We posted on SSRN a working paper by Heng Xu and Nan Zhang, addressing challenges facing the theorization of complex privacy phenomena.

Abstract: People’s perceptions about privacy are marked by complexity, as they often respond to privacy issues in idiosyncratic and occasionally contradictory ways. In this paper, we explain how such complexity transgresses the limits of “net-effect thinking”, which permeates the canonical paradigm of variable-/variance-centric theorizing in privacy research. We also explore the promise of configurational theorizing, a “people-centric” approach that captures how multiple concepts combine in distinct configurations for different individuals. Noting the inadequacy of extant quantitative methods for configurational theorizing in capturing the distinct characteristics of privacy phenomena, we leverage recent advances in machine learning to develop metric-LPA, a novel metric-learning based method for configurational theorizing. Empirical evidence demonstrates the value of configurational theorizing in general, and metric-LPA in particular, for advancing theory development in privacy research.

Here is the link to the working paper.