While the sharing economy provides flexible and low-barrier jobs for millions of workers globally, a lack of both organization identity and social bonds contributes to the high attrition rate experienced by the sharing-economy platforms. Understanding how to incentivize and engage workers is thus critical. To improve worker engagement, we propose a human-centered data science framework that synergizes strengths across machine learning, causal inference, field experiment, and social science theories. In this talk, I will present our empirical work in collaboration with a leading ride-sharing platform. Inspired by social identity theory and contest theory, we first use a large-scale field experiment to examine the impact of organizing drivers into teams and engaging teams into team contests. We then combine machine learning and causal inference to unpack the heterogeneous treatment effect of team contests at the individual level. Our work identifies directly actionable insights for contest design. More future directions will be discussed to showcase the effectiveness and flexibility of applying the human-centered data science framework to improve worker performance at large.

Speaker: Ms Teng YE
Date: 19 February 2021 (Wed)
Time: 10:00am – 11:00am
Poster: Click here