NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Back to History
Maasai Watchman Field Experiment
Initial registration date
February 14, 2018
November 03, 2019 10:55 AM EST
University of Toronto
Other Primary Investigator(s)
Additional Trial Information
I create randomised variation in farm security by matching randomly selected Kenyan smallholding farmers with subsidized Maasai watchmen to protect farms during the main agricultural season. I explore the effect of theft and property crime by testing whether farmers engage in different types of production when their farms are more secure against theft. I further explore how social networks are used to substitute for imperfect protection of property in an environment where state institutions are unable to fully protect farms, and how theft is used to sanction those who neglect social obligations.
The intervention in this project is matching farming households to high-quality, trusted Maasai watchmen at a heavily subsidized rate. The intention of this intervention is to cause variation in the security of farms during the long rainy season.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
Share of land used for growing high-value or theft-prone crops, and decision to plant new crops.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
This experiment will consist of a core sample who will be the treatment and control group, answering the main research questions, and a supplementary spillover sample that will be used to estimate the spillovers on crime of the intervention. The core sample will be cluster randomized by village into treatment or control.
Experimental Design Details
Randomization done in office by computer. See Pre Analysis Plan for details.
Village. See Pre Analysis Plan for details.
Was the treatment clustered?
Sample size: planned number of clusters
60 Villages, 10 farming households per village.
Sample size: planned number of observations
Target is 600 farming households
Sample size (or number of clusters) by treatment arms
200 farmers treated, 400 control.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
See Pre Analysis Plan.
INSTITUTIONAL REVIEW BOARDS (IRBs)
University of Toronto Research Ethics Board
IRB Approval Date
IRB Approval Number
Protocol Reference # 34160