Citizens at the Council: A Field Experiment on Citizen Observers in Burkina Faso (Phase I)
Last registered on December 21, 2016

Pre-Trial

Trial Information
General Information
Title
Citizens at the Council: A Field Experiment on Citizen Observers in Burkina Faso (Phase I)
RCT ID
AEARCTR-0001283
Initial registration date
May 20, 2016
Last updated
December 21, 2016 9:31 AM EST
Location(s)

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

Request Information
Primary Investigator
Affiliation
Yale University
Other Primary Investigator(s)
PI Affiliation
World Bank
Additional Trial Information
Status
On going
Start date
2015-06-05
End date
2018-12-31
Secondary IDs
Abstract
We investigate (1) how the presence of ordinary citizens at municipal council meetings affects the behavior of municipal decision makers, and (2) how personalized invitations to attend a municipal council meeting as a "citizen observer" influence ordinary citizens' voluntary civic participation, as well as their attitudes towards municipal governance. In a first phase, from November 2015 through May 2016, the experiment is carried out on centrally appointed local councils (special delegations), which were established in the aftermath of the popular insurrection in 2014. In a second phase, following the 2016 municipal elections, the experiment will be continued on the newly elected municipal councils that will replace the special delegations.

A pre-analysis plan will be attached to this design registration prior to the investigators' access to outcome data.
External Link(s)
Registration Citation
Citation
Holmlund, Marcus and Malte Lierl. 2016. "Citizens at the Council: A Field Experiment on Citizen Observers in Burkina Faso (Phase I)." AEA RCT Registry. December 21. https://www.socialscienceregistry.org/trials/1283/history/12752
Experimental Details
Interventions
Intervention(s)
The experimental intervention consists of issuing personal invitations to 40 randomly selected citizens from typically five randomly selected villages within a municipality to attend a municipal council meeting. To implement this intervention, we partner with the Programme d'appui aux collectivit├ęs territoriales (PACT) in Burkina Faso, six regional NGOs that act as implementing partners for the PACT, and the mayors of the municipalities (in Phase I of the experiment: the presidents of the special delegations).

The targeted citizens receive personal invitation letters from the mayor/president of the special delegation, encouraging them to attend the upcoming municipal council/special delegation meeting as "citizen observers" and to share their views at a townhall meeting that will take place in conjunction with the deliberations of the council/special delegation. The invitation letters, signed by the mayor/president of the special delegation, are hand-delivered to each selected citizen. As "citizen observers", the invited citizens are formally welcomed by the mayor/president of the special delegation and are invited to attend the council session and, where applicable, a lunch break. Additionally, they are invited to attend a townhall meeting during which they will be asked to share and discuss their points of view and those of their co-villagers with the delegates.

In Phase I of the experiment, until the May 2016 municipal elections, the targeted citizens are invited to meetings of the externally appointed special delegations. Until the election of new municipal governments in 2016, these special delegations serve the same purpose and operate under the same legal framework as the elected municipal councils. However, the members of the special delegations are externally appointed and are not allowed to be candidates in the upcoming municipal elections. By law, the special delegations include representatives of the regional administration, of the municipal service providers, of women, youth and handicapped citizens in the municipality, as well as of local civil society organizations. Elected municipal councils will be reconstituted after the municipal elections in May 2016.

In Phase I, we implement the citizen observer intervention at one regular meeting of the externally appointed special delegation per municipality, either in Q4 of 2015 or in Q1 of 2016. In Phase II, we implement the citizen observer intervention at one regular meeting of the elected municipal council per municipality (i.e. a meeting after the council's constitutive session), either in Q3 or Q4 of 2016.

Intervention Start Date
2015-11-03
Intervention End Date
2016-12-31
Primary Outcomes
Primary Outcomes (end points)
*Outcomes of interest in Phase I* CITIZEN LEVEL: (1) Voluntary civic participation (2) Intermediate outcomes: See pre-analysis plan. MUNICIPAL LEVEL: (1) Delegate attendance: Proportion of delegates in a municipality attending the meeting. (2) Representation of village-specific concerns in the meeting: Difference between the number of delegate statements addressing a village-specific concern of the treatment/sampled villages and the mean number of delegate statements addressing a village-specific concern in the non-treatment/non-sampled villages of the municipality. (3) Scrutiny of the municipal administration: Proportion of delegate statements during the meeting that are critical of the municipal administration.
Primary Outcomes (explanation)
We construct an index of six indicators of voluntary civic participation: (1) intent to vote in the municipal elections, (2) interest in attending future municipal council meetings, (3) attendance of village meetings, (4) communicating with others about municipal affairs, (5) confidentially stated interest in running for local office, (6) voluntary donation to the municipal treasury. Details are provided in the pre-analysis plan.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The experiment is carried out in all of the 118 rural municipalities that are located within six regions of Burkina Faso: Cascades, Centre Nord, Centre Est, Centre Sud, Plateau Central, and Sahel. Of these 118 municipalities, 58 are assigned to treatment (invitation of citizen observers to municipal council meetings), 60 are assigned to control (blocked by region). The 58 treatment municipalities are the same for Phase I and Phase II of the experiment.

Phase I (November 2015 - May 2016):
(a) Treatment: Personalized invitation letters to 40 randomly sampled citizens per municipality to attend a regular meeting of the externally appointed special delegation of their municipality.
(b) Treatment assignment: The treatment is randomized at three levels: 1. municipalities (block randomized by region), 2. villages within municipalities, 3. citizens within villages.

Midline survey (May 2016): In three fifths of the treatment villages within 44 of the 58 treatment municipalities, the midline survey targets all citizens who were assigned to the citizen observer invitation, as well as a random sample of their co-villagers. Additionally, the midline survey targets a random sample of citizens in one control village per each of the 44 municipalities.

Phase II (July 2016 - December 2016):
(a) Treatment: Personalized invitations to 40 randomly sampled citizens per municipality to attend a meeting of the elected municipal council (a regular meeting, after the initial constitutive session).

Endline survey (Date to be determined)
Experimental Design Details
Not available
Randomization Method
Randomization was carried out by the PI, using an R script. To ensure compliance with the random assignment of the citizen observer invitations, the research team automatically pre-filled invitation letters with the names of the sampled individuals and handed over hard copies of the invitation letters to the implementation partners.
Randomization Unit
The treatment is randomized at three levels: 1. municipalities (block randomized by region), 2. villages within municipalities, 3. citizens within villages.
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
The treatment is randomized at the individual, village and municipal levels. At the individual level, the midline data collection at the end of Phase I will cover 2070 citizens within 44 treatment municipalities, 1029 of whom were assigned to treatment and 1041 to control. At the municipal level, 118 special delegation meetings will be observed in Phase I, covering all of the 58 treatment municipalities and all of the 60 control municipalities.
Sample size: planned number of observations
see above
Sample size (or number of clusters) by treatment arms
Number of citizens invited to council meetings:
Phase I: 2294 citizens in 58 municipalities
Phase II: 2294 citizens in 58 municipalities
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
See analysis plan.
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Innovations for Poverty Action (IPA) IRB
IRB Approval Date
2016-03-12
IRB Approval Number
14159
IRB Name
Yale University Human Subjects Committee
IRB Approval Date
2016-04-06
IRB Approval Number
1601017005
IRB Name
Yale University Human Subjects Committee
IRB Approval Date
2014-07-06
IRB Approval Number
1405013896
Analysis Plan
Analysis Plan Documents
Analysis plan (Phase 1, results-blind analysis)

MD5: 13612839b6b30fc0309cae33aaacd2e6

SHA1: a22a46a24c18ea2c27f99ce0f6cea3eb4d034143

Uploaded At: December 21, 2016