GNGTS 2015 - Atti del 34° Convegno Nazionale

During the above mentioned meeting at the DPC, a new aspect was considered: how to find real vulnerability data of private houses to be compared with perception data? For this reason it was decided to implement the questionnaire with a section dedicated to collect information about the vulnerability of the interviewee’s house. In collaboration with DPC, ReLUIS and INGVMilan, the following questions were included in the questionnaire: structural typology of the building; age of building; number of floors; structural context of the building; maintenance state. Our intent was to compare information on house received by the interviewee with his perception of vulnerability. The questionnaire was designed to be completed on the web or to be pencilled. For this reason it was necessary to adapt it for a telephone interview. The adaptation concerns: • the repositioning of the questions on personal data at the end of the interview, when attention could be lower than at the beginning; • the repetition of the Likert scale values for all items of the questionnaire during the interview; • the conversion of the web's informal language, into the formal language typical of telephone interviews. A preliminary test to check the understanding and the keeping of the questionnaire was carried out on over 50 people with good results, one month before the survey launching (December 2014). Once the telephone survey was completed (March 2015), the revised questionnaire was published on the web at www.terremototest.it . However, the responses to the online users were not yet implemented. The web answer page produces, as in the previous version, a comparison between the hazard perception score and the hazard by law for seismic zone, and does not consider the other risk factors. Furthermore, the answer page contains some general information about the vulnerability and seismic risk mitigation. The aspect, concerning the communication and education of risk, requires a thorough discussion between the different actors of the research, that are DPC, INGV and engineers. Once the process will be completed, we plan to replace the current answer page with a dedicated website containing results of perception and insights regarding the seismic risk reduction. Sampling. Basically, the types of surveys are broadly categorised into two groups: according to instrumentation and to the span of time involved. The types of surveys according to instrumentation include the questionnaire and the interview. On the other hand, the types of surveys of the second group contain both cross-sectional and longitudinal surveys. The most important methodological steps of a survey include to: • identify and select potential sample members; • contact sampled individuals and collect data from those who are hard to reach (or reluctant to respond); • evaluate and test questions; select the mode for posing questions and collecting responses; • train and supervise interviewers; • check data files for accuracy and internal consistency; • adjust survey estimates to correct for identified errors. In the first year of research (2012-2013) the survey was conducted via web on a random sample. In the second research year we executed a cross-sectional study on a statistical sample of the Italian population in a fixed-time, using the CATI technique. CATI technique is one of the most used in statistical surveys because it assures the acquisition of accurate and standardised data set, reducing the time needed to process data. In addition, the CATI technique, compared to other ones, makes possible to reach almost all population groups. 294 GNGTS 2015 S essione 2.3