Politician Charity and Corruption in Kenya
Last registered on August 07, 2017

Pre-Trial

Trial Information
General Information
Title
Politician Charity and Corruption in Kenya
RCT ID
AEARCTR-0002375
Initial registration date
August 05, 2017
Last updated
August 07, 2017 9:58 AM EDT
Location(s)
Region
Primary Investigator
Affiliation
Stanford University
Other Primary Investigator(s)
Additional Trial Information
Status
Completed
Start date
2017-07-12
End date
2017-08-05
Secondary IDs
Abstract
My research considers how incumbent politicians are viewed by the community as playing a critical role in providing cash to the citizens for expenses that they cannot afford and in contributing to local public goods initiatives. It suggests that although voters dislike corruption and value public goods provision, voters also tend evaluate candidate performance on the basis of observed cash transfers to individuals and to the community. This study has a series of survey, list, and behavioral experiments to test the efficacy of different campaign promises, the extent to which politician charity and corruption impacts voter evaluations of candidates, how poverty can increase voter expectations of cash, and how pessimism about corruption can undermine citizen valuations of public goods.
External Link(s)
Registration Citation
Citation
Zhang, Kelly. 2017. "Politician Charity and Corruption in Kenya." AEA RCT Registry. August 07. https://www.socialscienceregistry.org/trials/2375/history/20270
Experimental Details
Interventions
Intervention(s)
This study uses vignette, list, and behavioral experiments embedded within a survey.
Intervention Start Date
2017-07-12
Intervention End Date
2017-08-05
Outcomes
Outcomes (end points)
Voter support for a candidate, count of campaign promises that are important, expected cash assistance by politicians, expected fundraiser contributions by politicians, expected corruption, and participant donations to local hospital.
Outcomes (explanation)
Experimental Design
Experimental Design
This study uses vignette, list, and behavioral experiments embedded within a survey.
Experimental Design Details
Randomization Method
Randomization done within SurveyCTO.
Randomization Unit
Individual
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
450
Sample size: planned number of observations
450
Sample size (or number of clusters) by treatment arms
225 for treatment arm, 225 for control arm.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power is calculated to capture a drop in voter support for a candidate from 49% to 29%, assuming the standard deviation of the control group 0.50, with α = 0.05 and β = 0.80 for a two-sided t-test. The sample is also large enough to detect a difference of 23,885.71 for MP harambee contributions (assuming the standard deviation of the sample at 86,463.09 with α = 0.05 and β = 0.80 for a two-sided t-test) and a difference of 118,632.14 for expected assistance from an MP (assuming the standard deviation of the sample at 302,032.89 with α = 0.05 and β = 0.80 for a two-sided t-test).
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Study has received IRB approval. Details not available.
IRB Approval Date
Details not available
IRB Approval Number
Details not available
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers
</