In the last 10-15 years, use of social networking sites has exploded in the United States and globally. Users range from tech-savvy young adults to baby boomers and older adults seeking ways to reconnect with family and friends (Anderson, 2009). In this study, we examine one of the important user groups, college students, and their attitudes toward using social networking. Drawing upon the Theory of Reasoned Action, Theory of Planned Behavior, the Composite Model of Attitude Behavior Relations and the Technology Acceptance Model, we develop and test a model to explain college students' intentions to use social networking. Our findings shed light on factors that have contributed to the rapid increase in social networking.
Social media allows users to go from simply content consumers to content producers by publishing information. Kaplan and Haenlein (2010) define social media as "a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0 (1), and that allow the creation and exchange of user generated content." According to Kaplan and Haenlein (2010) there are six types of social media: collaborative projects, blogs and microblogs, content communities, social networking sites, virtual game worlds, and virtual social worlds. Our focus in this study is social networking sites, which are applications that enable users to connect by creating personal information profiles, inviting friends and colleagues to have access to those profiles, and sending e-mails and instant messages between each other Kaplan and Haenlein (2010).
Popular examples of social networking sites are MySpace (created in 2003) and Facebook (created in 2004). Facebook is ranked as the third most popular online brand in the world, with over 54% of the world's internet population visiting Facebook (Neilsen, 2010). In April 2010, social networking sites were visited by three-quarters of global consumers who went online, which is an increase of 24% over April 2009 (Neilsen, 2010). The average visitor spends 66% more time on these sites than a year ago, almost 6 hours in April 2010 versus 3 hours, 31 minutes in April 2009 (Neilsen, 2010). In July, 2010, Facebook surpassed having 500 Million users worldwide (Zuckerberg, 2010). It took the site about three months to climb from 300 to 350 million users and only about two months to gain another 50 million, then another three months to make it to 500 million. MySpace is still a top 10 website in the United States, with about 57 Million unique visitors and over one-quarter of the US internet population still interacting with MySpace on a daily basis (Prescott, 2010). Among college students, Anderson Analytics (Anderson, 2009) in their annual American College Student Survey found that Facebook was viewed as "cool" by 82% of males and 90% of females. According to Anderson, Facebook is the hands-down winner with the 18-25 year olds.
The following sections of the paper outline model development, methodology, results, and discussion.
We developed the model in Figure 1 to provide a framework for examining factors influencing consumer usage of social networks. Drawing upon the Theory of Reasoned Action (Ajzen & Fishbein 1980; Fishbein & Ajzen 1975), Theory of Planned Behavior (Ajzen 1991), and Composite Model of Attitude Behavior Relations (Eagly & Chaiken 1996), this model illustrates how beliefs are expected to influence a user's attitude toward a social network and how that attitude is expected to influence the user's intentions to engage in different social networking behaviors. A similar model, the Technology Acceptance Model and has proven useful in understanding consumer usage of technology (Davis, Bagozzi, and Warshaw (1989) and consumer usage of technology-based customer interfaces (Curran, Meuter, and Surprenant 2003; Curran and Meuter 2007).
The model includes five beliefs which are hypothesized to influence attitude toward the social network. These five beliefs are: ease of use, usefulness, enjoyment, social influence, and drama. Each of these factors is discussed more fully in the following sections.
[FIGURE 1 OMITTED]
Ease of Use
Ease of use has been defined as the degree to which a user would find the use of a particular technology to be free from effort on their part (Davis, Bagozzi and Warshaw 1989). This construct is central to the Modified American Customer Satisfaction Model which has been used to study technology (Tung, 2010). Ease of use is also important in the technology acceptance model (Davis, Bagozzi and Warshaw 1989) and has been used in many studies since (Awa, Nwibere & inyang, 2010; Bagozzi, 2007; Thompson, Compeau & Higgins, 2006; Kuo, et al., 2005; igbaria, Guimaraes and Davis 1995; and Taylor and Todd 1995). Therefore, it is hypothesized:
H1a: Perceived ease of use of the social network will be positively related to attitude toward the social network.
Usefulness is the subjective probability that using the technology would improve the way a user could complete a given task (Davis, Bagozzi and Warshaw 1989). Usefulness is the second central construct for the technology acceptance model (Davis, Bagozzi and Warshaw 1989), and has also received a great deal of attention in adoption literature (Bell, 2009; Choi, Lee and Soriano, 2009; Kamis, Koufaris and Stern 2008; igbaria, Parasuraman and Baroudi 1996; Jackson, Chow and Leitch 1997; Taylor and Todd 1995). Therefore, it is hypothesized:
H1b: Perceived usefulness of the social network will be positively related to attitude toward the social network.
Curran and Meuter (2007) showed that enjoyment can be an important influence in the adoption of a self-service technology. Dellaert and Dabholkar (2009) found that enjoyment was enhanced when using an on-line mass customization process. Eighmey and McCord (1998) identified enjoyment as a key construct in consumer patronage of websites. Wolfinbarger and Gilly (2001) provided qualitative evidence that fun was an important desired outcome when choosing to use technology to shop. Koufaris (2002) found enjoyment to be an important direct antecedent to intentions to return to an online retailer, while Sun (2009) found enjoyment to be an important antecedent of retention. While research has shown that people will use technologies because they are more pleasurable, fun, or entertaining than more traditional ways of doing things, more work is needed to clarify its place in understanding social network adoption. Therefore, it is hypothesized:
H1c: Perceived enjoyment of the social network will be positively related to attitude toward the social network.
Social influence relates to the approval or disapproval of others when the consumer decides to adopt and use products and services. The idea that people will purchase goods or services primarily to make a favorable impression on other people has been documented in many different contexts. For example, Steenkamp and Gielens (2003) found that social influences can have a pronounced negative effect on the chances of new product trial as products are perceived as more novel. Trocchia and Janda (2002) show that consumers may cease to use a product because usage may not portray them to others in the fashion desired. Yoh, Damhorst, Sapp, and Laczniak (2003) showed that people who had more social support for internet shopping had greater intentions to purchase on line. Howcroft, Hamilton and Hewer (2002) found that an important factor in consumer adoption of banking was the recommendation of a friend or family member. Agarwal, Animesh and Prasad (2009) showed that social influence helps to explain internet usage. Boyd (2008) states that the popularity of sites like MySpace among young adults is due to the "sociality" aspect. Therefore, it is hypothesized:
H1d: Perceived social influence to use the social network will be positively related to attitude toward the social network.
Focus groups conducted in the preliminary phases of this research revealed a recurring theme that some users of social networks engaged in dramatic reactions to social network postings and behaviors. These focus groups also provided evidence that some social network users will engage in behaviors online that they would not ever consider if they were face to face with others from the social network. Boyd (2008) found that the simple act of listing who were best friends versus next best friends on MySpace created pure social drama. When discussing how to enhance creativity in the workplace Burke (2009) recommends that using mainstream social networking sites, such as Facebook simply has too much drama. Thelwall, Wilkinson and Uppal (2010) found that MySpace is an extraordinarily emotion-rich environment, with 86% of the comments they researched in MySpace containing some type of emotion. According to Thelwal et al. (2010, p. 198), "... emotion is apparently the norm in social networking sites." They conclude, "MySpace users should not be afraid of emotional statements: These are the norm." Therefore, it is hypothesized:
H1e: Perceived level of drama found on the social network will be positively related to attitude toward the social network.
Attitudes and Intentions
An attitude is defined as "a psychological tendency that is expressed by evaluating a particular entity with some degree of favor or disfavor" (Eagly & Chaiken 1993, p.1). in fact, research regarding the relevance of attitudes has found that attitudes, and ultimately behavioral intentions, develop sequentially or in a hierarchical fashion (Eagly & Chaiken 1993). The notion that attitudes influence behavioral intentions (Ajzen & Fishbein 1980; Fishbein & Ajzen 1975) has been researched extensively and this relationship has been well established in the marketing literature in areas such as loyalty (Auh, Bell, McLeod, and Shih, 2007), advertising (Karson and Fisher, 2005), and technology adoption (Curran, Meuter, and Surprenant, 2003). Given the support for this attitude-behavioral intention relationship, the following relationships are hypothesized:
H2a: Attitudes toward a social network will positively influence intentions to continuing to utilize social networks.
H2b: Attitudes toward a social network will positively influence intentions to recommend that social network to others.
H2c: Attitudes toward a social network will positively influence intentions to join other social networks.
H2d: Attitudes toward a social network will positively influence intentions to stop using some social networks.
The primary objectives for this study are to develop and test the proposed structural model (Figure 1) and the hypothesized relationships. in order to complete this effectively, a survey approach was utilized targeting frequent users of Social Media. The survey consisted of 56 questions, seven of which were demographic, the rest covering respondents' feelings, opinions and intentions related to Social Media.
The survey was administered to students enrolled in various business courses at two campuses of a major US southeastern university over a two month period. Each potential respondent was given the option of completing the questionnaire or opting out of it.
A convenience sample of 495 useable questionnaires was collected. 55.6% of the respondents were male and 44.4% were female. Respondents ranged in age from 18 to 64 with 61% being 22 years of age and under. 40% of the respondents use social networks several times each day, 24.1% use social networks once each day, and another 17% characterize their use of social networks as a few times each week. 94.7% of the respondents use Facebook, 64.9% of the respondents use MySpace, 19.7% of the respondents use Twitter, and 8.1% of the respondents use Linkedin.
Independent and Dependent Variables
The five independent constructs (ease of use, usefulness, enjoyment, social influence, and drama) were all measured using multiple item scales utilizing a seven-point Likert structure with the endpoints being "strongly disagree" to "strongly agree." The exact items used can be seen in Table 2. The attitude construct was measured using a three-item, seven-point bipolar semantic differential scale with endpoints of good/bad, like/dislike, and pleasant/unpleasant (Curran, Meuter & Surprenant 2003; Allen, Machleit & Kleine 1992; Dabholkar 1996). Behavioral intentions were measured using single-item, semantic differential measures for each behavior with the end points being "extremely likely" and "extremely unlikely."
The model proposed in Figure 1 was tested using structural equation modeling and several criteria are used to assess the overall fit of the models. The first criterion is the ratio of the chi square value to the degrees of freedom for the model, also referred to as the normed chi-square. Although there is some debate about the acceptable value of this ratio, it is generally held that a value between 1 and 3 is desirable (Hair, Black, Babin, and Anderson 2010) and anything above 5 would indicate a poor fit for the model (Marsh & Hocevar 1985; Kline 2005). The second criterion to evaluate fit is the comparative fit index (Bentler 1990), which is an incremental fit index and should have a value larger than .90 (Hair, Black, Babin, and Anderson 2010) and approaching .95 (Hu & Bentler 1999) in order for the model to be deemed reasonable. The third criterion of model adequacy is the root mean square error of approximation (RMSEA), which is based on a population discrepancy function. A RMSEA value of .05 or less would indicate a close fit of the model in relation to the degrees of freedom, while .08 or less would indicate a reasonable level of error of approximation and any model with a RMSEA above .10 would be unacceptable (Browne & Cudeck 1993).
The Measurement Model
A measurement model was constructed to determine the correlations among the constructs included in the model and assess the convergent and discriminant validity of the various constructs. The fit of the correlation model was good with a Chi-Square value of 310.576, degrees of freedom of 137, and an acceptable ratio Chi-Sq/df of 2.267 (Marsh & Hocevar 1985). The CFi for this model is .975 and the RMSEA is .051, and both are indicative of good fit (Hu & Bentler 1999; Browne & Cudeck 1993). Table 1 shows that a total of 15 correlations were calculated and two correlations were greater than 0.5.
The average variance extracted was calculated for each of the six multi-measure constructs. Fornell and Larker (1981) wrote that the average variance extracted for each construct should exceed 0.5 to establish convergent validity. The average variance extracted for each of the constructs can be found in Table 2, each exceeded the 0.5 threshold and thus convergent validity is established. in order to examine discriminant validity, it is necessary to compare the squared correlations between each pair of constructs with the average variance extracted for the individual constructs in each pair. if the average variance extracted for both constructs in the pair exceeds the square of the correlation between them, discriminant validity is demonstrated (Fornell & Larker, 1981). These conditions were met for every pair of constructs and discriminant validity is established.
The Structural Model Comparison
in order to fully understand the data that was collected a series of three nested structural models were evaluated to see which provided the best explanation of the hypothesized relationships. The first model tested was a non-moderated model in which all of the independent constructs included are assumed to directly influence the dependent variables without any moderating influence from the attitude construct. The second model is the hypothesized model in which the attitude construct moderates all influences that the independent constructs have on each of the dependent variables and the independent constructs have no direct effects on the dependent variables at all. The third and final model will examine a combination of effects from the first two models where the dependent variables will be influenced by both attitude as a moderating construct and directly by each of the independent constructs. The fit indices for all three models can be seen in Table 3 and the paths values for each model can be seen in Table 4.
Model 1 (Direct Effects Only)
The first model tested contained only the direct effects from the five independent constructs to the four dependent variables and the attitude construct plays no role at all. The Chi-Square for this model is 1262.589, the degrees of freedom are 204 and this results in a ratio Chi-Square/df of 6.189 which is well above any acceptable limit for this model to be acceptable (Marsh & Hocevar 1985). The CFi value is .865, which is below the minimum level of acceptability for this index of .90 (Hair, Black, Babin, and Anderson 2010). The Root Mean Square Error of Approximation (RMSEA) was found to be .102, which also exceeds the .10 limit for acceptability (Browne & Cudeck 1993). Given that none of these fit indices are within acceptable limits, it can be concluded that the direct effects only model does not represent an acceptable explanation for the relationship among these constructs and variables.
Model 2 (Moderated Effects Only)
The moderated effects model is a direct test of the model that was shown in Figure 1 where the independent constructs all influence attitude toward the social network and attitude is the only influence on the dependent variables representing behavioral intentions. The results for the moderated effects model can be seen in Figure 2. For this model the Chi-Square is 871.334 and the degrees of freedom are 215, yielding a ratio Chi-Square/df of 4.05. The ratio is within the desirable range of 1 to 5 to demonstrate acceptable performance (Marsh & Hocevar 1985). The CFi value is .916, which is above the minimum level of acceptability of .90 (Hair, Black, Babin, and Anderson 2010). The Root Mean Square Error of Approximation (RMSEA) was found to be .079 which is within acceptable limits (Browne & Cudeck 1993). The combination of these results demonstrates that this is an acceptable model.
[FIGURE 2 OMITTED]
In this model, 70.2% of the variation in attitude toward the social network is explained and three of the five hypothesized relationships with antecedent factors were found to be significant. The significant paths were from Enjoyment to Attitude (.79, p<.001), Social influence to Attitude (-.103; p<.01), and Drama to Attitude (-.067; p<.05). All of the paths from Attitude to the four behavioral intention variables were significant at p<.001. The standardized coefficients for the paths were .603 for the path from Attitude to intention to Continue, .659 for the path from Attitude to intention to Recommend, .378 for the path from Attitude to Join other Networks, and -.25 for the path from Attitude to intention to Stop Using Some Social Networks. This model explains 36% of the variation in intention to Continue Using the Social Network, 44% of the variation for intention to Recommend Social Networks to Others, 14% of the variation for intention to Join Other Social Networks and 6% of the variation for intention to Stop Using Some Social Networks. These results support hypotheses H1c, H1d, H1e, H2a, H2b, H2c, and H2d. Hypotheses H1a and H1b are not supported.
Model 3 (Moderated and Direct Effects)
The final model combines all of the paths that were included in the other two models, both direct from the independent constructs to the dependent variables but also the paths through the attitude construct. The significant results for this model can be seen in Figure 3. The Chi-Square for this model is 786.736 and the degrees of freedom are 195, yielding a ratio Chi-Square/df of 4.035 which is acceptable (Marsh & Hocevar 1985). The CFi value is .924, which is above the minimum level of acceptability of .90 (Hair, Black, Babin, and Anderson 2010) and closer to the desirable level of .95 (Hu & Bentler 1999). The Root Mean Square Error of Approximation (RMSEA) was found to be .078 which is again within acceptable limits (Browne & Cudeck 1993). These results demonstrate that this is an acceptable model. A comparison of the Chi-Square (AChi-Square = 84.598) and Degrees of Freedom ([DELTA]df = 20) for this model and the Moderated Effects Model shows that this Combination Model is significantly better at explaining this data than the Moderated Effects Model.
There are a number of differences in the significant paths between the Moderated Effects Model and the Combination Model. There are only two paths that significantly influence Attitude in this model and those paths are Enjoyment (.770; p<.001) and Social influence (-.097; p<.01) and only 66.3% of the variation in Attitude is explained. The path from Drama to Attitude was not significant in this model as it was in the Moderated Effects Model. Attitude is significantly related to only three of the four dependent variables in this model where it was related to all four in the Moderated Effects Model. The significant paths in this model were from Attitude to intention to Continue (.306; p<.001), from Attitude to intention to Recommend (.331; p<.001), and from Attitude to intention to Stop Using Some Social Networks (-.235; p<.01). The path from Attitude to intention to Join other Networks was not significant in this model.
Several of the direct paths from the independent constructs to the dependent variables were also found to be significant and this model explains 36% of the variation in intention to Continue Using the Social Network, 43% of the variation for intention to Recommend Social Networks to Others, 17% of the variation for intention to Join Other Social Networks and 11% of the variation for intention to Stop Using Some Social Networks. None of the direct paths from Ease of Use or Usefulness to any of the dependent variables were found to be significant. The direct paths from Enjoyment to intention to Continue Using the Social Network (.312; p<.001), intention to Recommend Social Networks to Others (.305; p<.001), and intention to Join Other Social Networks (.230; p<.01) were all found significant. Three of the direct paths from Drama to the dependent variables were found to be significant and those paths were to intention to Continue Using the Social Network (-.093; p<.01), intention to Recommend Social Networks to Others (-.133; p<.01), and intention to Stop Using Some Social Networks (.227; p<.001). The direct path from Social influence to intention to Continue Using the Social Network (-.123; p<.01) was the only direct path from Social influence to be found significant.
The comparison of the three nested structural model clearly demonstrates that the best explanation for this data is offered by the model that contains a combination of direct and moderated effects on the behavioral intentions related to social networks. The combination model shows that there are significant direct and indirect effects on intentions to use social networks, intentions to recommend social networks, and intentions to stop using social networks. including the direct paths had a particularly dramatic effect on the variance explained for intention to Stop Using Social Network as the Drama construct had a strong direct effect on this dependent variable. The variance explained for the other variables and construct remained comparable between the two better fitting models.
In our research, we found that enjoyment is the strongest factor influencing Attitude and has a significant direct effect on intention to Continue Using Social Networks, intention to Recommend Social Networks, and intention to Join Other Social Networks. There can be little doubt from these results that social network users derive a great deal of enjoyment from the use of social networking and it would be critical for anyone seeking to use social networks to reach out to others of interest to them to be keenly aware that if the experience they provide on the social network is less than enjoyable they may be disappointed with the outcome.
Social influence has a significant positive effect on Attitude and a significant but a negative influence on intentions to Continue Using Social Networks. This can be interpreted as social network users feeling positive about being part of the network but disliking the pressure they may feel to use it beyond their level of comfort.
Drama has no direct effect on Attitude but it does have significant and negative influence on intention to Continue Using Social Networks and intention to Recommend Social Networks. There is also a strong positive effect on intention to Stop Using Social Networks. These results show that social network users do not like certain behaviors from others with whom they connect through social networks and these behaviors may result in a termination of use of the network not just the individual relationship. Most of the research on social networking does not look at this particular factor, therefore, this is a very interesting finding that could have major implications for marketers.
[FIGURE 3 OMITTED]
Interestingly from a theoretical perspective, the two constructs that formed the foundation of the Technology Acceptance Model, namely Ease of Use and Usefulness, were not found to have a significant relationship with any of the dependent constructs in any of the models tested in this research. This may indicate that social network users are no longer concerned about the functionality of the technology itself but more concerned with the outcomes derived from its use.
Allen, C. T., K. A. Machleit & S. S. Kleine (1992). A Comparison of Attitudes and Emotions as Predictors of Behavior at Diverse Levels of Behavioral Experience. Journal of Consumer Research, 18 (March), 493-504.
Agarwal, R., A. Animesh & K. Prasad (2009). Social interactions and the "Digital Divide": Explaining variations in internet use. Information Systems Research, 20(2), 277-294.
Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human Decision Process, 50, 179-211.
Ajzen, I. and M. Fishbein (1980). Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, N J: Prentice-Hall.
Anderson, T. H.C. (December 6, 2009). College Students say Facebook is the only Social Networking site that really matters. Retrieved June 30, 2010 from: www.prweb.com/releases/ 2009/12/prweb3301654.htm
Auh, S., Bell, S.J., McCloud, C.S., & Shih, E. (2007). Co-production and customer loyalty in financial services. Journal of Retailing, 83(3), 359-370.
Awa, H. O., B. M. Nwibere, & B. J inyang. (2010). The Update of Electronic Commerce by SMES: A Meta Theoretical Framework Expanding the Determining Constructs of TAM and TOE Frameworks. Journal of Global Business and Technology, 6(1),1-28.
Bagozzi, R. P. (2007). The Legacy of the Technology Acceptance Model and a Proposal for a Paradigm Shift. Journal of the Association for Information Systems, 8(4), 243-255.
Bell, J. E. (2009). Consumers' willingness to depend on user generated content and producer generated content within e-commerce. Ph.D. Dissertation, Purdue University.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238-246.
Boyd, D. (2008). Why Youth love Social Network sites: The role of networked publics in teenage social life, in D. Buckingham (ed.) Youth, Identity and Digital media (pp.119-142). Cambridge, MA. MiT Press.
Browne, M. W., & R. Cudeck (1993). Alternative Ways of Assessing Model Fit, Testing Structural Equation Models. in K. A. Bollen & J. S. Long, (eds.), Testing structural equation models (pp.136-162). Newbury Park, CA: Sage.
Burke, P. S. (2009). Creative Networking. Macworld, 26(8),72-73.
Choi, J., Sang M. L. & D. R. Soriano (2009). An Empirical Study of User Acceptance of Fee-Based online content. The Journal of Computer Information Systems, 49(3), 60-71.
Curran, J. M. & M. L. Meuter (2007). Encouraging Existing Customers to Switch to Self-Service Technologies: Put a Little Fun in their Lives. Journal of Marketing Theory and Practice, 15(4),283-298.
Curran, J. M., M. L. Meuter & C. F. Surprenant (2003). intentions to use Self-Service Technologies: A Confluence of Multiple Attitudes. Journal of Service Research, 5(3), 209-224.
Dabholkar, P. A. (1996). Consumer Evaluations of New Technology-Based Self-Service Options: An investigation of Alternative Models of Service Quality. International Journal of Research in Marketing, 13(1), 29-51.
Davis, F. D., R. P. Bagozzi & P. R. Warshaw (1989). User acceptance of computer technology: A comparison of two theoretical models. Management Science, 35, 982-1003.
Dellaert, B. G.C. & P. A. Dabholkar (2009). increasing the Attractiveness of Mass Customization: The Role of Complementary On-line Services and Range of Options. International Journal of Electronic Commerce, 13(3), 43-70.
Eagly, A. A. & S. Chaiken (1993). The Psychology of Attitudes. Fort Worth TX: Harcourt Brace College Publishers.
Eighmey, J. & L. McCord (1998). Adding value in the information age: Uses and gratifications of sites on the World Wide Web. Journal of Business Research, 41(3), 187-194.
Fishbein, M. & i. Ajzen (1975). Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research. Reading, MA: Addison-Wesley.
Fornell, C., & D. F. Larcker (1981). Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. Journal of Marketing Research, 18(February), 39-50.
Hair, Jr. J. F., Black, W., B. Babin & Rolph E. Anderson (2010). Multivariate Data Analysis 7th edition. Englewood Cliffs: Prentice Hall.
Howcroft B., R. Hamilton, P. Hewer (2002). Consumer attitude and the usage and adoption of home-based banking in the United Kingdom. The International Journal of Bank Marketing, 20(2/3), 111-122.
Hsu C. & Lu, H. (2004). Why do People play online games? An extended TAM with social influences and flow experience. Information and Management, 41(7), 853-868.
Hu, Li-Tze, & P. M. Bentler (1999). Cutoff Criteria for Fit indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives. Structural Equation Modeling, 6(1), 1-55.
Igbaria, M., Parasuraman, S. & J. J. Baroudi (1996). A motivational model of microcomputer usage. Journal of Management Information Systems, 13(1), 127-144.
Igbaria, M., Guimaraes, T. & G. B. Davis (1995). Testing the determinants of microcomputer usage via a structural equation model. Journal of Management Information Systems, 11(4), 87-114.
Jackson, C. M, Chow, S. & R. A Leitch (1997). Toward an understanding of the behavioral intention to use an information system. Decision Sciences, 28(2), 357-390.
Kamis, A., Koufaris, M. & T. Stern (2008). Using an Attribute-based decision support system for user-customized products online: An experimental investigation. MIS Quarertly, 3(1), 159-177.
Kaplan, A. M., & Haenlein, M. (2010). Users of the world unite! The challenges and opportunities of social media. Business Horizons, 53(1), 59-68.
Karson, E.J., & Fisher, R.J. (2005). Reexamining and extending the dual mediation hypotheses in an on-line advertising context. Psychology & Marketing, 22(4), 333-351.
Kline, R. B. (2005). Principles and practice of structural equation modeling. New York: Guildford Press
Koufaris, M. (2002). Applying the technology acceptance model and flow theory to online consumer behavior. Information Systems Research, 13(2), 205-225.
Kuo, T., Iulan-yuan Lu, C. H., & Guo-chiang, W. (2005). Measuring users' perceived portal service quality: An
empirical study. Total Quality Management and Business Excellence, 16(3), 309-320. Marsh, H. W, & D. Hocevar (1985). Application of Confirmatory Factor Analysis to the Study of Self-Concept: First and Higher-Order Factor Models and Their invariance Across Groups. Psychological Bulletin, 97(3), 562-582.
Nielsenwire (June 15, 2010). Social Networks/Blogs now account for one in every four and a half minutes online. Retrieved June 27, 2010 from: http://blog.nielsen.com/nielsenwire/ online_mobile/social-media-accountsfor-22-percent-of-time-online/
Prescott, L. (February 10, 2010). Social Networking by the numbers: Facebook is eating up MySpace, 2nd tier networks led by Tagged. Retrieved June 29, 2010 from: http://www.research-write.com/2010/02/socialnetworking-by-the- numbers.html
Steenkamp, J. E. M. & K. Gielens (2003). Consumer and Market Drivers of the Trial Probability of new consumer packaged goods. Journal of Consumer Research, 30, 368-384.
Sun, H. (2009). Sellers' Trust and Continued use of online marketplaces. Journal of the Association for Information System, 11(4), 182-212.
Taylor, S. & P. A. Todd (1995). Understanding information Technology Usage: A test of Competing Models. Information Systems Research, 6, 144-176.
Thelwall, M., D. Wilkinson & S. Uppal (2010). Data Mining emotion in social network communication: Gender differences in MySpace. Journal of the American society for information science & technology, 61(1), 190-199.
Thompson, R., D. Compeau & C. Higgins (2006). intentions to Use information Technologies: An integrative Model. Journal of Organizational and End User Computing, 18(3), 25-47.
Trocchia, P. J. & J. Swinder (2002). An investigation of product purchase and subsequent non-consumption. The Journal of Consumer Marketing, 19(2/3), 188-205.
Tung, F. (2010). Exploring Customer Satisfaction, Perceived Quality and image: An Empirical Study in the Mobile Services industry. The Business Review, 14(2), 63-69.
Wolfinbarger, M. & M. Gilly (2001). Shopping online for freedom, control, and fun. California Management Review, 43(2), 34-56.
Yoh, E., M. L. Damhorst, S. Sapp, R. Laczniak (2003). Consumer Adoption of the internet: The Case of Apparel Shopping. Psychology & Marketing, 20(12), 1095-1117.
Zuckerberg, M. (July 21, 2010). 500 Million Stories. Retrieved August 4, 2010 from: http://blog.facebook.com/blog.php?post=40975335213
James M. Curran, University of South Florida Sarasota-Manatee
Ron Lennon, University of South Florida Sarasota-Manatee
Table 1: Correlations Between Constructs Construct Attitude Ease of Use Usefulness Attitude 1.00 Ease of Use .362 *** 1.00 Usefulness .500 *** 274 *** 1.00 Enjoyment 804 *** .438 *** .582 *** Drama -.071ns 149 ** -.063ns Social influence -.019ns -.138 ** 234 *** Construct Enjoyment Drama Social influence Attitude Ease of Use Usefulness Enjoyment 1.00 Drama -.024ns 1.00 Social influence .080ns -.011ns 1.00 Note: *: p < .05; **: p < .01; ***: p < .001; ns: not significant Table 2: Measurement Model Results Internal Consistency Construct and Scale Items Standardized Composite Loadings Reliability Attitude .906 Overall, how good or bad do you .855 feel about being a member of this social network? Do you like or dislike being a .924 member of this social network? How pleasant or unpleasant is your .842 time spent using this social network? Ease of Use .940 Learning how to participate in a .887 social network was easy for me. I find social networks easy to .928 use. It was easy for me to figure out .931 how to participate in social networks. Usefulness .895 Social networks make it easier for .880 me to keep up with issues that interest to me. Social networks make it easier for .807 me to keep up with businesses that interest to me. Social networks improve the way I .892 keep up with things that interest me. Enjoyment .943 I enjoy keeping up with people .799 using social networks. It's fun to be involved with .944 social networks. I enjoy being part of a social .962 network. I find social networks to be .878 entertaining. Social Influence .772 I participate in a social network .532 because someone I know wants me to. I joined a social network to fit .933 in with a group of people. I am part of a social network .693 because friends would think less of me if I was not Drama .760 People write things on social .674 networks that they would never say face to face. People get too emotional about .778 things that are put on social networks. There is too much drama dealing .696 with people on social networks. Internal Consistency Construct and Scale Items Coefficient Average Alpha Variance Extracted Attitude .903 .764 Overall, how good or bad do you feel about being a member of this social network? Do you like or dislike being a member of this social network? How pleasant or unpleasant is your time spent using this social network? Ease of Use .940 .839 Learning how to participate in a social network was easy for me. I find social networks easy to use. It was easy for me to figure out how to participate in social networks. Usefulness .895 .739 Social networks make it easier for me to keep up with issues that interest to me. Social networks make it easier for me to keep up with businesses that interest to me. Social networks improve the way I keep up with things that interest me. Enjoyment .941 .807 I enjoy keeping up with people using social networks. It's fun to be involved with social networks. I enjoy being part of a social network. I find social networks to be entertaining. Social Influence .734 .543 I participate in a social network because someone I know wants me to. I joined a social network to fit in with a group of people. I am part of a social network because friends would think less of me if I was not Drama .759 .514 People write things on social networks that they would never say face to face. People get too emotional about things that are put on social networks. There is too much drama dealing with people on social networks. Table 3: Comparison of Model Fit Measures Direct Moderated Moderated Effects Only & Direct Only Model Model Model Chi Square 1262.589 871.334 786.736 Degrees of Freedom 204 215 195 Chi Square/df 6.189 4.053 4.035 CFI .865 .916 .924 RMSEA .102 .079 .078 Variance Explained -- .702 .663 ([R.sup.2]): Attitude Variance Explained .349 .364 .360 ([R.sup.2]): Continue to Use Variance Explained .408 .435 .426 ([R.sup.2]): Recommend Variance Explained .167 .143 .168 ([R.sup.2]): Join Other Variance Explained .099 .063 .114 ([R.sup.2]): Stop Using Table 4: Comparison of Model Paths Direct Moderated Moderated Effects Only Model & Direct Only Model Model Ease of Use to Attitude -- .002ns .000ns Usefulness to Attitude -- .076ns .073ns Enjoyment to Attitude -- 790 *** 770 *** Social Influence to Attitude -- -.103 ** .097 ** Drama to Attitude -- -.067 * -.046ns Ease of Use to Continue .035ns -- .047ns Ease of Use to Recommend -.054ns -- -.043ns Ease of Use to Join Other -.053ns -- -.047ns Ease of Use to Stop -.060ns -- -.063ns Usefulness to Continue -.037ns -- -.060ns Usefulness to Recommend .095ns -- .071ns Usefulness to Join Other .096ns -- .089ns Usefulness to Stop -.045ns -- -.029ns Enjoyment to Continue .565 *** 312 *** Enjoyment to Recommend 576 *** -- .305 *** Enjoyment to Join Other .338 *** -- .230 ** Enjoyment to Stop -.130 * -- .058ns Social Influence to Continue -175 *** -- -.123 ** Social Influence to Recommend -.127 ** -- -.075ns Social Influence to Join Other .058ns -- .083ns Social Influence to Stop -.047ns -- -.075ns Drama to Continue -.116 ** -- -.093 * Drama to Recommend -157 *** -- -.133 ** Drama to Join Other -.087ns -- -.077ns Drama to Stop .242 *** -- 227 *** Attitude to Continue -- .603 *** .306 *** Attitude to Recommend -- 659 *** .331 *** Attitude to Join Other -- 378 *** .123ns Attitude to Stop -- -.250 *** -.235 ** Note: ***: p<.001; **: p<.01; *: p<.05; ns: not significant…