Adoption and Impact of Low Cost Solar Energy: A Randomized Field Experiment in Rural Kenya

Last registered on March 10, 2016

Pre-Trial

Trial Information

General Information

Title
Adoption and Impact of Low Cost Solar Energy: A Randomized Field Experiment in Rural Kenya
RCT ID
AEARCTR-0000912
Initial registration date
March 10, 2016

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
March 10, 2016, 5:07 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
ETH Zurich

Other Primary Investigator(s)

Additional Trial Information

Status
On going
Start date
2015-06-11
End date
2016-04-01
Secondary IDs
Abstract
Access to electricity is considered both an outcome and a driver of development. Most households in rural sub-Saharan Africa and over 90% of households in rural Kenya are not connected to the grid. These households typically rely on kerosene lanterns with high operational costs and low-quality light. Prices for solar power have fallen dramatically, raising the possibility that solar energy could provide an alternative to kerosene, especially for lightning needs. This may allow households to save on energy expenditure, increase the time household members can spend on productive activities, and improve safety, as well as health and schooling outcomes. Solar lights have therefore attracted the attention of investors, entrepreneurs, donors, NGOs, and governments. Still, solar lights only provide minimal access to energy and despite the high hopes for this technology, there is little empirical evidence of its cost-effectiveness, its impact on household welfare, or its potential effect on the environment. Even less is known about what drives adoption of solar energy and solar lightning, in particular. The proposed randomized field experiment aims to better understand current demand for solar lights and constraints to adoption, as well as the lanterns’ impact on rural households, focusing on their effect on energy expenditures, light use, time use, as well as health, education, safety, and happiness.
External Link(s)

Registration Citation

Citation
Rom, Adina. 2016. "Adoption and Impact of Low Cost Solar Energy: A Randomized Field Experiment in Rural Kenya." AEA RCT Registry. March 10. https://doi.org/10.1257/rct.912-1.0
Former Citation
Rom, Adina. 2016. "Adoption and Impact of Low Cost Solar Energy: A Randomized Field Experiment in Rural Kenya." AEA RCT Registry. March 10. https://www.socialscienceregistry.org/trials/912/history/7213
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
(1) Pure control group
(2) Opportunity to buy at market price
(3) Free solar product, half of which can only be used for lighting and half of which can be used for both lighting and mobile phone charging
(4) Solar product offered at a discounted price

During a follow-up experiment looking at adoption in particular, households from group (1) and (4) are randomly assigned to one of the following treatments:
(5) Opportunity to buy solar product at market price
(6) Information about solar product and opportunity to buy at market price
Intervention Start Date
2015-06-11
Intervention End Date
2016-04-01

Primary Outcomes

Primary Outcomes (end points)
A detailed table with the outcome of interest and how they are measured can be found in the pre-analysis plan attached to this document.
The survey instruments were built on standardized questions or survey, which have already been used in a similar setting, where possible. Standardized questions include:
- Questions from the World Values Survey are used to measure happiness and the CES-D SCALE is used to measure depression.
- For the health questions, an adapted version of the European Community Respiratory Health Survey will be used.

In addition to survey data we also collect the following non-survey data:
- Educational outcomes are measured by total study time, as well as children's test scores.
- Sensors are installed on a subsample of households measuring the use of solar and kerosene lights (more details in next section).
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
In this study we sample pupils from 20 schools in Nambale and Teso-South, two sub-counties in Busia, Western Kenya. In each of these 20 schools, 70 pupils from classes (grades) five, six, and seven are randomly selected and assigned to one of the following treatments:
(1) Pure control group
(2) Opportunity to buy at market price
(3) Free solar product, half of which can only be used for lighting and half of which can be used for both lighting and mobile phone charging
(4) Solar product offered at a discounted price

During a follow-up experiment looking at adoption in particular, households from group (1) and (3) are assigned to one of the following treatments:
(5) Opportunity to buy solar product at market price
(6) Information about solar product and opportunity to buy at market price

A subsample of 200 households in group (3) received a solar light with an electronic sensor that measures light use. In addition, those purchasing solar lights at the market price (subset of 2) and the smallest subsidy (subset of 4) received a sensor, too. The sensors record the time the solar light was used for up to 6 months. The research team installed kerosene sensors in 13 households in each school from group (1) and (4) to measure the use of the kerosene lights. All participants will be asked for consent before the data from these sensors is retrieved.

Data will be collected at baseline, as well as 3 months (midline), and 6 months (endline) after baseline.
In addition, since the randomization was stratified at the school level, the random variation in the number of households who received a free solar light at the village level will be exploited to see if variation in the share of solar light owners in a given household’s social network will have an impact on adoption.
Experimental Design Details
Randomization Method
Randomization is done at the office by a computer.
Randomization Unit
The randomization is stratified at the school level. The unit of randomization is the individual.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
No clusters are planned.
Sample size: planned number of observations
1,400 households. We will interview 1,400 pupils as well as 1,400 guardians.
Sample size (or number of clusters) by treatment arms
(1) Pure control group (N=400)
(2) Opportunity to buy at market price (N=200)
(3) Free solar product, half of which can only be used for lighting and half of which can be used for both lighting and mobile phone charging (N=400)
(4) Solar product offered at a discounted price (N=400)

During a follow-up experiment looking at adoption in particular, households from group (1) and (3) are assigned to one of the following treatments:
(5) Opportunity to buy solar product at market price (N=400)
(6) Information about solar product and opportunity to buy at market price (N=400)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Maseno University Ethics Review Committee
IRB Approval Date
2015-05-08
IRB Approval Number
MSU/DRPI/MUERC/00166/15
IRB Name
Ethics Commission of ETH Zurich
IRB Approval Date
2015-02-17
IRB Approval Number
EK 2015-N-08
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials