Labor Force Survey 2018

The sample design of the LFS is a two-stage stratified design according to which at the first stage of sampling, a stratified sample of enumeration areas from the latest population census is drawn with probabilities proportional to size measured in terms of the census number of households or census number of household members, and at the second stage of sampling, a fixed number of sample of households is selected with equal probability within each sample enumeration areas. Finally, all household members in the sample households are selected for survey interviewing.

The application of this principle with the choice of parameters described below leads to a sample size requirement of 18,691 households for measuring annual unemployment with margin of errors of +/- 0.3% at 95% confidence level. In these calculations, the main indicator is the ratio of unemployment to the working age population, set at r=0.024 according to the 2012 population census of Rwanda. The design effect (deff) is set at 3, within the range of values (3 to 4) generally used for labour force surveys. The margin of errors (ME = 0.0026) is twice the standard error of the estimate.13Similar calculations for employment gave a sample size requirement of 18,792 households.

Because the sample size should be a multiple of 4x16 for February 2018 survey round, four rotation groups with 146 Primary Sampling Unit per rotation and 16 sample households per Primary Sampling Unit (PSU) that give a sample size of 9,344 households. Since we are shifting from bi-annual to quarterly LFS, August 2018 survey round has been set as a period of transition, therefore 530 PSU has been used with the sample size of 9,248 households. Two rounds of 2018 LFS have been rounded to the sample size 18,592 households.

To ensure adequate geographical distribution of the sample over the different parts of the population, the sample is allocated among the 30 districts of the country. Prior to sample selection, the sampling frame is sorted according to urban and rural areas within districts. This provides an implicit stratification of the population by urban and rural areas. The resulting distribution of the sample of the February and August 2018 LFS rounds obtained on the basis of the square-root allocation is shown in the following tables. The sample design for February round provides a total of 9344 households in 584 enumeration areas, with 3248 urban households in 203 urban enumeration areas and 6096 rural households in 381 rural enumeration areas while the sample design for August round provides a total of 9248 households in 530 enumeration areas, with 3112 urban households in 180 urban enumeration areas and 6136 rural households in 381 rural enumeration areas.

The next step in sample design was the selection of the 584 sample enumeration areas. The sample was drawn in each district by probability proportional to size (pps) from the sampling frame. In practice, the sample of enumeration areas were selected from the start for an entire sequence of 18 survey rounds from August 2016 to August 2023 when in principle a new sampling frame constructed on the basis of the next population and housing census will become available. The selected sample was then assigned randomly to 18 rotation groups. Sample enumeration areas in rotation groups R1 to R4 formed the sample for LFS August 2016. The sample of enumeration areas assigned to rotation groups R3 to R6 were used for LFS August 2017. The sample enumeration areas assigned to R4 to R7 will be used for LFS February 2018, and so on.
Given that the August 2018 has been the transition to the quarterly design, a special composition of the August 2018 sample has been designed. It is composed of the old rotation group 5, the outgoing rotation group from the bi-annual design. It is also composed of the two other old rotation groups 6 and 7, randomly divided into three equal parts (6+7)/3, one serving as new rotation group r1, the other as new rotation group r2 and the third as new rotation group r3. Finally, the August 2018 sample include a randomly selected one-third of the old rotation groups 8 and 9, which was treated as the new households entering the sample. The other one-third samples (8+9)/3 will be used as the new households entering the sample in February 2019 and May 2019, respectively. The same procedure will be used for the new rotation groups entering the sample each quarter from August 2019 onward.

The sample enumeration areas were freshly listed prior to selection of the final sample of households. According to the February 2018 sample design, 16 sample households were selected from the list of households in each sample enumeration area by systematic sampling with equal probability. If the list contained 16 households or less all households in the sample enumeration area were drawn in the sample. If the list contained more than 16 households, a sample of fixed size (16 households) was drawn from the list by systematic random sampling. In order to keep almost the same number of households, the sample intake of 24 households in 96 new enumerations areas were used in August 2018 sample size which resulted to the sample size of 9,248 instead of 9,344 sample households for previous rounds.