Statistical evaluation of socio economic development in Kerala

Abstract

The study entitled “Statistical evaluation of socio economic development in Kerala” was conducted to construct a composite index of socio economic development for each district of Kerala, to classify the districts into homogeneous groups based on the development and to identify major factors affecting socio economic development across districts of Kerala. Data pertaining to the development of various sectors such as agriculture, demography, industry and infrastructure were collected from various departments for a period of five years, 2015-16 to 2019-20 for the purpose of the study. The average value of each indicator for the five year period was standardised and the weightage was assigned for each indicator based on coefficient of variation to obtain the overall composite indices of fourteen districts, and these districts were ranked based on the composite indices. The district with a low composite index value was ranked first and the district with a higher composite index value received a lower rank. Palakkad district (composite index of 0.34) was ranked first and Pathanamthitta (composite index of 0.63) was the last. In order to identify the homogenous groups, the districts were classified into low, medium and highly developed districts. The districts were categorised into three different groups using mean ± 0.5 standard deviation value of the composite index, as the indices followed a normal distribution pattern. The districts Pathanamthitta, Wayanad and Kasaragod were low developed districts, whereas Palakkad, Thrissur and Ernakulam were highly developed districts, and rest of the districts such as Thiruvananthapuram, Kollam, Alappuzha, Kottayam, Idukki, Malappuram, Kozhikode, and Kannur were classified as medium developed districts. The significant variables that contribute to the socio economic development of the districts were identified using principal component regression method. Principal component analysis was performed and first five principal components were selected, as this explained around 70 percentage variation of the total indicators. The multiple linear regression analysis was performed using principal components as explanatory variables and the composite indices as the dependent variable, the first three principal components were found to be significant and the major contributing variables from these significant components were identified based on the factor loadings obtained from the principal components. The variables such as number of hospitals, number of banks, number of higher secondary schools, number of government colleges, number of SHGs, per capita income, area under paddy and area under irrigation were found to be the major contributing variables. A correlation analysis was performed to study the inter-relationship between the various sectors. Each sector were found to be significantly and positively correlated with socio economic development. It was also observed that agriculture sector had the highest correlation with socio economic development. Model districts and the potential target variables were identified for each low developed districts. The variables such as net sown area, birth rate, death rate, infant mortality rate, per capita income, number of registered factories and number of banks were identified as potential target variables. The study suggests that by bringing the values of the observed variables in the low developed districts closer to that of the target variables in the model districts, would lead to the improvement in socio economic development of the low developed districts.