Download StudyGeneral

Study Overview

Title:
Assessing the Downstream Socioeconomic and Land Health Impacts of Agroforestry in Kenya
Study ID:
RIDIE-STUDY-ID-58f6ee8c9ddab
Initial Registration Date:
04/18/2017
Last Update Date:
11/30/2016
Study Status:
Ongoing
Location(s):
Kenya
Abstract:
Agroforestry is widely purported to improve the livelihoods of smallholder farmers, rehabilitate degraded landscapes, and enhance the provisioning of ecosystem services, including carbon sequestration. Yet the evidence base supporting these claims is weak. Using a quasi-experimental evaluation design based on a theory-based and mixed methods framework, this study investigates the longer-term and intermediate effects of an eight year effort led by Vi Agroforestry, a Swedish non-governmental organization (NGO), to promote agroforestry in large sections of Bungoma and Kakamega counties in western Kenya. Key longer-term outcomes of interest include consumption expenditure, asset accumulation, food security, shock resilience, soil organic carbon and erosion prevalence, and educational attainment.
Categories:
Agriculture and Rural Development
Environment and Disaster Management
Health, Nutrition, and Population
Additional Keywords:
Secondary ID Number(s):

Principal Investigator(s)

Name of First PI:
Karl Hughes
Affiliation:
The World Agroforestry Centre (ICRAF)
Name of Second PI:
Katherine Baylis
Affiliation:
The University of Illinois at Urbana-Champaign (UIUC)

Study Sponsor

Name:
The Standing Panel on Impact Assessment (SPIA) c/o The Food and Agricultural Organization of the United Nations (FAO)
Study Sponsor Location:
Italy
Funding Proposal:

Research Partner

Name of Partner Institution:
The University of Illinois at Urbana-Champaign, Department of Agricultural & Consumer Economics
Type of Organization:
Research institute/University
Website:
ace.illinois.edu
Location:
United States
Intervention

Intervention Overview

Intervention:
Vi promoted several interrelated, complementary agroforestry practices, including trees for (a) domestically consumable and marketable products, e.g. timber, fuel wood, and fruit; (b) soil health improvement practices, e.g. soil fertility enhancement, erosion control and increased water infiltration; and (c) livestock fodder. These promotional efforts were undertaken by targeting pre-existing farmer groups and providing their members with tree seeds, support to establish tree nurseries, and training and extension support on how to integrate trees with field crops on their farms. The implementation of Vi’s interventions in the program area started in 2008 through two different projects: the Kenya Agricultural Carbon Project (KACP) and the Farmer Organizations and Agroforestry (FOA) project. The KACP project focuses on increasing carbon sequestration in small-holder systems. The FOA project focuses on capacity building for farmer organizations. Vi staff describe the extension training and related support provided as being similar, but with two key differences: FOA does not provide carbon payments and does not monitor tree planting with the same rigor.
Theory of Change:
Multiple Treatment Arms Evaluated?
No

Implementing Agency

Name of Organization:
Vi Agroforestry
Type of Organization:
NGO (International)

Program Funder

Name of Organization:
World Bank BioCarbon Fund provides carbon credit funding for the KACP project. FOA funded by the Swedish International Development Cooperation Agency and private donors.
Type of Organization:
Foreign or Multilateral Aid Agency

Intervention Timing

Intervention or Program Started at time of Registration?
Yes
Start Date:
11/30/2016
End Date:
Evaluation Method

Evaluation Method Overview

Primary (or First) Evaluation Method:
Difference in difference/fixed effects
Other (not Listed) Method:
Additional Evaluation Method (If Any):
Matching
Other (not Listed) Method:

Method Details

Details of Evaluation Approach:
The causal effects of Vi’s program will be estimated by comparing outcomes between two sets of households: those belonging to (a) 226 targeted and pre-existing farmer groups operating in 60 villages; and (b) 61 non-targeted pre-existing farmer groups operating in 60 villages located outside the program area but within the same two Kenyan counties—Bungoma and Kakamega. Given the study’s non-experimental nature, and the fact that an appropriate baseline survey was never undertaken, several measures have and will be undertaken to counter both program placement and self-selection bias, as well as other internal validity threats. These are as follows: 1. Village matching based on selected geophysical and demographic variables 2. Sampling from all farmer groups that existed in both the intervention and comparison villages at baseline. 3. Reconstruction of baseline using recall data for difference-in-differences estimation 4. Doubly robust estimation and other appropriate econometric techniques to control for other observable differences between the two groups. 5. Use of Intention-to-treat (ITT) and local average treatment effect (LATE) estimation.
Outcomes (Endpoints):
We will measure various exposure, adoption, intermediate and final outcome variables: Exposure Variables: 1. Agroforestry group participation 2. Receipt and Implementation of agroforestry training 3. Receipt and Implementation of agroforestry extension 4. Other agricultural support Uptake of Promoted Practices: 1. Agroforestry Practice Index 2. Differenced fractional vegetative cover Intermediary Outcome Variables: 1. Cash earned from sale of agroforestry products 2. Estimated cash value of firewood harvested from farm 3. Estimated hours collecting firewood per month 4. Average % change in milk yields among dairy producers 5. Self-reported increase in income from dairy production 6. Estimated soil erosion prevalence and soil organic carbon Primary Outcome Variable: Differenced predicted consumption expenditure Other Outcome Variables: 1. Differenced asset wealth 2. Single-difference consumption expenditure 3. Adapted Coping Strategies Index 4. Minimum Dietary Diversity – Women (MDD-W) 5. Months of Adequate Food Provisioning 6. Education Progression 7. Education Spending 8. Perceived Change on Economic Ladder
Measurement:
Differenced Predicted Consumption Expenditure: Consumption expenditure data were collected during the survey using a one week recall food consumption module; a four week regular non-food spending module; and one year non-regular occasional spending module. These data will be used to compute the daily per adult equivalent consumption expenditure, adjusted for purchase power parity (PPP). Data on over 80 assets and other household wealth indicators were also collected during the household survey for both the endline and baseline periods. Stepwise regression will be used on the 2016 variables to determine the set that best predict household consumption expenditure. We will check to ensure that the resulting predictions are sufficiently correlated (r > 0.60) with the actual expenditure. We will then assign weights to each shortlisted asset/wealth indicator based on their respective coefficients. These weights will also be applied to the asset/wealth indicators for the baseline period. We will then difference the two to create the differenced predicted consumption expenditure measure. Details for other outcomes can be found in the attached pre-analysis plan document.
Unit of Analysis:
Household
Hypotheses:
Our primary hypothesis is that a participation in Vi’s program led to increased household consumption expenditure and asset accumulation on average, albeit with considerable variation among the participating households. Given that data were obtained from households belonging to non-Vi farmers from the program villages, we also hypothesis that the LATE estimates on consumption expenditure and asset accumulation will be greater than ITT effect estimates. However, the differences will not be that significant, given that these households represent only about one-quarter of those making up the program village sample. Details for secondary hypotheses can be found in the attached pre-analysis plan document.
Unit of Intervention or Assignment:
Farmer group
Number of Clusters in Sample:
432 farmer groups
Number of Individuals in Sample:
2860
Size of Treatment, Control, or Comparison Subsamples:
Data has already been collected from 432 farmer groups operating in 121 villages (making up 1,450 and 1,410 households in the program and non-program villages, respectively).

Supplementary Files

Analysis Plan:
pre_analysis_plan_SPIA_2016.pdf
Other Documents:
Data

Outcomes Data

Description:
Household survey administered using smart phones with Open Data Kit (ADK) software. GPS coordinates were also captured for each plot associated with the interviewed household’s main farming parcel. Satellite imagery was then analyzed to estimate various soil health parameters, including the three outcome measures presented above.
Data Already Collected?
Yes
Survey Name:
SPIA_Vi_AF_2016
Data Previously Used?
No
Data Access:
Not restricted - access with no requirements or minimal requirements (e.g. web registration)
Data Obtained by the Study Researchers?
Yes
Data Approval Process:
Approval Status:

Treatment Assignment Data

Participation or Assignment Information:
Yes
Description:
Data Obtained by the Study Researchers?
Data Previously Used?
Data Access:
Data Obtained by the Study Researchers?
Data Approval Process:
Approval Status:

Data Analysis

Data Analysis Status:
No

Study Materials

Upload Study Materials:

Registration Category

Registration Category:
Prospective, Category 3: Data for measuring impacts have been obtained/collected by the research team but analysis for this evaluation has not started
Completion

Completion Overview

Intervention Completion Date:
Data Collection Completion Date:
Unit of Analysis:
Clusters in Final Sample:
Total Observations in Final Sample:
Size of Treatment, Control, or Comparison Subsamples:

Findings

Preliminary Report:
Preliminary Report URL:
Summary of Findings:
Paper:
Paper Summary:
Paper Citation:

Data Availability

Data Availability (Primary Data):
Date of Data Availability:
Data URL or Contact:
Access procedure:

Other Materials

Survey:
Survey Instrument Links or Contact:
Program Files:
Program Files Links or Contact:
External Link:
External Link Description:
Description of Changes:

Study Stopped

Date:
Reason: