Computers in Human Behavior 52 (2015) 190–199

Contents lists available at ScienceDirect

Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh

Channels matter: Multimodal connectedness, types of co-players and social capital for Multiplayer Online Battle Arena gamers Jingbo Meng a,⇑, Dmitri Williams b, Cuihua Shen c a

Department of Communication, Michigan State University, United States Annenberg School for Communication and Journalism, University of Southern California, United States c Department of Communication, University of California at Davis, United States b

a r t i c l e

i n f o

Article history: Available online 15 June 2015 Keywords: Online gaming Social capital Multimodal connectedness Co-player

a b s t r a c t The study aimed to examine the roles and interactions of (1) multimodal connectedness and (2) three types of co-player networks in online gamers’ social capital acquisition. Over 17,000 players of the popular game League of Legends were surveyed on their playing partners, the media channels used, and social capital. Combined with behavioral data from server logs, the results showed that multimodal connectedness (i.e., the number of communication channels used for social interaction among players) was positively associated with one’s bridging and bonding social capital. The frequency of playing with an existing offline friend was positively associated with one’s bridging and bonding social capital; the frequency of playing with an online friend first met in the game was positively associated with one’s bridging social capital; the frequency of playing with a family member was not a significant predictor of one’s social capital outcomes. Moreover, multimodal connectedness magnified the positive relationships found between social capital outcomes and playing with online and offline friends. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction The rapid growth of the Internet has made the world a global networked society (Castells, 2000), and created new forms of sociability to facilitate both online and offline interactions (Katz & Rice, 2002). Online games have emerged as a new space where people can interact with one another, often bridging the offline context with the online one. In recent years, a growing body of research has examined social aspects of online games such as World of Warcraft (WOW), EverQuest, or the virtual world of Second Life (Cole & Griffiths, 2007; Guitton, 2012a,b, 2015; Williams, Yee, & Caplan, 2008). Following the tradition of research on impacts of Internet usage (e.g., Kraut et al., 1998, 2002), the present study aims to examine social effects of playing online games, with a particular interest in the accumulation of social capital. Although quite a few studies have investigated social capital outcomes of online gamers (Steinkuehler & Williams, 2006; Williams et al., 2006; Zhong, 2011), these have several limitations. First, it is nearly definitional in communication research that the medium should matter (McLuhan, 1995), yet very few studies so far treat games as various ways people communicate through the game (Guitton, 2012b, 2015). Online games often integrate media modalities (e.g., text, voice, image, etc.) so that players are able to ⇑ Corresponding author. http://dx.doi.org/10.1016/j.chb.2015.06.007 0747-5632/Ó 2015 Elsevier Ltd. All rights reserved.

interact with one another through multiple communication channels. It is important to understand if and what extent online game players use a spectrum of communication channels for social interaction (Rice, 1992). This paper introduces the concept of multimodal connectedness among online gamers and examines its relation with players’ social capital. Second, previous research has less often explicitly investigated and compared the effects of playing with different relational categories of co-players. It is not until recently that game researchers have begun to consider play partners as a central part of an integrated gaming experience (Shen & Chen, 2015; Shen & Williams, 2011; Waddell & Peng, 2014). Pew research has shown that 47% of teens play online games with people they know in their offline lives (Lenhart, Jones, & Macgill, 2008); A study of over 5000 EverQuest II players found that 69% played the game with their friends or families (Shen & Williams, 2011). As co-play is a significant phenomenon in online games (Zhong, 2011), it becomes essential to understand and compare social capital outcomes of playing with different types of relational partners. This paper borrows the concept of tie strength and examines social capital acquisition of playing with three types of relational partners: family members, pre-existing offline friends, and online friends meet for the first time in the game. Moreover, the current study is situated in an emerging genre of online games, the so-called Multiplayer Online Battle Arena

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199

(MOBA), which has experienced a rapid rise in popularity over the past five years (Pereira, 2014). MOBA is an eSports (Kain, 2014), and it is characterized by competition between two teams of players. Even though this genre shares important features with MMOGs, such as 3D representation and multiplayers interacting online, MOBAs are distinct in their game mechanisms in several aspects. It will be worthwhile to examine if this genre leads to different social capital processes and outcomes. The findings will contribute to the literature on online gaming and social capital in general. 2. Conceptual framework, hypotheses and research questions 2.1. Social capital and online games Social capital is a theoretical framework that traces back to the work of Bourdieu (1985) and Coleman (1988), with further extension by Burt (1992), Putnam (2000) and Lin (2001). Each theorist defined social capital somewhat differently (see Adler & Kwon, 2002). The present study will employ the definition widely used in research on online gaming and social media. Social capital is broadly defined as the benefits (e.g., informational and support-based resources) obtained through the social relationships in which one is embedded (Ellison, Steinfield, & Lampe, 2007; Reer & Kramer, 2014; Shen & Cage, 2013). Online gaming researchers often refer to Putman’s delineation of two basic forms of social capital: bridging and bonding (Reer & Kramer, 2014; Williams, 2006a). Bridging social capital comprises weak ties wherein individuals from different backgrounds exchange diverse information, open up opportunities for new resources and broaden social horizons (Williams, 2006a). Bonding social capital comprises strong ties wherein a tightly knit group such as families and close friends share intimacy and provide emotional and substantive support over a longer period of time (Williams, 2006a). Since the appearance of Putnam’s (2000) pessimistic view on mass media and degrading social capital in the United States, a series of projects have found mixed results with respect to the Internet’s effects on people’s social relationship development and maintenance (e.g., Kraut et al., 1998; Kraut et al., 2002; Wellman, Haase, Witte, & Hampton, 2001). The inconsistent findings called for more scholarly attention to specific services that people use on the Internet (Steinkuehler & Williams, 2006), as some services are social and conducive to social capital acquisition (e.g., chat rooms, virtual communities) (Guitton, 2015), and others are not (e.g., video streaming, online shopping). Similar to specific services such as social network sites (Ellison et al., 2007), online games have been examined in relation to socialization and social capital outcomes (Guitton, 2012a,b; Williams, 2006a). Several studies found positive effects of playing online games on bridging social capital. For example, MMOGs provided a virtual place where a range of social interactions could occur (Steinkuehler & Williams, 2006). Playing MMOGs facilitated bridging social capital by providing more opportunities to develop new friendships, expand network diversity and nurture a sense of global community (Kobayashi, 2010; Williams, 2006a). However, findings diverge on the impact on bonding social capital. Early studies on MMOGs reported that building up strong relationships among players was possible but very rare (Williams et al., 2006). More recent studies showed that playing MMOGs either positively (Skoric & Kwan, 2011) or negatively predicted bonding social capital (Zhong, 2011). The mixed findings may be partially explained and resolved by filling two gaps in the literature. First, players’ social capital outcomes may vary depending on the extent to which they use multiple media to communicate with one another (Chan, 2014). However, research often focuses on the impacts of the game itself rather than the ways people use the game as a true medium for communication. Online games can be considered as

191

additional communication channels in players’ daily personal communication systems. Players have greater chances to develop new relationships and strengthen their existing relationships if more media are employed for social interaction (Wellman et al., 2001). Second, players’ social capital outcomes may vary depending on different types of co-players. Although co-play with others positively predicted bridging and bonding social capital (Trepte, Reinecke, & Juechems, 2012; Zhong, 2011), fewer studies explicitly examined the relations between different types of co-players and social capital outcomes. A couple of recent studies found that playing with pre-existing ties reduced players’ levels of loneliness (Shen & Williams, 2011), and examined co-play patterns in relation to players’ network sizes (Shen & Chen, 2015). This paper will directly examine different social capital outcomes based on specific relational categories of co-players. 2.2. Multimodal connectedness and social capital Game researchers have found that the development and maintenance of social relationships among players vary by in-game media use. For example, guildmates who used voice chat had higher levels of relationship strength and trusted one another more than guildmates who used text chat (Ratan, Chung, Shen, Williams, & Poole, 2010; Williams, Caplan, & Xiong, 2007). While these studies are valuable and informative, they typically focus on the effects of a single communication channel. In reality, most online games such as MOBAs and MMOGs provide and are embedded in multimodal communication environments. That is to say, an individual player has a personal communication system composed of a variety of communication channels to interact with other players (Boase, 2008). Multimodal connectedness, defined as ‘‘the various modalities through which people maintain their connections with each other’’ (Schroeder, 2010, p. 79), manifests the fact that individual players have a repertoire of communication channels (i.e., text, voice, images, etc.) for social interaction. The present study does not examine the effect of any specific communication modality. Instead, it focuses on the effect of multimodal connectedness (i.e., the number of communication channels a player uses to communicate with other players) on social capital. Without using multiple communication channels, individuals’ social capital acquisition can be very limited in online games. According to media niche theory (Dimmick, Kline, & Stafford, 2000), each communication channel has its own niche to meet the gratifications of communication actors (Dimmick et al., 2000). As multimodal connectedness increases, more gratification niches will be fulfilled and created, thereby leading to more means and opportunities for communication. For example, in-game text chat may serve to initiate new connections among players. However, given the lack of cues in computer-mediated text communication (Culnan & Markus, 1987), diverse information may be rarely exchanged or the horizons may be rarely broadened during the relatively short and intensive game sessions in MOBAs. If two players are of different cultures or political ideologies, but never know this about each other, the practical effect of their diversity is negated (Williams et al., 2007) and gains in social capital are in doubt. Multiple communication channels reintroduce communication cues that are important to enable the exchange of diverse information and the horizon-broadening function. For example, the use of game-related discussion forums facilitates communication at a more flexible pace and across a broader range of topics, while the use of social network sites allows acquainted players to know more about each other beyond game settings (Ellison et al., 2007). Moreover, the use of voice chat increases sociability and trust among players, as they know the person talking (Ratan et al., 2010). As trust is being built among players, the use of more private communication channels, such as instant messaging, helps to secure practical and

192

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199

emotional support (Haythornthwaite, 2005). Based on these arguments, it is hypothesized that, H1. Multimodal connectedness is positively associated with a player’s (a) bridging, and (b) bonding social capital.

H2. The frequency of playing with family members is positively associated with one’s bonding social capital.

H3. The frequency of playing with pre-existing offline friends is positively associated with one’s (a) bonding social capital, and (b) bridging social capital.

2.3. Types of co-players and social capital The type of social relation is an important indicator of tie strength (Granovetter, 1973). Tie strength is defined as ‘‘a (probably linear) combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which characterize the tie’’ (Granovettter, 1973, p. 1361). The concept ranges from weak ties at one extreme to strong ties at the other. By definition, the concept of tie strength serves as a heuristic guide to the assumption that interactions with strong ties facilitate bonding social capital acquisition whereas interactions with weak ties facilitate bridging social capital acquisition. Tie strength is traditionally indicated by several variables such as closeness (Marsden & Campbell, 2012), and type of social relation (e.g., family, close friends, acquaintance; Granovetter, 1973). Online gaming research has focused on different relational categories of co-players, demonstrating that the people with whom gamers play are associated with different social and psychological outcomes (Shen & Chen, 2015; Shen & Williams, 2011). For example, playing MMOGs with family members increased family communication time and quality (Coyne, Padilla-Walker, Stockdale, & Day, 2011); playing MMOGs with existing offline friends predicted lower levels of loneliness (Shen & Williams, 2011). These studies suggest that playing with pre-existing social ties may make a difference in social capital outcomes. Positive social effects could emerge when players play with families and pre-existing offline friends because online games, just like other new media technologies such as social network sites, provide additional interaction opportunities for interaction (Ellison et al., 2007), which stimulates shared activities and conversation topics (Nardi & Harris, 2006). A less studied yet important relational category is online friends who meet for the first time in the game. Online gamers have opportunities to develop new friendships in the game by either being randomly assigned into combat teams or actively seeking other gamers as teammates. Acquisition of bridging social capital is possible in that social interactions and conversations are as important as in-game combat activities. Teammates can communicate combat strategies, share information, comment on players’ individual or shared experiences, and talk about ‘‘off-topic’’ issues such as politics, culture and offline mundanities (Murphy, 2011). While family members are traditionally considered strong ties (Gilbert & Karahalios, 2009; Granovetter, 1973; Vetere et al., 2005), pre-existing offline friends may range from strong ties such as close friends to weak ties such as acquaintances (Gilbert & Karahalios, 2009; Granovetter, 1973). Therefore, we hypothesize that playing with family members predicts one’s bonding social capital whereas playing with pre-existing offline friends predicts both bonding and bridging social capital. It is generally believed that online relationships are weak ties (Chan & Cheng, 2004; Cummings, Butler, & Kraut, 2002; Mesch & Talmud, 2007; Williams, 2007), therefore, we hypothesize that playing with online friends first met in the game predicts one’s bridging social capital. Although some studies suggested that game-related online friendships may not be as weak and shallow as assumed (Haythornthwaite, 2000; Shen & Chen, 2015), given the lack of extensive evidence, we will explore the relationship between playing with online friends first met in the game and one’s bonding social capital.

H4. The frequency of playing with online friends first met in the game is positively associated with one’s bridging social capital. RQ1. What is the relationship between the frequency of playing with online friends first met in the game and one’s bonding social capital? 2.4. Interaction between multimodal connectedness and types of co-players This paper will also investigate the potential interaction effects between multimodal connectedness and different types of co-players on social capital outcomes. According to media multiplexity theory (Haythornthwaite, 2005), the strength of interpersonal ties is associated with a diverse array of communication channels employed. Strong tie relational partners tend to use multiple communication channels for relationship maintenance (Baym, Zhang, & Lin, 2004). Empirical studies found that college students used more means of communication, such as face-to-face, telephone, social network sites and instant messaging, to stay in touch with those who were perceived as friends than those perceived as less close (Ledbetter, 2009). Pre-existing strong ties benefit from the integration of communication channels in that they provide new forms of multimodal connectedness, and in turn, facilitate more frequent and diverse interactions, as well as greater levels of intimacy relative to those with fewer channels of communication (Chan, 2014). In other words, media multiplexity fosters bonding social capital because of more communication opportunities with strong ties afforded by multiple communication channels. According to media multiplexity (Haythornthwaite, 2005), weak tie relational partners tend not to use multiple communication channels, instead relying on one primary medium for interaction and relationship maintenance. The addition of a new medium in communication helps to preserve weak tie relationships by counteracting the disruptive effect of losing primary communication channel. Therefore, multimodal connectedness prevents overreliance on a single primary communication medium, so that weak ties have a better chance to sustain and develop. Therefore, we hypothesize that multimodal connectedness will moderate the relationship between bonding social capital and playing with strong tie relationships, as well as the connection between bridging social capital and playing with weak tie relationships. H5. The relationship between bonding social capital and playing with (a) family members, and (b) pre-existing offline friends will be stronger, for those who have higher levels of multimodal connectedness.

H6. The relationship between bridging social capital and playing with (a) pre-existing offline friends, and (b) online friends first met in the game will be stronger, for those who have higher levels of multimodal connectedness.

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199

Lastly, literature has pointed out a few important control variables to be included in the study. Socio-demographic variables, including age, gender, education and income, and personality trait such as extroversion are related to uses of social applications on the Internet and social benefits gained from using communication technologies (Correa, Hinsley, & Zuniga, 2010; Kraut et al., 2002). Total number of game sessions played and game performance also influences social outcomes of playing online games (Shen & Williams, 2011). Therefore, these variables are taken into consideration in our study design and analysis. 3. Material and methods 3.1. Study site: League of Legends League of Legends (LoL) is one of the most-played PC game in the world (Levy, 2014). Released on October 27, 2009, LoL had 70 million registered users by 2013, with 27 million playing on a daily basis (Pereira, 2014). LoL is a typical MOBA featuring strategic combat between teams of three or five players. It shares important features with MMOGs, such as 3D representation, multiple players interacting online, and multimodal communication channels enabled. One major difference between LoL and a typical MMOG is that there is not a persistent world in LoL; instead, it is consisted of a series of independent combat matches, which typically last between 25 and 45 min. There is a persistent account identity for players, and players may mix and match their teammates, or play alone and be placed on a team by the system. Players are allowed to label other players as ‘‘friends’’ so that they can form teams with them more easily in the future. Team members are crucial to individual players because the success of a player—both for their overall account (their ‘‘summoner’’) and for the particular session—depends on them. Sessions begin when each player chooses a character, or ‘‘champion’’ to play. Playing as the champion, players strive to destroy their enemy’s forces (e.g., towers and minions), and kill enemy champions. When a team wins a match, there is a much greater boost to individual summoners on the team than when the team loses. LoL therefore puts an emphasis on who to play with in determining the success in the game. To avoid the uncertainty of random strangers and the frequently poor behavior found in anonymous environments, many players arrange teams with their offline friends and families, or look for potential online friends on LoL’s discussion board to recruit teams. Clans are formed among players voluntarily to facilitate pairing with active players to play together at a regular time. Given the feature of multimedia and the importance of teammates while playing the game, LoL is an adequate testbed for the hypotheses and research questions proposed in this study. Moreover, as a different genre of online games, LoL is worth studying in that it manifests different game mechanics and it serves as a good contrast to social capital processes and outcomes found in MMOGs. 3.2. Data collection This study collected data through an online survey within LoL. Responses were collected via an online web-based system during a one-week period in October 2010. Survey respondents were randomly selected for participation if they had played a match within the month prior to the survey window. These players received an email directly from the developer, Riot Games, inviting them to participate and linking them to a unique web page. In exchange, Riot offered the players a ‘‘boost’’ to their account, which allowed players to more quickly collect a particular kind of points they

193

can only earn through playing matches. These points are used for improving either the player’s in-game abilities or their aesthetic appearance, making them appealing to all types of players. The survey data were linked with the behavioral log data by using players’ unique account numbers. All personally identifying information was removed, making the analysis data set totally anonymous. The response rate can be calculated either based on the number of emails sent, or the number actually opened. The ‘‘open rate’’ is the percentage of email recipients who opened the email and thereby downloaded the graphics files in them. 113,579 emails were sent, and 25,996 (22.9%) of these were opened. Of these, 22,521 complete responses were collected during the one-week window. Based on the raw number of emails sent, this was a response rate of 19.8%. Based on the number of emails opened, this was a response rate of 86.6%. The measures used in this study were embedded in a larger instrument. Pre-tests of the survey instrument suggested a mean completion time of 18 min. Subjects who completed the study too fast were suspected of simply clicking through in order to receive the incentive. Moreover, some participants tried to take the survey twice to gain double incentives. For these cases, the first survey entry was kept, whereas the second entry was excluded from the analysis. Therefore, after excluding duplicates and subjects who completed the study too fast, we had a total of 17,995 cases left.

3.3. Measurement 3.3.1. Frequency of playing with different types of co-players Participants were asked how often they played LoL with each of the following: a friend who they knew offline before joining the game, a family member, and a friend they first met online in LoL. The response options were never (1), rarely (2), sometimes (3), often (4) and always (5).

3.3.2. Multimodal connectedness The first question about media use asked participants how often they communicated with other players through each of the following, while playing LoL: in-game text chat, voice chat (e.g., Ventrilo, TeamSpeak, Skype), and face-to-face (e.g., in the same room). The second question about media use asked participants how often they communicated with other players through each of the following, while not playing LoL: voice chat (e.g., Ventrilo, TeamSpeak, Skype), face-to-face, discussion board, instant messaging (e.g., AIM, MSN), email, telephone/mobile phone (including SMS), and social network sites. The response options for both questions were never (1), rarely (2), sometimes (3), often (4) and always (5). Multimodal connectedness was computed for each participant in the following steps: first, measures of media use were recoded into either 0 (if ‘‘never’’ was chosen) or 1 (if other responses were chosen); second, multimodal connectedness was computed by adding up the number of unique communication channels used, ranging from 0 to 7.

3.3.3. Social capital This study adopted Williams’ (2006b) Internet Social Capital Scales (ISCS) to measure individual-level bridging and bonding social capital. Participants were asked to rate the items on a five-point Likert scale (1 = strongly disagree and 5 = strongly agree). 10 items focused on weak-tie relationships (e.g., ‘‘Interacting with people makes me interested in things that happen outside of my town’’), and 10items focused on strong-tie relationships (e.g., ‘‘There are several people I trust to help solve my problems’’). Cronbach’s alpha for bridging social capital was .89, and for bonding social capital, .84.

194

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199

3.3.4. Control variables Taking advantage of unobtrusively collected behavioral log data provided by the game company, the amount of play was measured by the total number of sessions a player had played and finished. The range of total sessions played was from 1 to 1749 (M = 328.03, SD = 242.22). The amount of play was scored by multiplying the original number with 5/1749 to make it on the consistent scale with other variables. Game performance was measured by average experience points earned per session and the percentage of wins (i.e., sum of wins/total games played). Data on players’ personality traits were also collected. Participants were asked how well the following words described them: competitive, affectionate, analytical, artistic, bold and shy (Diekman & Eagly, 2000; Saucier, 1994). The response options were not at all (1), slightly (2), moderately (3), very much (4) and extremely (5). Two items ‘‘bold’’ and ‘‘shy’’ were used as proxy measures of extroversion in the analysis (Saucier, 1994). In addition, basic socio-demographic information including age, gender, education and household income was collected in the survey. They were all included in the analysis as control variables. Specifically, participants were asked to indicate their highest level of schooling with response options of (1) less than high school, (2) high school diploma, (3) some college, (4) associates degree, (5) Bachelor’s degree, (6) some graduate training, and (7) graduate degree. Income was measured by asking for the household annual income of participants. Response options started with (1) 6 5000, incremental by denominations of $5,000 through (31) = $150,000 or more.

step, control variables were entered into two regression models with bridging (in Model 1) and bonding (in Model 2) social capital as the dependent variable respectively. The results showed that age was negatively associated with bridging social capital, b = .05, p < .001, but positively associated with bonding social capital, b = .03, p < .05; Being female was positively associated with bridging (b = .03, p < .01) and bonding social capital (b = .03, p < .01); higher levels of education was positively associated with bridging (b = .03, p < .01) and bonding social capital (b = .04, p < .001); higher levels of income was positively associated with bridging (b = .04, p < .001) and bonding social capital (b = .11, p < .001). Players’ boldness was positively associated with bridging (b = .23, p < .001) and bonding social capital (b = .15, p < .001), whereas the shyness was negatively associated with bridging (b = .13, p < .001) and bonding social capital (b = .18, p < .001). In addition, the average point earned per session was positively associated with bonding social capital, b = 0.05, p < .01. Models 1 and 2 explained 8% and 9% of the variance in players’ general bridging and bonding social capital respectively.

4.2. Hypotheses testing In the second step, four independent variables (i.e., multimodal connectedness, frequencies of playing with three different types of co-players) were added into the regression models with bridging (in Model 3) and bonding (in Model 4) social capital as the dependent variable respectively. This step aimed to test H1–H4 and answer RQ1. The results showed that multimodal connectedness was positively associated with bridging (b = .08, p < .001) and bonding (b = .05, p < .001) social capital. Therefore, H1 was supported. The frequency of playing with an existing offline friends was positively associated with bridging (b = .04, p < .001) and bonding (b = .11, p < .001) social capital, while the frequency of playing with a family member was not a significant predictor of one’s social capital outcomes. Therefore, H2 was not supported but H3 was supported. The results also showed that the frequency of playing with a friend first met in LoL was positively associated with bridging social capital (b = .09, p < .001), but it was not a significant predictor to bonding social capital. Therefore, H4 was supported, and the answer to RQ1 was that the frequency of playing with a friend first met in LoL was not related to players’ bonding social capital. Models 3 and 4 explained 11% of the variance in players’ general bridging and bonding social capital respectively. Models 3 and 4 with the independent variables included explained significantly more variance in bridging social capital than Model 1,

4. Results 4.1. Descriptive statistics Preliminary analyses showed that 95.9% of the participants were male, 56.9% were White, 20.6% were Asian, 8.9% were Hispanic, 1.6% were Black and the rest were of other ethnicities. Of these participants, 56.9% were from the United States and the rest were from 150 different countries, ranging from Canada (13.2%), Australia (4.4%) and Brazil (4%) to Poland (2%), China (1.2%) and Germany (1.4%). The average age of participants was 21.90 (SD = 5.22), and over 90% of respondents were between 16 and 30 years old. Table 1 presents zero-order correlations among all variables, as well as their means and standard deviations. Hierarchical multiple regression was used to test H1–H6 and answer RQ1. In the first

Table 1 Descriptive statistics of and zero-order correlations between all important variables. 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Age Education Income Trait: Bold Trait: Shy Number of sessions played Avg. points per session Pct. of win Multimodal connectedness Play with an offline friend Play with a family member Play with an online friend Bridging social capital Bonding social capital Mean SD

2

3

4

5

6

7

8

1 0.45 0.03 0.03 0.10 0.03 0.08 0.03 0.06 0.10 0.02 0.05 0.01 0.05

1 0.03 0.00 0.08 0.01 0.07 0.01 0.00 0.04 0.02 0.04 0.02 0.07

1 0.07 0.09 0.01 0.03 0.01 0.01 0.06 0.03 0.00 0.07 0.16

1 0.25 0.03 0.05 0.03 0.08 0.08 0.06 0.10 0.26 0.22

1 0.02 0.02 0.03 0.03 0.04 0.02 0.01 0.16 0.22

1 0.31 0.03 0.06 0.00 0.05 0.17 0.00 0.01

1 0.61 0.02 0.01 0.05 0.04 0.01 0.04

21.9 5.22

3.38 1.65

11.76 9.04

3.23 1.07

2.86 1.19

0.94 0.70

119.68 19.38

Note: non significant (p > .05) correlations are in italic.

9

10

11

12

13

14

1 0.03 0.03 0.01 0.04 0.02 0.02

1 0.37 0.20 0.23 0.15 0.11

1 0.15 0.09 0.10 0.15

1 0.10 0.05 0.02

1 0.13 0.04

1 0.56

1

0.51 0.07

4.04 2.50

3.71 1.19

1.94 1.28

3.13 1.24

3.80 0.69

3.80 0.71

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199

F(4, 10587) = 63.42, p < .001, and in bonding social capital than Model 2, F(4, 10434) = 44.23, p < .001. In the third step, three interaction terms between multimodal connectedness and three types of co-players were added into the regression models with bridging (in Model 5) and bonding (in Model 6) social capital as the dependent variable respectively. This step aimed to test hypotheses of interaction effects, H5 and H6. The results showed that the interaction between multimodal connectedness and the frequency of playing with a family member was not a significant predictor of one’s social capital outcomes. Therefore, H5a was not supported. The results also revealed that the interaction between multimodal connectedness and the frequency of playing with an existing offline friend was positively associated with bridging (b = .05, p < .001) and bonding social capital (b = .04, p < .001); the interaction between multimodal connectedness and the frequency of playing with a friend first met in LoL was positively associated with players’ bridging social capital only, b = .03, p < .01. Therefore, H5b, H6a&b were supported. Models 5 and 6 explained 12% of the variance in players’ general bridging and bonding social capital respectively. Models 5 and 6, with the addition of interaction terms, explained significantly more variance in bridging social capital than Model 3, F(3, 10584) = 12.60, p < .001, and in bonding social capital than Model 4, F(3, 10431) = 8.53, p < .01. Figs. 1 and 2 depict the interaction effects between multimodal connectedness and playing with an existing offline friend on players’ bridging and bonding social capital respectively. Table 2 presents the results of hierarchical multiple regression models.

4.3. Post-hoc analysis To further explore the relationships between the key variables examined in this study, multidimensional scaling was conducted using a Smallest Space Analysis (SSA) in Statistica 12. A distance matrix was generated based on the zero-order correlation matrix in Table 1. The result was visualized in Fig. 3. It showed a few patterns such that playing with family, offline friends and online friends first met in the game were all clustered around multimodal connectedness; number of sessions played and boldness were the next closest to multimodal connectedness; bonding and bridging social capital were associated with each other, with the latter being closer to multimodal connectedness and playing with different types of co-players than the former; income was the variable closest to bonding social capital; boldness was the variable closest to bridging social capital; and age, education, average points earned per session, percentage of wins and shyness were relatively far from social capital outcomes and multimodal connectedness.

195

5. Discussion The present study aimed to examine the roles and interactions of multimodal connectedness and three types of co-players in online gamers’ social capital outcomes. Multimodal connectedness referred to the number of communication channels that players used to interact with other players. We found that multimodal connectedness was positively associated with players’ bridging and bonding social capital (H1a&b). The result of multimodal connectedness among players was consistent with the augmentation perspective of Internet use. The displacement perspective states that the Internet makes people more isolated and reduces social capital, whereas the augmentation perspective states that the Internet enhances existing social relationships and enlarges one’s social networks by bringing new people (e.g., Boase, Horrigan, Wellman, & Rainie, 2006; Quan-Haase, Wellman, Witte, & Hampton, 2002). Our results indicated that the use of multiple media to communicate with other players played a significant role in the process of social capital acquisition for online gamers. Moreover, as multimodal connectedness was operationalized as various communication channels used both while playing the game and while not playing the game, the result suggested the importance of both using the game as a true medium (i.e., in-game communication channels) and the larger media ecosystem in which the game was embedded (i.e., out-of-game communication channels). While in-game communication channels contribute to relationship building (Ratan et al., 2010; Williams et al., 2007), communication venues outside of the actual game system provide complementary means for players to interact with each other. For example, official and unofficial discussion forums are online spaces where LoL players interact in a wide range of activities, varying from sharing game-related information, casual talk about mundane life, to recruiting team members and exchanging Skype and email accounts information for further interactions. Among the three types of co-players, the frequency of playing with existing offline friends was positively associated with one’s bonding social capital (H2b), whereas the frequency of playing with family member (H2a) and of playing with online friends first met in LoL (RQ1) were not associated with one’s bonding social capital. The findings demonstrated the important role of different types of co-players in online gamers’ accumulation of bonding social capital. They were consistent with most findings in a recent study in a MMOG (Shen & Chen, 2015), and revealed a similar pattern as in studies of the impacts of social network sites (Ellison, Steinfield, & Lampe, 2011) and the Internet on the development and maintenance of social relationships (Shklovski, Kraut, & Rainie, 2006). The similar findings all suggested that the ties with family members may have been strong and stable enough so that

5

Bridging social capital

4.5 4

Low multimodal connectedness

3.5

Moderate multimodal connectedness

3

High multimodal connectedness

2.5 2 1.5

Low

Mean

High

Frequency of playing with a pre-existing offline friend

Fig. 1. The effect of interaction between multimodal connectedness and the frequency of playing with existing offline friends on bridging social capital. Note: For horizontal axis – the frequency of playing with an existing offline friend, Low = Mean SD, High = Mean + SD.

196

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199 5

Bonding Social Capital

4.5

Low multimodal connectedness

4

Moderate multimodal connectedness

3.5

High multimodal connectedness

3 2.5 2

Low

Mean

High

Frequency of playing with an existing offline friend Fig. 2. The effect of interaction between multimodal connectedness and the frequency of playing with existing offline friends on bonding social capital. Note: For horizontal axis – the frequency of playing with an existing offline friend, Low = Mean SD, High = Mean + SD.

Table 2 Hierarchical multiple regression models with bridging and bonding social capital as separated dependent variables (standardized coefficients are reported in the table). Bridging social capital (DV)

Age Gender Education Income Trait being bold Trait being shy Number of sessions played Avg points per session Percentage of wins Multimodal connectedness (H1) Play with a family member (H2) Play with an existing offline friend (H3a&b) Play with a friend first met in LoL (H4&RQ1) Multimodal connectedness Play with a family member (H5a) Multimodal connectedness Play with an existing offline friend (H5b and 6a) Multimodal connectedness Play with a friend first met in LoL (H6b) Adjusted R2

***

Model 3

Model 5

Model 2

Model 4

Model 6

0.05*** 0.03** 0.03** 0.04*** 0.23*** 0.13*** 0.01 0.02 0.02

0.04*** 0.03** 0.04** 0.04*** 0.21*** 0.13*** 0.01 0.02 0.03* 0.08*** 0.01 0.04*** 0.09***

0.04*** 0.03** 0.04** 0.04*** 0.21*** 0.12*** 0.01 0.02 0.03* 0.10*** 0.01 0.05*** 0.09*** 0.02

0.03* 0.03** 0.05*** 0.11*** 0.15*** 0.18*** 0.01 0.05** 0.01

0.02* 0.02** 0.05*** 0.12*** 0.15*** 0.18*** 0.01 0.05** 0.01 0.05*** 0.01 0.11*** 0.01

0.02* 0.02* 0.05*** 0.12*** 0.15*** 0.18*** 0.01 0.05** 0.01 0.05*** 0.01 0.11*** 0.01 0.02

0.05***

0.04***

0.03** 0.08

***

***

0.11

***

0.12

0.01 0.09

***

***

0.11

0.12***

p < .05. p < .01. p < .001.

shy

session

Dimension 2

* **

Bonding social capital (DV)

Model 1

ratio of win online friends avg. points

family

multimodal offline friends age

education bonding

bridging bold

income

Dimension 1 Fig. 3. Multidimensional scaling visualization (Alienation = .19; Stress = .15).

playing with them did not have a significant impact on one’s bonding social capital. In contrast, ties with existing offline friends may wane if communication and interactions rarely take place. Playing with existing offline friends could possibly create more opportunities and supply new means of social interaction, thereby strengthening the existing relationships. It is worth noting that we did not find any relationship between the frequency of playing with online friends first met in LoL and one’s bonding social capital, whereas Shen and Chen (2015) found that gamers who played frequently with friends first met in a MMOG had more strong ties in their core networks. The different finding may be attributed to the game mechanics of LoL as a MOBA, which is less a social game than MMOGs because it does not provide a persistent world. Since MOBAs do not facilitate sociability and nurture long-term sustained communities in the way that MMOGs do, it may be more difficult to build up strong connections among players who initiate their relationships in the game. With respect to bridging social capital, the frequencies of playing with existing offline friends and online friends first met in LoL

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199

were significant predictors. The findings were consistent with studies on MMOGs, such that online games could provide additional opportunities for existing friends to interact (Shen & Williams, 2011), and foster new connections with those beyond one’s existing social networks to broaden one’s worldview and create social tolerance (Kobayashi, 2010; Williams, 2007). It was also echoed by the finding that playing with one’s existing friends made within and outside of a MMOG had a positive impact on both the size and diversity of one’s social network (Shen & Chen, 2015). This tells us that although MOBAs do not provide a persistent world as MMOGs do, players can possibly acquire bridging social capital by playing with their existing offline friends, or playing with new friends met in the game. It may be because MOBAs provide interactive environments in the form of diverse in-game communication channels, and are embedded in a highly interactive media ecosystem in the form of diverse out-of-game communication channels. Alternatively, it may be that the game and its ecology is not particularly social, but that people are undeniably so. The active social impulses suggested by Social Information Processing theory (Walther, 1992) or, on the cultural side by Jenkins (2006) suggest that humans are simply social. The study also found interesting interaction effects between multimodal connectedness and playing with three types of co-players on one’s social capital outcomes. It found that the relationship between the frequency of playing with pre-existing offline friends and both bridging and bonding social capital was stronger for those who had higher levels of multimodal connectedness; the relationship between the frequency of playing with online friends first met in the game and bridging social capital was stronger for those who had higher levels of multimodal connectedness. The findings illustrated that the use of multiple communication channels played a role in magnifying the relationships between specific types of co-players and one’s social capital outcomes. They highlighted the importance of providing multiple communication channels in online games, and leveraging a variety of media options surrounding online games, as communication via multiple media is critical for one’s social capital to grow. We could also encourage online gamers to communicate via multiple channels in order for them to build and maintain more meaningful relationships. Post-hoc analysis showed a few more observations. Multimodal connectedness was associated with the frequencies of playing with offline friends, family members and online friends first met in the game, indicating that in-game tie strength (i.e., frequency of playing with others) may foster using multiple forms of communication. Number of sessions played was associated with multimodal connectedness, meaning that in-game experience may also encourage using multiple forms of communication. Boldness is an important indicator of extraversion (Saucier, 1994), and it was close to using multiple forms of communication, bridging and bonding social capital. In contrast, shyness was fairly far from multimodal connectedness and social capital variables. These observations are consistent with the ‘‘rich-get-richer’’ phenomenon found in previous literature, such that extroverts tend to communicate and socialize more than introverts in online games (Shen & Williams, 2011). The study has a few limitations. First, the one-time survey data means that no causal relationship can be inferred regarding multimodal connectedness, three types of co-players and social capital outcomes. Players’ existing social capital may influence their choices of types of game, game mechanisms, with whom they play games and media use in- and out-of-game. Future research should design controlled experiments and surveys with multiple time points to disentangle the causal relationships of these variables. Second, as the study contributes to the literature of online gaming by examining a new genre MOBAs, its results may not be able to be

197

generalized to other genres of online games. As described in the method section, LoL players likely enjoy themselves more when teaming up with reliable players and as a result, over 80% of players play with pre-existing social ties. Therefore, future research should address the question that how different types of game influence the social context of playing, and the according social capital gained or lost in the process of playing. Third, three types of co-players served as major relationship categories, which were used to illustrate the concept of tie strength. Relationship category is only one of several ways to operationalize tie strength (Marsden & Campbell, 2012), and it is not the best choice. As discussed in the literature review, wide variations in tie strength exist within each relationship category, such as a player feeling emotionally close to an online friend first met in LoL. The present study chose relational categories due in part to making references to the existing literature on online games (e.g., Shen & Williams, 2011). Future research should incorporate more accurate measures of tie strength, such as closeness of co-players, to examine its relations to social capital outcomes. Future research should consider other types of co-players, such as strangers, or more refined types of co-players, such as existing offline close friends and acquaintances as a replacement of the broader category of existing offline friends. The use of more refined types of co-players will yield more informative findings. Finally, the response rate was low as 19.8% based on the raw number of emails sent. This could possibly be due to players not logging into their emails, or using a different email account for the game than their regular email account, and not checking it during the one week when data were collection. Nonetheless, the sample might not be representative of all LoL players. The sample was also disproportionately male, over-representing male players. Interpretations of the findings should take these into consideration. The data in this study were collected in 2010. Over the past 5 years, LoL has not changed its basic game structure, although it has added some social feature and maps. These are largely about avoiding the negative effects of strangers with ‘‘toxic’’ behavior, rather than different social infrastructure for known friends. So, it is unlikely that the findings here would be very different if replicated today. Still, future researchers may want to investigate how Riot’s ongoing social adjustments may alter the underlying social dynamics.

6. Conclusion In conclusion, this study considers multimodal connectedness and relational types of co-players as central variables associated with online gamers’ social capital acquisition. Situated in a popular MOBA game League of Legends, the study surveyed over 17,000 players on their uses of communication channels to interact with co-players, frequency of playing with different relational partners, and social capital outcomes. Our results indicate that the number of communication channels used for social interaction was related to players’ bridging and bonding social capital. Frequency of playing with preexisting offline friends was related to players’ bridging and bonding social capital, while frequency of playing with online friends first met in the game was related to players’ bridging social capital. Moreover, multimodal connectedness magnifies the relationships between social capital outcomes and frequency of playing with offline and online friends.

References Adler, P. S., & Kwon, S.-W. (2002). Social capital: Prospects for a new concept. The Academy of Management Review, 27(1), 17–40. http://dx.doi.org/10.5465/ AMR.2002.5922314.

198

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199

Baym, N. K., Zhang, Y. B., & Lin, M.-C. (2004). Social interactions across media: Interpersonal communication on the Internet, telephone and face-to-face. New Media & Society, 6(3), 299–318. http://dx.doi.org/10.1177/1461444804041438. Boase, J. (2008). Personal networks and the personal communication system: Using multiple media to connect. Information, Communication & Society, 11, 490–508. http://dx.doi.org/10.1080/13691180801999001. Boase, J., Horrigan, J. B.,Wellman, B., & Rainie, L. (2006). The strength of Internet ties: Pew Internet & American Life Project. Retrieved at 12/15/2014 from . Bourdieu, P. (1985). The forms of social capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). New York: Greenwood. Burt, R. S. (1992). Structural holes: The social structure of competition. Harvard University Press. Castells, M. (2000). The rise of the network society, the information age: Economy, society and culture (vol. I). Cambridge, MA: Blackwell. Chan, M. (2014). Multimodal connectedness and quality of life: Examining the influences of technology adoption and interpersonal communication on wellbeing across the life span. Journal of Computer-Mediated Communication. http:// dx.doi.org/10.1111/jcc4.12089. Chan, D. K.-S., & Cheng, G. H.-L. (2004). A comparison of offline and online friendship qualities at different stages of relationship development. Journal of Social and Personal Relationships, 21(3), 305–320. http://dx.doi.org/10.1177/ 0265407504042834. Cole, H., & Griffiths, M. (2007). Social interactions in massively multiplayer online role-playing gamers. CyberPsychology & Behavior, 10(4), 575–583. http:// dx.doi.org/10.1089/cpb.2007.9988. Coleman, J. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, S95–S120. Correa, T., Hinsley, A. W., & Zuniga, H. G. (2010). Who interacts on the Web? The intersection of users’ personality and social media use. Computers in Human Behavior, 26(2), 247–253. http://dx.doi.org/10.1016/j.chb.2009.09.003. Coyne, S. M., Padilla-Walker, L. M., Stockdale, L., & Day, R. D. (2011). Game on. . .girls: Associations between co-playing video games and adolescent behavioral and family outcomes. Journal of Adolescent Health, 49(2), 160–165. http://dx.doi.org/10.1016/j.jadohealth.2010.11.249. Culnan, M., & Markus, M. (1987). Information technologies. In F. Jablin, L. Putnam, K. Roberts, & L. Porter (Eds.), Handbook of organizational communication: An interdisciplinary perspective (pp. 420–443). Newbury Park, CA: Sage. Cummings, J., Butler, B., & Kraut, R. (2002). The quality of online social relationships. Communications of the ACM, 45(7), 103–108. http://dx.doi.org/10.1145/ 514236.514242. Diekman, A. B., & Eagly, A. H. (2000). Stereotypes as dynamic constructs: Women and men of the past, present, and future. Personality and Social Psychology Bulletin, 26, 1171–1188. http://dx.doi.org/10.1177/0146167200262001. Dimmick, J., Kline, S., & Stafford, L. (2000). The gratification niches of personal email and the telephone. Communication Research, 27(2), 227–248. http:// dx.doi.org/10.1177/009365000027002005. Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook ‘‘friends:’’ Social capital and college students’ use of online social network sites. Journal of Computer-Mediated Communication, 12, 1143–1168. http://dx.doi.org/10.1111/ j.1083-6101.2007.00367.x. Ellison, N. B., Steinfield, C., & Lampe, C. (2011). Connection strategies: Social capital implications of Facebook-enabled communication practices. New Media and Society. http://dx.doi.org/10.1177/1461444810385389. Gilbert, E., & Karahalios, K., 2009. Predicting tie strength with social media. In CHI proceedings of the SIGCHI conference on human factors in computing systems (pp. 211–220), Boston, Massachusetts, U.S. doi:http://dx.doi.org/10.1145/1518701. 1518736. Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, 78(6), 1360–1380. Guitton, M. J. (2012a). The immersive impact of meta-media in a virtual world. Computers in Human Behavior, 28, 450–455. http://dx.doi.org/10.1016/ j.chb.2011.10.016. Guitton, M. J. (2012b). Living in the Hutt space: Immersive process in the star wars role-play community of second life. Computers in Human Behavior, 28, 1681–1691. http://dx.doi.org/10.1016/j.chb.2012.04.006. Guitton, M. J. (2015). Swimming with mermaids: Communication and social density in the Second Life merfolk community. Computers in Human Behavior, 48, 226–235. http://dx.doi.org/10.1016/j.chb.2015.02.004. Haythornthwaite, C. (2000). Online personal networks. New Media & Society, 2(2), 195–226. Haythornthwaite, C. (2005). Social networks and Internet connectivity effects. Information, Communication & Society, 8, 125–147. http://dx.doi.org/10.1080/ 13691180500146185. Jenkins, H. (2006). Fans, bloggers, and gamers: Exploring participatory culture. New York, NY: NYU Press. Kain, E., 2014. The future of the MMO is casual, accessible and free-to-play. Retrieved at 12/15/2014 from . Katz, J., & Rice, R. (2002). Social consequences of Internet use: Access, involvement, and interaction. Cambridge, MA: The MIT Press. Kobayashi, T. (2010). Bridging social capital in online communities: Heterogeneity and social tolerance of online game players in Japan. Human Communication Research, 36(4), 546–569. http://dx.doi.org/10.1111/j.1468-2958.2010.01388.x.

Kraut, R., Kiesler, S., Boneva, B., Cummings, J., Helgeson, V., & Crawford, A. (2002). Internet paradox revisited. Journal of Social Issues, 58(1), 49–74. Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T., & Scherlis, W. (1998). Internet paradox: A social technology that reduces social involvement and psychological well-being? American Psychologist, 53(9), 1017–1031. Ledbetter, A. M. (2009). Patterns of media use and multiplexity: Associations with sex, geographic distance and friendship interdependence. New Media & Society, 11, 1187–1208. http://dx.doi.org/10.1177/1461444809342057. Lenhart, A., Jones, S., & Macgill, A. R. (2008). Adults and video games. Washington, DC: Pew Internet & American Life Project. Levy, K., 2014. What it’s like to play ‘League of Legends’, the game where the best players have millions on the line. Business Insider. Retrieved at 12/15/2014 from . Lin, N. (2001). Social capital: Theory and research. New York: Aldine de Gruyter. Marsden, P. V., & Campbell, K. E. (2012). Reflections on conceptualizing and measuring tie strength. Social Forces, 91, 17–23. http://dx.doi.org/10.1093/ sf/sos112. McLuhan, M. (1995). Understanding media: The extensions of man. Cambridge, MA: The MIT Press. Mesch, G. S., & Talmud, Ilan. (2007). Similarity and the quality of online and offline social relationships among adolescents in Israel. Journal of Research on Adolescence, 17(2), 455–466. http://dx.doi.org/10.1111/j.15327795.2007.00529.x. Murphy, D., 2011. Tencent acquires riot games (and league of legends) for $400M. Retrieved at 12/12/2011, from . Nardi, B., & Harris, J., 2006. Strangers and friends: Collaborative play in World of Warcraft. In Proceedings of the 20th anniversary conference on Computer Supported Cooperative Work (pp. 149–158), Banff, Alberta, Canada. doi:http:// dx.doi.org/10.1145/1180875.1180898. Pereira, C., 2014. League of Legends now has 27 million players daily. IGN Entertainment. Retrieved at 12/15/2014 from . Putnam, R. (2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon and Schuster. Quan-Haase, A., Wellman, B., Witte, J., & Hampton, K. (2002). Capitalizing on the net. In B. Wellman & C. Haythornthwaite (Eds.), The Internet in everyday life (pp. 291–324). Malden, MA: Blackwell. Ratan, R. A., Chung, J., Shen, C., Williams, D., & Poole, M. (2010). Schmoozing and smiting: Trust, social institutions, and communication patterns in an MMOG. Journal of Computer Mediated Communication, 16(1), 93–114. http://dx.doi.org/ 10.1111/j.1083-6101.2010.01534.x. Reer, F., & Kramer, N. C. (2014). Underlying factor of social capital acquisition in the context of online-gaming: Comparing World of Warcraft and Counter-Strike. Computers in Human Behavior, 36, 179–189. http://dx.doi.org/10.1016/ j.chb.2014.03.057. Rice, R. (1992). Task analyzability, use of new media, and effectiveness: A multi-site exploration of media richness. Organization Science, 3(4), 475–500. Saucier, G. (1994). Mini-markers: A brief version of Goldberg’s unipolar big-five markers. Journal of Personality Assessment, 63(3), 506–516. Schroeder, R. (2010). Mobile phones and the inexorable advance of multimodal connectedness. New Media Society, 12, 75–90. http://dx.doi.org/10.1177/ 1461444809355114. Shen, C., & Cage, C. (2013). Exodus to the real world? Assessing the impact of offline meet ups on community participation and social capital. New Media and Society. http://dx.doi.org/10.1177/1461444813504275. Shen, C., & Chen, W. (2015). Gamers’ confidants: Massively multiplayer online game participation and core networks in China. Social Networks, 40, 207–214. http:// dx.doi.org/10.1016/j.socnet.2014.11.001. Shen, C., & Williams, D. (2011). Unpacking time online: Connecting internet and massively multiplayer online game use with psychosocial well-being. Communication Research, 38(1), 123–149. http://dx.doi.org/10.1177/ 0093650210377196. Shklovski, I., Kraut, R., & Rainie, L. (2006). The Internet and social participation: Contrasting cross-sectional and longitudinal analyses. Journal of ComputerMediated Communication, 10(1), Article 1. http://dx.doi.org/10.1111/j.10836101.2004.tb00226. Skoric, M. M., & Kwan, G. C. E. (2011). Platforms for mediated sociability and online social capital: The role of Facebook and massively multiplayer online games. Asian Journal of Communication, 21(5), 467–484. http://dx.doi.org/10.1080/ 01292986.2011.587014. Steinkuehler, C., & Williams, D. (2006). Where everybody knows your (screen) name: Online games as ‘‘third places’’. Journal of Computer Mediated Communication, 11(4), 885–909. http://dx.doi.org/10.1111/j.10836101.2006.00300.x. Trepte, S., Reinecke, L., & Juechems, K. (2012). The social side of gaming: How playing online computer games creates online and offline social support. Computers in Human Behavior, 28(3), 832–839. http://dx.doi.org/10.1016/ j.chb.2011.12.003. Vetere, F., Gibbs, M. R., Kjeldskov, J., Howard, S., Mueller, F., Pedell, S., Mecoles, K., & Bunyan, M., 2005. Mediating intimacy: Designing technologies to support strong-tie relationships. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 471–480), Portland, Oregon. doi:http://dx.doi. org/10.1145/1054972.1055038. Waddell, J. C., & Peng, W. (2014). Does it matter with whom you slay? The effects of competition, cooperation and relationship type among video game players.

J. Meng et al. / Computers in Human Behavior 52 (2015) 190–199 Computers in Human Behavior, 38, 331–338. http://dx.doi.org/10.1016/ j.chb.2014.06.017. Walther, J. (1992). Interpersonal effects in computer-meditated interaction: A relational perspective. Communication Research, 19(1), 52–90. Wellman, B., Haase, A., Witte, J., & Hampton, K. (2001). Does the Internet increase, decrease, or supplement social capital. American Behavioral Scientist, 45(3), 436–455. Williams, D. (2006a). On and off the ‘Net: Scales for social capital in an online era. Journal of Computer Mediated Communication, 11(2), 593–628. http://dx.doi.org/ 10.1111/j.1083-6101.2006.00029.x. Williams, D. (2006b). Why game studies now? Gamers don’t bowl alone. Games Culture, 1, 13–16. http://dx.doi.org/10.1177/1555412005281774. Williams, D. (2007). The impact of time online: Social capital and cyberbalkanization. CyberPsychology & Behavior, 10(3), 398–406. http:// dx.doi.org/10.1089/cpb.2006.9939.

199

Williams, D., Caplan, S., & Xiong, L. (2007). Can you hear me now? The impact of voice in an online gaming community. Human Communication Research, 33(4), 427–449. http://dx.doi.org/10.1111/j.1468-2958.2007.00306.x. Williams, D., Ducheneaut, N., Xiong, L., Zhang, Y., Yee, N., & Nickell, E. (2006). From tree house to barracks: The social life of guilds in world of Warcraft. Games and Culture, 1(4), 338–361. http://dx.doi.org/10.1177/1555412006292616. Williams, D., Yee, N., & Caplan, S. E. (2008). Who plays, how much, and why? Debunking the stereotypical gamer profile. Journal of Computer-Mediated Communication, 13, 993–1018. http://dx.doi.org/10.1111/j.10836101.2008.00428.x. Zhong, Z.-J. (2011). The effects of collective MMORPG (Massively Multiplayer Online Role-Playing Games) play on gamers’ online and offline social capital. Computers in Human Behavior, 27, 2352–2363. http://dx.doi.org/10.1016/ j.chb.2011.07.014.