DETERMINANTS OF RURAL HOUSEHOLDS’ POVERTY WITH SELECTED LINK FUNCTIONS: THE CASE OF SORO DISTRICT, HADIYA ZONE, SNNPR, ETHIOPIA

Abstract

Poverty is a general feature in Ethiopia causing many sufferings to the largest proportion of the population. This study assessed the determinants of rural household poverty in five selected kebeles in Soro district by using the generalized linear modeling approach. With the specific objectives of estimating rural households’ poverty status, identifying appropriate link function, and identifying determinants of rural households’ poverty. Primary data were obtained through structured questionnaire interview. Secondary data were also obtained from published, unpublished materials and journals. A total of 184 selected sample households were identified by proportional allocation and sample respondents were selected by systematic sampling technique. Based on the primary data whereby set of consumption food energy-intake method the probability of being poor was assessed. By using national poverty line of 2200 kcal, out of total of 184 sample households surveyed 65.76% were found to be poor. Log-log link function is found to be more appropriate to fit the data. Model adequacy diagnostic tests of the cook’s distance and GLM residuals shows that there were no outliers and influential values that had significant impact on the model results. Based on generalized linear model results, the major determinants of rural households’ poverty were age of household heads, family size per adult equivalent, access to credit service, dependency ratio, TLU, access to health service, and number of oxen ownership. Hence, promoting equitable economic growth, family planning, increasing land productivity, increasing credit service, increasing health service, increasing TLU and promoting research extension farmer linkage are indispensable policy interventions to better reduce rural poverty