A Binary Logit regression model of Farmers' participation in Fadama II project in Niger State, Nigeria

Abstract

This paper is aimed at determining the factors that characterized the farmers’ participation in Fadama II project as a way of improving their livelihood. The study was conducted in 2009 in Niger State in the North central Nigeria, situated on longitude 9036'N and latitude of Greenwich meridian. A two -stage sampling plan was adopted in this study. Three local governments in which Fadama II project is being implemented were randomly selected in the state. The sampling frame of all farmers in the 3 LGAs were stratified into two : Fadama project beneficiaries (farmers) and Non Beneficiary (farmers) who lived in Fadama II project LGAs but did not participate in the project. In each stratum, 75 questionnaires were administered each (25 per LGA); totaling 150 respondents. A binary logistic (logit) regression model was employed to analyze the data. The model indicated that the goodness of fit of observed data was adequate. 87.3% of the (respondents) farmers were correctly classified by the model. The model chi-square value for goodness of fit was 133.657. The results showed level of education and age of the respondents were significant (p<0.05) and (p <0.01) respectively. They both had inverse (negative) relationship on log of odds of farmers’ participation in Fadama II project. However, income and membership of cooperative society had positive relationship with the odds of farmers’ participation in Fadama II project and they were both significant (p< 0.01). Others factors such as sex, household size, farm size and major occupation of the respondents were not significant (p < 0.05) in this study.