Survey ID Number
NPL_2010_LSS-III_v01_M
Title
Living Standards Survey 2010-2011
Sampling Procedure
Sampling Frame
The initial frame for the NLSS-III survey was based on the frame prepared for the 2008 Nepal Labour Force Survey (NLFS-II). This was done “to take advantage of the cartographic segmentation and household listing operations” conducted by the CBS for the NLFS-II. Number of households at ward level was obtained from the 2001 Population Census. CBS has constructed a data set with basic information (number of households, total population, and male and female population) at the ward level. This data set was used to construct the frame for the selection of primary sampling units (PSUs).
The PSU for the NLFS-II were either individual wards or sub-wards or groups of contiguous wards in the same VDC. A group of wards was considered as single PSU, to ensure that each unit continued at least 30 households.
For the ultimate sample selection of households in the NLFS-II, a frame was prepared in each ward selected at the earlier stage of sampling. A list of all households was prepared in all the wards selected for the survey. Selection of households was carried out from these listings using systematic sampling with a random start. Before the listing, an intensive cartographic work was undertaken (in the urban areas and some of the rural areas) to form appropriate enumeration block having around 200 households.
Stratification
For the NLFS-II sample selection, 75 districts along with the urban and rural areas were grouped into six strata – mountains, urban areas of the Kathmandu valley, other urban areas in the hills, rural hills, urban hills, urban Tarai and rural Tarai. These six strata of the NLFS-II were further regrouped into 14 strata for the NLSS-III purposes. The “explicit” strata formed for the NLSS-III were as follows: mountains, urban areas of the Kathmandu valley, other urban areas in the hills, rural eastern hills, rural central hills, rural western hills, rural mid-western hills, rural far-western hills, urban Tarai, rural eastern Tarai, rural central Tarai, rural western Tarai, rural mid-western Tarai, and rural far-western Tarai.
Sample Design
The sample design adopted in NLSS-III was modified sub-sample of the sample adopted in NLFS-II. For the NLFS-II, 800 PSUs were selected – 400 PSUs each from urban and rural areas. As mentioned earlier, the PSU for the NLFS-II was a ward or a sub-ward or a combination of wards. The PSUs were selected with probability proportional to size, the measure of size being the number of households.
For the NLSS-III, two independent samples were selected: the first was a cross sectional sample and the second was a panel. The panel sample consisted of PSUs and households previously enumerated in one or both of the past two rounds of the survey.
Cross-section Sample
The cross-section sample was selected in three stages:
• In the first stage, the enumeration area units (primary sampling units, PSUs) are identical to that of NLFS-II. NLFS selected a sample of 800 PSUs allocated into six explicit strata. Within each stratum, the PSUs were selected with probability proportional to size (pps), the number of households being the measure of size (mos).
• In the second stage, 500 of the 800 NLFS PSUs were selected to form the NLSS-III PSUs. As mentioned earlier, for this selection purpose the six strata of NLFS-II were further divided into 14 sub-strata (Box 1.2). Within each sub-stratum, the sample was selected with equal probability.
• In the third stage, 12 households were selected with equal probability in each PSU (using the computerized household listings generated by the NLFS-II as a sample frame). An additional 6 households were selected in each PSU, to be used as replacement household for non-response among the 12 originally selected households.
Panel Sample
The panel sample of the NLSS-III is composed of all households visited by the NLSS-II in 100 of its primary sampling units (PSU). Fifty of them were taken from the cross-sectional component of the NLSS-II, and the remaining fifty from its panel component. In other words, one half of the NLSS-III panel households were households visited for the first time during the NLSS-II, whereas the other half were the households visited during both the NLSS-I and the NLSS-II.
Sample Size
The sample size for the survey was estimated at 7200 households in 600 PSUs. Among them, 100 PSUs with 1200 households interviewed in the NLSS-I or NLSS-II were selected for reinterviewing in the NLSS-III. And, 500 PSUs with 6000 households were selected as the crosssection sample. The PSUs were selected with probability proportional to size, the measure of size being the number of households in each ward. As mentioned earlier, twelve households were selected for the enumeration from each of the selected PSU.