GNGTS 2015 - Atti del 34° Convegno Nazionale

Discussion. According to our interpretation the risk perception seems be underestimated for all factors. In particular seismic Vulnerability factor scores are extremely low (2.77 per PGA<0.15; 2.83 per PGA>0.15) and moreover they are contrariwise than expected. When we will have processed house information data of the respondents, we will comment better these scores of (physical) vulnerability. We will be able to know if these scores are low because the respondents’ houses are safe, or on the contrary, these scores show a low vulnerability perception. Research on risk perception aims to aid risk analysis and policy-making by providing a basis for understanding and anticipating public responses to hazards and improving the communication of risk information among people, technical experts, and decision-makers. Those who deals with seismic risk need to understand how people think about and respond to this risk. Without such understanding, well-intended policies may be ineffective. (Slovic, 1987). For these reasons we believe that comparing the perception factors with the “real factors” of seismic risk, is a crucial point. Risk perception research without a comparison with reality is a pure intellectual exercise. Since the first year of our research we produced a comparison between seismic hazard and perception of hazard. In this research, we are committed to complete the comparison with the factors of Exposure and Vulnerability. About the considered difficulties to obtain data on the vulnerability of private buildings, we have introduced in the questionnaire a new section dedicated to Vulnerability, with the aim to collect real data on interviewee’s houses (see Paragraph Questionnaire Update). These data are being processed by INGV-Milan, aiming to compare them with the vulnerability perception data. INGV-RM1 is working on tangible and intangible assets data in order to obtain a synthetic Exposure value to compare with Exposure perception data. Other data collected by the questionnaire regard Earthquake information level, Sources of information, Earthquake occurrence with respect to other natural hazards, participation at risk reduction activities and level of involvement. The results relating to Earthquake Information show that 6% of the sample thinks to be informed “very much”, 33% “somewhat”, the remaining 61% is divided between “slight” (38%) and “not at all” (23%). Fig. 3 – Comparison of data collected on the web and CATI survey data (mean values). GNGTS 2015 S essione 2.3 297