If we want to get an accurate idea of what people consume, researchers often use food frequency questionnaires (FFQs). This means that respondents are asked how often they have consumed a particular type of food (e.g., dairy or chicken) over a certain period (e.g., over the last three months or in the past day). 

However, this method has some shortcomings. An individual’s consumption of certain animal products varies a lot, which makes it hard to predict or even measure consumption changes in a statistically significant way. Similar problems arise with products that are rarely consumed (e.g., liver). Finally, people aren’t always reliable in reporting the actual amount of animal products they eat.

In this study, the researchers address these issues by proposing a novel alternative: measuring animal product consumption as a trait. Rather than measuring specific quantities of products eaten, such questionnaires would indicate general patterns (e.g., being a high-meat-consumer or low-meat-consumer). To do this, the researchers refined a questionnaire by taking three FFQ studies and reducing the food items to those providing the most information about the “trait” of animal product consumption. 

The authors first used an existing FFQ that measured people’s consumption of 25 different animal products over the past three months and over the past 24 hours. Through a statistical method known as factor analysis, the authors determined that some animal products were eaten too infrequently to provide helpful information. From there, the authors conducted two FFQs of their own, removing rarely-eaten animal products from the survey questions as well as other animal products that didn’t do a good job of measuring different levels of the consumption “trait.” 

They ended up identifying 6 items for the three-month FFQ and 13 items for the 24-hour measure that could be considered good estimates of the trait of eating animals. All of these items were related to each other and to other measures that have been shown to correlate with animal product consumption (e.g. meat-eating rationalizations). Because of this, the authors were confident that their survey accurately measured what it was supposed to be measuring (in other words, that it was valid). 

The authors emphasize that more research is needed to fully explore the idea of animal product consumption as a trait. However, their initial results are promising — rather than splitting hairs trying to understand precisely how much of a certain animal product people are eating, advocates may eventually save themselves resources by studying general consumption traits instead. As a result, it’s worth keeping an eye on this topic moving forward.