IMPROVING THE VALIDITY OF REGULATORY FOCUS SCALES FOR CULTURAL TARGETING AND TAILORING OF DECISION MAKING TOOLS ACROSS DIVERSE PATIENT POPULATIONS
Method: Two U.S. online panel samples (n=208, age 21-53, 60% female, 77% White / n=222, age 27-48, 57% female, 71% White) completed the Higgins scale as part of larger cross-sectional studies. Two other U.S. online samples (n=406, age 40-69, 51% female, 73% White / n=289, age 40-69, 54% female, 100% White) completed the Haws scale. Self-identified White participants were targeted in the last study to determine whether subcultural values influenced responses to the Haws scale.
Result: A CFA measurement model was tested for all four studies. Fit measures for the Higgins scale in studies 1/2 were below accepted standards (e.g., significant χ²’s, CFI < .90, RMSEA > .08, SRMR > .05). Several factor loadings were below .50 and AVE estimates indicated lack of factor item convergent validity. The Haws composite scale, published in a leading journal and based on multiple RF scales, produced even poorer fit in study 3 (e.g., CFI < .70 and RMSEA > .11). Model fit was equally poor for the low diversity sample in study 4 (e.g., CFI < .70. RMSEA > .11). Modification indices in all four studies indicated numerous problems with correlated errors and cross-loaded items.
Conclusion: Improving effective medical decision making across diverse patient populations necessitates deeper understanding of how cultural values moderate communications between patients and providers. Without valid measures of mindsets that vary across cultures, evidence-based progress will be difficult. Researchers should be aware of the limitations of current RF measures. Continued efforts to produce valid measures of this important construct are critical to developing culturally appropriate decision making communications across diverse patient populations.
See more of: The 36th Annual Meeting of the Society for Medical Decision Making