Sampling Procedure
Sample Design
Outline of sample design
A stratified two-stage sample design was adopted for the 52nd round .The first-stage units were the census villages for the rural areas (panchayat wards in case of Kerala) and the NSSO Urban Frame Survey(UFS) blocks for the urban areas. The second-stage units were the households in both cases.
Sampling Frame for First Stage Units
The lists of census villages of the1991 census (1981 census list for Jammu & Kashmir) constituted the sampling frame for the rural areas ( panchayat wards for kerela / villages within 5kms of bus route for nagaland / list of accesible villages for andamand and nicobar islands ). For the urban areas, the list of NSSO Urban Frame Survey(UFS) blocks was used as the sampling frame.
Stratification
Within each district of a State/ UT, generally speaking, two basic strata was formed: i) rural stratum comprising of all rural areas of the district and (ii) urban stratum comprising of all the urban areas of the district. In all, there were 78 regions covering the entire geographical area of the country.
Sub-stratification
Rural sector
In the rural areas, each district within a region was treated as a separate stratum. However, if the 1991 census population of the district was found to be greater than or equal to 2 million (1.8 million population as per 1981 census for Jammu & Kashmir), the district was split into two or more strata, by grouping contiguous tehsils to form strata.
Urban sector
In the urban areas, strata were formed, within each NSS region, by grouping towns on the basis of the population of towns. All towns with population less than 50,000 formed the first stratum, all towns with population 50,000 -1,99,999 formed 2nd stratum, all towns with population 200,000 -9,99,999 formed 3rd stratum and each city with population 10,00,000 or above formed rest of the starta and so on.
Total sample size (FSUs)
16,480 villages were planned to survey in this round. Of these, 7888 were allocated to the Central sample and the rest to the State sample . In the urban sector, the allocations for the Central and State samples were respectively, 5112 and 6320 blocks. The number of villages and that of urban blocks actually surveyed as the Central sample were 7,663 and 4,991 respectively.
Total sample size(SSUs)
10 households were planned to survey in each selected FSU. In the Central sample, the actual numbers of households surveyed in the rural and urban areas were 71,284 and 49,658, respectively.
Allocation of State/ UT level sample to rural and urban sectors
The sample size of FSUs (rural & urban combined) for the Central sample for a State/U.T. was allocated to its rural and urban areas considering the relative sizes of the rural and urban population with double weightage to the urban areas. The State level rural sample size was allocated to the rural strata in proportion to their rural population figures as per the census and urban sample size of the State/U.T. was allocated to the urban strata in proportion to urban population figures as per the census. All the stratum-level allocations were adjusted to multiples of 8.
Allocation to strata
Within each sector of a State/ UT, the respective sample size was allocated to the different strata in proportion to the stratum population as per census 2001. Allocations at stratum level was adjusted to a multiple of 4 with a minimum sample size of 4.
Selection of FSUs
The sample FSUs in the rural and urban areas were selected circular systematically with equal probability in the form of two independent sub samples (cluster sampling for rural areas of arunachal pradesh).
Selection of SSUs
In each of the selected FSUs, three different enquiries, viz. "Survey on Health Care", "Survey on Participation in Education" and "Survey on Consumer Expenditure", were conducted on three independent samples of households. For the present enquiry, i.e. the survey on "HealthCare", a sample of 10 households was selected for the detailed enquiry. However, before selection, the listed households were first grouped into two second-stage strata.
Formation and selection of hamlet-groups/ sub-blocks
A large FSU was divided into a suitable number of hamlet-groups/ sub-blocks having approximately equal population content.Two hamlet-groups were selected from each large FSU in the rural areas and only one sub-block was selected from each large FSU of the urban areas.
Selection of households (second-stage units)
From each selected FSU (or the selected hamlet-group/sub-block of the FSU for large FSUs), a sample of 4 households for schedule 1.0, 10 households for schedule 25.0 and 6 households for schedule 25.2 were selected for the detailed enquiry. However, before selection, the listed households were first grouped into second-stage strata for sch. 25.0 and 25.2 .
For any schedule type (i. e 25.0 or 25. 2), if there was a shortfall in the required number of households in the frame of any particular second-stage stratum, the quota for the other second-stage stratum/strata had to be increased within the same schedule type so that the required total number of households ( i.e 10 for schedule 25.0 and 6 for schedule 25.2 ) was selected for the particular schedule type from the FSU (or hamlet-groups/sub-block for large FSUs ) for detailed enquiry.
After determining the number of households to be finally selected from different second-stage, the required number of sample households for schedules 25.0 & 25.2 were selected independently from each second-stage stratum by circular systematic sampling with a random start.