Masters Theses

Date of Award


Degree Type


Degree Name

Master of Science


Agricultural Economics

Major Professor

Jsmes G. Snell

Committee Members

M. B. Badenhop, Charles M. Cuskaden


The general objectives of this study were to develop statistical consumption functions from sample data for 14 categories of beef and to estimate income elasticities and consumer unit elasticities for each of these categories for quantity and expenditure using selected social and economic factors of households. The 14 categories of beef include quarters, classes, and wholesale cuts. The data were obtained in Knox County, Tennessee, from 215 families for a 10-week period, September 4 through November 12, 1967. Earl R. Williams collected the data for his Ph.D. dissertation. Least squares regression was used to fit three linear models to the beef consumption data for a continuous income range for each beef item. Model I designated quantities purchased or expenditures for each beef item to be a function of annual disposable household income. Model II designated quantities purchased or expenditures to be a function of annual disposable household income and consumer units. Model III denoted the quantities purchased or expenditures for each beef category to be a function of annual disposable household income, consumer units, race, age of head of household, number of social groups to which the homemaker belongs, primary food purchaser, religion, occupation, number of guest meals served, number of children's meals eaten at school, number of meals eaten out, education of the homemaker, education of the husband, whether the family owned a deep freeze, and whether the family owned a charcoal grill. Dummy variables were used to integrate qualitative variables into the regression equations. On the basis of R2 values, Model III was selected as the model which best explained the variation in the dependent variables. This model was the object of the analysis. The variable Income was found to be most related to beef cut quantity and expenditure. The Income coefficient tested significant 19 out of 28 equations computed. The Catholic coefficient was tested significant in 15 equations; however, its validity may be questionable due to the small sample size. Consumer units was found significant in only 9 equations. Charcoal grill was found significant in none of the equations. A comparison of the income elasticities computed from the models were in general agreement with economic theory. The relatively inexpensive items were inelastic with several items having negative signs. The more expensive items were inelastic with positive signs, while the most expensive were elastic. The consumer unit elasticities were less than one except for ground quantity, 1.04, and brisket expenditure, 1.21, both of Model III. Significance tests conducted on the consumer unit elasticities were not fruitful. However, among those elasticities which were significant there was general agreement with economic theory.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."