Exploring heterogeneity in meat consumption and eating out by using a latent class model

Abstract

PurposeWhile it is incontestable that eating in restaurants leads to a higher energy intake than eating at home, this paper explores the even more environmentally relevant connection between meat intake and the location of eating.Design/methodology/approachBased on secondary data from the US National Health and Nutrition Examination Survey (NHANES), the authors apply a latent class model (LCM), combining latent profile analysis (LPA) and regression analysis. Different (latent) consumer classes are modeled based (1) on share of meat consumption and (2) share of eating out by means of LPA, while class-specific socio-demographic characteristics are estimated by means of ordered logistic regression.FindingsResults of the LPA reveal four (latent) consumer classes with regard to the share of meat consumption and the share of eating out. One class consists mostly of male meat lovers with a high share of eating out, which, however, only represents 7% of the sample. A much larger class represents an affluent social group that consumes the majority of food outside of the home but does not consume significantly more meat than the large group of moderates who mostly eat at home. The fourth class mostly consists of children with a very low intake of meat.Originality/valueBy applying a LCM, the authors shed some light on the relation between meat consumption and eating out. The authors demonstrated that commonly assumed relations, such as men eating more meat than women, do not necessarily apply. Similar findings apply to factors potentially influencing meat consumption, such as education, marital status and income.