Supplemental Material for Does Statistical Evidence Change Case-Based Perception of the Effectiveness of Earthquake Early Warning?

A large number of studies have reported that the perceived effectiveness of a hazard mitigation measure is one of the major determinants of an individual’s adoption of the measure. However, few studies have examined perceived effectiveness in the context of a communication technology such as Earthquake Early Warning (EEW). Two types of information can influence the effectiveness of a measure: case-based evidence focusing on effectiveness in the context of a particular person and statistical evidence that represents effectiveness for the greater population. This study investigated whether an individual’s perceived effectiveness of EEW, formed with case-based evidence, could be revised following provision of statistical evidence (and vice versa). The results of our experiment (N = 109) using actual statistics and person-focused cases showed that statistics did not revise participants’ perceptions of EEW effectiveness that they had already formed based on the cases. The statistics failed to revise not only perceived effectiveness formed by a negative case where EEW was presented as less effective (in line with the identifiable victim effect) but also perceptions formed by a positive case. When reversed, it was also found that perceived effectiveness based on statistics presented first can easily be overturned by case-based information. These findings of this study add new empirical evidence to the literature that shows specific cases influence people’s judgment and decision making more than abstract statistics. Practitioners involved in risk communication should understand that once people’s effectiveness perceptions are formed by a particular case, it is difficult to revise them by presenting statistics.