Applying Situational Theory of Problem Solving in Cancer Information Seeking: A Cross-Sectional Analysis of 2014 HINTS Survey

Abstract

Informed by the Situational Theory of Problem Solving (STOPS), this study used data from the Health Information National Trends Survey, a large and representative national sample, to examine predictors of information seeking and information accessing of health information, including cancer-related information. We found that the independent variables in STOPS—problem recognition, involvement recognition, and referent criterion— well predicted people’s information seeking of cancer-related information and accessing of health information on line. However, the impact of trust in online health information was more complicated than anticipated. Our study demonstrated the utility of the STOPS in the health information context. Theoretical and practical implications are discussed.

Notes

¹ HINTS 2014 analytics recommendations document noted that small sample sizes tend to result in unstable estimates. Our final sample was not of a small size (N = 2,293), after excluding missing data and cases with zero weights.

² HINTS 2014 analytics recommendations document stated that if inferential statistical testing is to be performed, statistical programs such as SUDAAN, STATA, SAS and Wesvar should be used. Data analysis should appropriately use weights and replicate weights. We used SAS and performed a two-step Structural Equation Modeling analysis (Confirmatory Factor Analysis as a first step to check for measurement validity and reliability). The commonly accepted practice to account for survey weighting in Structural Equation Modeling is to scale the weights to reflect the sample size and include them in the analysis, which was implemented in our study.