A Long-Run Case Study of College Football Attendance
Abstract: Regular season game-day data for a single college football team over 33 years are used to estimate attendance and attendance as a percent of stadium capacity. Two estimation methods with stationary variables are employed: ordinary least squares and two-stage least squares with instrumental variables. Results are consistent across estimation methods. Short-term team performance for home and visiting teams and the temperature all increase attendance measures. Some game-specific characteristics are also significant: conference games, televised games and non-Football Bowl Subdivision opponents reduce attendance while rivalry and homecoming games increase it. Economic variables, with the exception of travel cost, are insignificant.
Full-Text Available: PDF
Authors: Paul A. Natke