The Malawi Rural Livelihood Survey (MRLS) is a panel survey of households, interviewed 4 times from November 2012-2013. The survey was designed in order to conduct a randomized control study of Malawi Malawi's large-scale public works program (PWP) which operates under the Malawi Social Action Fund (MASAF).
Kind of data
Sample survey data [ssd]
Among the 28 districts which were included in the PWP at the time of this study, the MRLS was conducted in 12 districts which were randomly selected across the three strata of the country's three geographic regions to ensure that the study was representative of the country's population.
Producers and sponsors
Malawi Government National Statistics Office
The World Bank
In order to have pre-program baseline data that predated the vetting of communities for the PWP, the national Integrated Household Survey (IHS3) collected in 2010/2011 by Malawi's National Statistics Office was used as the basis for the MRLS. Within the 12 selected districts, the list of PWP-eligible (pre-screened) communities from the District Council and Traditional Authorities was compared with the 786 enumeration areas (EAs) in the IHS3. The sampling frame for the MRLS was the subset of IHS3 enumeration areas (EAs) in the 12 selected districts which were also in pre-selected communities for the PWP. This resulted in a total of 182 villages (EAs) out of the 786 EAs in the IHS3 to be included in the MRLS. In each community, the MRLS would re-survey the 16 randomly surveyed households in the IHS3.
Deviations from sample design
In practice, of the 182 EAs selected for the MRLS, 23 EAs were incorrectly identified as IHS3 EAs. These non-IHS3 EAs then have no 2010/2011 data and as a result, a new household listing was conducted in these EAs in order to randomly select 16 households in each EA.
The final target sample for the MRLS was 2,912 households (16 households in 182 communities). This includes replacement households drawn at random when required. Households in EAs that were covered by the IHS3-panel survey in the fall 2013 were excluded from Round 4 field work (resulting in a smaller sample in Round 4). One can merge in those data from the IHS3 panel data (publicly available).
The 182 villages in the MRLS sample were randomly divided into five categories with regards to the PWP (four with PWP treatment and one with no PWP). Within each PWP treatment village, a random sample of households was offered an opportunity to work in the public works program. The details of this are described in Beegle, Galasso, and Goldberg (2017), “Direct and Indirect Effects of Malawi’s Public Works Program on Food Security.” Journal of Development Economics 128: 1-23.
Dates of collection
After the first cycle of the PWP and during the lean season
Mode of data collection
In each of the four rounds of the MRLS, a Household Questionnaire and a Community Questionnaire was fielded. The subsections of the household questionnaire are summarized in Table 2 of the Basic Information Document provided under Related Materials. The questionnaires are drawn from the IHS3 questionnaire to ensure comparability. After Round 1, modules were updated in some places to reflect improvements or capture transitions from the previous round.The subsample of the MRLS which is from IHS3 can be linked to the IHS3 data through the variable case_id in the data set pwp_hhstatus_public.dta.
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download
Malawi Government National Statistics Office., The World Bank. 2013. The Malawi Rural Livelihood Survey (MRLS-R2) 2013, Round 2. Ref. MWI_2013_MRLS-R2_v01_M. Downloaded from [url] on [date]
Disclaimer and copyrights
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.