to dual-factor and single-factor models. In a third study, we assessed concurrent validity by examining the impact of cancer diagnosis on (a) levels of deliberative, affective, and experiential risk perception, and (b) the strength of relations among risk components, and tested predictive validity by assessing relations with behavioral intentions to prevent cancer. RESULTS: The tripartite factor structure was supported, producing better model fit across diseases (studies 1 and 2). Inter-correlations among the components were significantly smaller among participants who had been diagnosed with cancer, suggesting that affected populations make finer-grained distinctions among risk perceptions (study 3). Moreover, all three risk perception components predicted unique variance in intentions to engage in preventive behavior (study 3). CONCLUSIONS: The TRIRISK model offers both a novel conceptualization of health-related risk perceptions, and new measures that enhance predictive validity beyond that engendered by unidimensional and bidimensional models. The present findings have implications for the ways in which risk perceptions are targeted in health behavior change interventions, health communications, and decision aids.
BACKGROUND: Although risk perception is a key predictor in health behavior theories, current conceptions of risk comprise only one (deliberative) or two (deliberative vs. affective/experiential) dimensions. PURPOSE: This research tested a tripartite model that distinguishes among deliberative, affective, and experiential components of risk perception. METHOD: In two studies, and in relation to three common diseases (cancer, heart disease, diabetes), we used confirmatory factor analyses to examine the factor structure of the tripartite risk perception (TRIRISK) model and compared the fit of the TRIRISK model to dual-factor and single-factor models. In a third study, we assessed concurrent validity by examining the impact of cancer diagnosis on (a) levels of deliberative, affective, and experiential risk perception, and (b) the strength of relations among risk components, and tested predictive validity by assessing relations with behavioral intentions to prevent cancer. RESULTS: The tripartite factor structure was supported, producing better model fit across diseases (studies 1 and 2). Inter-correlations among the components were significantly smaller among participants who had been diagnosed with cancer, suggesting that affected populations make finer-grained distinctions among risk perceptions (study 3). Moreover, all three risk perception components predicted unique variance in intentions to engage in preventive behavior (study 3). CONCLUSIONS: The TRIRISK model offers both a novel conceptualization of health-related risk perceptions, and new measures that enhance predictive validity beyond that engendered by unidimensional and bidimensional models. The present findings have implications for the ways in which risk perceptions are targeted in health behavior change interventions, health communications, and decision aids.
Authors: Rebecca A Ferrer; Isaac M Lipkus; Jennifer L Cerully; Colleen M McBride; James A Shepperd; William M P Klein Journal: Soc Sci Med Date: 2017-11-14 Impact factor: 4.634
Authors: Annette R Kaufman; Jenny E Twesten; Jerry Suls; Kevin D McCaul; Jamie S Ostroff; Rebecca A Ferrer; Noel T Brewer; Linda D Cameron; Bonnie Halpern-Felsher; Jennifer L Hay; Elyse R Park; Ellen Peters; David R Strong; Erika A Waters; Neil D Weinstein; Paul D Windschitl; William M P Klein Journal: Nicotine Tob Res Date: 2020-10-29 Impact factor: 4.244