Individuals are known to lie and/or provide untruthful data when providing information online as a way to protect their privacy. Prior studies have attempted to explain when and why individuals lie online. However, no work has examined into how people lie online, i.e. the specific strategies they follow to provide untruthful data, or attempted to predict whether people would be truthful or not depending on the specific question/data. To close this gap, we present a large-scale study with over 800 participants. Based on it, we show that it is possible to predict whether users are truthful or not using machine learning with very high accuracy (89.7%). We also identify four main strategies people employ to provide untruthful data and show the factors that influence the choices of their strategies. We discuss the implications of findings and argue that understanding privacy lies at this level can help both users and data collectors.
https://doi.org/10.1145/3411764.3445625
The ACM CHI Conference on Human Factors in Computing Systems (https://chi2021.acm.org/)