Crime Perceptions, Urban Transport Choices and Support for Climate Change Initiatives

How to cite

Martini, Santiago De; Gonzalez, Juan; Perez-Vincent, Santiago. 2024. "Crime Perceptions, Urban Transport Choices and Support for Climate Change Initiatives." AEA RCT Registry. AEARCTR-0013745. https://doi.org/10.1257/rct.13745-1.0

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

Santiago De Martini* , Juan Gonzalez , Santiago Perez-Vincent

* Primary investigator

Registry ID
AEARCTR-0013745
DOI
https://doi.org/10.1257/rct.13745-1.0
First registered
Last updated
Status
Completed
Trial dates
2024-06-04 – 2024-06-14
Countries
Argentina (Latin America), Chile (Latin America), Colombia (Latin America), Guatemala (Latin America), Jamaica (Latin America), Mexico (Latin America), Peru (Latin America), Trinidad and Tobago (Latin America)
JEL codes
R41, Q56, H23
Registry record
http://www.socialscienceregistry.org/trials/13745

Abstract

This study aims to find causal evidence on the relationship between safety in the public transport and commuters' transport mode choice. This study has three main research questions to be answered. First, we want to get an estimate of what is the willingness to pay for safety in the public transport. Second, for a given trip, up to what extent would people substitute their choice of transport mode from private taxis (say, Uber or Lyft) to bus if safety in the latter increases. Finally, we want to understand whether higher insecurity in the public transport system crowds out support for other policies (in particular, environmental policies). We study this through a non-incetivized lab experiment carried on in eight Latin American countries. The study is divided in three different experiments, where participants are randomized in different groups where each group is presented with a different set of information. While in the first two experiments participants are asked to choose between different transport modes, in the last one we ask the participants to allocate a budget between four policies as if they were policymakers. We hope that the results of this study would contribute to understand whether increasing the safety in the public transport system plays a role in tackling climate change by substituting from private to public transport modes.

Experimental design

The experiment is divided into three sub-experiments which we denote as parts 1-3. Morover, agents answer questions about other control variables (such as demographics).

In part 1, subjects will have to choose between two types of buses (Bus A and Bus B) to do a given trip. The information of each bus is presented in two columns, but each piece of information is hidden behind a box that has the title of the information contained in each of them. To see the information behind each box, the participants have to click in the corresponding box. There are boxes (pieces of information) about 4 attributes: estimated trip duration (in minutes), estimated emissions (in g of CO2), safety (in number of crimes in the bus) and price (in local currency). Regarding the safety information of Bus B, subjects will be randomized into different levels (+30%, +10%, -10%, -30% in the number of crimes committed on public transportation with respect to the average crime level in buses in their city), and the crime level of Bus A will always be equal to 'same number of crime of buses as the average in your city'. For groups +10% and +30% the price presented of Bus B will be 50% lower than the one of Bus A, and for groups -10% and -30% the price is 50% higher. We proced through the Gabor-Granger method to capture the indifference price between each transport mode.

In part 2, will face a choice between two options: an Uber ride at its usual local price, and an equivalent bus ride at its usual local price; we will present the same attributes as in part 1 The attributes will be presented in two columns, all at once (no need to click on boxes in this part). Moreover, subjects will be randomized between the same levels of bus safety as in part one: +30%, +10%, -10%, -30%. Then, participants will choose whether they would use the private or public transportation option. Following the Gabor-Granger method again, participants who choose public (private) transport will then face the same choice but with public transportation price X% higher (cheaper).

In part 3, we will randomly allocate the participants in one of three groups: two treatments and one control. In one of the treatments, we will present a newspaper article that suggests that the public transport in their city is safe and in the other treatment an article that suggests that it is not safe (and we present a 'neutral' or 'placebo' article the control group). Then, we will tell the participants to think that they are the policymakers of their city and that they have a budget of 120 thousand dollars to allocate between four policies: environmental, efficiency, price and safety. The agents then give their allocation choice and we are interested in studying whether being exposed to different newspaper articles (that aims to vary the safety perception in the public transport system of their city) changes the allocation of the budget.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual
Clusters
0
Observations
5400
Arms
Evenly distributed. For the first two experiments there are 4 arms, so we expect to have approximately 1350 individuals in each one of them. In the last experiment, there are 3 arms, so we expect to have approximately 1800 participants in each arm.
Primary outcomes
First, willingness to pay to use a bus in a given trip under different safety levels. Probability of choosing a bus over a private taxi for different bus safety levels. How the allocation of a budget changes when participants are exposed to different newspaper articles that suggest that the public transport system is safe, not safe or a newspaper article that says nothing about safety. Finally, we are interested in capturing the order in which participants click the 'tags' of different attributes of transport modes (mousetracking). Participants are asked to click on different boxes that have a tag that correspond to the topic of the information hidden in each one of them to see such information.

Cited by

No citation data available for this registration yet.

Keywords