Rwanda - Agricultural Household Survey 2020
Reference ID | RWA-NISR-RAHS-2020-V0.1 |
Year | 2020 |
Country | Rwanda |
Producer(s) | National Institute of Statistics of Rwanda - Ministry of Finance and Economic Planning |
Sponsor(s) | Government of Rwanda - GoR - Funding |
Metadata | Documentation in PDF |
Created on
Apr 19, 2022
Last modified
Apr 19, 2022
Page views
1028907
Sampling
Sampling Procedure
To ensure a good representation of data at the district level, the survey was conducted using a sample size of 900 Enumeration areas (EAs). A sample frame used was composed of a list of enumeration areas (EAs) retrieved from the 2012 Rwanda General Population and Housing Census (RGPHC-2012). A stratified two-stages sample design was used.
The first stage focused on a stratified sample of enumeration areas from the latest RGPHC-2012. Given that rural areas are dominated by many households practicing agriculture, the frame of EAs was sorted by urban and rural areas within districts. This provides an implicit stratification of the households by urban and rural areas. Out of 900 EAs, 860 were selected from rural EAs while the rest was selected from urban EAs, where there are fewer but not negligible numbers of households practicing agriculture.
To ensure adequate geographical distribution of the sample and given that the results were analysed up to district level, the sample of 860 rural EAs was allocated equally among 30 districts, while a sample of 40 urban EAs was allocated proportionally throughout the country. At this first stage, the sampled EAs in each district were selected systematically with a probability proportional to size (PPS) measured in terms of the total number of households in each EA from the RGPHC-2012.
The second stage looked at a random selection of a fixed number of 12 households who only did agriculture in each sampled EA. 12 households were selected from an updated number of households listed in each sampled EA, in the first phase of AHS 2020 data collection.
It is important to note that there are EAs with less than 12 agricultural households and these have been directly taken as a sample at the second stage. Finally, the sampling process conveyed a total sample of 10,666 agricultural households.
Weighting
The weight of a given sample-households is the inverse of the probability of selection of households. This probability is computed as the product of two probabilities