Replication of Village Savings and Loans Groups in Africa

How to cite

Ramey, Rachel. 2021. "Replication of Village Savings and Loans Groups in Africa." AEA RCT Registry. AEARCTR-0007421. https://doi.org/10.1257/rct.7421-1.0

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

Rachel Ramey*

* Primary investigator

Registry ID
AEARCTR-0007421
DOI
https://doi.org/10.1257/rct.7421-1.0
First registered
Last updated
Status
In development
Trial dates
2021-06-15 – 2022-12-15
Countries
Malawi (), Tanzania, United Republic of ()
Registry record
http://www.socialscienceregistry.org/trials/7421

Abstract

This pre-registry outlines pre-specified hypotheses for analyzing data collected from the study “Replication of Village Savings and Loans Groups in Africa.” The primary research question is to examine the factors that influence the quality and quantity of replication of Village Savings and Loans Associations (VSLA’s). We plan to test the implications of social and individual identity-based motivation training on replication quality and quantity Differences between social motivation and the control group as well as the difference between the individual motivation and control group will be tested. By studying replication and examining treatments designed to increase replication, NGO’s can have more accurate estimations of their impact and reach as well as learn how to help motivate groups to replicate thereby expanding their global impact. Data collection is scheduled to take place in Mchinji and Kasungu, Malawi from July 2021-December 2022 and is expected to be expanded into Tanzania.

Experimental design

Our research focuses on members of Village Savings and Loans Associations (VSLA’s) in Malawi and Tanzania. These are local community groups created to promote savings. We examine the long-term effects of motivational replication training on the replication rates and quality of VSLA groups in an emerging market setting.

We will use a train-the-trainer model. Village Agents (VA’s) are responsible for creating VSLA groups and train them with modules about savings practices and how to operate a VSLA. We will add an additional module to the training that VA’s give their VSLA’s. The new module focuses on teaching VSLA groups how to replicate and will include either social or individual identity motivation for the treatment groups. The control group will receive a placebo training module about an unrelated topic.

All VA’s for the region will be recruited and invited to a training session to be taught how to administer the new replication module as well as how to count replication. They will be randomly assigned to one of three treatment arms determining the type of training they will receive to give to their VSLA group. The three treatment arms are 1) social motivation, 2) individual motivation, and 3) placebo training.
The VA's will then take their modules to their VSLA's and train the VSLA's. 3rd party enumerators will collect data on manipulation checks and measures of intention to replicate at the VSLA replication training and 3-4 weeks after the training.

We will use messaging reinforcement of the concepts in the replication module to send to the VSLA group members throughout the year. When the VLSA groups share out (this will be on different schedules based on their cycle length and creation date, but we expect a large concentration of them to be at the end of the year (November 2021), some are expected to form their own replicated groups or expected to encourage their friends, family, and acquaintances to form new groups. We will count and measure the quality of these replicated groups to determine the effectiveness of social and individual motivation for self-replication.

Design details

Randomization method
For the VA training we randomly assign VA’s to one of the three treatment arms in office by a computer.
Randomization unit
VA Level
Clusters
100 VA’s are in the region and we will invite all of them to the additional training using the intention to treat method.
Observations
Observations are at the VSLA level. For data collection, replication will be counted for all VSLA groups (745+) and in depth surveys will be conducted on a random sample of 3 VSLA’s per VA for a total of 300 VSLA’s
Arms
There will be 33 VA’s per treatment arm, and the VSLA’s supervised by each VA will be the observations for each treatment arm. VA’s are expected, on average to supervise 10 VSLA’s each.
Primary outcomes
For the Replication Motivation Training Experiment, the primary outcomes relate to the quantity and quality of replicated VSLA groups created by each VSLA that received replication training.

Cited by

No citation data available for this registration yet.

Keywords