Most studies on consumers’ food waste use models that increase the likelihood of false positives. Here, using EU-level Eurobarometer data from 2013, we use alternative analytical methods that avoid these problems (Bayesian Networks) to identify the impact of household characteristics and other variables on self-assessed food waste. Our analysis confirmed that the country, the age of the respondent, the status (student/non-student), and a belief that the family wastes too much are related to the level of self-assessed food waste.

The paper explores the use of decision models rather than traditional frequentist statistics in determining the drivers of food waste. This approach has an advantage over other more traditional approaches in that the probability for false positives is effectively zero meaning that, for policy makers, there is a reduced probability of wasting resources on in-correctly targeted interventions.

Our analysis confirmed that the country, the age of the respondent, the status (student/non-student), and a belief that the family wastes too much are related to the level of self-assessed food waste. But we found no evidence that waste behaviours differ between people living in urban and rural areas, and little support of a difference between genders. Households from lower-income EU countries (e.g. Portugal, Greece, Bulgaria, Cyprus and Latvia), as well as students and young adults tend to report higher levels of food waste. Hence, the adoption of an EU strategy based on the concept of subsidiarity, and of country-level policy measures targeting different age groups is suggested. Furthermore, our analysis shows that policy makers need to be wary of relying on analysis based on large datasets that do not control for false-positives, particularly when sample sizes are small.