India - Household Expenditure on Services Durable Goods : NSS Round 72, Schedule 1.5, July 2014 - June 2015

1.4 Sample Design

1.4.1 Outline of sample design: A stratified multi-stage design has been adopted for the 72nd round survey. The first stage units (FSU) will be the Census villages (Panchayat wards for Kerala) in the rural sector and Urban Frame Survey (UFS) blocks in the urban sector. The ultimate stage units (USU) will be households in both the sectors. In the case of large FSUs, one intermediate stage of sampling will be the selection of two hamlet-groups (hgs)/ sub-blocks (sbs) from each rural/ urban FSU.

1.4.2 Sampling Frame for First Stage Units: For the rural sector, the list of 2011 Population Census villages will constitute the sampling frame. However, for Kerala, the latest available updated list of Panchayat wards will constitute the sampling frame. For the urban sector, the list of UFS blocks (2007-12) will be considered as the sampling frame.

1.4.3 Stratification:

(a) Each district will be a stratum. Within each district of a State/UT, generally speaking, two basic strata will be 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. However, within the urban areas of a district, if there are one or more towns with population 4 lakhs or more as per Census 2011, each of them will form a separate basic stratum and the remaining urban areas of the district will be considered as another basic stratum.
(b) A special stratum in the rural sector only will be formed at State/UT level before district-strata are formed in each of the States/UTs. This stratum will comprise all the villages of the State/UT with population less than 150 as per Census 2011. However, this special stratum will be formed if at least 50 such villages are available in a State/UT.

(c) In case of rural sectors of Nagaland another special stratum will be formed within the State consisting of all the villages which are difficult to access.

1.4.4 Sub-stratification:

1.4.4.1 Rural sector: If 'r' be the sample size allocated for a rural stratum, 'r/4' sub-strata will be formed in that stratum. The villages within a stratum (district) as per frame will be first arranged in ascending order of population. Then sub-strata 1 to 'r/4' will be demarcated in such a way that each sub-stratum will comprise a group of villages of the arranged frame and have more or less equal population.

1.4.4.2 Urban sector: Each urban stratum will be divided into two parts - 'Affluent part' consisting of UFS blocks identified as 'Affluent Area' and 'Non-Affluent part' consisting of the remaining UFS blocks within the stratum. Sample allocation of a stratum will then be allocated to Affluent and Non-Affluent parts in proportion to total number of households in the UFS blocks with double weightage to Affluent part subject to a maximum of 8 FSUs in 'Affluent part' of any stratum. Let 'u' be the sample size allocated for an urban stratum consisting of both affluent area UFS Blocks and non affluent area UFS Blocks. 'u/4' sub-strata will be formed within each stratum. Out of these 'u/4' sub-strata, the first two sub-strata '01' and '02' will be earmarked for those UFS blocks which are identified as 'Affluent Area' and the remaining sub-strata, '03', '04',…... and so on, will be assigned to the non-affluent UFS blocks. If any stratum does not have any Affluent Area UFS block then also the sub-stratum number will start from '03' for that stratum. For all strata, if u/4 >1, implying formation of 2 or more sub-strata, all the UFS blocks within the stratum will be first arranged in ascending order of total number of households in the UFS Blocks as per UFS phase 2007-12. Then sub-strata 1 to 'u/4' will be demarcated in such a way that each sub-stratum will have more or less equal number of households. This procedure will be done separately for Affluent-part and Non-Affluent part of the stratum (if two sub-strata are required to be formed in the Affluent part).
The following three cases may arise while doing the sub-stratification:

i) If there is no 'Affluent Area' UFS block in the stratum, all the UFS blocks will be divided into 'u/4' sub-strata and will be numbered as '03', '04', ….., '(u/4) 2'.
ii) If only one sub-stratum is formed with the 'Affluent Area' UFS blocks, then all the remaining non-affluent blocks will be divided into '(u-4)/4' sub-strata and will be numbered as '03', '04', ….., '(u/4) 1'.
iii) If two sub-strata are formed with the 'Affluent Area' blocks, then all the remaining non-affluent UFS blocks will be divided into '(u-8)/4' sub-strata and will be numbered as '03', '04', ….., 'u/4'.

1.4.5 Total sample size (FSUs): About 14088 FSUs will be surveyed for the central sample at all-India level. State wise allocation of sample FSUs is given in Table 1 of page A-27.

1.4.6 Allocation of total sample to States and UTs: The total number of sample FSUs will be allocated to the States and UTs in proportion to population as per Census 2011 subject to a minimum sample allocation to each State/UT. While doing so, the resource availability in terms of field investigators will be kept in view.

1.4.7 Allocation of State/ UT level sample to rural and urban sectors: State/ UT level sample size will be allocated between two sectors in proportion to population as per Census 2011 with double weightage to urban sector in general. A minimum of 16 FSUs (minimum 8 each for rural and urban sector separately) will be allocated to each State/ UT.

1.4.8 Allocation to strata: Within each sector of a State/ UT, the respective sample size will be allocated to the different strata in proportion to the population as per Census 2011. Stratum level allocation will be adjusted to multiples of 4 with a minimum sample size of 4.

For special stratum formed in rural areas of Nagaland as discussed in para 1.4.3 (c), 12 FSUs will be allocated.

1.4.9 Allocation to sub-strata: Allocation for each sub-stratum will be 4 in both rural and urban sectors.

1.4.10 Selection of FSUs:

For the rural sector, from each stratum/sub-stratum, required number of sample villages will be selected by Probability Proportional to Size With Replacement (PPSWR), size being the population of the village as per Census 2011.

For the urban sector, from each stratum/sub-stratum, FSUs will be selected by Probability Proportional to Size With Replacement (PPSWR), size being the number of household of the UFS Block.

Both rural and urban samples are to be drawn in the form of two independent sub-samples and equal number of samples will be allocated among the four sub rounds.

1.4.11 Selection of hamlet-groups/ sub-blocks - important steps

1.4.11.1 Proper identification of the FSU boundaries: The first task of the field investigators is to ascertain the exact boundaries of the sample FSU as per its identification particulars given in the sample list. For urban samples, the boundaries of each FSU may be identified by referring to the map for the UFS Phase 2007-12.

1.4.11.2 Criterion for hamlet-group/ sub-block formation: After identification of the boundaries of the FSU, it is to be determined whether listing will be done in the whole sample FSU or not. In case the approximate present population of the selected FSU is found to be 1200 or more, it will be divided into a suitable number (say, D) of 'hamlet-groups' in the rural sector and 'sub-blocks' in the urban sector by more or less equalising the population as stated below.

approximate present population of the sample FSU no. of hg's/sb's to be formed
less than 1200 (no hamlet-groups/sub-blocks) 1
1200 to 1799 3
1800 to 2399 4
2400 to 2999 5
3000 to 3599 6
…...and so on -

For rural areas of Himachal Pradesh, Sikkim, Andaman & Nicobar Islands, Uttarakhand (except four districts Dehradun, Nainital, Hardwar and Udham Singh Nagar), Punch, Rajouri, Udhampur, Reasi, Doda, Kishtwar, Ramban, Leh (Ladakh), Kargil districts of Jammu and Kashmir and Idukki district of Kerala, the number of hamlet-groups will be formed as follows:


approximate present population of the sample village no. of hg's to be formed
less than 600 (no hamlet-groups) 1
600 to 899 3
900 to 1199 4
1200 to 1499 5
1500 to 1799 6
.….and so on -


1.4.11.3 Formation and selection of hamlet-groups/ sub-blocks: In case hamlet-groups/ sub-blocks are to be formed in the sample FSU, the same should be done by more or less equalising population. Note that while doing so, it is to be ensured that the hamlet-groups/ sub-blocks formed are clearly identifiable in terms of physical landmarks.
Two hamlet-groups (hg)/ sub-blocks (sb) will be selected from a large FSU wherever hamlet-groups/ sub-blocks have been formed in the following manner - one hg/ sb with maximum percentage share of population will always be selected and termed as hg/ sb 1; one more hg/ sb will be selected from the remaining hg's/ sb's by simple random sampling (SRS) and termed as hg/ sb 2. Listing and selection of the households will be done independently in the two selected hamlet-groups/ sub-blocks. The FSUs without hg/ sb formation will be treated as sample hg/ sb number 1. It is to be noted that if more than one hg/ sb have same maximum percentage share of population, the one among them which is listed first in block 4.2 of Schedule 0.0 will be treated as hg/ sb 1.
1.5 Listing of households: Having determined the hamlet-groups/ sub-blocks, i.e. area(s) to be considered for listing, the next step is to list all the households (including those found to be temporarily locked after ascertaining the temporariness of locking of households through local enquiry). The hamlet-group/ sub-block with hg/sb number 1 will be considered for listing first, to be followed by the listing of households of the hg/sb number 2.

There was no deviation from the original sampling design.