Elsevier

Computers in Human Behavior

Volume 84, July 2018, Pages 58-67
Computers in Human Behavior

Social motivations of live-streaming viewer engagement on Twitch

https://doi.org/10.1016/j.chb.2018.02.013Get rights and content

Highlights

  • Live-stream viewer engagement was explored from a socio-motivational perspective.

  • Social and community interaction motivated all types of live-stream engagement.

  • Viewers who lacked external support spent more time watching live-streams.

  • Small channel viewers were more socially motivated than larger channel viewers.

Abstract

Little is known about the motivations underlying viewer engagement in the rapidly growing live-streaming multimedia phenomenon. This study trialled an eight-factor socio-motivational model, based on Uses and Gratifications Theory, to explain four aspects of live-stream viewer engagement. Cross-sectional data was collected through an international, online self-report survey of Twitch users (N = 2227). Multiple and ordinal linear regression analyses identified six motivations which helped to explain live-stream engagement: social interaction, sense of community, meeting new people, entertainment, information seeking, and a lack of external support in real life. Compared to mass media, viewer motivations to engage in live-stream entertainment appear to have a stronger social and community basis. Furthermore, live-stream viewers who preferred smaller channels (<500 viewers) were more motivated by social engagement than viewers who preferred larger channels. These findings offer insight into the motivations for live-stream engagement, and help to lay a foundation for further research.

Introduction

Live-streaming is an interactive form of internet-based multimedia entertainment that has grown rapidly in popularity worldwide since 2011 (Needleman, 2015, Twitch, 2017). Live-streaming has become so popular that, in some cases, there are more people watching others do activities, such as play computer games, than doing the activity themselves (Kaytoue, Silva, Cerf, Meira Jr, & Raïssi, 2012). Unlike previous streaming services, such as television and YouTube, live-streaming offers real-time human interaction between the streamer and viewers, facilitating their ability to interact with each other. While online technology has made this development possible, what is driving the rapid growth in live-streams, is not well understood from a psychological perspective. Despite the popularity of live-streaming consumption, little psychological research has been conducted. Some past research has explored live-streaming consumption motivations including tension release, escapism, and acquiring knowledge (e.g., Sjöblom and Hamari, 2017, Sjöblom et al., 2017), however social-based motivations such as needs for interaction, connection, and community, have not been studied within the context of live-streaming. This study seeks to address this gap by exploring the relationship between live-stream viewer socio-motivations and their psychological and behavioural live-stream engagement.

Live-streaming is a rapidly growing internet-age phenomenon. Nearly 10 million people tuned in to Twitch per day in 2016 to stream, watch, and interact with others (Twitch, 2016) Twitch is currently the largest live-streaming platform (Twitch, 2016). Live-streamers broadcast themselves playing video games, eating food, painting, dancing, and so on, in real time, to a public audience. Viewers watch and listen to the streamer, but are also able to directly interact with, and responding to, what the streamer says and does. This facilitates a two-way connection where streamers can directly acknowledge and respond to viewers, and viewers can actively participate in, and influence, live-stream broadcasts (Sjöblom & Hamari, 2017).

Twitch has grown rapidly in popularity. For example, Twitch exceeded Facebook in peak internet traffic in 2014 (Fitzgerald and Wakabayashi, 2014, Gamezone and May 15, 2014), with monthly viewers doubling from 45 million in December 2013 to 100 million in December 2014 (Leslie, 2015). Other sites such as YouTube, Dailymotion, and Facebook have introduced live-streaming services with similar features, however, Twitch is currently the most popular (D’onfro, 2015, O'Neill, 2014, Wilhelm, 2013). Live-streaming began as a niche, gaming-oriented domain, but is diversifying and growing into a broader social media trend. Despite its rapidly growing trajectory, there is limited research about streamer and viewer motivations.

Uses and Gratification Theory (UGT) evolved from theories about why people choose different types of media consumption (Rubin, 2009), and has been used to analyse consumer engagement in online social networking sites (Ku et al., 2013, West and Turner, 2010, Whiting and Williams, 2013), video sharing (Chiang & Hsiao, 2015), live-streaming (Sjöblom & Hamari, 2017), and eSports (Hamari & Sjöblom, 2017). According to UGT, people actively seek out, consume, and participate in media that fulfil their individual needs and match their preferences better than other media options (Ruggiero, 2000, Shao, 2009). UGT offers a functional approach which can help to understand how different forms of media engagement relate to the fulfilment of various psychological needs. Thus, the UGT provides a framework for understanding consumer choices and roles in media engagement as behaviours which are aimed at the fulfilment of individual psychological needs.

Building on UGT, theory and research about social media consumption suggests at least eight socio-motivations for live-stream engagement: entertainment, information seeking, meeting new people, social interactions, social support, sense of community, social anxiety, and external support. These motivations are explained in the following sections.

To capture live-stream engagement across various domains, four indicators of live-stream engagement were conceptualised. Emotional connectedness to the live-stream platform as an indicator of psychological engagement, time spent (per week) on the live-stream platform as a behavioural indicator of engagement. Time subscribed (in months) to channels on the live-stream platform and amount donated (in USD) to channels on the live-stream platform as financial indicators of engagement. Fig. 1 provides a visual representation of the proposed motivators of live-stream engagement and indicators of engagement.

Entertainment and information seeking have been previously identified as important motivators for consuming streaming services (Cheung and Huang, 2011, Hamilton et al., 2014, Sjöblom and Hamari, 2017). An important aspect of streaming is to provide an entertaining media experience. The most popular Twitch live-streams host eSport tournaments and professional gamer streams which allow viewers to watch and learn from the best players in the world (Cheung & Huang, 2011). Live-streams also commonly conduct product reviews either intentionally or by showing games being played. Therefore, streams are also used to provide previews of gameplay, allowing potential buyers to ask questions or request opinions of the game. Thus, entertainment and information seeking are proposed as two key reasons for live-stream engagement.

Social motivations, such as meeting new people, social interaction, and sense of community are important to live-stream viewers. Hamilton et al. (2014) found that live-streams serve as virtual “third places” where communities form and grow, with viewers using chat rooms to converse, laugh, and joke with each other about content they are watching. Chat rooms foster sociability, with regular viewers and moderators playing specific roles in welcoming and engaging newcomers (Hamilton et al., 2014). Thus, live-stream environments facilitate meeting new people, a common motivator for participation in online communities (Brandtzæg and Heim, 2009, Ridings and Gefen, 2004, Zhu and Chang, 2014).

A sense of community is also important in live-streaming (Hamilton et al., 2014). Sense of community includes experiences of membership, influence, fulfilment of needs, and belonging to the channel, streamer, and other participants (McMillan & Chavis, 1986). This suggests that live-stream viewers are attracted to channels where they feel noticed, important, and influential. Live-stream viewers can have their membership needs fulfilled by participating in community success and sociability. Shared experiences and continued participation in live-streams helps to develop connectedness with other stream participants and fosters a sense of community in the channel.

People may participate in online communities to compensate for a lack of community in real life (Miller, 2011). Online communities can be particularly beneficial for the psychological well-being of participants who lack external support from their family, friends, and local community (Bargh & McKenna, 2004). Online social interactions have been reported to reduce loneliness by presenting low-risk opportunities for self-disclosure, online social support, and involvement (Valkenburg & Peter, 2009).

Online interaction can also help individuals with social anxiety, who find it hard to socially engage with others in real life (Baumeister and Leary, 1995, Desjarlais and Willoughby, 2010, Mazalin and Klein, 2008). Live-stream environments can provide low-threat alternatives to real life socialising, removing barriers socially anxious individuals may experience in real life.

This research seeks to capture live-stream engagement on four factors: emotional connectedness to the service (psychological attachment), amount of time watching live-streams, number of channel subscriptions, and amount of money donated (behavioural attachments).

Emotional Connectedness. Emotional connectedness refers to how psychologically attached the viewer is to the live-stream platform and specific live-stream communities. Emotional connectedness to social media platforms, such as Facebook, has been explored (Ellison, Steinfield, & Lampe, 2007), however emotional connectedness to live-stream platforms has not been previously researched.

Time Spent. A second indicator of live-stream engagement is time spent watching live-streams. Twitch viewers reported spending an average 106 min every day watching live-streams and nearly half of the surveyed users reported spending over 20 h per week on Twitch, equivalent to a half-time job (Twitch, 2016).

Time Subscribed. A third indicator of live-stream engagement is subscription. Subscribing involves direct financial investment to support a streamer's channel. Subscribing to a Twitch channel costs the viewer USD 4.99 per month. In return, subscribers receive a badge next to their name, and use of customised emoticons to express feelings or intended tone. Subscribing often provides other social-oriented benefits, such as access to a live-stream voice server to engage verbally with other subscribers and invitations to participate in games or activities being broadcasted by the streamer. Viewers tend to subscribe for social integration reasons, essentially to enhance connections with the channel community (Sjöblom & Hamari, 2017).

Donations. A fourth indicator of live-stream engagement is financial donation to the streamer (via PayPal) or charities (Twitch, 2017). Unlike subscriptions, donations do not provide access to additional content. Instead, the donator is usually thanked personally by the streamer, celebrated in the chat room, and may have their name displayed on the stream as the “Top Donator” or a “Recent Donator”.

Some viewers prefer small streams, as they facilitate meaningful social interactions that are harder to achieve in large, fast moving chat rooms (Hamilton et al., 2014). Streamers have also reported that they can interact effectively with only 100 to 150 participants and that with more people the personal interaction between the streamer and the viewer breaks down (Hamilton et al., 2014). This is consistent with the cognitive limit on the number of meaningful social relationships that can be sustained, as suggested by Dunbar (1992). Smaller channels may therefore provide viewers with greater individual attention, participation, and influence in content; factors that increase overall belonging and community (Hamilton et al., 2014, McMillan and Chavis, 1986). UGT suggests that viewers will actively seek and choose media environments that fulfil their individual psychological needs (Ruggiero, 2000, Shao, 2009). From this perspective, viewers who prefer smaller channels are likely to be more driven by social needs than viewers who prefer larger channels. Viewers who prefer larger channels are more likely to be motivated by less social needs, such as entertainment and information seeking. However, this has not been empirically tested.

This study investigates what motivates people to engage in live-streams on the internet, based on a UGT framework. More specifically, the current study investigates the extent to which eight socio-motivators (entertainment, information seeking, meeting new people, social interaction, social support, sense of community, social anxiety, and external support) explain four types of Twitch live-stream engagement (emotional connectedness, time, subscriptions, and donations). The study also aims to investigate whether relationship between live-stream viewer motivations and indicators of engagement vary by channel sizes, as has been suggested by past research applying a UGT framework (Hamilton et al., 2014). Accordingly, two hypotheses are proposed.

Firstly, it is hypothesised that entertainment, information seeking, meeting new people, social interaction, sense of community, social support, social anxiety, and a lack of external support, will explain live-stream engagement. Secondly, it is hypothesised that social motivators (social interaction, sense of community, social support, meeting new people, social anxiety and external support) will better explain live-stream engagement behaviours for viewers who prefer small channels than for viewers who prefer larger channels.

Section snippets

Participants

There were 3611 responses to an online survey, however 38.3% of these responses were removed because they reported typically watching Twitch for 0 h per week, missed 50% or more of the survey questions, provided unrealistic answers (e.g., watching for more hours than possible in a week), or finished the survey in under 5 min, leading to a final sample of 2227 participants. The main referral sources were Reddit (n = 1949; 87.5%), direct referrals (n = 91; 4.1%), Twitch (n = 77; 3.5%), Twitter (n

Descriptive statistics

There was less than 5% data for missing for each item. Missing value analysis using Little MCAR's test was non-significant, suggesting that data was missing completely at random and thus listwise deletion was appropriate. An exploratory factorial analysis with one factor (no rotation) revealed that the variance explained by a single factor was less than 50% (32.67%), suggesting the data was free from common method bias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Due to positive skewness of

Discussion

This study examined viewer motivations for engaging in Twitch live-stream communities. It sought to identify factors that explain psychological and behavioural live-stream engagement and whether such motivations vary per channel size.

The utility of the proposed socio-motivational model was partially supported. Six of the eight motivators (social interaction, sense of community, meeting new people, entertainment, information seeking, and external support) significantly explained at least one

Conclusions

This study investigated how motivations of users of live-streams explain their psychological and behavioural engagement with live-stream media. Socially and community-motivated engagement was associated with greater emotional connection, time spent, and financial contribution to live-streams, particularly for smaller channels. Engagement with larger, more populated channels was less socially driven. Regardless, live-stream engagement across all channel sizes was motivated by a desire for social

Acknowledgements

This research has been carried out as part of research projects (40009/16) funded by the Finnish Funding Agency for Technology and Innovation (Tekes) and project partners.

References (58)

  • R.P. Mattick et al.

    Development and validation of measures of social phobia scrutiny fear and social interaction anxiety

    Behaviour Research and Therapy

    (1998)
  • T.L. Rodebaugh et al.

    The reverse of social anxiety is not always the opposite: The reverse-scored items of the Social Interaction Anxiety Scale do not belong

    Behavior Therapy

    (2007)
  • K.L. Satow

    Social approval and helping

    Journal of Experimental Social Psychology

    (1975)
  • M. Sjöblom et al.

    Content structure is king: An empirical study on gratifications, game genres and content type on Twitch

    Computers in Human Behavior

    (2017)
  • J.A. Bargh et al.

    The internet and social life

    Annual Review of Psychology

    (2004)
  • R.F. Baumeister et al.

    The need to belong: Desire for interpersonal attachments as a fundamental human motivation

    Psychological Bulletin

    (1995)
  • P.B. Brandtzæg et al.

    Why people use social networking sites

    Online Communities and Social Computing

    (2009)
  • G. Cheung et al.

    Starcraft from the stands: Understanding the game spectator

  • H.S. Chiang et al.

    YouTube stickiness: The needs, personal, and environmental perspective

    Internet Research

    (2015)
  • W.W. Chin

    The partial least squares approach to structural equation modeling

  • S. Cohen et al.

    Positive events and social supports as buffers of life change stress

    Journal of Applied Social Psychology

    (1983)
  • J. D’onfro

    Facebook takes on streaming apps live Periscope and Meerkat with its new ‘live video’ feature. Business Insider

    (2015, December 4)
  • N.B. Ellison et al.

    The benefits of Facebook “friends”: Social capital and college students' use of online social network sites

    Journal of Computer-mediated Communication

    (2007)
  • D. Fitzgerald et al.

    Apple quietly builds new networks

    The Wall Street Journal

    (2014, February 3)
  • C. Fornell et al.

    Evaluating structural equation models with unobservable variables and measurement error

    Journal of Marketing Research

    (1981)
  • Gamezone

    The rise of Twitch.tv: How this giant came to be. Gamezone

    (2014, May 15)
  • D. Gros et al.

    World of streaming. Motivation and gratification on Twitch

  • J. Hamari et al.

    What is eSports and why do people watch it?

    Internet Research

    (2017)
  • W.A. Hamilton et al.

    Streaming on Twitch: Fostering participatory communities of play within live mixed media

  • Cited by (473)

    • The Streamer's sales strategy choice considering sales effort

      2024, Journal of Retailing and Consumer Services
    View all citing articles on Scopus
    View full text