Esports Online Live Streaming vs. On-site Consumption: A Comparison of Spectator Motivations and Market Segmentation

Citation

Kim S & Hong HJ (2024) Esports Online Live Streaming vs. On-site Consumption: A Comparison of Spectator Motivations and Market Segmentation. European Association for Sport Management 2024, Paris, France, 03.09.2024-06.09.2024.

Abstract
Aim and Research Questions: The focal point of this research is the domain of electronic sports (commonly known as "Esports") event management and consumer behaviours (i.e., esport fans in our context). As a distinct category of sporting events, esports has garnered immense popularity, drawing millions of spectators both in person to host cities and online via live-streaming platforms such as YouTube Live and Twitch. A larger market not only signifies more opportunities but also presents challenges that demand more nuanced and sophisticated marketing strategies. In addition, there is a gap in knowledge in understanding online viewership and on-site spectatorship in esports event management and consumer behaviours. The existing knowledge is limited to a comparison between traditional and esports spectator motivations (e.g., Pizzo et al., 2018) or streamer and audience behaviour (Li et al., 2020). To fill the gap, the first objective is to identify and categorise the different segments of esports fans as effective marketing often begins with customer segmentation. The second objective is to explore the relationship between spectators' motivational desires—performance dimensions, social dimensions, and performer dimensions—and their flow experience and behaviour intention (to continue attend the event). RQ1. Who attended the esports event (market segmentation)? RQ2. How are spectators' motivational desires ([a] performance dimensions, [b] social dimensions, and [c] performer dimensions) related to flow experience and subjective wellbeing? RQ3. Are their differences between online live streaming and on-site spectatorship in terms of market segmentation and the structural model? Theoretical Background and Literature Review: The current body of literature reveals notable gaps in our understanding of esports spectatorship. Rietz and Hallnamm (2022) posited that research on this topic should pivot towards a more qualitative and explanatory direction. Comprehending the motives of esports spectators is crucial, as these motivations can effectively predict behavioural outcomes (Hamari & Sjöblom, 2017). Much of the research on consumer behaviour and motivation in esports centres on player consumption, leaving spectatorship relatively unexplored (Lee & Schoenstedt, 2011). Consequently, this study's primary aim is to shed light on the psychology of consumer behaviour in esports attendance for both on-site and online, with an emphasis on understanding the mechanism of the flow experience. We put the flow experience as a dependent variable as it has many positive outcomes such as subjective wellbeing, behavioural intention, and loyalty (e.g., Kim & Kim, 2020). Thus, this study establishes hypothetical relationships based on the empirical studies among the variables. Research Design, Methodology and Data Analysis: The data collection for this study will be conducted in collaboration with the Korea eSports Association (KeSPA) by utilising their official website and social media channels during May. Convenience sampling will be employed to facilitate data collection. This study assesses ten esports spectator motivations, drawing on the Motivation Scale for Sports Consumption (MSSC) developed by Trail et al. (2003). Additionally, the measurement of flow experience is adapted from Kim and Kim (2020). The research methods employed in this study are twofold: (1) K-means clustering technique (to address RQ1 and to fulfil the first research objective) and (2) structural equation modelling (to answer RQ2 and RQ3, thereby achieving the second and third research objectives). K-means clustering is a widely adopted unsupervised learning algorithm in data mining, designed to detect and group distinct clusters based on specified attributes. K-means clustering aims to segment data into groups where respondents are as close to each other as possible while ensuring these groups remain distinct from one another. In our research, we will analyse the four attributes: (a) age (b) gender (c) annual income and (d) the length of being a esports fan. The partial least square structural equation modelling (PLS-SEM) is to use to test direct and indirect effect of the established hypotheses among motivations, flow, and behaviour intention. The research employs established measurement scales that have been psychometrically validated to assess the variables. Then, the model comparison between two different consumer groups is conducted: (a) online live streaming spectators and (b) on-site (venue) spectators.

Keywords
esports