Responsiveness to Information on Environmental Risk

How to cite

Korb, Mira. 2024. "Responsiveness to Information on Environmental Risk." AEA RCT Registry. AEARCTR-0014166. https://doi.org/10.1257/rct.14166-1.0

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

Mira Korb*

* Primary investigator

Registry ID
AEARCTR-0014166
DOI
https://doi.org/10.1257/rct.14166-1.0
First registered
Last updated
Status
In development
Trial dates
2024-08-23 – 2024-11-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/14166

Abstract

Existing work has documented that interventions lead by non-governmental organizations (NGOs) tend to be more effective
than those lead by other actors, such as government agencies (Vivalt, 2020). Various observed and unobserved implementer-specific factors, such as implementer location, organizational capacity and beneficiaries’ prior experiences with the implementer, can influence the success of the implementer’s intervention. I define an “implementer” as the organization, institution or group involved in planning and executing an intervention. Many public health interventions in developing countries involve sharing information about health risks to encourage people to adopt protective or collective action behaviors. Whether communities adopt these protective behaviors could be affected by whether participants trust the implementer, which in turn may be influenced by prior experiences with the implementer. In this paper, my primary aim is to examine how making the identity of the implementer salient affects the extent to which people update their beliefs towards the provided information. In addition, I explore how belief formation in response to pre-existing perceptions and the information provision can affect willingness to participate in a future intervention. I study these questions in the context of providing information on water pollution to mining communities in Ghana.

Intervention

In my intervention, I provide participants with two measures of water quality for their community borehole: a water quality index that summarizes physiochemical indicators (e.g. heavy metals, pH, turbidity, nitrates, sulfates) and the E-coli count, which summarizes microbiological indicators of water quality. I refer to the physiochemical water quality index as "chemical pollution" and the E-coli count as "bacteria pollution" for simplicity. These water quality measures come from tests of samples of the community's borehole, where the sampling was conducted jointly by researchers at the Ghana Environmental Protection Agency (EPA) and a local NGO, the Centre for Social Impact Studies (CeSIS). During the provision of the test results, I experimentally vary the salience of the implementer of the sampling, holding constant the test result. In one treatment arm, respondents are told that the sampling was conducted by researchers from the EPA, a government agency that works to address issues of environmental quality, while in the other treatment arm, respondents are told that the sampling was conducted by researchers from CeSIS, an NGO that works on environmental advocacy. I identify treatment effects on measures of belief updating about pollution levels.

Experimental design

I study how community members update their beliefs about water pollution in response to information about pollution risk, specifically examining how this belief updating may differ depending on who participants perceive that the information is coming from. In addition, I study the extent to which information can improve perceptions of or change preferences for the implementer of an intervention, as well as influence future interest in engagement with the implementer. I share two measures of water quality that summarize the results of tests conducted on samples collected from community boreholes by researchers at the Ghana Environmental Protection Agency (EPA) and a local NGO, CeSIS. I collect incentivized point beliefs for the levels of chemical and bacteria pollution in a community's borehole water, to understand how belief updating may be affected by making an implementer salient when the test results are shared. I also measure perceptions of accuracy of government and NGO sources, as well as the extent to which perceptions can be changed by the information provided. Finally, I examine whether the provided information can influence interest in engagement with the implementer as part of a future project.

Design details

Randomization method
Randomization is stratified by community, on R using a computer, to ensure that there is balance between the government and NGO treatment arms within a community.
Randomization unit
Individual
Clusters
990 individuals
Observations
990 individuals
Arms
Government (EPA) - 495 individuals NGO (CeSIS) - 495 individuals
Primary outcomes
1. Belief updating about chemical pollution levels in water 2. Belief updating about bacteria pollution levels in water
Secondary outcomes
All secondary outcomes are listed in the pre-analysis plan, which will be made publicly available at the end of my experiment.

Cited by

No citation data available for this registration yet.

Keywords