India - Employment and Unemployment Survey, July 1999 - June 2000, NSS 55th Round

One salient feature of the 55th round is that the rotation sampling scheme has been adopted for the first time in the NSS (central sample only) for the purpose of collection of employment-unemployment data. Under this scheme, 50 per cent of the sample first stage units (fsu's) of each sub-round will be revisited in the subsequent sub-round. From each such fsu, sample households visited in the previous sub-round for collecting data on employment-unemployment will be revisited in the subsequent sub-round for collecting employment-unemployment details. In addition, for the purpose of collecting employment-unemployment data, a thin sample of 2 households will be selected during the revisit from the frame of newly formed households in the fsu. It may be noted that the above scheme of rotation sampling scheme for collecting employment-unemployment data will be followed only for the fsu's belonging to the central sample. For state samples, the fsu's are to be visited only once as they appear in the sample list for canvassing various schedules in the selected households/enterprises.



















Sampling frame for first stage units:

The frame used for selection of first stage units in the rural sector was the 1991 census list of villages for all the four sub-rounds for 8 states/u.t.s viz. Andhra Pradesh, Assam, Kerala, Madhya Pradesh, Orissa, Uttar Pradesh, West Bengal and Chandigarh. However for Agra district of U.P. and the three districts, viz.Durg, Sagar, and Morena of M.P., samples were drawn using 1981 census list of villages. For Jammu & Kashmir samples for all the 4 sub-rounds were drawn using the 1981 census list as the 1991 census was not conducted in the st ate. For the remaining 23 states/u.t.s, the frame was 1991 census list for sub-rounds 2 to 4 and 1981 census list for sub-round 1 as the 1991 census list was not available for use at the time of drawing the samples.
As usual, for Nagaland the list of villages within 5 kms. of the bus route and for Andaman and Nicobar Islands the list of accessible villages constituted the frame. In the case of urban sector the frame consisted of the UFS blocks and, for some newly declared towns where these were not available, the 1991 census enumeration blocks were used.

Region formation and stratification:
States were divided into regions by grouping contiguous districts similar in respect of population density and cropping pattern. In rural sector each district was treated a separate stratum if the population was below 2 million and where it exceeded 2 million, it was split into two or more strata. This cut off point of population was taken as 1.8 million ( in place of 2 million ) for the purpose of stratification for districts for which the 1981 census frame wa s used. In the urban sector, strata were formed, within each NSS region on the basis of population size class of towns. However for towns with population of 4 lakhs or more the urban blocks were divided into two classes viz. one consisting of blocks inhabited by affluent section of the population and the other consisting of the remaining blocks.

Selection of first stage units :

Selection of sample villages was done circular systematically with probability proportional to population and sample blocks circular system-atically with equal probability. Both the sample villages and the sample blocks were selected in the form of two or more independent sub-samples. In Arunachal Pradesh the procedure of cluster sampling has been followed. Further large villages/blocks having present population of 1200 or more were divided into a suitable number of hamlet- groups/ sub-blocks having equal population content. Two hamlet- groups were selected from the
larger villages while one sub-block was selected in urban sector for larger blocks.

Selection of households :

While listing the households in the selected villages, certain relatively affluent households were identified and considered as second stage stratum 1 and the rest as second stage stratum 2.
A total of 10 households were surveyed from the selected village/hamlet-groups, 2 from the fi rst category and remaining from the second.
Further in the second stage stratum-2, the households were arranged according to the means of livelihood. The means of livelihood were identified on the basis of the major source of income as i) self-employed in non-agricultu re, ii) rural labour and iii) others. The land possessed by the households was also ascertained and the frame for selection was arranged on the basis of this information.
The households were selected circular systematically from both the second stage strata.

In the urban blocks a different method was used for arranging the households for selection. This involved the identification means of livelihood of households as any one of a) self-employed, b)regular salaried/wage earnings, c) casual labour, d) others. Further the average household monthly per capita consumer expenditure (mpce) was also ascertained. All households with MPCE of (i) Rs. 1200/- or more (in towns with population less than 10 lakhs or (ii) Rs. 1500/- or more (in towns with population 10 lakh or more) formed second-stage stratum 1 and the rest, second-stage stratum 2.The households of second-stage stratum 2 were arranged according to means of livelihood class and MPCE ranges before selection of sample households. A total of 10 households were selected from each sample block as follows
(i) For affluent strata/classes : 4 households from second- stage stratum 1 and 6 households from second-stage stratum 2
(ii) For other strata/classes : 2 households from second-stage stratum 1 and 8 from second-stage stratum 2.
Households were then selected circular systematically w ith a random start.

Shortfall in the required number of household in any second-stage st ratum was made up by ncreasing the quota for the other second stage stratum.

A total of 7284 sample villages (Rural) and 4792 sample blocks (Urban) were allotted in central sample. 6983 sample villages and 470 sample blocks were successfullly surveyed covering 356351 persons in sample villages and 208389 persons in sample blocks.


There was no deviation from the original sampling design


Three different weightsare provided in each record of filess in the data set. Deatils are as follows:-

1. Weight for each sub-round (Sub-round wise weight) is stored in Variable name : WGT_SubSsample
2. Weight for each subsample (Sub-sample wise weight) is stored in Variable name : WGT_Subround
3. Weight for all Subrounds pooled and combined subsample weight is stored in Variable name : WGT_Pooled