Unraveling Heterogeneity in Online Shopping and Travel Behavior Through Latent Class Modeling

Abstract

While existing literature has extensively explored the impact of online shopping on travel behavior, few studies have undertaken segmentation analysis to uncover hidden behavioral heterogeneity. This study fills this gap by addressing heterogeneity and identifying distinct shopper segments based on online shopping and shopping travel behaviors, with a focus on product types. Data collected in November and December 2021 from 1,747 shoppers in Florida were analyzed using Latent Class Analysis (LCA) with covariates. Sociodemographic and residential characteristics, COVID-19 influences, attitudes, and perceptions of channel-specific factors served as active and inactive covariates to predict class membership. Our model identified six classes of shoppers, with short-distance dual-channel shoppers representing the largest class (28.4%) and exclusive online shoppers the smallest (6.2%). Dual-channel shopaholics, overrepresented by Gen Zers, Millennials, Blacks, and workers, exhibited high average monthly vehicle miles traveled (VMT) across all product types and a strong potential for complementary shopping behavior. Conversely, exclusive online shoppers overrepresented by members of the silent generation, those who live alone, have no vehicle, and do not enjoy shopping, demonstrated potential substitutive shopping behavior. In general, single-channel shoppers showed lower monthly VMT than their dual-channel counterparts across all product types. These findings contribute to a deeper understanding of shopping behavior, offering insights for a more accurate quantification of the net traffic and environmental impacts of e-commerce. Additionally, they provide valuable considerations for designing segment-specific policies aimed at minimizing complementary shopping and maximizing substitutive shopping.