Land Husbandry, Water Harvesting and Hillside Irrigation Project Impact Evaluation 2012-2016
Rwanda, 2012 - 2016
Florence Kondylis, Maria Jones, Saahil Karpe
Identification
IDNO
RWA_2012-2016_LWHIE-BL-F2_v01_M
Title
Land Husbandry, Water Harvesting and Hillside Irrigation Project Impact Evaluation 2012-2016
Subtitle
Baseline, First Follow-up and Second Follow-up Surveys
Countries
| Name | Code |
|---|---|
| Rwanda | RWA |
Study notes
Agriculture is a major engine of the Rwandan economy and remains a priority sector in the Government of Rwanda's goals of reducing poverty and achieving food security through commercialized agriculture. Sustainable improvements to agricultural productivity is the only way to achieve this target, calling for investments in participatory land management, water harvesting and intensified irrigation of the hillsides. The Land Husbandry, Water Harvesting and Hillside Irrigation (LWH) project has been working to meet these goals.
Evaluating the overall impact of LWH is important to allow Ministry of Agriculture and Animal Resources (MINAGRIs) to effectively plan for its future activities. LWH covers a relatively small area of 30,250 ha, eventually affecting approximately 20 watersheds. From the perspective of MINAGRI, LWH can be seen as a pilot program for comprehensive agricultural overhaul. LWH includes major infrastructure investments such as hillside terracing, irrigation dams, and post-harvest storage. The project aims to operationalize MINAGRIs strategy to encourage mono cropping of cash crops, as opposed to the traditional system of inter-cropping for household consumption.
LWH has been rolled out in three phases: implementation in the four Phase 1A sites began in 2010, in the three Phase 1B sites in 2012, and in the Phase 1C sites in late 2013.
There have been five surveys for this project including baselines and four follow-ups. There were two sample groups: 1B and 1C. For the 1B sample group, the baseline was conducted in 2012, and follow-up surveys in 2013, 2014, and 2016. For the 1C sample, the baseline was implemented in 2014, and a follow-up in 2016.
The data files documented here are from the baseline, the first follow-up, and the second follow-up surveys.
The datasets from the third and the fourth follow-up surveys are also published in the Microdata Library with the following study IDs:
1) The third follow-up: RWA_2014_LWHIE-F3_v01_M
2) The fourth follow-up: RWA_2016_LWHIE-F4_v01_M
Evaluating the overall impact of LWH is important to allow Ministry of Agriculture and Animal Resources (MINAGRIs) to effectively plan for its future activities. LWH covers a relatively small area of 30,250 ha, eventually affecting approximately 20 watersheds. From the perspective of MINAGRI, LWH can be seen as a pilot program for comprehensive agricultural overhaul. LWH includes major infrastructure investments such as hillside terracing, irrigation dams, and post-harvest storage. The project aims to operationalize MINAGRIs strategy to encourage mono cropping of cash crops, as opposed to the traditional system of inter-cropping for household consumption.
LWH has been rolled out in three phases: implementation in the four Phase 1A sites began in 2010, in the three Phase 1B sites in 2012, and in the Phase 1C sites in late 2013.
There have been five surveys for this project including baselines and four follow-ups. There were two sample groups: 1B and 1C. For the 1B sample group, the baseline was conducted in 2012, and follow-up surveys in 2013, 2014, and 2016. For the 1C sample, the baseline was implemented in 2014, and a follow-up in 2016.
The data files documented here are from the baseline, the first follow-up, and the second follow-up surveys.
The datasets from the third and the fourth follow-up surveys are also published in the Microdata Library with the following study IDs:
1) The third follow-up: RWA_2014_LWHIE-F3_v01_M
2) The fourth follow-up: RWA_2016_LWHIE-F4_v01_M
Kind of data
Sample survey data [ssd]
Version
Version
v01
The data files are from the baseline, the first follow-up, and the second follow-up surveys.
The datasets from the third and the fourth follow-up surveys are also published in the Microdata Library with the following study IDs:
1) The third follow-up: RWA_2014_LWHIE-F3_v01_M
2) The fourth follow-up: RWA_2016_LWHIE-F4_v01_M
The data files are from the baseline, the first follow-up, and the second follow-up surveys.
The datasets from the third and the fourth follow-up surveys are also published in the Microdata Library with the following study IDs:
1) The third follow-up: RWA_2014_LWHIE-F3_v01_M
2) The fourth follow-up: RWA_2016_LWHIE-F4_v01_M
Coverage
Geographic coverage
The study covers six sites the Eastern and North Eastern regions of the country: Nyamuziga, Gicumbi, Muyanza, Cyonyonyo, Rwamagana and Kayonza.
Unit of analysis
Household
Producers and sponsors
Authoring entity
| Name | Affiliation |
|---|---|
| Florence Kondylis | The World Bank |
| Maria Jones | The World Bank |
| Saahil Karpe | The World Bank |
Funding agencies
| Name | Abbreviation |
|---|---|
| Global Agriculture Food Security Program | GAFSP |
Sampling
Sampling procedure
There were two baseline surveys, one for the 1B sample group and another for the 1C sample group.
1B sample:
A baseline survey was implemented in three 1B LWH project sites. Rwamagana-34, Rwamagana-35 and Kayonza-4, and three control sites were selected by pairwise-matching. The sample was designed to meet three criteria: geographic representation at the site-level, inclusion of multiple members of farmer groups within sites, and sufficient size to power tests of variations in treatment within the Phase 1B sites.
To meet these criteria, sampling was done through a two-stage process. In the first stage, researchers randomly sampled "seed" households, stratified by village. The total number of "seed" households varied by treatment status. For treatment sites, the number of "seeds" depended on the estimated number of Self-help Group (SHG), calculated based on the site population and the typical size of SHGs, which is 20 households. For control sites, the number of "seeds" was set at 25, which provided sufficient power for the measurement of overall project impact.
In the second stage, researchers asked each "seed" farmer to list four "plot neighbors", i.e. people who have contiguous or near-contiguous plots to his own. These four people were added to the survey sample, and the five respondents together constitute a synthetic SHG. Since the SHGs are formed based on proximity of agricultural land, households in synthetic SHGs are likely to be assigned into a single farmer group. Comparison sites were elected using pair-wise matching, from a list of sites considered eligible to receive LWH but that will not receive the project.
1C sample:
A baseline survey was implemented in four 1C LWH project sites. The LWH Project works at a site level (the valley and surrounding hills chosen for the intervention). The LWH project implemented different activities at different areas within the site: Command Area, Command Area Catchment, and Water Catchment areas. Because Phase 1C site Gicumbi was only receiving land husbandry interventions, the sample was taken only from the Water Catchment area of the site. At Muyanza site, the full portfolio of LWH interventions to be implemented, as reflected in the sampling across the different areas. Comparison sites were selected using pair-wise matching, from a list of sites considered eligible for the LWH project but not to receive the project. Extensive data was recorded for the sites considered eligible for the project, including data on geography, weather and land use patterns, making the identification of such matching sites possible.
1B sample:
A baseline survey was implemented in three 1B LWH project sites. Rwamagana-34, Rwamagana-35 and Kayonza-4, and three control sites were selected by pairwise-matching. The sample was designed to meet three criteria: geographic representation at the site-level, inclusion of multiple members of farmer groups within sites, and sufficient size to power tests of variations in treatment within the Phase 1B sites.
To meet these criteria, sampling was done through a two-stage process. In the first stage, researchers randomly sampled "seed" households, stratified by village. The total number of "seed" households varied by treatment status. For treatment sites, the number of "seeds" depended on the estimated number of Self-help Group (SHG), calculated based on the site population and the typical size of SHGs, which is 20 households. For control sites, the number of "seeds" was set at 25, which provided sufficient power for the measurement of overall project impact.
In the second stage, researchers asked each "seed" farmer to list four "plot neighbors", i.e. people who have contiguous or near-contiguous plots to his own. These four people were added to the survey sample, and the five respondents together constitute a synthetic SHG. Since the SHGs are formed based on proximity of agricultural land, households in synthetic SHGs are likely to be assigned into a single farmer group. Comparison sites were elected using pair-wise matching, from a list of sites considered eligible to receive LWH but that will not receive the project.
1C sample:
A baseline survey was implemented in four 1C LWH project sites. The LWH Project works at a site level (the valley and surrounding hills chosen for the intervention). The LWH project implemented different activities at different areas within the site: Command Area, Command Area Catchment, and Water Catchment areas. Because Phase 1C site Gicumbi was only receiving land husbandry interventions, the sample was taken only from the Water Catchment area of the site. At Muyanza site, the full portfolio of LWH interventions to be implemented, as reflected in the sampling across the different areas. Comparison sites were selected using pair-wise matching, from a list of sites considered eligible for the LWH project but not to receive the project. Extensive data was recorded for the sites considered eligible for the project, including data on geography, weather and land use patterns, making the identification of such matching sites possible.
Data Collection
Dates of collection
| Start | End | Cycle |
|---|---|---|
| 2012 | 2012 | Baseline for sample group 1B |
| 2013 | 2013 | Follow-up 1 and 2 for sample group 1B |
| 2014 | 2014 | Follow-up 3 for sample group 1B |
| 2016 | 2016 | Follow-up 4 for sample group 1B |
| 2014 | 2014 | Baseline for sample group 1C |
| 2016 | 2016 | Follow-up for sample group 1C |
Mode of data collection
Computer Assisted Personal Interview [capi]
Questionnaires
The baseline household survey used a multi-module questionnaire, with a specific focus on agricultural production, access to agricultural extension services, rural finance, and food security. In addition, the questionnaire contains modules on housing, labor, education, health, income and household assets.
Data access
Citation requirements
The use of the datasets must be acknowledged using a citation which would include:
- the identification of the Primary Investigator (including country name);
- the full title of the survey and its acronym (when available), and the year(s) of implementation;
- the survey reference number;
- the source and date of download (for datasets disseminated online).
Example:
Florence Kondylis, Maria Jones, Saahil Karpe, World Bank. Rwanda Land Husbandry, Water Harvesting and Hillside Irrigation Project Impact Evaluation 2012-2016, Baseline, First Follow-up and Second Follow-up Surveys, Ref. RWA_2012-2016_LWHIE-BL-F2_v01_M. Dataset downloaded from [URL] on [date].
- the identification of the Primary Investigator (including country name);
- the full title of the survey and its acronym (when available), and the year(s) of implementation;
- the survey reference number;
- the source and date of download (for datasets disseminated online).
Example:
Florence Kondylis, Maria Jones, Saahil Karpe, World Bank. Rwanda Land Husbandry, Water Harvesting and Hillside Irrigation Project Impact Evaluation 2012-2016, Baseline, First Follow-up and Second Follow-up Surveys, Ref. RWA_2012-2016_LWHIE-BL-F2_v01_M. Dataset downloaded from [URL] on [date].
Disclaimer and copyrights
Disclaimer
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.
Contacts
Contact(s)
| Name | Affiliation | |
|---|---|---|
| Florence Kondylis | The World Bank | fkondylis@worldbank.org |
| Maria Jones | The World Bank | mjones5@worldbank.org |
| Saahil Karpe | The World Bank | skarpe@worldbank.org |
Metadata production
IDNo
DDI_RWA_2012-2016_LWHIE-BL-F2_v01_M_WB
Producers
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| Development Data Group | DECDG | World Bank | Preparing of DDI |
| Development Impact Evaluation | DECIE | World Bank | Preparing of metadata |
Production date
2017-09-28
Version
v01 (September 2017)