Using Artificial Intelligence to Detect the Relationship Between Social Media Sentiment and Season Ticket Purchases

Abstract

Sport marketing researchers and practitioners have suggested that sport organizations that effectively engage in social media conversations with fans are likely to influence fan behavior. Few prior studies have empirically examined the relationship between social media engagement and sport product purchases, particularly event tickets. The current study utilized artificial intelligence to examine eight user sentiments on official sport organizations’ Twitter accounts, then determine if those sentiments were related to season ticket sales. Three years of season ticket data were obtained from 62 NCAA Division I men’s basketball teams and utilized in a regression model, which also identified Twitter sentiment scores from 176,439 posts captured from the official Twitter account of those programs. A final model, which included several control variables, explained 65.7% of the variance in season ticket sales, with the lagged sentiments of “joy” (positive) and “sadness” (negative) having a statistically significant relationship with season tickets sold.