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.
Dressing for Success: Evaluating a Governance Training and Uniform Distribution Program in Malawi
Initial registration date
November 21, 2017
December 01, 2017 4:23 PM EST
University of Notre Dame
Other Primary Investigator(s)
University of Notre Dame
Additional Trial Information
This project explores the effect of a governance training and uniform distribution program on governance by Village Development Councils (VDC) in rural Malawi. The project leverages a randomized control design to evaluate to interventions, a governance training and a uniform distribution program to committee members, to understand how the two treatments affect a) general governance outcomes b) VDC members' efficacy and job performance, and c) villagers' perceptions of VDC members. Working in partnership with Catholic Relief Services, the PIs will conduct a baseline survey in all Group Village Heads (administrative unit for GVHs) and then an endline in communities, which have received treatment 1 (governance training), treatment 2 (governance training and uniform distribution), and control villages (no training or distribution in the period of study). The project conducts surveys with villagers, VDC members, other committee members, and GVH heads to evaluate the outcome variables of interest. In addition, we will conduct functionality assessments on the performance of VDCs.
Baul, Tushi and Jaimie Bleck. 2017. "Dressing for Success: Evaluating a Governance Training and Uniform Distribution Program in Malawi." AEA RCT Registry. December 01.
In partnership with Catholic Relief Services, we use a phased out, stratified, clustered randomized design to assign village development committees in Malawi to receive the following treatments:
Treatment 1: 1 day governance training run by CRS
Treatment 2: uniform (hat, badge, shirt) distribution to VDC members plus the 1 day governance training
Control: no treatments in the period of study; slated to receive trainings after the endline in November 2017.
We assess the impact of these interventions on governance outcomes in these communities.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
1. Governance: this includes a VDCs performance (as perceived by villagers and the Group Village Head), VDCs internal procedures (as collected in functionality assessment), and VDC performance (based on reporting of social service provision and community coordination)
2. VDC Members' Efficacy and Performance: VDC members' assessment of their own efficacy and competence, VDC members' commitment to their job, and VDC members' job knowledge and performance
3. Community Perceptions of VDC and VDC Members: Villagers assessments of VDC member performance and desirable valence characteristics (eg competence, confidence, trustworthiness), GVH heads (traditional leaders)' assessments of the VDC members
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
This intervention is designed to study the effectiveness of Catholic relief service’s governance training program (Treatment 1) on functioning of village development committees (VDCs) in Malawi. A second intervention (Treatment 2) will comple- ment the trainings with a distribution of uniforms to VDC members. The research teams assess the effect of Treatment 1 and Treatment 2 on a) VDC governance out- comes; b) members’ assessment of themselves and their performance; c) villagers’ perceptions of the VDC members and their performance.
This idea for the second treatment came directly from focus groups with village committee members during the needs assessment phase, in which the research team asked about things that would help with their job performance. Village Committee Members had cited as uniforms as items that could help them to better do their job. This prompted the research team to consider literature from political science that suggests that ”professional dress” could influence members’ self-assessments and constituents preferences.
In partnership with CRS, the research team used a list of 207 administrative units (GVHs) to implement a phased out stratified cluster randomized design. It assigned all GVHs participating in the CRS program into a Treatment 1, Treat- ment 2 and Control Groups. VDC members in Treatment 1 received the 1 day CRS governance training, while VDC members in Treatment 2 receive the training as well as a uniform (hat, badge, and shirt). VDC members in control villages do not receive any treatments before the endline is conducted (end of year 2017). The research team will use a series of longitudinal surveys (collected before and after the interventions) that target the following populations: VDC members, Other Com- mittee Members, villagers, and traditional leaders in the communities. In addition, the team will conduct a functionality assessment of VDC organization and perfor- mance during the endline. Finally, the endline will include an embedded survey experiment, in which villagers are randomly assigned to images of VDC members (male/female/uniform/no uniform) to further test the mechanisms that could drive Treatment 2.
Experimental Design Details
We will use a phase-out design to evaluate the UBALE program. We will use a stratified design and divide GVHs in three district randomly into three groups: Treatment 1 : GVHs with training and no uniform (T NU) Treatment 2: GVH with training and uniform Control : GVH with no training and no uniform Treatment 1 and Treatment 2 will receive CRS governance training in 2016-2017 whereas the control group will receive CRS training module in 2018. We obtained the list of GVHs in this study from CRS and randomized the list into 3 groups using a STATA randomizer.
We conduct the baseline survey in May-June 2016 and, after the treatment is implemented in year 2016-2017, we will conduct the endline survey in November 2017. Within each GVH we will survey all the VDC members, a random sample of the other village committee members, a random sample of villagers, and the GVH head (a traditional leader ). In 2018, we will conduct another around of survey to whether there is a significant long term effect of the intervention.
Group Village Head (GVH) - administrative unit that is comprised of a group of villages
Was the treatment clustered?
Sample size: planned number of clusters
Sample size: planned number of observations
1600 VDC Members
1890 Other Committee Members
187 GVH Heads (traditional leader)
207 Functionality Assessments
Sample size (or number of clusters) by treatment arms
Treatment 1 (only governance training): 69 GVHs
Treatment 2 (governance training and uniform distribution): 69 GVHs
Control (no treatment in period of study): 69 GVHs
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We have conducted Power calculations using Optimal Design as fixed effect multi- site (blocked) clustered analysis to determine our sample size. The assumptions for the power calculations are as follows: Significance level (α) of .05; Number of blocks (K) of 3; Number of clusters (J) of 75; Standardized effect size () of .2; Intra-class correlations () of .01 and .08; Proportion of the explained variance by the blocking variable (B) of .2 or 0; R-squared of a covariate ((RL2)2) of 0 or .1 ; Fixed Effects or effect size variability (2) of 0; Expected attrition - 20
INSTITUTIONAL REVIEW BOARDS (IRBs)
University of Notre Dame
IRB Approval Date
IRB Approval Number
Post Trial Information
Is the intervention completed?
Is data collection complete?