This study analyze effect of tractorization on cropping intensity, crop yields and adoption of major agricultural inputs, on human labor employment, determine the utilization and per unit cost of tractor power according to the farm size, compare the different costs and profits for Draught animal power and Tractor operated farms according to the farm size. Data were collected from the stratified sample of 345 farmers from three purposively selected districts, Hexosa, Asasa and Sinana, where the agricultural mechanization operations are becoming increasingly practiced. The Tobit model showed that wheat farm land size, tractor use, and labor for the adoption of the chemical fertilizers, Sex of the household, land allocated for wheat land and tractor use for the adoption of improved seed, and Age of the households and total cultivated land for the adoption of agro chemicals were affect positively and significantly. Tractorization has positively and statistically significant affect the adoption of chemical fertilizers and improved seed rate, except the adoption of agro chemical application. Average of wheat production in all farms of tractor operated farms were higher (34.67 qt ha-1) than other operated farms. There were the significance differences of wheat yield among different types of farms. There has been a reduction of total human labor employment to the extent of about 76 % on the tractor farms as compared to both the draught animal and mixed operated farms. The net income was higher on tractor-operated farms than both mixed and draught animal operated farms. Therefore, the tractor-operated farms were economically more efficient than the draught animal power and mixed operated farms especially in the case of farms of small and large farm sizes.
The bounded nature of the fractional dependent variables, for instance in corporate finance leverage ratio clustering with a substantial number of observations at unit interval raises some important issues in estimation and inference. Ordinary Least Square (OLS) regression with Gaussian distributional assumption has been the main choice to model fractional outcomes in many business problems. Nevertheless, it is conceptually flawed to assume Gaussian distribution for a response variable in the interval [0,1]. Tobit model which is a Single-component method for modelling proportional outcome also share properties with OLS. Two-part Fractional regression models have been shown as the most natural way of modelling bounded, proportional response variables. Beta regression method has been used to achieve the objective in this paper.