A Novel Data Analytics Methodology for Discovering Behavioral Risk Profiles: The Case of Diners During a Pandemic

Abstract

Understanding tourist profiles and behaviors during health pandemics is key to better preparedness for unforeseen future outbreaks, particularly for tourism and hospitality businesses. This study develops and applies a novel data analytics methodology to gain insights into the health risk reduction behavior of restaurant diners/patrons during their dining out experiences in a pandemic. The methodology builds on data relating to four constructs (question categories) and measurements (questions and attributes), with the constructs being worry, health risk prevention behavior, health risk reduction behavior, and demographic characteristics. As a unique contribution, the methodology generates a behavioral typology by identifying risk profiles, which are expressed as one- and two-level decision rules. For example, the results highlighted the significance of restaurants’ adherence to cautionary measures and diners’ perception of seclusion. These and other factors enable a multifaceted analysis, typology, and understanding of diners’ risk profiles, offering valuable guidance for developing managerial strategies and skill development programs to promote safer dining experiences during pandemics. Besides yielding novel types of insights through rules, another practical contribution of the research is the development of a public web-based analytics dashboard for interactive insight discovery and decision support.

Keywords

• data analytics
• machine learning
• risk profiling
• behavioral typology
• worry
• risk preventive behavior