This paper summarise findings from farmers of 3 different provinces of Afghanistan. The survey sample included 1503 farmers owners/operators. Questionnaires were asked as an structured interview with the farmers.
1 Summary In conclusion, based on the mean and standard deviation uni-variate anal- ysis of our continues and discrete variables through sub-sampling we can conclude that in rst province which has lower average temperature and comparatively low average rainfall rate as well. Both factors has a ected the average livestock income of farmers to be less than the average of the second and third provinces. Furthermore, in sub-sample analysis in-dept comparison cross analysis of di erent variables has been done which illus- trates numerical variables di erences through critical measures of dispersion analysis.
2 Introduction This paper summarizes ndings from farmers of 3 di erent provinces of Afghanistan. The survey sample included 1503 farmers owners/operators. Questionnaires were asked as an structured interview with the aforemen- tioned farmers. Moreover, The survey project was motivated by a number of questions. And all questions asked respondents to comment on farm char- acteristics, practices, tools, facilities and personal conditions that latter on encoded as number of categorical and numerical variables in accumulated data-set. Hence, categorical variables of our data-set are as follow: Province, Solar Panel, Access to Internet, Generator Main Occupation Farming and Literacy. Furthermore, the numerical variables of our data-set is of eight variables that are Livestock Net Income, Temperature, Rainfall, Mobile, TV , Motor-Bike, Car and No Hrs with Electricity.
3 Methods In this paper for analysis of categorical and numerical variables we are go- ing to use di erent numerical measures such as central tendency, variation, distribution and also we have used simple frequency table for categorical data. Since, frequency tables are useful for analyzing categorical data and for screening data for data entry errors we measured our categorical variables with them. Subsequently, in order to better communicate our results we have used visualization tools such as pie-charts for categorical variables as well as histograms for our continues and discrete numerical variables. Furthermore in sub-sample analysis we have used measures of dispersion (mean/standard deviation) in order for assessing our numerical data spread and dispersion in order for better comparison.