REPORT ON POOLING OF CENTRAL AND STATE SAMPLE DATA OF NSS 66 th ROUND (July June 2010)

Transcription

1 REPORT ON POOLING OF CENTRAL AND STATE SAMPLE DATA OF NSS 66 th ROUND (July June 2010) HOUSEHOLD CONSUMER EXPENDITURE (TYPE 1 & 2) & EMPLOYMENT & UNEMPLOYMENT DIRECTORATE OF ECONOMICS AND STATISTICS GOVERNMENT OF UTTAR PRADESH LUCKNOW dsdesd@up.nic.in

2 PREFACE The 66 th round (July 2009-June 2010) of NSS was earmarked for survey on Household Consumer Expenditure and Employment and Unemployment. The survey on household consumer expenditure and employment and unemployment is the eighth quinquennial survey in the series, the last one being conducted in the 61 st round (July 2004-June 2005) of NSS. Uttar Pradesh State is a partner in these surveys since the 9 th round (1955), generally on equal matching basis. National Statistical Commission constituted a professional committee under the Chairmanship of Prof. R.Radhakrishna, to identify the preconditions for pooling of Central and State Sample NSS data, to suggest appropriate methodology for pooling the data and to bridge the data gaps and in turn strengthen the database for decentralized planning and governance. The necessity for pooling the Central and State sample data arose due to the growing need for improving the precision of estimates of policy parameters such as the labour force participation, level of living and well being, incidence of poverty, State Domestic Product (SDP), District Domestic Product (DDP) etc., and for strengthening the database at district level required for decentralized governance. Accordingly the NSC professional committee has recommended certain poolability tests and the methodologies for pooling of Central and State sample data of NSS. In the workshop which was held on 2 nd and 3 rd January 2013 at New Delhi, it was demonstrated about the poolability tests and pooling the two sets of data and estimating the parameters based on two methods- (1) Matching ratio method and (2) Inverse Weight of the Variance of estimates method. But, to apply methodologies for pooling of Central and State sample data of NSS, main problem which encountered to UP was differences in layout of data of Central and State sample and recasting State sample data on Central sample layout. Recasting of State sample data on Central sample layout and thereafter pooling became possible by the meticulous planning and execution of various activities of DES officials. I am thankfull to the officers of DPD, Kolkata for their valuable technical guidance provided at different stages. Uttar Pradesh State has prepared present report on pooled estimates of Central and State sample data of NSS 66 th round on Household Consumer Expenditure (Type I & II) and Employment and Un-Employment data. Comments and Valuable Suggestions from the Researchers and Scholars on this report are most welcome. Date: Lucknow: March 19, 2015 (Girja Shanker Katiyar) Director

3 OFFICERS AND STAFF ASSOCIATED WITH THE REPORT Mr. Girja Shanker Katiyar Mr. Arvind Kumar Pandey Mr. Rajesh Kumar Dhusia Dr. Shri Nath Yadav Director Joint Director Joint Director(Computer) Deputy Director Mr. Sanjeev Kumar Mr. Surya Prakash Smt. Shakti Additional Statistical Officer Additional Statistical Officer Additional Statistical Officer Mr. Rahul Pathak Assistant Statistical Officer Mr. Mohan Singh Senior Assistant

4 TABLE OF CONTENTS CHAPTER-1 Introduction and Background 0-3 CHAPTER-2 Summary Findings 4-11 Table 2.1 to 2.14 CHAPTER-3 Testing Poolability and Methodology for pooling CHAPTER-4 Poolability Test Results Table 4.1 Districtwise results of test of MPCE (URP, MRP, (R&U): Table 4.2 (R&U): Table 4.3 (R& U) : Table 4.4 (R&U): Table 4.5 (R&U): Table 4.6 (R&U): Table 4.7 (R&U): Table 4.8 (R&U): Table 4.9 (R&U): Table 4.10 (R&U): Table 4.11 (R&U) : Table 4.12 (R&U): Table 4.13 (R&U) : Table 4.14 (R&U) : Table 4.15 (R&U) : Table 4.16 (R&U) : MMRP) for pooled sample-run TEST Districtwise results of test of persons over worker,unemployed and out of labour force for pooled sample- Chi-square Test Results of Sch. 1.0 and 10 and their RSEs Districtwise estimated number of households(00), persons(00), sex ratio and their RSE's (%) for pooled samples (Schedule type-1) Districtwise estimated number of households(00), persons(00), sex ratio and their RSE's (%) for pooled samples (Schedule type-2) Districtwise estimate of MPCE (URP) for sample Districtwise estimate of MPCE (MRP) for sample Districtwise estimate of RSE (%) of MPCE (URP) for sample Districtwise estimate of RSE (%) of MPCE (MRP) for sample Districtwise estimate of MPCE (MMRP) for sample Districtwise estimate of RSE (%) of MPCE (MMRP) for sample Districtwise WPR per 1000 (PS+SS) for pooled samples Districtwise RSE (%) of WPR (PS+SS) for pooled samples Districtwise LFPR per 1000 (PS+SS) for pooled samples Districtwise RSE (%) of LFPR (PS+SS) for pooled samples Districtwise LFPR per 1000 (CWS and CDS) for pooled samples Districtwise RSE of LFPR (CWS and CDS) for pooled samples

5 ANNEXURE-1 DISTRICTWISE SAMPLE LIST- UTTAR PRADESH STATE 83 Table 1.6a (R/U/State): Table 1.6b (R/U/State): Table 2.6 (R/U/State): Table (21) (R/U/State- M/F/P): Table (34) (R/U/State- M/F/P): Table (36) (R/U/State- M/F/P): DETAILED TABLES Value of consumption (Rs.0.00) of food and non-food per person for a period of 30 days for each quantile class of MPCE(URP). (for pooled samples of Rural, Urban and State seperately) Value of consumption (Rs.0.00) of food and non-food per person for a period of 30 days for each Quantile class of MPCE(MRP). (for pooled samples of Rural, Urban and State seperately) Value of consumption (Rs.0.00) of food and non-food per person for a period of 30 days for each quantile class of MPCE(MMRP) (for pooled samples of Rural, Urban and State seperately) Per thousand distribution of persons by usual activity category taking also into consideration the subsidiary economic status of persons categorized 'not working' in the principal status (for pooled samples) Per thousand distribution of persons by Current Weekly Activity (for pooled samples) Per thousand distribution of persons-days by Current Daily Activity (for pooled samples)

6 CHAPTER-1 Introduction and Background 1.0 Background The National Sample Survey Office (NSSO) conducts nationwide household consumer expenditure surveys at regular intervals as part of its rounds, each round normally of a year s duration. The NSS surveys are conducted through interviews of a random sample of households selected through a Multistage Stratified Random Sampling design and cover practically the entire geographical area of the country. The household consumer expenditure survey (CES) is generally covered as one of the main subjects of the NSS survey in quinquennial rounds. This provides a series of CES s. The 66 th round survey (July 2009-June 2010) was the eighth such survey of the quinquennial series, the seventh had been conducted during the 61 st round (July 2004-June 2005). 1.1 State s participation in NSS surveys Uttar Pradesh state has been participating in the NSS surveys from 9th round onwards by using the same concepts, definitions and procedures and by adopting the same sample design based on independently drawn sample as that of NSSO called as the state samples. The sample size of the state sample in Uttar Pradesh was double to that of central samples for 66 th round. The sample list of the state sample was drawn and supplied by the NSSO. District wise sample size of 66 th round is given in Annexure-1. One of the objectives of states participation in the NSS programme is to provide a mechanism by which sample size will be increased and the pooling of the two sets of data would enable better estimates at lower sub state level, particularly at district level. This resulted in increased precision of the estimates at disaggregated level. But the major benefit will be derived in the case of estimates are generated at sub-state level like NSS regions/districts. 1.2 Emerging needs for pooling of Central and State samples The 73 rd and 74 th constitutional amendment (1992) has brought into existence the democratically elected grassroots institutions of local self governance, with respective delegated functions, both in rural and urban areas. This has enhanced the demand for local level statistics and necessitated requirement of developing basic capabilities at grass root levels to organize such statistics in a harmonious manner. In this context, it is envisioned that the survey resources in overall NSS programme both by Central and State Agencies can be more effectively utilized to generate lower level estimates of key indicators at district level. 13 th Finance

7 Commission, in Para of its report, noted that Comparable estimates of district income are extremely relevant for measuring intra-state income disparities. This will enable State Governments to effectively plan policy and programme interventions. They could also be used as a parameter for horizontal distribution of fiscal transfers. The Commission also recommended for granting finance to State Governments, which should be utilized by them for strengthening statistical infrastructure at the district level. These requirements are subsequently brought in institutional framework in the implementation of the 13 th Finance Commission. The National Statistical Commission (NSC) in its reports has to also observed the importance of pooling in the statement: The statistical agencies of different State governments have been participating in the NSS programme and canvassing the same questionnaires in matched samples of households in their respective States following identical concepts, definitions and procedures. Results from the central samples and state sample(s) have occasionally been compared. The main purpose of the programme is to pool the two samples and obtain dependable estimates for regions within the States. 1.3 Harmonization of data processing process The harmonization of data processing process is one of the key essences for pooling the different sets of data. The state sample data should be processed using the same set of validation rules as in the case of central sample data. Accordingly, it is essential that the state sample data is processed, ensuring the use of same data entry layout as in the case of central sample. If the states are evolving their own data layout, as per their convenience, then the state data should be put in the layout, harmonized of that with the central data for using the same software developed for central samples. In this concerning (66th)round Uttar Pradesh has used his own developed Data Entry, Validation and Tabulation Softwares for their reports but in case of pooling Uttar Pradesh has recasted their data on the layout according to the central layout and then after pooling exercise is done by using the software supplied by the NSSO (DPD), Uttar Pradesh is using the data entry, validation and tabulation software supplied by the NSSO(DPD) for data entry, validation and tabulation of data from 70th round onwards. so therefore the question of harmonization of data processing process will not arise from 70 th round and onward in UP. 1.4 Methodology of pooling: Two alternate methods are used in pooling the central and state sample data. a) Weighting by Matching ratio: Building aggregate estimate of pooled sample in proportion matching ratio m:n of central and state sample aggregate estimate where m and n are the allotted sample for central and state sample separately for rural and urban sector. Building ratio estimate of pooled sample as ratio of aggregate estimates. 1

8 b) Weighting by inverse of variance: Ratio estimates are built by weighting the ratio estimate of central and state sample in proportion to inverse of variance of ratio of the central and state sample. 1.5 Parameters considered for pooling: Considering the smaller sample size at district level following broad parameters were considered for pooling. a) MPCE of FOOD, Non-FOOD, and Total MPCE derived from detail item for URP, MRP and MMRP b) Household size, sex ratio c) Activity status principal, subsidiary, weekly, daily and their intensity d) District level quantile class computation 1.6 Tables generated Following tables were generated for testing the poolability of State and Central samples; a. District wise results of run test of MPCE (URP, MRP, MMRP) for pooled sample for rural and urban sectors. b. District wise results of chi-square test of persons over worker, unemployed and out of labour force for pooled sample for rural and urban sectors Following Tables were generated pertaining to the estimation of different parameters and their RSEs (%); a. District wise estimated number of households and their RSEs for pooled samples for rural and urban sectors for schedule type-i and schedule type-ii b. District wise estimated number of persons and their RSEs for pooled samples for rural and urban sectors for schedule type-i and schedule type-ii c. District wise estimated sex ratio and their RSEs for pooled samples for rural and urban sectors for schedule type-i and schedule type-ii d. District wise estimate of MPCE (URP, MRP, MMRP) for food, non-food and total for pooled samples (Matching ratio & Inverse variance methods) for rural and urban sectors. e. District wise RSE of estimate of MPCE (URP, MRP, MMRP) for food, non-food and total for pooled samples (Matching ratio & Inverse variance methods) for rural and urban sectors. f. District wise estimated number of male, female and total persons for pooled samples (Matching ratio & Inverse variance methods) for rural and urban sectors (schedule-10). 2

9 g. District wise WPR per 1000(PS+SS) of male, female and total persons for pooled samples (Matching ratio & Inverse variance methods) for rural and urban sectors. h. District wise RSE of WPR (PS+SS) of male, female and total persons for pooled samples (Matching ratio & Inverse variance methods) for rural and urban sectors. i. District wise LFPR per 1000(PS+SS, CWS, CDS) for pooled samples (Matching ratio & Inverse variance methods) for rural and urban sectors. j. District wise RSE of LFPR (PS+SS, CWS, CDS) for pooled samples (Matching ratio & Inverse variance methods) for rural and urban sectors. 1.7 Sample Size of Uttar Pradesh: The sample size of Central and State samples for Uttar Pradesh state is given below; Table 1.1: Sample size of Central and State samples RURAL Schedule Central Sample State Sample FSU Surveyed House holds Surveyed Persons Surveyed FSU Surveyed House holds Surveyed Persons Surveyed (1) (2) (3) (4) (5) (6) (7) 1.0 Type-I Type-II URBAN Schedule Central Sample State Sample FSU Surveyed House holds Surveyed Persons Surveyed FSU Surveyed House holds Surveyed Persons Surveyed (1) (2) (3) (4) (5) (6) (7) 1.0 Type-I Type-II

10 CHAPTER-2 Summary Findings 2.0 Introduction The theoretical framework of pooling and the methodology as well as issues concerning operational and technical aspects of pooling of NSS central and state sample data are given in chapter 3. This chapter focus the results obtained after performing poolability test for schedule 1.0 (type-1 and type-2) and schedule 10. Run test (one sided) was done over MPCE (URP, MRP, MMRP) for testing the poolability of central and state sample data for urban and rural. Number of districts for which poolability was rejected by both the above mentioned tests is given below in table 2.1; Table: 2.1 Number of Districts for which Poolability was rejected over MPCE by run test using Z-statistics (one sided) Type Sector Total number of Rural Urban Districts Run Test URP MRP MMRP In the above statement run test at 1% critical error the null hypothesis is rejected for 8, 9 and 11 rural districts for URP, MRP and MMRP respectively and rejection is 7, 3 and 8 in case of urban districts respectively. Similarly number of districts for which poolability was rejected over 'Worker', 'Unemployed' and 'Out of Labour Force' for pooled Sample by Chi-Square test is given below in table 2.2; Table: 2.2 Number of Districts for which Poolability was rejected over 'Worker', 'Unemployed' and 'Out of Labour Force' for pooled Sample by Chi-Square test Type Sector Total number of Districts Rural Urban Chi-square Test PS + SS CWS CDS

11 2.2 Divergence between the estimates of central and state sample: Household consumer expenditure Considering the district as domain of pooling, the divergence is worked out as absolute percentage difference between central and state sample estimates. The distribution of districts by absolute percentage range of divergence of MPCE (Food, Non-food and Total) of central and state sample in Uttar Pradesh for rural and urban sector and for MPCE (URP, MRP, MMRP) is given in Table: 2.3 to 2.5 below: 1 MPCE: Food Group Rural MPCE: Non Food Group MPCE: Total MPCE: Food Group Urban MPCE: Non Food Group MPCE: Total MPCE: Food Group Rural MPCE: Non Food Group MPCE: Total MPCE: Food Group Urban MPCE: Non Food Group MPCE: Total Table 2.3: Distribution of districts by range of percentage divergence of MPCE(URP) of central and state sample estimates Sl. No. Item sector <=5 % 5-10% 10-15% 15-20% 20-25% 25-30% >30 % Total Districts Table 2.4: Distribution of districts by range of percentage divergence of MPCE(MRP) of central and state sample estimates Sl. No. Item sector <=5% 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Table 2.5: Distribution of districts by range of percentage divergence of MPCE(MMRP) of central and state sample estimates Sl. No. Item sector <=5% 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts 1 MPCE: Food Group Rural MPCE: Non Food Group MPCE: Total MPCE: Food Group Urban MPCE: Non Food Group MPCE: Total

12 The districts with more than 20 percent divergence in total MPCE(URP) were 33 (47%) in the rural sector and 36(51%) in the urban sector out of 70 districts in both sectors. Similarly in Total MPCE (MRP) it is 27 (39%) in rural and 33(47%) in urban sector. In case of MPCE (MMRP) it is 25(36%) in the rural sector and 35(50%) in the urban sector. The distribution of district by range of RSE of MPCE (URP, MRP, MMRP) of central and state sample for rural and urban sector is given in table 2.6 to table Table 2.6: Distribution of districts by range of RSE of MPCE(URP) of central, state and pooled sample of Uttar Pradesh Rural Item <=5 % 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total State Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample(Matching Ratio) MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample (Inverse Variance) MPCE: Food Group MPCE: Non Food Group MPCE: Total

13 Table 2.7: Distribution of districts by range of RSE of MPCE(URP) of central, state and pooled sample of Uttar Pradesh Urban Item <=5 % 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total State Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample(Matching Ratio) MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample (Inverse Variance) MPCE: Food Group MPCE: Non Food Group MPCE: Total Table 2.8: Distribution of districts by range of RSE of MPCE(MRP) of central, state and pooled sample of Uttar Pradesh Rural Item <=5 % 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total State Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample(Matching Ratio) MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample (Inverse Variance) MPCE: Food Group MPCE: Non Food Group MPCE: Total

14 Table 2.9: Distribution of districts by range of RSE of MPCE(MRP) of central, state and pooled sample of Uttar Pradesh Urban Item <=5 % 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total State Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample(Matching Ratio) MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample (Inverse Variance) MPCE: Food Group MPCE: Non Food Group MPCE: Total Table 2.10: Distribution of districts by range of RSE of MPCE(MMRP) of central, state and pooled sample of Uttar Pradesh Rural Item <=5 % 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total State Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample(Matching Ratio) MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample (Inverse Variance) MPCE: Food Group MPCE: Non Food Group MPCE: Total

15 Table 2.11: Distribution of districts by range of RSE of MPCE (MMRP) of central, state and pooled sample of Uttar Pradesh Urban Item <=5 % 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total State Sample MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample(Matching Ratio) MPCE: Food Group MPCE: Non Food Group MPCE: Total Sample (Inverse Variance) MPCE: Food Group MPCE: Non Food Group MPCE: Total From the above tables which shows the distribution of districts by RSE level, it can be seen that the pooled estimates of MPCE on Food, Non-Food and Total have lower RSE (inverse variance) when compared to central and state sample RSE. The RSE of estimates of MPCE (URP, MRP, MMRP) for both rural and urban districts are within 20% for pooled samples (Matching ratio and Inverse Variance methods) except for 3-4 districts. 2.3 Divergence between the estimates of central and state sample : Employment and Un-employment Considering the district as domain of pooling, the divergence is worked out as absolute percentage difference between central and state sample estimates. The distribution of districts by absolute percentage range of divergence of WPR and LFPR (PS+SS, CWS, CDS) of central and state sample in Uttar Pradesh for rural and urban sector is given in table 2.12 below: 9

16 Table 2.12: Distribution of districts by range of percentage divergence of WPR and LFPR (PS+SS, CWS, CDS) of central and state sample estimates Item <=5% 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Rural WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS Urban WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS It can be seen from the above table that the divergence level of approx. 50 percent districts are above 20% for PS+SS and 30 percent districts are above 20% for CWS & CDS in the rural sector. The divergence level of urban sector is much lesser when compared to rural sector of the state. The distribution of districts by range of RSE of WPR and LFPR (PS+SS, CWS, CDS) of central and state sample in Uttar Pradesh for rural and urban sector is given in table: 2.13 & 2.14 respectively. Pooling estimates by using inverse variance method gives more precise values as compared to matching ratio method. More than 90% districts have found RSE of less than 10% with inverse variance method of pooling of estimates. Table 2.13: Distribution of districts by range of RSE of WPR and LFPR (PS+SS, CWS, CDS) of central and state sample of Uttar Pradesh Rural Item <=5% 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS State Sample WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS Sample(Matching Ratio) WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS

17 Item <=5% 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Sample(Inverse Variance) WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS Table 2.14: Distribution of districts by range of RSE of WPR and LFPR (PS+SS, CWS, CDS) of central and state sample of Uttar Pradesh Urban Item <=5% 5-10% 10-15% 15-20% 20-25% 25-30% >30% Total Districts Central Sample WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS State Sample WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS Sample(Matching Ratio) WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS Sample(Inverse Variance) WPR - PS+SS LFPR - PS+SS LFPR - CWS LFPR - CDS

18 CHAPTER-3 Testing poolability and methodology for Pooling 3.1 Testing poolability Though the central sample and state sample are drawn independently following identical sampling design with same concepts, definitions and instructions to collect the state sample data but due to lack of adequate training of field and processing staff of State DESs, unit level data in some cases may not be properly validated. There may also expected agency bias in the two sets of data generated by different agencies. As such they cannot be merged for generating pooled estimate without testing that the samples are realized from identical distribution function. Since the parametric distribution of the sample mean is unknown one may adopt non-parametric tests such Run test, Median test, chi-square test etc to test that the samples are coming from identical distribution function Median test In statistics, the median test is a special case of Pearson's Chi-square test. It tests the null hypothesis that the medians of the populations from which two samples are drawn, are identical. Observations in each sample are assigned to two groups, one consisting of data whose values are higher than the median value in the two groups combined, and the other consisting of data whose values are at the median or below. A Pearson's Chi-square test is then used to determine whether the observed frequencies in each group differ from expected frequencies derived from a distribution combining the two groups. Let m * be the median of the pooled sample data. Construct 2 X 2 contingency table as below and use chi-square test if State sample and Central sample have identical median. Sample-type no of sample observation <= m * > m * State Sample N 11 N 12 N 1. Central Sample N 21 N 22 N 2. Total N.1 N.2 N.. Observed frequency of each cell O ij = N ij where i= 1 to 2, j= 1 to 2. Total Expected frequency of each cell E ij = (N i. * N.j )/N.. where i= 1 to 2, j= 1 to 2. 2 Value = O E O i ( ) / 1 j with degrees of freedom = (2-1)*(2-1) = 1 1 ij ij ij The statistical power of this test may sometimes be improved by using a value other than the median to define the groups say quintile classes that is, by using a value which divides the groups into more nearly equal groups than the median would. 12

19 3.1.2 Multinomial distribution test or 2 test For discrete data such as status of activity, educational level and categorical variable such as land possedetc, standard tests of equality of sample proportions of two sets of data based on multinomial distributions, relevant chi-square tests may be used after grouping the attributes/categorical variables in to a suitable number of classes so that each class contains adequate number of sample observations. Construct 2 X k contingency table for k classes at the domain where two sets of data are to be pooled as below and use chi-square test if State sample and Central sample have identical distribution. Sample-type no of sample observation Class-1 Class-2... Class-k-1 Class-k Total State Sample N 11 N N 1k-1 N 1k N 1. Central Sample N 21 N N 2k-1 N 2k N 2. Total N.1 N.2... N.k-1 N.k N.. Observed frequency of each cell O ij = N ij where i= 1 to 2, j= 1 to k. Expected frequency of each cell E ij = (N i. * N.j )/N.. where i= 1 to 2, j= 1 to k. 2 Value = O E O i ( ) / 1 j with degrees of freedom = (2-1)*(k-1) = k-1 1 ij ij ij Wald-Wolfowitz run test Suppose X and Y are independent random samples with cumulative distribution function (CDF) as F s (x) and F c (y). Null Hypothesis to be tested is H 0: F s (x) = F c (x) for all x against alternative Hypothesis is H 1 : F s (x) <= F c (x) for all x and F s (x) < F c (x)for some x. Let x 1,x 2,..,x m be iid observation from state sample with distributive function F s andy 1,y 2,..,y n be iid observation from central sample with distribution function F c. Pool the data and order them with respect to comparable characteristic under consideration say monthly per capita expenditure (MPCE). In the pooled order sequence put 1 for X and 0 for Y. Let U be the total runs observed where 'run' is a sequence of adjacent equal symbols. For example, following sequence: is divided in six runs, three of them are made out of 1 and the others are made out of 0. The number of runs U is a random variable whose distribution for large sample can be treated as normal with: 2mn mean: 1 m n variance: 2mn(2mn m n) 2 ( m n) ( m n 1) 13

20 After normalizing the variable U one may use one sided z-test for testing the Null hypothesis. In extreme case the value of U will be 2 meaning by observed characteristic of all the observation of one sample is less than the other samples One of the limitations of this test is when there is a tie between two samples in the observed value. One has to resolve ties in usual manner. However if there is large number of ties which is bound to occur specially for qualitative attributes like education level, activity status etc, this test is not recommended. This test can be well applied for a continuous variable such as MPCE which are less prone to ties. For discrete variable chi-square test is recommended Parametric test Aggregate estimate: Let t yc andt ys be theestimate of Y at domain level of pooling based on central and state sample respectively with corresponding variances V(t yc ) and V(t ys ). For large sample, making all assumption of parametric test, one may use Z-Statistic to test the null hypothesis H 0 E(t yc ) = E(t ys ) where E stands for expectation. Z= ( t ( V ( t yc yc t ys ) ) V ( t ys )) V(t yc ) and V(t ys ) could be estimated as ^ ^ 2 2 V( tyc) ( tyc 1 tyc 2) /4, V( tys) ( tys 1 tys2 ) / 4based on sub-sample 1 & 2 l l estimates where stands for summing over stratum x sub-stratum level variance at the l domain of pooling Estimate of rate: Let r c andr s be the estimate of population rates R c and R s iey/x based on central and state sample respectively with corresponding mean square error MSE(r c ) and MSE (r s ). For large sample, making all assumption of parametric test, one may use Z-Statistic to test the null hypothesis H 0 : E(r c )=E(r s ) where E stands for expectation. Z= ( r c r ) ( MSE( r ) MSE( r )) c s s MSE(r c ) and MSE(r s ) are estimated as follows: mse(r c ) = ( V ^ ^ 2 ^ 2 (t yc ) 2 *r c Cov (t yc, t xc ) + r c * V (t xc ))/ t xc mse (r s ) = ( V ^ ^ 2 ^ 2 (t ys ) 2 * r s Cov (t ys, t xs ) + r s * V (t xs ))/ t xs 14

21 where ^ V( t ^ V( t yc xc ^ 2 2 ) ( t t ) /4, V( t ) ( t t ) / 4 l yc1 yc2 ys l ys1 ys2 ^ 2 2 ) ( t t ) /4, V( t ) ( t t ) / 4 l xc1 xc2 xs ^ Cov (t yc, t xc )= ( tyc 1 tyc2 )( txc 1 txc2 )/ 4 where l domain of pooling. l xs1 based on sub-sample 1 & 2 estimates. l stands for summing over stratum x sub-stratum level variance, covariance at the 3.2 Methodology for pooling Pooling by inverse weight of the variance of the estimates Aggregate estimate: For any characteristic, consider the state sample [s] in the form of two independent sub- sample s1 and s2 and the central sample [c] in the form of two independent sub- sample c1 and c2. Based on this, the respective estimates for state and central can be computed as: xs2 t s = l t c= l (t s1 + t s2 )/2 and (t c1 + t c2 )/2 estimate leading to optimum combination of these two estimates is given by weighing with inverse of the variance of the estimate. Thus the pooled estimate is given by: V ( tc) ts V ( ts ) t T p = V ( t ) V ( t ) c ( t c s c withv(t p ) = V ( tc) V ( ts ) V ( t ) V ( t ) In general V ) and V t ) are unknown and can be estimated as ( s ^ ^ 2 2 V( t ) ( t t ) /4, V( t ) ( t t ) / 4 c where l pooling. c s c1 c2 s s1 s2 l l stands for summing over stratum x sub-stratum level variance at the domain of Thus pooled estimate and estimate of pooled variance is given by 15

22 t p = ^ V ( t c ^ ) t V ( t c s ^ V ( t ^ ) V ( t s s ) t ) c ^ (, V ) = t p V ( t ^ ^ V ( t c c ^ ) V ( t ^ s ) V ( t ) s ) By virtue of weighing the two estimates at the domain level at which two estimates are pooled, the pooled estimate will always lie between the central and state sample estimates Estimate of rate: Let r c and r s be the estimate of R c and R s ie Y/X based on central and state sample respectively with corresponding estimated mean square error mse(r c ) and mse (r s ). The pooled estimate and estimate of variance of pooled ratio estimate may be given by: mse( rc ) rs mse( rs ) r r p = mse( r ) mse( r ) c s c, mse r ) = ( p mse( r ) mse( r ) mse( r ) mse( r ) Where mse(r c ) and mse(r s ) are calculated using formula given in para above. Alternatively one can generate the pooled estimate of aggregate by inverse weight of estimate of variance obtained from central and state sample using formula given in para for the characteristics x as well as y and obtain the pooled estimate of ratio as ratio of pooled estimate of aggregate. This will ensure consistency between pooled estimates of aggregate and the pooled estimate of ratio. Let t xp and t yp be the pooled estimate of aggregate for the parameter X and Y. The pooled estimate of R (i.e Y/X) is given by r p= t yp / t xp where t yp = at yc + bt ys and t xp = ct xc + dt xs and (a, b), (c, d) are the estimated inverse variance weight pair of the characteristic x and y respectively. The estimated mse of pooled ratio estimate r p is given by: mse(r p ) = ( ^ V (t yp ) 2 r p ^ Cov (t yp, t xp ) + r p 2 ^ V (t xp ))/ t xp 2 ^ ( where V ) = t yp ab a b, ^ V ( t ) xp = cd c d and ^ Cov (t yp, t xp )= ac Cov ^ (t yc, t xc ) +bd Cov ^ (t ys, t xs ). ^ Cov (t yc, t xc )= ( tyc 1 tyc2 )( txc 1 txc2 )/ 4 l c c s s based on sub-sample 1 & 2 estimates. 16

23 Similarly, where l pooling. ^ Cov (t ys, t xs )= ( tys 1 tys2 )( txs 1 txs2 )/ 4 l stands for summing over stratum x sub-stratum level covariance at the domain of Method laid down in para and requires calculation of estimate of variance of the estimates before pooling them. Reliability of estimate of variance should be ascertained with due consideration of sample size.besides the complex calculations of variances and covariances for each cell of the table, one needs to address the issue of nonadditivity of the component estimates with the estimate of marginal total. For e.g. pooled estimate of MPCE of FOOD and NON-FOOD may not add up to MPCE of TOTAL. To obviate this problem one may generate the pooled estimates of components first and then derive the estimate of total as sum of estimates of components Pooling by simple average of the estimates Many of the States are not fully equipped with complex calculation of estimate of variance especially when cells of the table contains ratio of two characteristics which is usually presented in the NSS reports. When the State s participation is equal matching of central samples, the simple average of two estimates may be a way of combining the estimates considering central and state samples as independent samples. The pooled estimate will always lie between the estimates based on central and state sample separately When the State s participation is of unequal matching of central samples, the weighted average of two estimates with weights being matching ratio of central and state sample may be a better way of combining the estimates considering central and state samples as independent samples. For any characteristic, consider the state sample [s] in the form of two independent sub-sample s1 and s2 and the central sample[c] in the form of two independent sub- sample c1 and c2. Let matching ratio of state and central sample be m : n. Based on this, the respective estimates for state and central can be computed as: t s = l (t s1 + t s2 )/2 and t c= l (t c1 + t c2 )/2 estimate of these two estimates is given by weighing with matching participation rate m:n. Thus the pooled estimate is given by: t p = mts m nt n c withv(t p ) = m 2 2 ( ts ) n ( m n) V V ( t 2 ^ 2 V( t ) ( ) / 4 In general V ( t c ) and V ( t s ) can be estimated as c tc 1 tc 2 l ^ 2 V( ts) ( ts 1 ts 2) /4and thus ^ ( ) l V = t p m 17 c ) ^ 2 2 ( ts) n ( m n) V 2 ^ V ( t c ),

24 The pooled estimate will always lie between the estimates based on central and state sample separately Summing up: For those characteristics which are known to be distributed as Normal, poolability of the two sets of central and state data may be tested by standard parametric tests such as Z-test. For those characteristics for which transformation makes them Normal, such methodology may be adopted. In most of the situations where the distribution is non-normal and unknown, the two sets of data may be tested through various non-parametric tests such as those laid down in para 1 of above. For discrete data, standard tests of equality of proportions based on binomial distribution may be used and for multinomial distributions relevant chisquare tests may be used. 18

25 CHAPTER-4 Poolability Test Results Poolability test were undertaken and the test results are given below; Table 4.1(R): Districtwise results of test of MPCE (URP, MRP, MMRP) for pooled sample State : Uttar Pradesh Sector: Rural RUN TEST Z 0.01 = [one sided test] reject if z-value<z 0.01 District URP MRP MMRP District name code z-value accept z-value accept z-value accept (1) (2) (3) (4) (5) (6) (7) (8) 1 SAHARANPUR 0.27 Y Y 0.00 Y 2 MUZAFFARNAGAR Y Y 0.50 Y 3 BIJNOR Y Y Y 4 MORADABAD 1.42 Y Y 1.07 Y 5 RAMPUR 0.76 Y Y Y 6 JYOTIBAPHULENAGAR Y Y 0.27 Y 7 MERRUT Y Y Y 8 BAGPAT Y Y 0.60 Y 9 GAZIABAD Y Y Y 10 GAUTAMBUDHNAGAR 0.11 Y 0.43 Y Y 11 BULANDSHAHAR Y Y Y 12 ALIGARH 0.27 Y Y Y 13 HATHRAS 0.43 Y 0.27 Y Y 14 MATHURA 0.11 Y Y Y 15 AGRA Y 1.41 Y Y 16 FIROZABAD 0.06 Y 0.06 Y Y 17 ETAH Y Y N 18 MAINPURI 1.25 Y 0.76 Y Y 19 BADAUN Y 0.15 Y 1.19 Y 20 BAREILLY 0.38 Y Y Y 21 PILIBHIT 0.43 Y 0.11 Y 0.76 Y 22 SHAHJAHANPUR Y Y Y 23 KHERI 0.84 Y Y Y 24 SITAPUR Y N N 25 HARDOI Y Y 1.99 Y 26 UNNAO 0.80 Y 1.86 Y 1.86 Y 27 LUCKNOW Y 0.43 Y Y 28 RAIBAREILLY Y N N 29 FARRUKHABAD N 0.19 Y Y 30 KANNAUJ Y Y 0.11 Y 31 ETAWAH Y Y Y 32 AURAIYA Y Y 0.11 Y 19

26 District code District name URP MRP MMRP z-value accept z-value accept z-value accept (1) (2) (3) (4) (5) (6) (7) (8) 33 KANPUR DEHAT 0.43 Y Y Y 34 KANPUR NAGAR 0.11 Y 0.11 Y 0.76 Y 35 JALAUN Y Y N 36 JHANSI Y 0.27 Y Y 37 LALITPUR N N N 38 HAMIRPUR Y Y Y 39 MAHOBA N N N 40 BANDA Y N Y 41 CHITRAKOOT Y 0.27 Y Y 42 FATEHPUR Y 0.11 Y 1.09 Y 43 PRATAPGARH 0.84 Y Y 0.37 Y 44 KAUSHAMBI Y Y N 45 ALLAHABAD Y Y Y 46 BARABANKI Y 0.27 Y Y 47 FAIZABAD 0.11 Y Y Y 48 AMBEDKARNAGAR Y Y Y 49 SULTANPUR 0.15 Y Y 1.07 Y 50 BAHRAICH Y N Y 51 SHRAWASTI Y Y Y 52 BALRAMPUR N Y N 53 GONDA Y Y 1.19 Y 54 SIDDHARTHNAGAR N Y Y 55 BASTI 0.43 Y Y Y 56 SANTKABIRNAGAR 0.60 Y Y Y 57 MAHARAJGANJ Y Y Y 58 GORAKHPUR Y Y Y 59 KUSHINAGAR 0.73 Y Y 0.38 Y 60 DEORIA Y N N 61 AZAMGARH Y Y Y 62 MAU N N N 63 BALLIA Y Y 0.15 Y 64 JAUNPUR 0.15 Y N Y 65 GAZIPUR Y Y Y 66 CHANDAULLI Y Y Y 67 VARANASI 0.09 Y Y 0.42 Y 68 SANTRAVIDASNAGAR Y 0.76 Y 0.43 Y 69 MIRZAPUR N 0.08 Y Y 70 SONBHADRA N Y N 20

27 Table 4.1(U): Districtwise results of test of MPCE (URP, MRP, MMRP) for pooled sample State : Uttar Pradesh Sector: Urban RUN TEST Z 0.01 = [one sided test] reject if z-value<z 0.01 District code District name URP MRP MMRP z-value accept z-value accept z-value accept (1) (2) (3) (4) (5) (6) (7) (8) 1 SAHARANPUR 0.60 Y 0.11 Y Y 2 MUZAFFARNAGAR 0.60 Y Y N 3 BIJNOR Y Y Y 4 MORADABAD Y 0.15 Y Y 5 RAMPUR Y Y Y 6 JYOTIBAPHULENAGAR 2.39 Y Y 0.54 Y 7 MERRUT Y Y Y 8 BAGPAT 1.70 Y Y 0.52 Y 9 GAZIABAD Y Y Y 10 GAUTAMBUDHNAGAR Y Y Y 11 BULANDSHAHAR Y Y Y 12 ALIGARH 0.43 Y 0.43 Y 0.27 Y 13 HATHRAS Y 0.77 Y Y 14 MATHURA Y Y Y 15 AGRA 0.27 Y Y Y 16 FIROZABAD 0.43 Y Y Y 17 ETAH 2.90 Y Y 1.23 Y 18 MAINPURI Y 1.46 Y N 19 BADAUN Y Y Y 20 BAREILLY Y Y Y 21 PILIBHIT Y Y Y 22 SHAHJAHANPUR Y Y Y 23 KHERI 0.08 Y 0.54 Y Y 24 SITAPUR N Y Y 25 HARDOI 0.77 Y Y Y 26 UNNAO 1.00 Y Y N 27 LUCKNOW Y 0.38 Y 0.50 Y 28 RAIBAREILLY Y Y Y 29 FARRUKHABAD Y 1.00 Y Y 30 KANNAUJ 0.08 Y Y Y 31 ETAWAH Y Y Y 32 AURAIYA 1.70 Y 0.31 Y N 33 KANPUR DEHAT Y Y 1.00 Y 34 KANPUR NAGAR Y 0.38 Y 0.27 Y 35 JALAUN Y Y Y 36 JHANSI Y Y Y 37 LALITPUR 1.70 Y 0.08 Y 0.08 Y 21

28 District code District name URP MRP MMRP z-value accept z-value accept z-value accept (1) (2) (3) (4) (5) (6) (7) (8) 38 HAMIRPUR Y Y 0.08 Y 39 MAHOBA Y Y Y 40 BANDA Y 0.08 Y Y 41 CHITRAKOOT 0.77 Y Y Y 42 FATEHPUR N Y 0.52 Y 43 PRATAPGARH 0.31 Y 1.23 Y Y 44 KAUSHAMBI Y Y Y 45 ALLAHABAD Y Y Y 46 BARABANKI 1.46 Y 2.39 Y 1.23 Y 47 FAIZABAD 1.23 Y 0.29 Y 0.08 Y 48 AMBEDKARNAGAR 0.77 Y 0.31 Y Y 49 SULTANPUR Y Y Y 50 BAHRAICH Y Y 0.08 Y 51 SHRAWASTI 0.77 Y 0.77 Y Y 52 BALRAMPUR N Y N 53 GONDA Y 0.08 Y Y 54 SIDDHARTHNAGAR Y 0.77 Y Y 55 BASTI 0.08 Y 0.08 Y 1.93 Y 56 SANTKABIRNAGAR N N Y 57 MAHARAJGANJ N N N 58 GORAKHPUR 0.02 Y 1.06 Y 0.19 Y 59 KUSHINAGAR Y Y Y 60 DEORIA Y Y Y 61 AZAMGARH 0.54 Y Y Y 62 MAU N N N 63 BALLIA Y Y Y 64 JAUNPUR Y Y Y 65 GAZIPUR 0.54 Y Y 0.08 Y 66 CHANDAULLI Y Y Y 67 VARANASI Y Y Y 68 SANTRAVIDASNAGAR N Y N 69 MIRZAPUR 0.54 Y Y Y 70 SONBHADRA Y Y 1.00 Y 22

29 Table 4.2(R) : Districtwise results of test of persons over worker,unemployed and out of labour force for pooled sample State: Uttar Pradesh Sector: Rural (Sch 10) Chi-square Test χ 2.01=9.21 df=2 [one sided test] reject if χ 2 -value > χ 2.01 District PS+SS CWS CDS District name code χ 2 _ value accept χ 2 _ value accept χ 2 _ value accept (1) (2) (3) (4) (5) (6) (7) (8) 1 SAHARANPUR N 9.64 N 5.22 Y 2 MUZAFFARNAGAR 3.83 Y 2.81 Y 1.97 Y 3 BIJNOR 4.05 Y 5.82 Y N 4 MORADABAD 5.28 Y N N 5 RAMPUR 1.90 Y 2.98 Y N 6 JYOTIBAPHULENAGAR 0.28 Y 0.28 Y N 7 MERRUT N N N 8 BAGPAT 1.44 Y 1.94 Y 3.74 Y 9 GAZIABAD N N N 10 GAUTAMBUDHNAGAR 1.11 Y 0.83 Y N 11 BULANDSHAHAR 0.15 Y 2.83 Y 1.56 Y 12 ALIGARH 8.77 Y N N 13 HATHRAS 6.92 Y 6.02 Y 6.88 Y 14 MATHURA 0.78 Y 0.45 Y 4.46 Y 15 AGRA 5.06 Y 5.06 Y 6.60 Y 16 FIROZABAD 7.30 Y 7.05 Y 6.90 Y 17 ETAH 6.81 Y 3.42 Y 3.62 Y 18 MAINPURI 0.74 Y 0.72 Y 5.04 Y 19 BADAUN 0.29 Y 0.27 Y 7.50 Y 20 BAREILLY 0.12 Y 2.31 Y 8.40 Y 21 PILIBHIT 0.09 Y 0.11 Y 0.80 Y 22 SHAHJAHANPUR 7.37 Y 9.52 N 7.34 Y 23 KHERI 2.63 Y 4.51 Y N 24 SITAPUR 0.01 Y 6.73 Y 2.87 Y 25 HARDOI 2.65 Y 4.96 Y 1.58 Y 26 UNNAO N 9.54 N N 27 LUCKNOW 1.47 Y 0.80 Y 1.05 Y 28 RAIBAREILLY 0.14 Y 1.90 Y 0.79 Y 29 FARRUKHABAD 2.59 Y 8.09 Y N 30 KANNAUJ 1.68 Y 1.33 Y 0.11 Y 31 ETAWAH 5.22 Y 6.67 Y N 32 AURAIYA 1.45 Y 3.02 Y N 33 KANPUR DEHAT 3.57 Y 1.07 Y 5.59 Y 34 KANPUR NAGAR 2.51 Y 0.77 Y 0.89 Y 35 JALAUN N N N 36 JHANSI 6.37 Y N N 37 LALITPUR N N N 23

30 District PS+SS CWS CDS District name code χ 2 _ value accept χ 2 _ value accept χ 2 _ value accept (1) (2) (3) (4) (5) (6) (7) (8) 38 HAMIRPUR N N N 39 MAHOBA N 3.64 Y N 40 BANDA N 2.37 Y 4.67 Y 41 CHITRAKOOT 2.16 Y 1.75 Y 2.16 Y 42 FATEHPUR N N N 43 PRATAPGARH 5.32 Y 1.79 Y 4.15 Y 44 KAUSHAMBI N N N 45 ALLAHABAD N N N 46 BARABANKI 1.46 Y 6.23 Y N 47 FAIZABAD N N N 48 AMBEDKARNAGAR N 2.15 Y 4.99 Y 49 SULTANPUR N 5.55 Y 1.59 Y 50 BAHRAICH 0.63 Y 0.26 Y 6.53 Y 51 SHRAWASTI 5.72 Y 5.26 Y 8.95 Y 52 BALRAMPUR 2.10 Y 7.17 Y 1.60 Y 53 GONDA 0.17 Y 0.31 Y 4.72 Y 54 SIDDHARTHNAGAR 4.36 Y N N 55 BASTI N N N 56 SANTKABIRNAGAR 0.57 Y 0.56 Y 1.62 Y 57 MAHARAJGANJ 2.78 Y 4.27 Y 4.83 Y 58 GORAKHPUR N 9.20 Y N 59 KUSHINAGAR 0.16 Y N 6.49 Y 60 DEORIA N 1.80 Y 2.56 Y 61 AZAMGARH N N N 62 MAU N N N 63 BALLIA 4.13 Y 4.57 Y N 64 JAUNPUR N 2.61 Y N 65 GAZIPUR N N N 66 CHANDAULLI 4.43 Y 1.00 Y N 67 VARANASI N N N 68 SANTRAVIDASNAGAR 0.90 Y 3.39 Y 4.59 Y 69 MIRZAPUR N 3.77 Y N 70 SONBHADRA N 2.78 Y N 24

31 Table 4.2(U) : Districtwise results of test of persons over worker,unemployed and out of labour force for pooled sample State: Uttar Pradesh Sector: Urban (Sch 10) Chi-square Test χ 2.01=9.21 df=2 [one sided test] reject if χ 2 -value > χ 2.01 District PS+SS CWS CDS District name code χ 2 _ value accept χ 2 _ value accept χ 2 _ value accept (1) (2) (3) (4) (5) (6) (7) (8) 1 SAHARANPUR 1.05 Y 1.98 Y 7.47 Y 2 MUZAFFARNAGAR 2.01 Y 2.04 Y 2.02 Y 3 BIJNOR 3.72 Y 3.21 Y 5.38 Y 4 MORADABAD 9.26 N 9.26 N 8.03 Y 5 RAMPUR 7.05 Y 7.05 Y N 6 JYOTIBAPHULENAGAR 1.22 Y 2.39 Y 2.03 Y 7 MERRUT 0.20 Y 0.28 Y 0.74 Y 8 BAGPAT 0.89 Y 2.23 Y N 9 GAZIABAD 7.06 Y 8.44 Y N 10 GAUTAMBUDHNAGAR 0.40 Y 0.40 Y 1.60 Y 11 BULANDSHAHAR 0.87 Y 2.04 Y 2.78 Y 12 ALIGARH N N N 13 HATHRAS 1.35 Y 2.72 Y 3.26 Y 14 MATHURA 9.18 Y 9.18 Y 6.78 Y 15 AGRA 1.06 Y 1.06 Y 3.58 Y 16 FIROZABAD N 9.78 N N 17 ETAH 2.84 Y 3.98 Y 2.75 Y 18 MAINPURI 2.54 Y 3.34 Y N 19 BADAUN 0.62 Y 0.74 Y 5.56 Y 20 BAREILLY 1.05 Y 1.05 Y 1.43 Y 21 PILIBHIT 2.61 Y 4.47 Y 3.57 Y 22 SHAHJAHANPUR 1.29 Y 1.40 Y 1.35 Y 23 KHERI 2.65 Y 2.65 Y 4.09 Y 24 SITAPUR 1.10 Y 0.98 Y 3.77 Y 25 HARDOI 1.10 Y 1.65 Y 1.71 Y 26 UNNAO 1.25 Y 1.43 Y 8.30 Y 27 LUCKNOW 9.98 N N N 28 RAIBAREILLY 6.94 Y 4.99 Y 4.98 Y 29 FARRUKHABAD 1.67 Y 1.67 Y 3.61 Y 30 KANNAUJ 0.99 Y 1.08 Y 0.75 Y 31 ETAWAH 0.36 Y 0.36 Y 0.36 Y 32 AURAIYA 0.41 Y 0.44 Y 0.43 Y 33 KANPUR DEHAT 1.46 Y 2.88 Y 3.37 Y 34 KANPUR NAGAR 4.37 Y 4.27 Y 6.81 Y 35 JALAUN 3.07 Y 2.27 Y 2.54 Y 36 JHANSI 8.82 Y 8.82 Y 7.58 Y 37 LALITPUR 0.28 Y 0.36 Y 0.43 Y 25