Multinomial Logistic Regression Model in Identifying Factors of m4agriNEI in CSA Innovations

Abstract

The impact of climate change has been evidence, creating a threat not only to environment but also human being too and it is imperative to act before it’s too late. An array of mitigation and adaptability practices are headed under Climate-Smart Agriculture (CSA) practices which are agricultural approaches that sustainably increases productivity, adaptation and reduces greenhouse gases. The study was, henceforth developed to identify factors for m4agriNEI in
enhancing CSA innovations. The Agro-Advisory Services (AAS) of mobile based agroadvisory system in Northeast India (m4agriNEI) consents the registered farmers to solve their farming related problems which also directly helped them in decision making to choose the best alternatives available. A sampled of 65 registered farmers were selected from the four project districts viz. Ri-bhoi, East Khasi Hills, West Khasi Hills and West Jaintia Hills districts of Meghalaya, based on the criteria of proactive average calls of 5 times a week and above made by them. Multinomial logistic regression was administered and consequently the factors viz. ‘Timeliness’, ‘Relevancy’, ‘Economy’ and ‘Accuracy’ of m4agriNEI were found to be statistically significant in influencing the CSA innovations by registered farmers of m4agriNEI. The model revealed that having the Cox & Snell R2 and the Nagelkerke R2 values of 0.557 and 0.633 respectively, determined the variability between 55.70 and 63.30 per cent of the
dependent variable i.e. ‘Adaptation Intention to enhancing CSA innovations’ in explaining the factors. The overall predictive accuracy for the present model was 72.30 per cent, signifying that the model was expedient.