Aug 20, 2015
This study reports public preferences for security, surveillance and privacy across 27 European Member States measured using a stated preference survey. It focuses on three real life contexts: train/metro travel, internet use and storage of health records — each exploring different dimensions of privacy.
Over 26,000 responses were collected using internet and face-to-face surveys in autumn of 2013. The questionnaire included questions on respondent's demographics, questions measuring trust in institutions, general distrust, and attitudes to risk taking. Depending on familiarity to travel by train/metro and internet, each respondent was presented with five stated preference questions from two out three of the above mentioned contexts. The stated preference experiments were designed to understand preferences relating to surveillance, amount of data collected, access to data, storage of data, and cost of security, surveillance and data handling. In the travel context experiment also collected preferences for presence and type of security personnel and physical security checks. Preferences in all three contexts were analysed using discrete choice modelling.
Clear differences in preferences for privacy, security and surveillance are found, depending on the context. The study finds that preferences for security and privacy are surprisingly consistent across the EU. Attitudes and demographic characteristics also influence preferences. This study which is one of the largest applications of discrete choice modelling in this domain, provides an important missing element on public perceptions to the debate on security and privacy.
Preferences relating to security, surveillance and privacy
The role of attitudes in determining the preferences relating to security, surveillance and privacy
Conclusions and policy recommendations
Modelling the effects of country and socio-economic factors on preferences relating to security, surveillance and privacy
Modelling the effects of country-specific, socio-economic and attitudinal factors on preferences relating to security, surveillance and privacy
Structural equation model results