The moderating role of virtual community cohesion and critical mass on the link between online-game website service quality and player satisfaction

Wu, Y.-L., Tao, Y.-H., Lee, C.-P., Yang, P.-C., and Huang, G.-S.,The moderating role of virtual

community cohesion and critical mass on the link between online-game website service quality and play

satisfaction, in Proceedings of Advances in Social Network Analysis and Mining, July 25-27, Kaohsiung,

Taiwan, R.O.C., 2011.

The moderating role of virtual community cohesion and critical mass on the link

between online-game website service quality and player satisfaction

Yu-Lung Wu

Yu-Hui Tao

Ching-Pu Lee

Pei-Chi Yang

Guo-Shin Huang

Department of Information Management, I-Shou University, Kaohsiung County, Taiwana Department of Information Management, National University of Kaohsiung, Kaohsiung, Taiwanb

Department of Information Engineering, I-Shou University, Kaohsiung County, Taiwanc

E-mail address: isu9803005d@isu.edu.tw Tel numbers: 886-933696069

postal address: No.1, Sec. 1, Syuecheng Rd., Dashu District, Kaohsiung City 84001,Taiwan, R.O.C.

Abstract

-

moderating role while critical mass does not.

Discussions and implications are also provided in this paper.

Keywords: Online Game, E-Service Quality, Virtual Community Cohesion, Critical Mass

I. INTRODUCTION

With the rapid development of the Internet and related applications, computer games also began to evolve. Due to its innovative business model, online games integrated with

new network technologies successfully opened up the existing online game market. According to a Foreseeing Innovative New Digiservice (FIND) estimate, the online population in Taiwan is now more than 1,200 million people. Among them, 4 to 5 million are online game players, which makes online games one of the booming leisure activities. The online game industry has not been affected by the recent unstable domestic economy but instead it continues to grow (FIND, 2008).

Three trends can be identified from the FIND (2008) study. First, the upsurge of online games in Taiwan always follows a swarm effect, causing online game companies to adopt a "short-term speculation" business model. Second, the type of players changed from long-term commitment in Massively Multiplayer Online Role-Playing Game (MMORPG) and fierce battle games into casual games. The most obvious example is the Happy Harvest game on Facebook. Third, most players will focus on playing one particular game after trying a variety of online games. To summarize the above points, although the online gaming industry may have unlimited potential business opportunities and profit, in recent years, the game market has gradually become saturated, and in addition to high homogeneity of the game, changes in the type of players, and changes in the environment of the online game market. How to enhance the competitiveness of the industry has become an important issue.

This study is based on Parasuraman et al.’s (2005) network quality of service (E-Service Quality) and further develops a measurement scale for online game service quality to explore the impact of various online game service items on player satisfaction. Then critical mass and virtual community cohesion are further explored for their moderating effects on the relationship of players’perceived service quality and satisfaction.

II. LITERATURE REVIEW

A. Online game

Yang et al. (2009) proposed that experience value, transaction cost, and SERVQUAL will affect the satisfaction of online gaming users, and thereby affecting their loyalty. The results showed that four factors of service quality are the key elements affecting the satisfaction of online game users, followed by transaction costs, and lastly, the experience value.

Hsu and Lu (2007) combined the technology acceptance model (TAM) and the theory of reasoned action (TRA) as the intrinsic motivation, replaced the usefulness construct in TAM with perceived enjoyment, and added perceived cohesion and social norms of extrinsic motivation to explore consumer behavior of the online gaming community. The study found that TAM in intrinsic motivation and perceived cohesion and social norms in extrinsic motivation had significant impact on the behavior of consumers. A few years earlier, Hsu and Lu (2004) also used TAM, social norms, critical mass, and flow experience to investigate the acceptance of online gaming. Their research showed that social norms had a direct impact on the acceptance of online gaming, critical mass had a direct impact on user intention to play online games, and flow experience was a successful key to managing the online game community.

From the behavioral science point of view, Koo (2009) explored factors of experiential motives influencing the intention to adopt online games, including concentration, perceived enjoyment, escape, epistemic curiosity and social affiliation, and added the external locus of control in the social cognitive theory as a moderator factor. The research outcomes indicated that perceived enjoyment, escape and social affiliation affected the consumers’intention to use, while the external locus of control as a moderator factor indirectly impacted the relationship between concentration, perceived enjoyment, escape, and intention to use.

Yang et al. (2009) addressed the online game industry by proposing a customized SERVQUAL framework. However, certain revisions are needed by basing the scales of E-SQ due to the characteristics of different industries. To address the above issues, the specificity of online games for the entertainment industry needs to be emphasized, which

can be achieved by building a conceptual model and consolidating the steps and methods of the measurement scale for online gaming service quality

.

B. Virtual Community Cohesion

Virtual community is defined as a group of individuals who have common goals or ideas, can exchange messages through electronic media such as the Internet or other media, and are not bound by geographical location and ethnic restrictions (Kardaras et al., 2003; Romm et al., 1997).

Koh and Kim (2004) pointed out that community awareness composed of the sense of belonging, influence and immersion experience, and discussed the enthusiasm of leaders, offline activities and the impact of entertainment community awareness. This study confirmed that sense of belonging was affected by the enthusiasm of leaders, offline activities and the impact of entertainment community awareness; influence was affected by offline activities; and immersion was impacted by offline activities and entertaining experience. Hsu and Lu (2007) also found that "community cohesion" (group cohesion) in extrinsic motivation influenced the behavior of online games consumers, indicating an unignored power in virtual community cohesion awareness among team members. Dholakia et al. (2004), from the social point of view, explored factors that the consumer participated in a community. Of the group norms, individual needs directly affected the users’ willingness to participate in a community. In addition, users’needs were subject to social identity and mutual agreement influence. Blanchard and Markus (2002) found that community awareness existed in the virtual community, but the composition of factors differed from traditional communities. Furthermore, community members’mutual supportiveness, identification, recognition, and trust are important factors to maintain community awareness.

To conclude, virtual community involved significantly different perspectives, and research directions, objectives and results are all varied. But it is not difficult to see that the community's sense of identity and cohesion are important factors to maintain a virtual community in operation. Therefore, virtual community cohesion deserves further investigation into its moderating effect between the service quality of online games and player satisfaction.

C. Critical Mass

In the clustering effect study of Rogers (1995), it was noted that different individuals have different cognitive capacity with respect to the acceptance threshold of innovative technology. Innovative technology spreads slowly for users with a low acceptance threshold, and when

the spread of new technology to the number of people under the clustering effect of cognition, the speed for adopting new technology will increase rapidly. Lou et al. (2000) pointed out that when the individual formed a cognitive cluster effect, a tendency to use the system was generated due to the influence of information and norms. In particular, information influence comes from giving more positive objective facts to most users, and specifications of normative influence is to gain recognition by meeting the expectation of others or a group.

Online games are entertainment-oriented information systems; therefore, if the user has a higher critical mass of awareness, he/she will be more willing to use online games. (Heijden, 2004; Hsu and Lu, 2004). However, in the related research field of critical-mass, previous studies have used a cluster effect in communication research, such as interactive media (Markus, 1987), shared databases (Lou et al., 2000), and online games (Hsu and Lu, 2004), and so on. A mixed result exists in the relationship between the clustering effect and the users, which may imply an indirect influence. Therefore, the impact of critical mass on users may have a moderating function.

III. RESEARCH METHODOLOGY

The main purpose of this study is to investigate whether the relationship of service quality of online games and player satisfaction are influenced by virtual community and critical mass. Since this study is targeted at online games mediated by the network as its business model, this research is based on the service quality of a Web-based conceptual model, and virtual communities and critical mass in previous studies. This research has established two hypotheses:

H1: Virtual community cohesion moderates the relationship between service quality and customer satisfaction.

H2: Critical mass moderates the relationship between service quality and customer satisfaction.

This research initially constructs the service quality dimensions of online gaming based on e-Service Quality (e-SQ) as the theoretical basis, and derives questionnaire items of virtual community cohesion and the cluster effect in the literature. Questionnaire items were designed as a seven-point Likert-type scale, ranging from 'strongly disagree' (1) to 'strongly agree' (7). Formal questionnaires were distributed to two samples for exploratory and confirmatory analysis.

The targeted users are those who are playing or used to play online games. The questionnaire was distributed via the Internet. The first survey collected 237 valid questionnaires, and after excluding 32 invalid questionnaires, 205 valid questionnaires were retained. The

second survey resulted in 250 valid and 24 invalid questionnaires, respectively.

In the first data set, the male to female ratio was 59% to 41%, respectively, and the 18 to 25 age group accounted for 59.2%, followed by the 26 to 35 age group, which accounted for 3.6%. Students accounted for 54.1%, and the most common online game type was role-playing games (MMORPG), which accounted for 65.8%. Those with more than 5 years of online gaming experience accounted for 47.8%, followed by 26.8% for those with 3 to 4 years of experience.

In the second questionnaire, the male to female ratio was 52% and 48 %, respectively, and the 18 to 25 age group accounted for 58.9%, followed by the 26 to 35 age group, which accounted for 32.2%. Again, students accounted for the major group with 53.9%, and the most common online game type was MMORPG, which accounted for 64.2%. Those with more than 5 years of online gaming experience accounted for 49%.

IV. DATA ANALYSIS AND RESULTS Data analysis methods in the EFA phase used principal component analysis (PCA) to extract the eigenvalue, and Equamax orthogonal rotation to analyze dimensions of factors, and both methods were implemented in SPSS 15. In the CFA phase, Partial Least Square (PLS), as implemented in SmartPLS 2.0 M3, was used to process the hypothesis model. Satisfaction items were added in the second questionnaire to verify the research model, and thus no satisfaction items were in the analytical part of EFA.

Part of the EFA results can be seen on the left section of TABLE I. Each item factor loading exceeds 0.5, a hurdle value suggested by Hair et al. (1998), and each dimensional reliability (the coefficient alpha) is over 0.6, an unacceptable level as suggested by Murphy and Davidshofer (1988). Both results demonstrate considerable reliability in both dimensions and items. Through 1,000 times of PLS re-sampling on the measurement model, we can see that the factor loadings for all latent variables exceed 0.5, the suggested standard. In the reliability, the composite reliability (CR) values show that the reliability of each dimension is higher than the suggested value of 0.7 by Nunnally (1994), which indicates a good construct reliability. The average variation extracted (AVE) values for all dimensions were greater than 0.5, which also indicates a model with convergent validity (Heir et al., 1998).

In TABLE II we can find that no pair of measures had correlations exceeding the criterion of 0.9 as suggested by Heir et al. (1998), which implies that no multicolinearity existed among these constructs. Meanwhile, each of the diagnal square root of the AVE value is greater than the other related coefficients of the non-diagonal construct

values, which shows that all latent constructs satisfied the discriminant validities.

According to the results of the structural model, hypothesis 1 (H1: Virtual community cohesion moderates the link between service quality and customer satisfaction)

was significant (p<0.05) while hypothesis 2 (H2: Critical mass moderates the link between service quality and customer satisfaction) was non-significant, and the final CFA results are shown in Fig. 1.

TABLE I Analysis results of EFA (n=205) and CFA (n=250)

EFA Loadings CFA

1 2 3 4 5 6 7 Loading AVE C.R.

Content (CO) (coefficient alpha = 0.879) 0.557 0.629 0.91

CO01 0.727 0.714 CO02 0.724 0.775 CO03 0.681 0.795 CO04 0.771 0.836 CO05 0.806 0.820 CO06 0.641 0.814

Efficiency (EF) (coefficient alpha = 0.903) 0.691 0.674 0.925

EF01 0.851 0.722 EF02 0.580 0.562 0.901 EF03 0.638 0.866 EF04 0.595 0.517 0.879 EF05 0.693 0.671 EF06 0.536 0.533 0.861

Fulfillment (FU) (coefficient alpha = 0.905) 0.660 0.788 0.937

FU01 0.788 0.923

FU02 0.791 0.875

FU03 0.753 0.852

FU04 0.726 0.899

Compensation (CP) (coefficient alpha = 0.682) 0.518 0.753 0.859

CP01 0.771 0.857

CP02 0.603 0.879

System Availability (SA) (coefficient alpha = 0.882) 0.731 0.709 0.924

SA01 0.596 0.800

SA02 0.827 0.758

SA03 0.680 0.909

SA04 0.679 0.829

SA05 0.640 0.902

Privacy (PR) (coefficient alpha = 0.938) 0.397 0.900 0.964

PR01 0.875 0.954

PR02 0.911 0.962

PR03 0.905 0.930

Responsiveness & Contact (RC) (coefficient alpha = 0.958) 0.220 0.599 0.816

RC01 0.970 0.827

RC02 0.968 0.819

RC03 0.905 0.665

Moderator

Virtual Community Cohesion (VCC) (coefficient alpha = 0.941) 0.895 0.962

VCC01 0.961 0.946

VCC02 0.928 0.963

VCC03 0.849 0.929

Critical Mass (CM) (coefficient alpha = 0.881) 0.831 0.937

CM01 0.880 0.927

CM02 0.939 0.907

CM03 0.766 0.900

Perceived value

Satisfaction (Sat) (coefficient alpha = 0.935) 0.596 0.839 0.954 0.948

0.945 0.846 0.920

TABLE II Discriminant validity of all dimensions

Latent Variable Correlations CO EF FU CP SA PR RC VCC CM Sat

CO 0.793 EF 0.537 0.821 FU 0.488 0.537 0.888 CP 0.515 0.793 0.438 0.868 SA 0.489 0.632 0.712 0.472 0.842 PR 0.38 0.336 0.531 0.323 0.535 0.949 RC 0.349 0.283 0.357 0.227 0.377 0.244 0.774 VCC 0.461 0.627 0.697 0.554 0.71 0.482 0.421 0.946 CM 0.341 0.467 0.492 0.353 0.416 0.257 0.314 0.576 0.912 Sat 0.471 0.505 0.636 0.47 0.687 0.337 0.475 0.733 0.518 0.916

TABLE III Effects of online game service quality on satisfaction (n=250)

Hypothesis path Coefficients T value Supported H0 H1: OLG-SQ * VCC -> Sat 1.200 2.218* Yes

H2: OLG-SQ * CM -> Sat -0.199 0.362 ns No

* p < 0.05, ns: non-significant (based on t(250), one-tailed test)

Fig. I Results of structural modeling analysis

OLG-SQ Satisfaction CO CP EF RC FU PR SA CM VCC 0.470*** 0.812*** 0.630*** 0.855*** 0.831*** 0.719*** 0.746*** -0.097 -0.199 ns 1.200*

V. DISCUSSION

In this study, we processed exploratory factor analysis based on Parasuraman et al.’s (2005) website service quality items, and sourced items from various studies (Hernona and Calvert, 2005; Hsu and Lu, 2004, 2007; Koo, 2009; Loiacono, et al., 2002; Wolfinbarger and Gilly, 2002, 2003; Yang et al., 2009; Yoo and Donthu, 2001).

The seven dimensions derived include content, efficiency, compensation, system availability, privacy, fulfillment, and enjoyment. Among them, efficiency, system availability, privacy, and fulfillment are also part of the original core e-SQ scale dimensions, which implies that even in different industries, such as the online game industry in this study, these four dimensions are generally considered important.

Responsiveness & contact is equivalent to the combination of two dimensions of the recovery service quality in the original E-SQ scale, which implies that when online game players encountered obstacles or problems, the game makers are able to provide effective communication channels and responsive solutions to the problems of online games. In addition to responsiveness & contact when players encountered problems, compensation is another relevant factor, in that the game maker will reimburse players with losses incurred as a result of online game website problems.

The last dimension, content, can be seen to arise from different characteristics of the industry. That is, in the gaming industry, players will consider game content, including graphic design and entertainment, in the measure of game quality.

Currently, online game players have been paid high attention in the in-game community. Meanwhile, many online games have also emphasized the cooperative efforts in passing game challenges or the level of game difficulty in order to continue the game. Thus, the perceived virtual community cohesion in the game or whether players have enough friends in an online game will be a critical factor for joining the game. Accordingly, virtual community cohesion and critical mass were assumed to have a moderating role on the link between online gaming service quality and player satisfaction.

The results show that virtual community cohesion does play a moderating role between the relationship of the online game service quality and satisfaction while critical mass does not. This could be interpreted as the higher the centripetal force of players in online games, the higher the degree of player satisfaction from the same gaming service quality. It could also be said that since most of the player’s game companions may not be his/her

friends, family or colleagues, critical mass does not play a substantial moderating role on the relationship between the online game service quality and player satisfaction.

In facing this issue, online game providers can consider adding communication and contact channels, or adding interactive elements into game levels to strengthen the link between players and the overall virtual community cohesion.

Although the results of this study show that there is no moderating role of critical mass, online game providers can deploy strategies to bring in players’friends and families, which not only enlarges the game player pool, but also enhances the cohesion between players and new player retention. This way, either existing or new players may increase the impact of players’perceived service quality on perceived player satisfaction.

REFERENCES

[1] Blanchard, A. L. and Markus, M. L. (2002), “Sense of Virtual Community–Maintaining the Experience of Belonging,” Proceedings of the 35th AnnualHawaii International Conference on System Science, pp. 3566-3575.

[2] Dholakia, U. M., Bagozzi, R. P. and Pearo, L. K. (2004),“A Social Influence Model of Consumer Participation in Network- and Smaill-Group-Based VirtualCommunities,” International Journal of Research in Marketing, Vol. 21, No. 3, pp. 241-263.

[3] Foreseeing Innovative New Digiservices, 2008; 2009, http://www.find.org.tw/find/home.aspx

[4] Hair, J.F, Anderson, R., Tatham, R., & Black, W. (1998). “Multivariate data analysis (5th Ed.),”Upper Saddle River, NJ: Prentice-Hall.

[5] Heijden, H. V. D. (2004), “User Acceptance of Hedonic Information Systems,” MIS Quarterly,Vol.28,No.4,pp. 695-704.

[6] Hernona, P. & Calvert, P. (2005). “E-service quality in libraries: Exploring its features and dimensions,”Library & Information Science Research, 27, 377-404.

[7] Hsu, C.L. & Lu, H.P. (2004). “Why do people play on-line games? An extended TAM with social influences and flow experience,”Information & Management , 41, 853 –868. [8] Hsu, C.L. & Lu, H.P. (2007). “Consumer behavior in online

game communities: A motivational factor perspective,” Information & Management 23, 1642–1659.

[9] Kardaras, D., Karakostas, B., and Papathanassiou, E. (2003), “The potential of virtual communities in the insurance industry in the UK and Greece,”International Journal of Information Management, 23, pp. 41-53.

[10] Koh, J. and Kim, Y. G. (2004), “Sense of Virtual Community: A Conceptual Framework and Empirical Validation,”InternationalJournalofElectronicCommerce, Vol. 8, No. 2, pp. 75-93.

[11] Koo, D. M. (2009). “The moderating role of locus of control on the links between experiential motives and

intention to play online games,”Computers in Human Behavior 25, 466 –474.

[12] Loiacono, E. T., Watson, R. T. & Goodhue, D. L. (2002). “WEBQUAL:A measureofwebsitequality,”In K. Evans, & L. Scheer (Eds.).

[13] Lou, H., Lou, W. and Strong, D., (2000), “Perceived CriticalMassEffecton GroupwareAcceptance,”European Journal of Information Systems, Vol. 9, No. 2, pp. 91-103. [14] Nunnally, I. C. & Bernstin, I. H. (1994).Psychometric

Theory (3rd ed.).New York: McGraw-Hill

[15] Markus, M. L. “Toward a 'Critical Mass' Theory of Interactive Media: Universal Access, Interdependence and Diffusion,”CommunicationsResearch,Volume14,1987, pp. 491-511.

[16] Murphy, Kevin R. and Charles 0. Davidshofer (1988), “Psychological Testing: Principles and Applications,” Engle- wood Cliffs, NJ: Prentice-Hall.

[17] Parasuraman, A., Zeithaml, Valarie A., Malhotra, A. (2005). “A Multiple-Item Scale for Assessing Electronic Service Quality,”Journal of Service Research, 7(3), 213-233. [18] Rogers, E.M. (1995). “Diffusion of Innovations,”New

York: Free Press.

[19] Romm, C., and Pliskin, N., Clarke, R. (1997). “Virtual Communities and Society: Towards an Integrative Three Phase Model,” International Journal of Information Management, Vol.17, No.4, pp. 261-270.

[20] Wolfinbarger, M. & Gilly, M. C. (2002). “.comQ: Dimensionalizing, Measuring and Predicting Quality of the E-tail Experience,”Marketing Science Institute, Cambridge, MA., No. 02-100.

[21] Wolfinbarger, M. & Gilly, M. C. (2003). “eTailQ: Dimensionalizing, measuring and predicting etail quality,” Journal of Retailing, 79(3), 183-198.

[22] Yang, H.E., Wu, C.C. & Wang, K.C.(2009). “An empirical analysis of online game service satisfaction and loyalty,” Expert Systems with Applications 36, 1816–1825.

[23] Yoo, B. & Donthu, N. (2001). “Developing a scale to measure the perceived quality of an Internet shopping site (SITEQUAL),”Quarterly Journal of Electronic Commerce, 2(1),