This study presents the empirical results of estimating the household broiler meat expenditure pattern in the western region using the 1977-78 United States Department of Agriculture Nationwide Food Consumption Survey. The effects of assuming homoscedasticity and heteroscedasticity in the tobit model on resulting estimates are discussed in terms of estimated coefficients, marginal effects, and elasticities. Based on the strength of the sample data, the results suggest that specification of the homoscedastic model should be rejected in favor of the heteroscedastic model, implying that the validity of homoscedasticity should not be routinely accepted without testing when applying tobit procedure to analyze survey data.