This story draft by @escholar has not been reviewed by an editor, YET.
Authors:
(1) Muhammad Zia Hydari, Katz Graduate School of Business, University of Pittsburgh and Corresponding author;
(2) Idris Adjerid, Pamplin College of Business;
(3) AAaron D. Striegel, Department of Computer Science and Engineering, University of Notre Dame.
2. Background and 2.1. Leaderboards
3. Effect of Leaderboards on Healthful Physical Activity and 3.1. Competition
3.3. Moderating Effects of Prior Activity Levels and Leaderboard Size
4. Data and Model
5. Estimation and Robustness of the Main Effects of Leaderboards
5.2. Robustness Check for Leaderboard Initiation
5.4. Fitbit Attrition, Leaderboard De-Adoption, and Additional Robustness Checks
6. Heterogeneous Effect of Leaderboards
6.1. Heterogeneity by Prior Activity Levels
6.2. Interaction of Leaderboard Size, Rank, and Prior Activity Levels
6.3. Summary of Findings from Heterogeneous Effect Analysis
7. Conclusions and Discussion, Endnotes, and References
The goal of our analysis is to estimate the effect of a user’s leaderboard adoption on their physical activity as measured by steps walked, using nonexperimental data.13 Thus, leaderboard adoption is the treatment in our observational study. In a randomized experiment, leaderboards could be randomly assigned to study participants, which would make the identification of treatment effect straightforward but would make the study treatment very different from naturally occurring leaderboards. In contrast, any Fitbit user in our study can opt into and construct their leaderboard, resulting in more natural leaderboards but making it more difficult to identify the treatment effect. The main empirical concern in identifying this treatment effect is the confoundedness of the leaderboard adoption with respect to users’ physical activity as measured by their daily step count.
We use a DID research design as Fitbit users are observed over multiple time periods, with roughly half of the users adopting leaderboards and the other half remaining untreated. Although a DID design controls for any time-invariant user characteristics and common shocks, it requires the identifying assumption that any uncontrolled time-varying user characteristics exhibit a common trend across the treated and untreated individuals. Under this identifying assumption, we can estimate the effect of leaderboards on steps walked for the Fitbit users who have adopted leaderboards. Our model specification is given here and its explanation follows:
This paper is available on arxiv under CC BY 4.0 DEED license.