Fear Appeals and Information Security Behaviors: An Empirical Study on Mechanical Turk

Abstract

This study aims to conduct a methodological replication of the information security study conducted by Johnston and Warkentin (2010). This study leveraged the use of the fear appeals model (FAM) in the context of information security as they pertain to the individual use of anti-spyware software. We adopt all measures, instruments, statistical tests, theory, and models from the original study, but apply them to the Amazon Mechanical Turk population. The results from this replication study are not consistent with the original study, in that two of the five posited hypotheses have opposite effects than those originally found; threat severity is shown to have a positive effect on both response efficacy and selfefficacy, where in the original study, this is shown to have a negative effect on both. The results imply that there may be differences in which populations the study was conducted, thus requiring additional samples and statistical tests.