Situated Food Safety Risk and the Influence of Social Norms

Previous studies of risk behavior observed weak or inconsistent relationships between risk perception and risk-taking. One aspect that has often been neglected in such studies is the situational context in which risk behavior is embedded: Even though a person may perceive a behavior as risky, the social norms governing the situation may work as a counteracting force, overriding the influence of risk perception. Three food context studies are reported. In Study 1 (N = 200), we assess how norm strength varies across different social situations, relate the variation in norm strength to the social characteristics of the situation, and identify situations with consistently low and high levels of pressure to comply with the social norm. In Study 2 (N = 502), we investigate how willingness to accept 15 different foods that vary in terms of objective risk relates to perceived risk in situations with low and high pressure to comply with a social norm. In Study 3 (N = 1,200), we test how risk-taking is jointly influenced by the perceived risk associated with the products and the social norms governing the situations in which the products are served. The results indicate that the effects of risk perception and social norm are additive, influencing risk-taking simultaneously but as counteracting forces. Social norm had a slightly stronger absolute effect, leading to a net effect of increased risk-taking. The relationships were stable over different social situations and food safety risks and did not disappear when detailed risk information was presented.     

ANOVA for unbalanced data: Use Type II instead of Type III sums of squares

Methods for analyzing unbalanced factorial designs can be traced back to Yates (1934). Today, most major statistical programs perform, by default, unbalanced ANOVA based on Type III sums of squares (Yates's weighted squares of means). As criticized by Nelder and Lane (1995), this analysis is founded on unrealistic models—models with interactions, but without all corresponding main effects. The Type II analysis (Yates's method of fitting constants) is usually not preferred because of the underlying assumption of no interactions. This argument is, however, also founded on unrealistic models. Furthermore, by considering the power of the two methods, it is clear that Type II is preferable.     

Analyzing Designed Experiments with Multiple Responses

This paper is an overview of a unified framework for analyzing designed experiments with univariate or multivariate responses. Both categorical and continuous design variables are considered. To handle unbalanced data, we introduce the so-called Type II* sums of squares. This means that the results are independent of the scale chosen for continuous design variables. Furthermore, it does not matter whether two-level variables are coded as categorical or continuous. Overall testing of all responses is done by 50-50 MANOVA, which handles several highly correlated responses. Univariate p-values for each response are adjusted by using rotation testing. To illustrate multivariate effects, mean values and mean predictions are illustrated in a principal component score plot or directly as curves. For the unbalanced cases, we introduce a new variant of adjusted means, which are independent to the coding of two-level variables. The methodology is exemplified by case studies from cheese and fish pudding production.