x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
Air Pollution SMS Messaging in Mexico City
Last registered on May 06, 2019
View Trial History

Pre-Trial

Trial Information
General Information
Title
Air Pollution SMS Messaging in Mexico City
RCT ID
AEARCTR-0003879
Initial registration date
February 11, 2019
Last updated
May 06, 2019 10:03 PM EDT
Location(s)

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Name
Bridget Hoffmann
Affiliation
Inter-American Development Bank
Contact Primary Investigator
Other Primary Investigator(s)
Additional Trial Information
Status
In development
Start date
2019-06-03
End date
2020-08-31
Keywords
Environment & Energy, Health
Additional Keywords
air pollution information, willingness to pay, avoidance behavior
JEL code(s)
Q53; Q51; D83; D12; O12; C93
Secondary IDs
Abstract
Many cities and countries invest in air pollution monitoring systems. The value of these systems depends crucially on how the information that they collect is used by the public. For example, Mexico City has a robust monitoring network of over 30 stations with hourly pollutant readings that are available online and through an app. However, the public is not regularly using this information to engage in protective "avoidance" behaviors to reduce their exposure to air pollution and mitigate its related health risks.

We develop a messaging service that provides location and pollutant specific (ozone and particulate matter) alerts through SMS message early in the day. We use an incentive compatible mechanism to elicit willingness to pay for the SMS alert service and we randomly assign households to receive or not receive the SMS alert service. Participants are also randomly assigned to two cross-cutting treatments: (1) monthly SMS reminders of air pollution trends and avoidance behaviors, and (2) provision of a free N95 certified mask. We will estimate the causal effect of the alert service and reminder information on avoidance behavior. Because the N95 mask is effective in reducing exposure to particulate matter but not ozone, receiving a free mask lowers the cost of avoidance for days with high particulate matter relative to days with high ozone. We will use this variation to investigate the causal effect of the cost of avoidance behavior on engaging in avoidance behavior.
External Link(s)
Registration Citation
Citation
Hoffmann, Bridget. 2019. "Air Pollution SMS Messaging in Mexico City." AEA RCT Registry. May 06. https://doi.org/10.1257/rct.3879-2.0
Former Citation
Hoffmann, Bridget. 2019. "Air Pollution SMS Messaging in Mexico City." AEA RCT Registry. May 06. https://www.socialscienceregistry.org/trials/3879/history/46023
Sponsors & Partners
Experimental Details
Interventions
Intervention(s)
Air quality alerts: SMS messaging service that sends location and pollutant specific air quality alerts to participants

Reminders: monthly SMS message with reminders of air quality trends and air pollution avoidance behaviors

Masks: free provision of a certified N95 mask
Intervention Start Date
2019-06-03
Intervention End Date
2020-08-31
Primary Outcomes
Primary Outcomes (end points)
Willingness to pay, air pollution information, avoidance behaviors
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Beliefs about air pollution trends and levels
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The sample consists of households in randomly selected blocks in Mexico City. Households are randomized into the pure control group or one of the 7 treatment arms described below.

Pure control: no intervention
Arm 1: free provision of mask
Arm 2: air quality alerts
Arm 3: reminders
Arm 4: air quality alerts + reminders
Arm 5: air quality alerts + mask
Arm 6: reminders + mask
Arm 7: air quality alerts + reminders + mask

Within each of the pure control and treatment arms, the participants will be split between a regular compensation group and a high compensation group.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by computer
Randomization Unit
Household
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
1,850 households
Sample size: planned number of observations
1,850 households
Sample size (or number of clusters) by treatment arms
Pure control and all arms: 1/8 of the sample. Within each cell of the treatment matrix, 70% of the participants will receive the regular compensation level and 30% of participants will receive the high compensation level.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
University of Southern California
IRB Approval Date
2017-04-14
IRB Approval Number
IRB17-0091
IRB Name
Harvard University
IRB Approval Date
2017-04-14
IRB Approval Number
IRB17-0091
Analysis Plan