This experiment aims to investigate sequence effects in the context of taxi driving. Specifically, it examines how the outcome of a previous booking (whether it was completed or canceled) influences taxi drivers’ beliefs about the likelihood of the next booking being canceled and their bidding on the next booking.
Intervention
[Hidden: Will disclose on completion of study]
Experimental design
[Hidden: Will disclose on completion of study]
Design details
Randomization method
Randomization by online Qualtrics system.
Randomization unit
Individual driver.
Clusters
Not clustered.
Observations
Target sample size: 2641 drivers. However, due to administrative constraints in scheduling surveyors and recruiting taxi drivers, we plan to stop at the collection of 500 responses or 1 March 2025, which ever is earlier.
Arms
Each subject will have an equal probability (25%) of being assigned to one of four conditions:
(a) Low fare, first booking cancelled.
(b) Low fare, first booking completed.
(c) High fare, first booking cancelled.
(d) High fare, first booking completed.
Primary outcomes
(1) Belief that next booking would be cancelled.
Subjects will be asked to estimate the likelihood that the next booking will be cancelled, using a 5-point Likert scale ranging from "much less likely than usual" to "much more likely than usual."
(2) Bidding.
Subjects will be asked to bid on the next booking from multiple choices based on the options in the taxi company booking system: 2 minutes, 4 minutes, 6 minutes, 8 minutes, or "do not bid."
Cited by
No citation data available for this registration yet.