Survey experiment on technology use among ride-hail drivers

How to cite

Ong, Pinchuan; Png, Ivan. 2024. "Survey experiment on technology use among ride-hail drivers." AEA RCT Registry. AEARCTR-0013809. https://doi.org/10.1257/rct.13809-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Pinchuan Ong* , Ivan Png

* Primary investigator

Registry ID
AEARCTR-0013809
DOI
https://doi.org/10.1257/rct.13809-1.1
First registered
Last updated
Status
In development
Trial dates
2024-07-01 – 2024-07-31
Countries
Private
JEL codes
O33, M5
Registry record
http://www.socialscienceregistry.org/trials/13809

Abstract

We study the deskilling effect of technology on the supply of labor. In previous work, we fielded a vignette experiment embedded in an online survey conducted on a sample of ride-hail drivers in Singapore. The study showed that ride-hail drivers value the use of map apps, and that its loss would lead to a decrease in supply of labor. In this study, we use a similar methodology to delve deeper and investigate valuations of the various capabilities of map and ride-hail apps---basic static route planning, provision of real-time traffic information, identification of where ride-hail demand is high, and facilitating translations of place names into different languages. Our survey also includes an information experiment on autonomous vehicles technology.

Experimental design

We field a vignette experiment embedded in an online survey, conducted on a sample of ride-hail drivers in Singapore. We show drivers multiple pairs of vignettes in a within-subject design. In each pair, respondents will be asked whether they would work in each scenario and which of the two scenarios they prefer. In the information treatment experiment, respondents are split into control and treatment groups and shown slightly different details.

Design details

Randomization method
Randomization is implemented using Qualtrics's randomization feature.
Randomization unit
Vignette experiments: Within-individual between-vignettes level randomization. Information treatment experiment: Individual-level randomization.
Clusters
1000 individuals.
Observations
Vignette experiments: 1000 individuals * 10 pairs of vignettes * 2 vignettes per pair = 20,000 observations Information treatment experiment: 1000 individuals
Arms
Vignette experiment: On average, 100 observations per each unique vignette scenario (i.e., each technology type by can/cannot use technology condition by commission rate). Information treatment experiment: 500 per treatment arm
Primary outcomes
Willingness-to-pay for the technology shown, and labor supply impact of taking away the technology

Cited by

No citation data available for this registration yet.

Keywords