Information Barriers to Public Service Delivery: Innovations in Strengthening Local Political Accountability in Uganda
Last registered on August 26, 2014


Trial Information
General Information
Information Barriers to Public Service Delivery: Innovations in Strengthening Local Political Accountability in Uganda
Initial registration date
August 26, 2014
Last updated
August 26, 2014 2:36 PM EDT
Primary Investigator
Innovations for Poverty Action
Other Primary Investigator(s)
PI Affiliation
Vanderbilt University
PI Affiliation
University of Pennsylvania
Additional Trial Information
On going
Start date
End date
Secondary IDs
There is a wide agreement among scholars of African politics that politicians feel little electoral pressure to improve vital public goods and social services. The prevalence of clientelism, ethnic voting and faulty elections are commonly held as the main culprits. Our research project focuses instead on two alternative barriers that weaken democratic accountability across Africa: (1) acute lack of politician performance information in the hands of citizens, which renders citizens ethnic or clientelistic voters; and (2) lack of information in the hands of politicians about service deficiencies in their constituency, which reduces their ability to monitor and sanction service providers. In collaboration with a leading Ugandan NGO – Advocates Coalition for Democracy and the Environment (ACODE) – IPA associated Principal Investigators Guy Grossman and Kristin Michelitch have designed two interventions to directly address these informational barriers: (1) a widespread dissemination of information regarding the performance of local government councilors' to citizens, and (2) opening a new channel of communication from citizens to local government councilors to report public service deficiencies via an innovative text message technology.

Through a randomized controlled trial (RCT) method, the research team is studying the extent to which those two interventions impact the behavior of voters and politicians, and ultimately the provision of public services, such as health and education. Our research design involves a panel survey of citizens, local councilors, and public service providers, pre and post-intervention implementation. This study has clear theoretical and policy implications. Theoretically, it will contribute to the growing micro-foundational body of work on political accountability in low-income countries with weak democratic institutions. Specifically it brings together and improves upon the largely disjointed literatures on political accountability and public service provision. From a policy perspective, the project will generate useful knowledge about the effectiveness of new innovations to improve the provision of education and health services to marginalized populations. More so, this study breaks new ground in the utilization of recent information technology innovations to affect politicians-constituents relations.
External Link(s)
Registration Citation
Garcia, Ana, Guy Grossman and Kristin Michelitch. 2014. "Information Barriers to Public Service Delivery: Innovations in Strengthening Local Political Accountability in Uganda." AEA RCT Registry. August 26.
Former Citation
Garcia, Ana et al. 2014. "Information Barriers to Public Service Delivery: Innovations in Strengthening Local Political Accountability in Uganda." AEA RCT Registry. August 26.
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
There are two interventions that are part of this project; disseminating politician performance information to citizens, and allowing citizens to communicate complaints to politicians via innovative text messaging software.

The intensive dissemination consist in community meetings held by ACODE, our partner organization, at parish level. They invite the community leaders of the community and explain them the roles of the councilors and their scores. In addition, the information on councilors' responsibilities and scores is shared with meeting attendees throughout the year via periodic text-messages.

The text message platform allows citizens to send messages to their councilors regarding problems or issues in their subcounties. The service is introduced through a series of community meetings. Reminders of the system are sent to meeting attendees throughout the year via text messages.

The RCT is implemented using a 2x2 factorial design.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
The research team is focusing on the following outcomes: (i) local councilors’ efforts with respect to public services, (ii) utility and quality of two key social services: health and education at the unit-level, and (iii) citizen turnout and vote-choice basis on politician performance versus clientelism or co-ethnicity (a long-term outcome).
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
We assigning constituencies to the different treatment conditions in a crosscutting research design. Operating in a total of 208 local councilor constituencies (“sub-counties”) in which ACODE issues scorecards, will yield four treatment groups. One quarter of the sub-counties will have the Intense LGCSC Dissemination with SMS communication service (Group A), one quarter will have the SMS communication service and Weak LGCSC Dissemination (Group C), one group will have solely the Intense LGCSC Dissemination (Group B), and one group will receive solely the Weak LGCSC Dissemination (Group D). We will compare change in key outcomes across these four treatment groups to establish the interventions' singular and joint effects.
Experimental Design Details
Randomization Method
Randomization was conducted using computer based software (Stata)
Randomization Unit
The units of randomization are the Special Women Councilor areas that are composed by two-three parishes.
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
218 special women councilors
Sample size: planned number of observations
471 councilors
Sample size (or number of clusters) by treatment arms
52 special women councilors in control, 55 in intensive dissemination, 52 in SMS platform and 52 in both treatments
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
0.25 standard deviations.
IRB Name
Innovations for Poverty Action IRB
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers