Maximizing visitors at college football bowl games

Abstract

Purpose The purpose of this paper is to isolate factors predictive of event attendees, and assist tourism professionals such as members of host committees, in maximizing the number of out-of-town visitors to their region and optimizing tourism-related revenue when hosting college football bowl games. Design/methodology/approach A total of 16 demand variables were entered into a hierarchical regression model, including the stature of the event and market-related variables, as well as team-related variables reflecting team or program stature and current season performance. Findings A final model containing seven variables (bowl age, market population, conference affiliation, bowl game stature, season wins, home attendance, and distance traveled) predicted 77.5 percent of the variance in bowl game attendance. Research limitations/implications This paper illustrates the use of predictive modeling for major sport event attendance with a unique sample and variables explored. Future research may build off the model to explore attendance for other populations or events. Practical implications The applied nature of this study allows practitioners working in the tourism and event management field to incorporate a predictive model to best select participants in sporting events to maximize event attendees. Originality/value Understanding the variables which predict event attendees in the context of college football bowl games provide useful data to practitioners. This study advances this area of research by treating event participants as unique observations (something which has not been done in prior studies), and looking at a new data set which incorporates the College Football Playoff era.

Examining valuation of corporate naming rights partnerships in collegiate sports and their impact on consumer behavior

“…In order to test the potential influence of a variety of variables on average annual deal value using a systematic approach, a hierarchical model building procedure was utilized (e.g., Popp et al ., 2017). This approach allows the researcher to determine whether each group of factors is explaining a statistically significant amount of the incremental variance in the dependent variable, as well as control for a variety of factors throughout the subsequent analysis.…”