Metrics on Restaurant Ordering Behavior

Abstract

The academic literature analyzing the behavior and interactions among commensals at a table generally resorts to experimental settings with volunteer decision makers or focuses on receipts issued at actual restaurants. The experimental approach widens the potential scope of the phenomena that can be analyzed but is subject to observer effects, with decision makers being aware of the fact that their actions are being monitored. The approach using receipts is not subject to observer effects but limited in its scope by lacking interactions with the commensals and the data that can be collected. In the current article, we make extensive use of a data set collected by restaurant personnel following specific instructions. They gathered information on a number of decisions made at the table throughout the whole meal without the commensals being aware that they are being monitored. As a result, we are able to examine empirically the importance that the choices of the first-person ordering (the leader) may have for the decisions made by the other commensals at the table. In particular, we study the similarity of orders—in terms of dishes, drinks, and prices—between the table leader and the other commensals. Our results reveal that table leaders, both male and female, have a considerable influence on the choices made by other commensals under a variety of different scenarios. We also describe the differences arising when males and females act as table leaders, as well as the influence that specific payment arrangements have on the ordering behavior of the commensals.