International Journal of Agriculture Sciences ISSN: 0975-3710 & E-ISSN: 0975-9107, Volume 11, Issue 4, 2019, pp.-7904-7909. Available online at https://www.bioinfopublication.org/jouarchive.php?opt=&jouid=BPJ0000217

Research Article REGIONAL DISPARITIES IN THE LEVELS OF DEVELOPMENT IN

VERMA S.1, TRIPATHI R.K.2 AND NITIN TANWAR*1 1Department of Mathematics and Statistics, Chaudhary Charan Singh Haryana Agricultural University, , 125004, Haryana, 2Department of Agricultural Statistics, Narendra Deva University of Agriculture & Technology, Kumarganj, Faizabad, 224 229, Uttar Pradesh, India *Corresponding Author: Email - [email protected]

Received: February 04, 2019; Revised: February 22, 2019; Accepted: February 23, 2019; Published: February 28, 2019 Abstract: As per socio-economic status of the Odisha state, it can be divided into two broad regions, i.e. the coastal region and inland districts. The former is fertile with high yielding capacity of agriculture comprising high proportion upper-caste population. The hilly and barren districts covering with forest are in primitive stage of economic comprising with ST and SC population. It is needless to mention that socioeconomic backwardness of a region is the root cause of mass poverty. In the present study, all the districts of Odisha were ranked on the basis of their levels of development obtained with the help of 30 indicators related to agriculture, social and industrial sectors. The district wise data in respect of the indicators published by Odisha government for the year 2014-15 have been used for all the districts of the Odisha state. The statistical technique composite index method has been used in addition to Principal Component Analysis (PCA) for ranking the districts. Cluster analysis has been used for classification or grouping the districts. The Composite Indices (C.I.) of development in respect of 30 developmental indicators for all the districts of Odisha has been computed for the year 2014-15. The districts of Ganjam, Mayurbhanj, , Sundargarh and Koraput were found to be most developed districts while the district of Deogarh was found to be most backward followed by the districts of , Boudh, Subarnapur and Nuapada in overall development. Keywords: Composite index, growing Developmental indicators, Socio-economic development, PCA, Cluster analysis Citation: Verma S., et al., (2019) Regional Disparities in the Levels of Development in Odisha. International Journal of Agriculture Sciences, ISSN: 0975-3710 & E-ISSN: 0975-9107, Volume 11, Issue 4, pp.- 7904-7909. Copyright: Copyright©2019 Verma S., et al., This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution and reproduction in any medium, provided the original author and source are credited.

Introduction The economy of Odisha state is one of the fastest state economies in India. and Andhra Pradesh toward the south. Odisha has 485 kilometres of coastline Odisha has a farming-based economy which is on the move towards an industry along the Bay of Bengal on its east, from to Ganjam. It is the ninth and administration-based economy. Development is a dynamic concept and has largest state by area, and the eleventh biggest by populace. It is likewise the third different meaning for different people. The notion of development in the context of most populous state of India in terms of tribal population. Orissa has changed regional development, which aims to enhance the levels of living of the people and from a genuinely slacking state to a state moving. From being the poorest state of general conditions of human welfare in a region [1]. Socio-economic improvement India in the mid-1990s, Orissa has turned into a state moving. The state's is the procedure of social and monetary advancement in a society. It is estimated economy has changed gear and is on a higher development direction. As per with economic variables, for example, gross domestic product (GDP), life socio-economic status of the state; it very well may be disconnected into two wide expectancy, literacy and levels of employment. Social advancement is a regions, for example the coastal area and inland areas. The previous is fruitful procedure which brings about the change of social organizations in such a way with high yielding limit of horticulture involving high extent upper cast population. which enhances the limit of the general public to satisfy its desires. It suggests a Then again, the inland regions are sloping and infertile land with covering of subjective change in the way the society shapes itself and completes its forest, including ST and SC population with crude phase of financial. In Odisha, exercises, for example, through more dynamic states of mind and conduct by the regions like the southern and northern are not well developed as compared to the populace, the appropriation of more viable procedures or further developed coastal region. One of the important reasons for this is the higher concentration of innovation. There is a nearby connection among conditions, methods for living and unfertile land and STs Population. About 44 percent of the total land area in innovation. Advancement in the economy of countries is the improvement of Odisha has been declared as Scheduled area. The total tribal population of the monetary abundance of nations or locales for the prosperity of their occupants. State is 8.15 million, which constitute 22.13 percent. According to the 2011 Economic enlargement is regularly accepted to point out the dimension of Census of India, 61.8 percent of the working population are engaged in monetary advancement of a country. The expression " economic growth " alludes agricultural activities. However, the agricultural contribution to the Gross State to the expansion (or development) of a particular measure, for example, real Domestic Produce (GSDP) was 16.3 percent in the fiscal year 2013-14 and it was national income, gross domestic product, or per capita income. It is the procedure estimated to be 15.4 percent in 2014-15. The area under cultivation was 5,691 by which a country enhances the economic, political, and social prosperity of its hectares in 2005-06 and it dropped to 5,424 hectares in 2013-14. Rice is the kin. Socio-economic development, in this manner, is a procedure of change in an dominant crop in Odisha. It is grown on 77 percent of the area under cultivation. assortment of ways. It needs to impact all parts of human life in a nation. Odisha Odisha has produced 8,360 metric huge amounts of rice in 2013-14, a drop from (once in the past known as Orissa) is one of the 29 states of India, located in 10,210 metric tons due to the violent . The industrial sector eastern India. It is surrounded by the conditions of West Bengal toward the north- contribution to the state's GSDP was estimated at 33.45 percent in 2014-15. east, toward the north, Chhattisgarh toward the west and north-west, Odisha has 25 percent of India's iron reserves.

International Journal of Agriculture Sciences ISSN: 0975-3710&E-ISSN: 0975-9107, Volume 11, Issue 4, 2019 || Bioinfo Publications || 7904 Regional Disparities in the Levels of Development in Odisha

It has 10 percent of India's production capacity in steel. Odisha is the top Total number of Secondary schools aluminium producing state in India. Mining contributed an estimated 6.31 percent Total number of government degree colleges to the GSDP [2]. The Green Revolution in the agriculture sector and excellent Total number of Community Health Centres advance on the industrial front has surely expanded the overall total production, Total number of Beds in Hospital yet there is no sign that these accomplishments have possessed the capacity to Total number of Primary Health Centres diminish generously the provincial disparities in the level of development [3]. Total number of Post Office Although resource transfers are being executed in backward region of country, it Total number of Villages electrified has been observed that the regional disparities exist in terms of socio-economic Total number of Commercial Banks development are not declining over time [4]. [5] developed composite indices (C.I.) Rural Road Length (km) of development in respect of 17 developmental indicators for 32 WSHGs doing Total literacy rate aquaculture activities in Keonjhar and Koraput districts of Odisha was estimated in three dimensions-economic, social and empowerment along with overall Method of Analysis development for the year 2008-09. An investigation of districts of Himachal Method of estimation of Composite Index of development [7]: Pradesh by [6] have attempted a few strategy measures to improve the levels of Let [Xij] be data matrix giving the values of the variables of ith district. Where i = 1, the advancement in horticulture, social and industrial segments utilizing auxiliary 2… n (number of districts) and j = 1, 2… k (number of indicators). information on various developmental indicators published by Himachal Pradesh For combined analysis [Xij] is transferred to [Zij] the matrix of standardized government for the year 2014-15. The present study is conducted in Odisha state indicators as follows 푋푖푗−푋̅푗 at district level, where the data on socio-economic variables for the year 20014-15 푍푖푗 = (1) were analysed for estimating the level of development. The level of development 푆푗

is estimated separately for agricultural sector, infrastructural facilities and overall Where, Sj = Standard deviation of jth indicator socio-economic field. It will be of interest to estimate the level of development at 푋̅ = mean of the jth indicator district level since there has been a growing consensus about the need of district From [Zij], identify the best value of each indicator. Let it be denoted as Zoj. The level planning in the country. Knowledge of level of development at district level best value will be either the maximum value or the minimum value of the indicator will help in identifying where a given district stands in relation to others. depending upon the direction of the impact of indicator on the level of

development. For obtaining the pattern of development Ci of ith districts, first Materials and Methods calculate Pij as follows: The present study comprised 30 districts of Odisha (Appendix 1). Each district Pij = (Zij –Zoj)2 (2) faced situational factors of development unique to it as well as common Pattern of development is given by administrative and financial factors. Each district confronted situational 1/2 C = ∑k P /(CV) (3) components of advancement one of a kind to it and also regular administrative i [ j=1 ij j] th and finance related variables. Factors common to all the districts were taken as Where, (CV)j = coefficient of variation in Xij for j indicator. the indicators of development. A total of 30 developmental indicators from which Composite index of development (C.I.) is given by 19 indicators related to agriculture sector to examine agricultural development and C.I. = Ci / C for i = 1, 2, …, n (4) 11 indicators from socio-economic sector to examine socio-economic C = C +3SDi (5) development. The district wise data in respect of these indicators published by Where C = mean of Ci and SDi = Standard deviation of Ci Odisha government were used for all districts in Odisha for the year 2014-15. Smaller value of C.I. will indicate high level of development and higher value of These indicators are the major interacting components of development. The C.I. will indicate low level of development. composite indices of development for different districts were obtained by using the data on the development indicators. Principal component analysis Principal component analysis (PCA) was invented by [8], as an analogue of the Developmental Indicators principal axis theorem in mechanics. Suppose we have a random vector The composite indices of development for different districts were obtained by population X, where T using the data on the following development indicators: X = (X1,X2,...... Xn ) (6) Total geographical area (square kilometres) And the mean of that population is denoted by Total cropped area (hectare) µx = E {X} (7) Net sown area (hectare) And the covariance matrix of the same data set is T Production of Paddy (quintal) CX = E(X − μX )(X − μX )  (8) Production of Maize (quintal) The components of Cx denoted by Cij represent the covariance between the Production of Groundnut (quintal) random variable components Xi and Xj. The component Cij is the variance of the Production of Mung (quintal) component Xi. The variance of a component indicates the spread of the Production of Biri (quintal) component values around its mean value. If two components Xi and Xj of the data Production of Kulthi (quintal) are uncorrelated, their covariance is zero, i.e. Production of Potato (quintal) Cij = C ji = 0 Production of Sugarcane (quintal) (9) Fertilizer consumption (Kg/hectare) The covariance matrix is always symmetric. Total Forest area under revenue village (hectare) From a sample of vectors X1, X2………, Xm, we can calculate the sample mean Total number of Livestock Hospitals and Dispensaries and the sample covariance matrix as the estimates of the mean and the Total number of Cattle covariance matrix. Total number of Buffalos From a symmetric matrix such as the covariance matrix, we can calculate an Total number of Goats orthogonal basis by finding its eigenvalues and eigenvectors. The eigenvectors ei Production of fish and the corresponding eigenvalues λi are the solutions of the equation Production of Poultry CX ei = λiei, i =1, 2, 3, ……, n (10) Total number of Primary schools For simplicity we assume that the λi are distinct.

International Journal of Agriculture Sciences ISSN: 0975-3710&E-ISSN: 0975-9107, Volume 11, Issue 4, 2019 || Bioinfo Publications || 7905 Verma S., Tripathi R.K. and Nitin Tanwar

Table-1 Composite index (C.I.) of development in different sectors Districts Agriculture Sector Socioeconomic Sector Overall (All Sectors) C.I Rank C.I Rank C.I Rank Nabarangpur 0.000 1 0.818 23 0.376 9 Koraput 0.100 2 0.650 13 0.307 5 Ganjam 0.102 3 0.000 1 0.000 1 Bolangir 0.189 4 0.592 11 0.317 6 Sundargarh 0.227 5 0.416 4 0.237 4 Mayurbhanj 0.233 6 0.241 2 0.160 2 Cuttack 0.284 7 0.258 3 0.196 3 Baragarh 0.345 8 0.684 15 0.458 12 Kalahandi 0.352 9 0.569 9 0.391 10 Keonjhar 0.406 10 0.459 6 0.361 7 Balasore 0.448 11 0.418 5 0.364 8 0.453 12 0.549 8 0.435 11 0.518 13 0.740 20 0.586 16 0.542 14 0.701 17 0.574 15 0.560 15 0.580 10 0.513 13 Rayagada 0.570 16 0.722 19 0.602 18 0.648 17 0.625 12 0.586 17 Nayagarh 0.671 18 0.774 22 0.689 23 Jagatsinghpur 0.676 19 0.756 21 0.680 22 Malkangiri 0.687 20 0.876 25 0.763 24 0.718 21 0.717 18 0.679 21 0.727 22 0.685 16 0.664 20 Kandhamal 0.732 23 0.681 14 0.664 19 Khurda 0.735 24 0.485 7 0.562 14 Nuapada 0.815 25 0.931 27 0.866 26 Gajapati 0.887 26 0.832 24 0.841 25 Subarnapur 0.892 27 0.926 26 0.903 27 Boudh 0.934 28 0.934 28 0.931 28 Deogarh 0.983 29 1.000 30 1.000 30 Jharsuguda 1.000 30 0.961 29 0.983 29

Table-2 Population at different stages of development Category Stage of Agricultural and socio-economic development of districts Population development (%) Category- I High Cuttack, Ganjam, Mayurbhanj, Sundargarh 25.65 Category- II High middle Balasore, Baragarh, Bolangir, Jajpur, Kalahandi, Keonjhar, Koraput, 35.63 Nabarangpur, Puri Category- III Low middle Angul, Bhadrak, Dhenkanal, Jagatsinghpur, Kandhamal, Kendrapara, Khurda, 31.26 Malkangiri, Nayagarh, Rayagada, Sambalpur Category- IV Low Boudh, Deogarh, Gajapati, Jharsuguda, Nuapada, Subarnapur 7.46

Table-3 Summary statistics of various clusters Cluster 1 2 3 4 Objects 15 6 4 5 Sum of weights 15 6 4 5 Within-class variance 948705515351.48 977943049117.18 4116131183378.74 2168977850428.24 Minimum distance to centroid 212148.78 591406.15 569684.81 538958.51 Average distance to centroid 788236.17 873380.20 1593480.36 1196374.97 Maximum distance to centroid 2218901.24 1266393.93 2650647.43 2176550.38

Dendrogram 2E+14 1.8E+14 1.6E+14 1.4E+14 1.2E+14 1E+14

8E+13 Dissimilarity 6E+13 4E+13 2E+13

0

Puri

Jajpur

Angul

Boudh

Khurda

Cuttack

Ganjam

Koraput

Gajapati

Bhadrak

Deogarh

Bolangir

Balasore

Nuapada

Baragarh

Keonjhar

Nayagarh

Rayagada

Kalahandi

Sambalpur Dhenkanal

Malkangiri

Jharsuguda

Kandhamal Sundargarh Subarnapur

Kendrapara

Mayurbhanj

Nabarangpur Jagatsinghpur

Fig-1 Dendrogram-Ward’s method

International Journal of Agriculture Sciences ISSN: 0975-3710&E-ISSN: 0975-9107, Volume 11, Issue 4, 2019 || Bioinfo Publications || 7906 Verma S., Tripathi R.K. and Nitin Tanwar

Scree plot

15 100 80 10 60 40

Eigenvalue 5 20

0 0 %) variability Cumulative F1 F2 F3 F4 F5 F6 F7 F8 F9 F10F11F12F13F14F15F16F17F18F19F20F21F22F23F24F25F26F27F28F29 axis

These values can be found, for example, by finding the solutions of the presented in [Table-1]. The perusal of [Table-1] reveals that for agricultural characteristic equation development, the district of Nabarangpur (C.I. = 0.000) was found to be most CX − λI = 0 (11) developed district in Odisha state followed by the districts of Koraput (C.I. = Where, I, is the identity matrix. 0.100), Ganjam (C.I. = 0.102), Bolangir (C.I. = 0.189) and Sundargarh (C.I. = 0.227) whereas the district of Jharsuguda (C.I. = 1.00) was found least developed Cluster Analysis followed by the districts of Deogarh (C.I. = 0.983), Boudh (C.I. = 0.934), Cluster analysis is a multivariate statistical technique, which identifies groups of Subarnapur (C.I. = 0.892) and Gajapati (C.I. = 0.887). Districts of Dhenkanal (C.I. samples that behave similarly or show similar characteristics. In common parlance = 0.518), Angul (C.I. = 0.542), Jajpur (C.I. = 0.560) and Rayagada (C.I. = 0.570) it is also called look-a-like groups. were found to be moderate districts regarding agricultural development. In reference to socio-economic development the district of Ganjam (C.I. = 0.000) Ward’s Method was found top most developed followed by the districts of Mayurbhanj (C.I. = Ward’s method, also called the incremental sum of squares method, uses within 0.241), Cuttack (C.I. = 0.258) Sundargarh (C.I. = 0.416) and Balasore (C.I. = cluster (squared) distances and the between-cluster (squared) distances [9, 10]. If 0.418) whereas the district of Deogarh has high composite index score (C.I. = AB is the cluster obtained by combining clusters A and B, then the sum of within- 100), was found least developed followed by the districts of Jharsuguda (C.I. = cluster distances (of the items from the cluster mean vectors) are 0.961), Boudh (C.I. = 0.934), Naupada (C.I. = 0.931) and Subarnapur (C.I. = nA 0.926). Districts of Koraput (C.I. = 0.650), Kandhamal (C.I. = 0.681), Baragarh  SSEA = (yi − yA ) (yi − yA ) (C.I. = 0.684) and Bhadrak (C.I. = 0.685) were found moderately developed i=1 (12) districts in case of socio-economic development. Regarding overall development nB  the district of Ganjam (C.I. = 0.000) was found most developed district followed by SSEB = (yi − yB ) (yi − yB ) i=1 (13) the districts of Mayurbhanj (C.I. = 0.160), Cuttack (C.I. = 0.196), Sundargarh (C.I. n AB  = 0.237) and Koraput (C.I. = 0.307) while the district of Deogarh (C.I. = 1.00) was SSEAB = (yi − yAB ) (yi − yAB ) i=1 (14) ranked last among all the districts in case of overall districts followed by the Where and nA, nB, and nAB=nA+ nB are the numbers of points in A, B, and AB, districts of Jharsuguda (C.I. = 0.983), Boudh (C.I. = 0.931), Subarnapur (C.I. = respectively. Since these sums of distances are equivalent to within-cluster sums 0.903) and Naupada (C.I. = 0.866). Districts of Jajpur (C.I. = 0.513), Khurda (C.I. = of squares, they are denoted by SSEA, SSEB, and SSEAB. 0.562), Angul (C.I. = 0.574) and Dhenkanal (C.I. = 0.586) were found moderately Ward’s method joins the two clusters A and B that minimize the increase in SSE, developed districts in case of overall development. A simple ranking of the districts defined as: on the basis of composite indices would be sufficient for classificatory purposes. A IAB= SSEAB − (SSEA+ SSEB) (15) suitable fractile classification of the districts from the assumed distribution of the It can be shown that the increase IAB in (15) has the following two equivalent mean of the composite indices will provide a more meaningful characterization of forms: different stages of development. The fractile groups can be used to classify the I = n (y − y )(y − y ) + n (y − y )(y − y ) various stages of development. For relative comparison, it appears appropriate to AB A A AB A AB B B AB B AB (16) assume that the districts having composite index less than or equal to (Mean-SD) n n = A B (y − y )(y − y ) are highly developed and these districts are classified in category-I of developed n + n A B A B A B (17) districts and the districts having composite index greater than (Mean+SD) are low Thus, by (17), minimizing the increase in SSE is equivalent to minimizing the developed and are classified in category-IV of low developed districts. Districts between -cluster distances. Dendrograms are used to display the cluster hierarchy with composite index lying between (Mean) and (Mean-SD) are high medium level and the distances at which the clusters were joined which can be useful when developed and these districts are put in category-II and the districts with selecting an appropriate number of clusters for the dataset. Dendrograms composite index lying between (Mean) and (Mean+SD) are classified in category- graphically present the information concerning which observations are grouped III or as low middle category developing districts. It may be seen that in case of together at various levels of (dis) similarity. A matrix tree plot i.e. dendrogram agricultural and socio-economic development, the four districts named Cuttack, visually demonstrates the hierarchy within the final cluster, where each merger is Ganjam, Mayurbhanj and Sundargarh were categorized in category-I of high represented by a binary tree. The most common distance measurement between developed districts category. About 25.65 percent population of the Odisha data points is the Euclidean distance. belongs to these districts. Nine districts were categorized in category-II of high middle developed districts and covering the population about 35.63 percent while Results and Discussion eleven districts were categorized in category-III of low middle developed districts. The composite indices of agricultural development were computed for different About 31.26 percent of the total population belongs to category-III. The six districts districts in respect of agricultural and socio-economic sector. The districts were named Boudh, Deogarh, Gajapati, Jharsuguda, Naupada and Subarnapur were ranked on the basis of developmental indices. The composite indices of categorized in category-IV of low developed districts, covering only 7.46 percent development along with the ranks of the districts for different sectors are population [Table-2].

International Journal of Agriculture Sciences ISSN: 0975-3710&E-ISSN: 0975-9107, Volume 11, Issue 4, 2019 || Bioinfo Publications || 7907 Regional Disparities in the Levels of Development in Odisha

In cluster analysis, we have used Ward’s hierarchical clustering method. The Table-6 Percentage of variation explained by the PCs for overall development graphical presentation of results with dendrogram given in [Fig-1] shows a fairly PC Eigen Value Variation (%) Cum. Variation (%) clear picture corresponding to Ward’s method and squared Euclidean distances. 1 13.53 45.11 45.11 [Table-3] shows the summary statistics of the clusters which were formed in 2 5.79 19.31 64.42 3 2.34 7.82 72.24 analysis. [Table-3] shows that the minimum variance within-class (Cluster) was 4 1.78 5.93 78.17 found cluster 1 i.e. 948705515351.48, while maximum variance within-class was 5 1.16 3.88 82.05 found in cluster 3 i.e. 4116131183378.74. It means a cluster having minimum 6 0.99 3.32 85.37 variance within cluster is more compact or homogeneous within itself. Cluster 1 The first 6 PCs explained 85.37 percent variation of the data set. Through PCA we had minimum and Cluster 3 had maximum distances to centroids among all the can also have a scree plot which plots Eigen values against the PCs. We can see four clusters. We focused on the solution with four groups (clusters) of districts. that after fourth component, there is not much difference in the height of Table-4 Districts of Odisha within similar groups or clusters successive bar of the graph, which shows that each successive PC is accounting Cluster 1 Cluster 2 Cluster 3 Cluster 4 for smaller and smaller variation to the total variance. Angul Balasore Baragarh Cuttack Table-7 Ranking of districts based on principal component analysis Bhadrak Kalahandi Bolangir Dhenkanal Districts PC scores Rank Districts PC scores Rank Boudh Keonjhar Ganjam Koraput Ganjam 6.748 1 Sambalpur -0.212 16 Deogarh Sambalpur Mayurbhanj Nabarangpur Mayurbhanj 5.872 2 Bhadrak -0.428 17 Gajapati Subarnapur Puri Sundargarh 4.980 3 Rayagada -0.610 18 Jagatsinghpur Sundargarh Cuttack 4.589 4 Kendrapara -0.805 19 Jajpur Keonjhar 4.338 5 Kandhamal -0.935 20 Jharsuguda Balasore 4.253 6 Nabarangpur -1.087 21 Kandhamal Bolangir 3.220 7 Jagatsinghpur -2.166 22 Kendrapara Kalahandi 3.170 8 Nayagarh -2.457 23 Khurda Koraput 1.958 9 Malkangiri -3.245 24 Malkangiri Puri 1.876 10 Gajapati -4.407 25 Nayagarh Baragarh 1.630 11 Nuapada -4.517 26 Nuapada Jajpur 1.600 12 Subarnapur -4.988 27 Rayagada Khurda 0.691 13 Boudh -6.055 28 Angul 0.019 14 Jharsuguda -6.332 29 The list of districts with four groups (clusters) is presented in [Table-4] in which Dhenkanal -0.086 15 Deogarh -6.613 30 cluster 1 has 15 objects (districts), cluster 2 has 6 objects, cluster 3 has 4 objects A ranking of districts on the basis of PCA [Table-7] represents that the districts and cluster 4 has 5 objects. First cluster contained 15 districts of Odisha (Angul, Ganjam, Mayurbhanj, Sundargarh, Cuttack, Keonjhar and Balasore have higher Bhadrak, Boudh, Deogarh, Gajapati, Jagatsinghpur, Jajpur, Jharsuguda, PC scores from 6.75 to 4.25 whereas Gajapati, Nuapada, Subarnapur, Boudh, Kandhamal, Kendrapara, Khurda, Malkangiri, Nayagarh, Naupada and Jharsuguda, Deogarh have low PC scores from -4.407 to -6.613 which indicates Rayagada), second cluster represented the 6 districts of Odisha (Balasore, that the high developed and low developed districts respectively. The results of Kalahandi, Keonjhar, Sambalpur, Subarnapur, Sundargarh) while third cluster PCA are closely in accordance with those of the composite indices as well as consists of 4 districts (Baragarh, Bolangir, Ganjam and Mayurbhanj) and 5 districts cluster analysis. namely Cuttack, Dhenkanal, Koraput, Nabarangpur and Puri formed a separate cluster known as cluster 4. Conclusion Table-5 Distances between the clusters centroids The results based on the individual ranking of each district gave us a clear idea Clusters 1 2 3 4 about the overall development of each district in Odisha. This study indicates that 1 0 4594448.58 8668799.82 2338272.85 there are inter-districts disparities in Odisha with reference to various majors of 2 4594448.58 0 4084292.83 2665797.94 development in agricultural and socio-economic sectors or in overall sector. On 3 8668799.82 4084292.83 0 6575219.42 4 2338272.85 2665797.94 6575219.42 0 the basis of composite indices, cluster analysis and principal component analysis, The number of districts in each group was different. [Table-5] shows the distance the districts of Ganjam, Mayurbhanj, Cuttack and Sundargarh, were observed to between cluster centroids i.e. the distance between cluster 1 and 4 was minimum be better off in overall development whereas the districts of Nuapada, i.e. 2338272.85 followed by the distance between cluster 2 and 4 i.e. 2665797.94, Subarnapur, Boudh, Jharsuguda and Deogarh were remained at the low level of whereas the maximum distance between clusters 1 and 3 was 8668799.82, it development. Out of the three most backward districts i.e. Boudh, Jharsuguda and means that the objects of clusters 1 and 3 were at different levels of development Deogarh the least developed in view of agriculture was Jharsuguda and on socio- and clusters 1 and 4 had minimum distance so the objects of both the clusters economic front Deogarh was the least developed district. To attain uniform were closely related with the level of development. development in Odisha individual indicators, need to be examined for making Classification of districts within clusters presented in the dendrogram [Fig-1] them at par with their values in the developed districts. Such information may help shows that cluster 3 represented the developed districts and cluster 2 and 4 the planners and administrators to readjust the resources allocation. contained the districts at developing stage, whereas cluster 1 contained districts with low development. Angul, Bhadrak, Boudh, Deogarh, Gajapati, Jagatsinghpur, Application of research: Statistical techniques to classify all the districts of Jajpur, Jharsuguda, Kandhamal, Kendrapara, Khurda, Malkangiri, Nayagarh, Odisha state according to level of development. Nuapada and Rayagada districts were found low developed districts. Districts Baragarh, Bolangir, Ganjam and Mayurbhanj were grouped in the developed Research Category: Socio-economic Development districts cluster, whereas Balasore, Kalahandi, Keonjhar, Sambalpur, Subarnapur, Sundargarh, Cuttack, Dhenkanal, Koraput, Nabarangpur and Puri were in grouped Abbreviations: in the cluster of developing stage. In present study, using principal component CI: Composite Indices analysis (PCA), six main components have been extracted with eigenvalue having PCA: Principal Component Analysis more than one or equal to one. The Eigen values represent the variation in the GDP: Gross Domestic product PCs. The eigenvalues and the percentage of variation explained by first six PCs GSDP: Gross State Domestic Produce for the 30 indicators of the both sectors (agriculture and socio-economic) are presented in [Table-6]. Acknowledgement / Funding: Authors are thankful to Chaudhary Charan Singh Haryana Agricultural University, Hisar, 125004, Haryana, India

International Journal of Agriculture Sciences ISSN: 0975-3710&E-ISSN: 0975-9107, Volume 11, Issue 4, 2019 || Bioinfo Publications || 7908 Verma S., Tripathi R.K. and Nitin Tanwar

*Principal Investigator or Chairperson of research: Dr Suman Verma University: Chaudhary Charan Singh Haryana Agricultural University, Hisar, 125004, Haryana, India Research project name or number: Research station trials

Author Contributions: All authors equally contributed

Author statement: All authors read, reviewed, agreed and approved the final manuscript. Note-All authors agreed that- Written informed consent was obtained from all participants prior to publish / enrolment

Study area / Sample Collection: 30 districts of Odisha

Cultivar / Variety name: Nil

Conflict of Interest: None declared

Ethical approval: This article does not contain any studies with human participants or animals performed by any of the authors. Ethical Committee Approval Number: Nil

References [1] Tanwar N., Sunil Kumar, Sisodia B.V.S. and Hooda B.K. (2016) Journal of Applied and Natural Science, 8(1), 5-9. [2] http//censusindia.gov.in/ [3] Narain P., Sharma S.D., Rai S.C. and Bhatia V.K. (2007) J. Ind. Soc. Agril. Statist., 61(3), 328-335. [4] Narain P., Sharma S.D., Rai S.C. and Bhatia V.K. (2003) Jour. of Ind. Soc. of Agril. Stat., 56, 52-63. [5] Panda N. and Dutta K.B. (2011) Jour. of Ind. Soc. of Agril. Stat., 65 (3), 285-289. [6] Tripathi R.K. and Tanwar N. (2017) Indian Journal of Economics and Development, 13(2a), 13-18 [7] Narain P., Rai S.C. and Shanti S. (1991) Jour. of Ind. Soc. of Agril. Stat., 43, 329-345. [8] Pearson K. (1901) Philosophical Magazine, Series, 6, 2(11), 559–572. [9] Ward J.H. (1963) Journal of the American Statistical Association, 58, 236–244. [10] Wishart D. (1969a) Biometrics, 22, 165.

International Journal of Agriculture Sciences ISSN: 0975-3710&E-ISSN: 0975-9107, Volume 11, Issue 4, 2019 || Bioinfo Publications || 7909