Action Filename Size Access Description License
Show more files...


The aim of this research is to determine the fish and meat consumption profiles of Turkish household by using multinomial logit model. The data of this study were gathered from 2009-2012 Household Budget Survey of Turkish Statistical Institute. A total of 40,033 observations were used in multinomial logit regression. According to the results of the final model, it is seen that red meat consumed 45.70%, fish meat consumed 20.16% and 34.14%, which did not consume red meat or fish meat. Dependent variable (1: Those who consume red meat, 2: Those who consume fish meat, 3: Those who do not consume red meat / fish) is classifiable qualitative variable. Independent variables explaining dependent variables; household size and structure, marital status, income, gender, education of the head of household, household with refrigerator and dishwasher. The assumption of multiple linear connectivity problem, model specification error, independence of irrelevant alternatives (IIA) is validated for the use of Multinomial Logit regression analysis. Statistical significance, effect levels, relative risk ratios, and marginal effects of the independent variables are given as the results of the analyzes. According to model result; household size, household income (medium-high and high), all marital levels of the householder, gender of householder, type of household (patriarchal, single adult, coexisting adult), the age of the householder and educational level of the householder factors have a statistically significant effect on the consumption of red meat in Turkey. Household size, household income, marital status of the head of household, gender of household head, household type (family with three or more children, single adult, cohabiting adults), presence of refrigerator and dishwasher in household independent variables have statistically significant effect on fish consumption in Turkey.


Downloads Statistics

Download Full History