Uncovering the predictors of unsafe computing behaviors in online crowdsourcing contexts

Abstract

The self-protective decisions of crowd workers are driven by the interplay of many factors, including the characteristics of the crowdsourced tasks, the trustworthiness of the task providers, and the perceived reliability of the crowdsourcing marketplace. In this paper, we report the results of an extensive empirical investigation into the factors driving threat avoidance intentions and behavior of workers. We utilize the Technology Threat Avoidance Theory (TTAT) to explore the decision determinants of crowd workers regarding whether to accept work on crowdsourced tasks involving potential security threats or privacy violations. The identification of the role of human factors in the present behavioral research is based on an analysis of the responses of 882 crowd workers to an online survey crowdsourced on a popular crowdsourcing-based marketplace. The results obtained after testing the TTAT-based behavioral model identify a perceived threat as the most important determinant of threat avoidance decisions of workers and shed light on key factors that can drive workers to discontinue their self-protection behaviors. Furthermore, we present platform design recommendations for helping crowd workers assure their privacy and security while working on tasks crowdsourced through online marketplaces. To the best of our knowledge, we conducted the first study highlighting the determinants of the self-protective behaviors of crowd workers and providing valuable insight into maximizing worker attention toward avoiding security- and privacy-related threats.