Social networks, incentives, and diffusion of house screening and push-pull technology interventions in rural Ethiopia

How to cite

Kassie, Menale; Asale, Abebe; Abro, Zewdu Abro; Balew, Solomon; Bulte, Erwin. 2020. "Social networks, incentives, and diffusion of house screening and push-pull technology interventions in rural Ethiopia." AEA RCT Registry. AEARCTR-0005642. https://doi.org/10.1257/rct.5642-1.0

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

Menale Kassie* , Abebe Asale , Zewdu Abro Abro , Solomon Balew , Erwin Bulte

* Primary investigator

Registry ID
AEARCTR-0005642
DOI
https://doi.org/10.1257/rct.5642-1.0
First registered
Last updated
Status
On going
Trial dates
2019-07-31 – 2022-07-31
Countries
Ethiopia ()
Registry record
http://www.socialscienceregistry.org/trials/5642

Abstract

In sub-Saharan Africa, smallholder agriculture transformation is constrained by multifaceted risks that transcend humans, livestock, and crops. Malaria is one of the top diseases with a high economic burden to farmers and governments. Invasive pests such as fall armyworm and stem borer reduce crop production and increase the cost of production considerably. Furthermore, shortage and/or unavailability of quality fodder hinder livestock production and draught power performance of animals. While uptake of available technologies remains a key challenge to policymakers and researchers, interventions addressing these challenges are often done in a piecemeal fashion. In this impact evaluation, we aim to study the role of social networks, incentives, and nudges in the diffusion of technologies that simultaneously reduce the burden of malaria and low productivity of crops and livestock. We use incentivized and nudged social networks to deliver extension information on the adoption of push-pull technology (PPT) for pest control and quality fodder production, and house screening (HS) for malaria control. The social networks we are studying are one-to-five groups, which are massively being institutionalized in the local government structure of Ethiopia. As an extension approach, the one–to–five groups can help disseminating information about new technologies and best practices. Even though the one–to–five group extension approach seems promising to transform the agriculture sector through the diffusion of new technologies and best practices, many accuse the government that it uses the one–to–five groups to control public life and mobilize support to the ruling party with little impact on technology diffusion. Evaluating the effectiveness of this unique social network informs policymakers to tailor policies towards extension approaches that deliver trusted information, which can enhance the adoption of new technologies. The findings also provide lessons to the rest of sub-Saharan Africa where the extension systems follow a top-down approach of reaching farmers rather than being participatory. Our study fits to the emerging literature on the role of incentives and nudges on various development outcomes in low-income countries.

Intervention

The focus of the impact evaluation is the Jabitehnan district of the Amhara Regional State of Ethiopia. Formally, the district is divided into 38 villages. The villages are also subdivided into sub-villages. For the sake of this impact evaluation, we use the sub-villages as the lowest subdivisions. The district has 113 sub-villages. We randomly assigned the sub-villages into five treatment arms and the control group. These are:
• Control
• HS (gain frame)
• HS (loss frame)
• PPT trainings only
• HS + PPT (loss frame for the PPT)
• HS + PPT (gain frame for the PPT)

Experimental design

The study focuses on 1052 chairpersons and their members (3240) in 113 sub-villages. Of the total 113 sub-villages, 86 sub-villages are in the treatment while the rest 27 sub-villages are controls. Each treatment arm has the following sample size.
Arms Sample size
Control 195
HS (gain frame) 125
HS (loss frame) 125
PPT trainings only 203
HS + PPT (loss frame for the PPT) 210
HS + PPT (gain frame for the PPT) 194
Total 1052

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
First, treatment was assigned at the sub-village level. Within control and treatment sub-villages, all households were assigned control and treatment, respectively.
Clusters
113 sub-villages
Observations
1052 chairpersons and 3240 their members
Arms
Arms Sample size Control 195 HS (gain frame) 125 HS (loss frame) 125 PPT trainings only 203 HS + PPT (loss frame for the PPT) 210
Primary outcomes
a) PPT knowledge score b) PPT and HS adoption Productivity of maize (kg/ha) c) Milk productivity (kg/animal) d) Cost of illness (USD/household) e) Malaria prevalence (%) f) Lost working days due to malaria g) Lost school days due to malaria for children
Secondary outcomes
a) Willingness to pay for PPT and HS b) Women and children Dietary diversity score c) Household Food Insecurity Access Scale (HFIAS

Cited by

No citation data available for this registration yet.

Keywords