Nudge Driving Safety: Evidence from a Field Experiment
How to cite
Chen, Fadong; Han, Jiatong. 2022. "Nudge Driving Safety: Evidence from a Field Experiment." AEA RCT Registry. AEARCTR-0009769. https://doi.org/10.1257/rct.9769-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
With the rise of the sharing economy, more and more drivers have joined online car-hailing platforms. On the one hand, this has brought new benefits to the platform. On the other hand, risky driving habits also bring a lot of safety hazards to platforms and society. Therefore, how to nudge drivers to develop good driving habits has become an urgent problem.
In this study, we plan to conduct a large-scale field experiment in China. We will select drivers with similar historical performance on an online car-hailing platform and randomly assign them to a course learning group, a voice broadcast group, a positive framing group, a negative framing group, a money incentive group, an honorary incentive group, team competition groups, and control groups. In the one-month experiment, by obtaining the GPS positioning information of the driver's mobile phone, we would take the zebra crossing overspeed ratio as the main indicator of the driver's driving behavior, and continue to observe the driver's driving behavior for another two weeks after the end of the experiment.
After the experiment, we expect to get the following results: (1) Compared with the control group, the zebra crossing overspeed ratio of the intervention group is lower; (2) Compared with the course learning group, the zebra crossing overspeed ratio of voice broadcast group is lower; (3) Compared with individual drivers participating in the safe driving competition, the zebra crossing overspeed ratio of drivers in teams is lower; (4) Compared with the positive framing group, the zebra crossing overspeed ratio of the negative framing group is lower; (5) Compared with the honorary incentive group, the zebra crossing overspeed ratio of the money incentive group is lower.
Intervention
Different groups have different interventions.
Control group: no intervention.
Course learning group: drivers need to learn safe driving courses and pass the exam before starting their work at the beginning of each week, 4 times in total.
Voice broadcast group: drivers will receive a voice broadcast before starting their work at the beginning of each week, 4 times in total.
Positive framing effect group: the staff will communicate with the drivers by telephone to emphasize the positive effect of safe driving, once every two weeks, twice in total.
Negative framing effect group: the staff will communicate with the drivers by telephone to emphasize the negative effect of risky driving, once every two weeks, twice in total.
Money incentive group: drivers will participate in the safe driving competition alone, and drivers who perform well will be rewarded with money. Drivers will be rewarded at the end of the second week, the third week, and the fourth week respectively.
Honorary incentive group: drivers will participate in the safe driving competition alone, and drivers who perform well will be rewarded with banners which can be decorated on the car. Drivers will be rewarded at the end of the second week, the third week, and the fourth week respectively.
Team competition group: drivers will participate in the safe driving competition in teams, and teams who perform well will be rewarded with money. Drivers will be rewarded at the end of the second week, the third week, and the fourth week respectively.
Team competition control group: drivers are in teams but have no intervention.
Experimental design
The purpose of the study is to find out the best way to nudge drivers to drive safely.
We plan to select drivers who worked 100-250 hours in the past month and had zebra crossing speeding behavior in the past three months. Then we would assign them to groups randomly by a computer. Due to the limitation of the total number of drivers, the experiment will be conducted in two cities A and B.
In city A, there would be one control group (200 drivers), one course learning group (200 drivers), one voice broadcast group (200 drivers), one positive framing group (200 drivers), one negative framing group (200 drivers), one money incentive group (200 drivers).In city B, there would be one control group (200 drivers), one honorary incentive group (200 drivers), one team competition group (20 teams, 400 drivers), and one team competition control group (20 teams, 400 drivers).
Details of the experimental intervention are described above. We are mainly interested in the zebra crossing overspeed ratio of drivers. The experiment would continue for four weeks and observe the driver's driving behavior for another two weeks after the experiment.
Design details
Cited by
No citation data available for this registration yet.