Age Sets and Accountability

How to cite

Lowes, Sara; Nunn, Nathan; Robinson, James; Montero, Eduardo. 2020. "Age Sets and Accountability." AEA RCT Registry. AEARCTR-0005370. https://doi.org/10.1257/rct.5370-1.6

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Sara Lowes* , Nathan Nunn , James Robinson , Eduardo Montero

* Primary investigator

Registry ID
AEARCTR-0005370
DOI
https://doi.org/10.1257/rct.5370-1.6
First registered
Last updated
Status
Completed
Trial dates
2016-08-01 – 2023-12-31
Countries
Congo (Democratic Republic of the) (Equateur)
JEL codes
O120, Z130, Z190
Registry record
http://www.socialscienceregistry.org/trials/5370

Abstract

This document provides a pre-analysis plan for a health intervention in the Northern region of the Democratic Republic of the Congo. Our study is interested in better understanding how variation in age sets practices -- initiation rituals for young men -- affect the accountability of local leaders. Anthropologists hypothesized that age sets increase accountability of leaders by creating cohesive groups of young men. We test this with an intervention providing village chiefs with grants to purchase health products, which are provided at subsidized rates and delivered to the village. As part of the project, village oversight committees are formed to monitor the chief's management of the project and its resources. The primary experimental manipulation is varying the composition of the oversight committees; half are composed of a random selection of volunteers and the other half over-represent young men. We examine how the committee composition affects chief performance and how this effect differs depending on whether age sets are present.

Experimental design

Chiefs in all villages receive funds (approximately $100) for the purchase of health products. Our partner will organize village oversight committees in each village to monitor the chief. Vendors will visit each village selling health products at a discounted price. The key experimental manipulation is that in some villages committees will be diverse, i.e. a random sample of those who volunteer. In other villages, the committee will comprise only young men. We examine how the committee composition affects chief performance and how this effect differs depending on whether age sets are present.

Design details

Randomization method
Randomization is done by a computer.
Randomization unit
Treatment is assigned at the village level.
Clusters
We will definitely implement in 100 villages. Budget allowing, we hope to have a final sample size of 300 villages.
Observations
Within each village, we intend to collect endline data from the chief, committee members, and a random sample of 14 to 16 villagers. For 100 villages this will be approximately 1400 observations.
Arms
Half of the sample will be randomized into a diverse committee; half of the sample will be randomized into a committee comprising only young men.
Primary outcomes
Our key outcome variables are the amount of missing project funds (i.e. funds that were not properly allocated to the purchase of health products) and the distribution of health products to village households.

Cited by

No citation data available for this registration yet.

Keywords