The aim of this research is to trace recent changes in family expenditures on food commodities between 2015 and 2017/18. The data is drawn from CAPMAS publications on family income, expenditure, and consumption for both years. Three food groups are explored in this paper: 1- The aggregate group of food and beverages; 2-The grains and bread group; and 3- The meat group. Two statistical models are utilized for the purposes of data analysis. The first model is known as Working's model and the second model is the well-Known double logarithmic model. The analysis shows that the expenditure elasticity of the aggregate group of food and beverages ranges between 0.423 for high-income urban families in 2017/18 to 0.763 for urban poor families in 2015. For rural families the same elasticity is about 0.313 for high-income families in 2017/18 and about 0.751 for poor families in 2015. These estimates are derived from the OLS estimates of Working's model. The double logarithmic model gives one single estimate for expenditure elasticity for the entire set of data. For example, the expenditure elasticity for all families in 2017/18 is 0.63 in urban areas and 0.59 in rural areas according to the double logarithmic model. The same estimates are 0.64 for urban areas and 0.56 for rural areas according to Working's model. The Working's model fits the family survey data quite well and gives multiple estimates of expenditure elasticity according to the level of family income. In contrast, the double logarithmic model gives one estimate of expenditure elasticity that does not change with the level of family income. That is Working's model would be more suitable for policy analysis purposes.