Modeling user perception and behavior when interacting with a mobile payment system
Niklas Kirschnick · DepositOnce · 2016
Today, security-relevant electronic systems like computers, smartphones, electronic access systems, or electronic banking, concern our whole life. Therefore, ordinary users, not only professionals, have to use such systems in a correct way to not harm their own security. Hence, these systems not only have to be secure, but also usable. System design has to balance security and usability at the same time, which is a challenge. In the usable security research community, it is common sense that both characteristics are mostly counter-directional: more security leads to less usability and vice versa. Hence, it is important to understand what influences the users’ perception, in terms of usability and security, of such systems, and what influences the decision to use a certain system. With this knowledge, more secure and at the same time more usable systems can be developed. Formalizing such knowledge in terms of models which can be used in the development process, ideally considering the usability engineering lifecycle, may lower the development costs and result in better (i.e., more secure and more usable) products. With this work, the author provides a basis to tackle the challenge to understand the users of security-relevant electronic systems. First, a theoretic examination of the user and possible influencing factors on his perception and usage behavior when interacting with security-relevant electronic systems was performed. With the gathered knowledge, a taxonomy of influencing factors was created. From this taxonomy, a theoretic model was derived. Based on this, an evaluation method to examine a mobile payment system was developed, which comprises a questionnaire and an experimental setup. With this evaluation method, five experiments with 88 participants were conducted. After every experiment, the questionnaire and the experimental setup were reviewed and potential enhancements were applied. Using the data of the experiments, statistical models were calculated to identify significant influencing factors. Furthermore, the theoretic model was adapted during the modeling process to reflect findings and inaccuracies of the initial model. Based on the calculated models and influencing factors, a final version of the evaluation method was constructed. To conclude, this thesis contributes a taxonomy of potential influencing factors of the user perception and user behavior. An empirical evaluation method comprising of a questionnaire and an experimental setup is presented. Furthermore, two models of user perception, and one model of user behavior have been calculated to identify significant influencing factors on the users’ perception and behavior. These factors are personality traits (openness, conscientiousness, risk perception, positive attitude) and service-related characteristics (usability perception, security perception, categorial price, service provider, and perceived financial loss magnitude).