13
ratio of goat meat consumption is 0.33 for African American, 1.42 for Black non African
American, and 1.18 for Hispanic. The likelihood of goat meat consumption for non-
White population is significantly higher. An exception is that the households of Asian
origin demonstrate the same likelihood as the White. This may be explained as bias due
to the small number of households of Asia origin (15) in the survey. Age is another
demographic contributor. The elder is more likely to consume goat meat, and the peak
consumption age group is 55-74. It may be explained as differences between generations,
or as differences between the young and the elder. Further clarification needs time series
or cohort data.
Among socioeconomic variables, no significant impact is observed among levels of
education, but the influences of gender, income and geographic are observed. The odds
ratio for males is significantly higher than that for females. Household income influences
consumption behavior. The household with $25,000 income represents a threshold.
Households earning less than the threshold are more likely to consume goat meat than
those earning more than the threshold. This signals an inferior good characterization of
consumer perception for goat meat. Different consumption patterns are observed among
the 11 Southern states. Nine of them have a similar tendency, but Florida and Texas
differ. Compared with other states, households in Florida are less likely to consume goat
meat; households in Texas are more likely to buy.
Results of Demand from per Capita Consumption Increase The model starts with a
similar variable set as model 1, ends up with variables quite different. Notably, the
goodness of fit is improved with some preference variables. The estimated results are
reported in table 3.