Assessing the mediating role of online social capital between social support
and instant messaging usage
Chieh-Peng Lin
⇑Institute of Business & Management, National Chiao Tung University, 4F, 118, Sec. 1, Jhongsiao W. Rd., Taipei 10044, Taiwan, ROC
a r t i c l e
i n f o
Article history:
Received 19 November 2009
Received in revised form 14 August 2010 Accepted 14 August 2010
Available online 8 September 2010 Keywords:
Commitment Reciprocity
Shared codes and language Shared narratives Centrality Network ties
a b s t r a c t
This study validates a research model that examines usage of instant messaging (IM) from the aspect of online social support. Drawing on the social capital theory, this study postulates that IM usage is indi-rectly affected by social support via the mediation of the following six dimensions of social capital: com-mitment, reciprocity, shared codes and language, shared narratives, centrality, and network ties. The model tests data obtained from business organizations in Taiwan, and the results suggest that the indi-rect influence of social support on IM usage through shared codes and language is significant, and the indirect influence of social support on IM usage through centrality is also significant. Managerial impli-cations and limitations of the empirical findings are provided.
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
Virtual communities in practice consist of large knitted and geographically distributed groups of online users engaged in shared activities. While e-mail has been the dominant communica-tion technology across virtual communities, another widely dif-fused and mature innovation is instant messaging (IM), which is well-known for its interactive online communication (Li et al. 2005). IM is a popular online, real-time, mobile computer-medi-ated communication technology (Zaman et al. 2010). Millions of people use IM with their friends and families for online social communication (Li et al. 2005). To date, IM usage has been further extended to business settings (Shiu and Lenhart 2004) such as communication among co-workers, sales promotions between buyers and sellers, and so on (Li et al. 2005).
Although IM users may not meet each other face-to-face, many are able to build a social relationship and foster interactions with one another, suggesting the importance of understanding social capital in IM virtual communities. Regardless of the increasing importance of online social capital in influencing individuals’ IM usage, little attention has been paid to such capital in the IM usage literature. Most of the contemporary research models cover IM usage, such as the technology acceptance model (Lu et al. 2009,
Wang et al. 2004), the motivational model (Lee et al. 2007, Li
et al. 2005), the theory of planned behavior (Lin et al. 2006, Lu
et al. 2009, Zhou 2007), the model of innovation diffusion theory
(Rouibah and Hamdy 2009), and the unified theory of acceptance
and use of technology (Lin and Bhattacherjee 2008, Park et al. 2007). These studies have all ignored the potential role of social capital in affecting IM usage.
Social capital is defined as an important resource embedded in a social structure, which is accessed and/or mobilized in deliberate action (Lin 2001, Song and Lin 2009). Social capital has been well applied to explain a variety of pro-social behaviors (such as online information exchanging and online experience sharing with oth-ers) that other forms of capital (e.g., human or financial capital) are unable to clarify (Bottrell 2009, Coleman 1990), suggesting its substantial influence on IM usage for pro-social behaviors. Whereas other forms of capital are established on the basis of financial assets or particular individuals, social capital resides in the interpersonal fabric of relationships embedded in the social realm (Putnam 1995, Wasko and Faraj 2005). With the continuous breakthrough of technology infrastructures, there is increasing evi-dence that users who have accumulated certain social capital are likely to use IM for maintaining frequent social interactions and relationships (Cummings et al. 2006), which is comparable to the mutual interactions in face-to-face settings (e.g.,Walther and Boyd 2002). Unfortunately, the key antecedents of social capital in IM contexts remain unknown in previous studies, and thus this study’s purpose is to explore IM usage caused by various social capital factors and their antecedents.
This study differs from previous research in two critical ways. First, while some literature links social capital and social support
1567-4223/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.elerap.2010.08.003
⇑ Tel.: +886 2 2381 2386x57662 (O). E-mail address:jacques@mail.nctu.edu.tw
Contents lists available atScienceDirect
Electronic Commerce Research and Applications
to IT usage which focuses on qualitative rationales (e.g.,Drentea
and Moren-Cross 2005), this study complements the previous
qualitative studies by providing empirical confirmations and by examining various dimensions of social capital in greater depth. Particularly, this study decomposes social capital into six specific constructs – commitment, reciprocity, shared codes and language, shared narratives, centrality, and network ties – that help e-com-merce marketers or corporate managers learn about efficient approaches in order to strengthen people’s IM usage. Second, although previous research finds that health-related Internet use is associated with social support among online individuals
(Kalichman et al. 2003), previous research rarely discusses how
the association between social support and IM usage may be fully mediated and by what factors. This study empirically confirms whether such associations are fully or partially mediated by social capital, further complementing the previous research.
The rest of the paper proceeds as follows. The next section describes the theoretical underpinnings of social capital and for-mulates a research model of IM usage based on social capital as the mediator. The third section presents our research methods, including our choice of empirical context, subject samples, and instrumentation. The fourth section describes the data analysis procedures and results. The fifth section outlines the implications of our findings for future IM usage research and practice, and then identifies the study’s limitations.
2. Social capital theory and hypotheses’ development 2.1. Social capital and IM usage
Social capital can be used for understanding individuals’ IT usage (e.g.,He et al. 2009), because social capital complements the med-ium theory as it explains what situations are important for individ-uals to voluntarily interact with online contacts to an extensive degree (Radin 2006). Particularly, online social capital influences the usage of IM that supports social networking. Since productivity tools such as Microsoft Office Word or EKR (electronic knowledge repositories) are primarily utilitarian in workplaces, it may be pos-sible that online social capital plays a less important role in its use than that of self-efficacy (Lin and Huang 2009, Kankanhalli et al. 2005). However, this does not suggest that social capital has no influence on usage behavior of information systems (e.g., knowledge management systems, KMS). In the studies byLin and Huang (2009) and Kankanhalli et al. (2005), the effects of social capital are signif-icant.He et al. (2009)conclude that employees are more dependent on a trustworthy social relationship based on a set of shared values and norms (p. 179). They further state that such a social relationship based on a social capital perspective (p. 176) evolves into even more communication and interaction, and thus ‘‘people exhibited a com-mitment to continue using the KMS” (p. 179). For instance, online individuals may easily apply IM to create knowledge or share infor-mation over open settings (e.g., online bulletin boards) via social interactions, given that online communities provide an environ-ment conducive to the developenviron-ment of social capital (e.g.,Nahapiest
and Ghoshal 1998, Wasko and Faraj 2005). Moreover, previous
re-search regarding employees’ relationships and their technology acceptance argues that users’ usage intention relies heavily upon the beliefs employees have about their social relationships with their team members (Magni and Pennarola 2008). Moreover, social norms (a form of social capital) have been confirmed to be a deter-minant of technology usage, but not vice versa (Hsu and Lu 2004, Wang et al. 2004). Collectively, individuals’ social capital substan-tially helps strengthen their subsequent technology usage (e.g., IM). An individuals’ IM usage is hypothetically driven by social cap-ital, including three dimensions: (1) structural links or connections
between online individuals, which are named structural capital; (2) individuals’ cognitive capabilities that help them to have a shared system of understanding among the individuals, also known as cognitive capital; (3) social online relationships that have strong, positive characteristics, which are named relational capital. Given that these three dimensions of social capital are pos-itively related with individuals’ technology usage, social support proposed as antecedents in this study hypothetically affects IM usage through the three dimensions. Specifically, subscribers’ usage is considered an outcome driven by social and individual fac-tors in the majority of previous e-commerce research (e.g.,Chau et al. 2007, Kim et al. 2010, Lee 2009, Rouibah 2008, Rouibah and
Hamdy 2009, Yoon and Kim 2007).
Many studies in the literature have argued that technology usage is the outcome of social networking (e.g.,Cheung and Lee 2010, Lin and Bhattacherjee 2008), and not vice versa. For example,
Zhang (2009)surveys users from popular social networking sites to
test the validity of the research model, confirming that there is a direct influence from social networking (i.e., online community) on users’ system usage. Based on the consensus mentioned above, the mediating role of social capital between social support and IM usage is discussed in the following section:
2.2. Social support and social capital
Social support is defined as ‘‘the exchange of verbal and non-verbal messages conveying emotion, information, or referral, to help reduce one’s uncertainty or stress” (Walther and Boyd 2002, p. 154). Social support represents a focal point around which social ecological models of human interaction and social actions can de-velop (Vaux 1988). A support group that provides social support is likely to offer relative social capital in which an embedded commu-nity is activated for purposeful action (Lin 2001). This implies a linkage from social support to social capital. Social capital can be seen as capital in which relations with friends, neighbors, relatives, and colleagues supply shared support, because it provides com-panionship, emotional support, information, and a sense of belong-ing (Wellman and Frank 2001).
Although social support has been traditionally examined in pre-vious research within the context of personal, face-to-face social capital, there is increasing evidence that social support obtained via IM helps derive social capital to a degree which is comparable to that found in face-to-face settings. In other words, people aggre-gate across communities to share valuable information, experi-ences, or empathy about a common cause (such as coping with terminal illnesses such as cancer or AIDS), overcoming personal crises (such drug or alcohol addiction), or sharing profit-making opportunities like stock tips or rumors. Taken as whole, these are activities that are likely to establish strong online social capital. For example, the ‘‘Systers” mailing list was originally intended for female computer scientists to provide online social support, but then it evolved into a forum for deriving online social capital. Online social support is indeed effective in fostering online social capital, even when those involved are virtual strangers.
The substantial influence of support on social capital has al-ready been indicated in a study covering an Internet mothers’ web-site (Drentea and Moren-Cross 2005). Particularly, online social support was provided by users who offered formal guidelines and informal information-sharing services as a resource to Internet mothers and was likely to boost the mothers’ technology usage
(Drentea and Moren-Cross 2005). This reveals that social capital
emerges through the diffusion of online social support.
The dynamics of IM-mediated social support and social capital remain quite different from those of face-to-face social support and social capital, given the geographically-dispersed nature of on-line networks, the willingness of network members to trust and
interact with virtual strangers, and the frequent participation of online members in multiple online communities to meet different social needs. Details about the influence of social support on social capital in online contexts are provided below.
2.3. Online relational capital
Online social support associated with an affective response based on the social relationship within online communities is re-ferred to as relational capital characterized by commitment and reciprocity (e.g.,Nahapiet and Ghoshal 1998). Social support im-pacts both commitment and reciprocity. First, reciprocity reflects the anticipation that online users’ collective efforts will be recipro-cated (Wasko and Faraj 2005). As a basic norm of reciprocity is a social cognition of interpersonal indebtedness, online users are likely to reciprocate the online social support they receive from others, ensuring ongoing social support exchanges (e.g.,Shumaker
and Brownell 1984) and resulting in a positive relationship
be-tween social support and reciprocity. Second, commitment is an obligation to engage in future social activities which arise from fre-quent social interaction (Coleman 1990). Online commitment is likely amplified with a social perception of duty to support others within the collective communities on the basis of shared online membership (Wasko and Faraj 2005), leading to a positive rela-tionship between social support and commitment.
Previous literature finds that online participation to provide so-cial support may intensify reciprocity and trust (Quan-Haase et al.
2002, Quan-Haase and Wellman 2004). Similarly, a previous
inves-tigation shows that half of those who belong to online communi-ties say that the Internet provides them with support to connect with people who reciprocally share their interests (Horrigan 2002), implying the influence of expressive and instrumental sup-port on reciprocity. As noted in a study of online social supsup-port for Japanese mothers (Miyata 2002), in a supportive online commu-nity the norms of generalized reciprocity are easily established and maintained. Collectively, the above phenomenon offers a posi-tive relationship between social support and social capital (e.g., reciprocity and commitment), and the hypotheses about IM usage are derived as follows.
H1. Users’ social support directly influences their commitment and indirectly influences IM usage via the mediation of the commitment.
H2. Users’ social support directly influences their reciprocity and indirectly influences IM usage via the mediation of the reciprocity.
2.4. Online cognitive capital
Online users’ cognitive capital, reflected by shared codes, lan-guage, and narratives, includes online resources which make pos-sible shared meanings, connotation, and stories among the users of a virtual community. Engaging in a meaningful sharing of useful information and benevolent advice (i.e., a form of social support) helps facilitate certain levels of shared understanding among indi-viduals who contain shared codes, languages, and narratives
(Nahapiet and Ghoshal 1998, Wasko and Faraj 2005).
The relationship between social support and cognitive social capital (e.g., shared codes, languages, and narratives) has been bol-stered by the communication accommodation theory, which ex-plains some of the cognitive reasons for code-switching and other exchanges (e.g., shared languages and narratives) in interper-sonal communications as individuals seek to emphasize or mini-mize the social differences between themselves and their
contacts (Buller and Aune 1992, Willemyns et al. 1997). Previous research studies indicate that meaningful communication, which is an essential part of social exchange and combination processes, requires at least some sharing of context between the parties to such an exchange (Boisot 1995, Boland and Tenaski 1995). This sharing comes from two major sources: shared codes and language and the shared narratives (Nahapiet and Ghoshal 1998, Swarbrick 2002).
Social support helps strengthen narratives told over time, be-cause narratives that are considered as war stories or workarounds provide insights into what individuals can do to support others when they need to resolve difficulties in life (e.g.,Brown and
Du-guid 1991). Previous literature finds emotional support,
instru-mental support, and community building all contribute to the creation and maintenance of shared codes, language, and narra-tives (e.g., women share their personal stories about miscarriages and implantation bleeding via the Internet) (Drentea and Moren-Cross 2005, Litt 2000). Given that shared codes, language, and nar-ratives jointly provide a frame of reference for interpreting the so-cial environment (Wasko and Faraj 2005) in which social support occurs, online cognitive capital turns out to be an important out-come of social support. In summary, individuals’ understanding of shared codes, language, and narratives is likely to be lifted by strong social support, leading to the following hypotheses.
H3. Users’ social support directly influences their shared codes and language and thus indirectly influences IM usage via the mediation of shared codes and language.
H4. Users’ social support directly influences their shared narra-tives and thus indirectly influences IM usage via the mediation of shared narratives.
2.5. Online structural capital
The social capital theory suggests that structural capital is com-prised of both centrality and network ties and is accumulated through interpersonal interactions in social communities. There-fore, structural capital is a critical outcome of collective action. The social support received from others is a typical initiator of col-lective action, suggesting its role in influencing structural capital.
The relationship between social support and structural social capital has been bolstered by the social network theory (Reagans
and Zuckerman 2001). Social network scholars have taken the lead
in formalizing and empirically testing issues related to structural social capital and regard network ties or relationships as the basic data for analysis (Seibert et al. 2001). Network ties among individ-uals of a social supportive group are often strong, while network ties that reach outside individuals’ social clique are weak ( Grano-vetter 1982, Seibert et al. 2001). This signifies a positive effect of social support on structural social capital.
When individuals’ collective actions such as social support, col-laboration, and contribution to others in online communities are achieved, they are likely to consider online communities as an important part of their life. To put it differently, the stronger the social support is that IM users receive from online contacts, the more strongly the online centrality is perceived by them. Specifi-cally, IM can be moved from a text to a voice dialogue, building a deeper communication as needed (Oliva 2003). This further facili-tates IM users to possess shared codes and languages and a sense of centrality.
In addition to centrality, individuals’ network ties that reflect online structural capital to a certain degree are relevant for the amount of social support available to individuals within online
communities. Previous literature suggests that IM and text mes-saging are both forms of technology-mediated communication that provides a tool for individuals to communicate in social support talks with one another and consequently helps to create and rein-force social ties and friendships (Boneva et al. 2006, Bryant et al. 2006). The more social support individuals receive from online contacts, the tighter ties they establish with others due to their strong social networking. Thus, the hypotheses are provided below. H5. Users’ social support directly influences their centrality and thus indirectly influences IM usage via the mediation of the centrality.
H6. Users’ social support directly influences their network ties and thus indirectly influences IM usage via the mediation of the net-work ties.
3. Method
3.1. Measures
This study’s hypotheses are empirically tested using a survey of instant messaging (IM) technology usage among employees in Taiwan. IM allows users to obtain social support via real-time com-munication on the Internet. IM was chosen for this study, because it is an advanced interactive information technology that lends it-self particularly well to a virtual world in which online individuals obtain online social support.
IM is a near-synchronous communication tool (Nardi et al. 2000). Similar to online chat rooms, IM allows users to type mes-sages into a window, but like the phone it is based on a dyadic ‘‘call” model (Nardi et al. 2000). IM users sometimes intentionally use IM (instead of e-mail) as a sticky note, because they know it is visible as soon as the person returns to his/her computer, and it is easier to retrieve and respond to in a timely manner than e-mail (or voicemail) (Isaacs et al. 2002). When comparing e-mail and IM, the previous literature indicates that people use IM more than e-mail for long-distance or social-related communication, but they use e-mail more than IM for communicating with people they know offline (Recchiuti 2003). For example, since e-mail allows people to have a chance to think about what they want to say, they feel more comfortable using e-mail to contact their supervisors or managers (i.e., task-related communication) (Recchiuti 2003). For social-related communication such as communicating with friends, IM is more influential than e-mail and online chat rooms (Recchiuti 2003).
During our investigation, a manager in a consulting company provided the necessary assistance for surveying subjects randomly drawn from seven large companies that are clients of the consult-ing company. Two of the seven large companies are from the gen-eral service industries and the other five companies are from high-tech industries (related to both servicing and manufacturing). Of the 500 questionnaires distributed to subjects, 364 usable ques-tionnaires were returned to the researcher for a response rate of 72.8%. Our respondents consisted of 62.6% males and 37.4% fe-males. Of these respondents, 38.5% are under 29 years old, while 45.1% are between 30 and 39 years old. The remaining 16.4% are 40 years old or older. Additionally, 10.4% have experience using IM for less than a year, 37.4% have experience for 1–2 years, and 52.2% have experience for more than two years. The sample also reveals that 11.0% are high school graduates, 57.7% are college graduates, and 31.3% are graduate school graduates.
The constructs in this study are measured using five-point Lik-ert scales drawn and modified from previous literature. A
univer-sity professor and four graduate students who are familiar with online behavior worked together as a focus group to help evaluate the appropriateness of the measurement items. The pilot test data collected from 50 student subjects were subjected to exploratory factor analysis (EFA) and reliability analysis in order to identify items that loaded poorly on their hypothesized scales, which were then re-worded. This process of instrument refinement led to con-siderable improvement in content validity and scale reliability. Note that the pilot test respondents were excluded in the subse-quent survey. Finally, tips from back-translations as indicated by
Reynolds et al. (1993)were applied in composing an English
ver-sion questionnaire and a Chinese one. A high degree of correspon-dence between the two questionnaires assured that the translation process did not introduce substantial translation biases in the Chi-nese version of the questionnaire.
Previous studies (e.g., Hirsch 1980, Cutrona and Suhr 1992) have proposed typologies of social support, including the follow-ing: (1) emotional support, such as expressions of caring, concern, and sympathy toward relieving pain and stress; (2) socializing sup-port, such as providing companionship or verbal reinforcement about one’s choices; (3) instrumental support, such as providing financial or practical assistance (e.g., job referrals) for a network member in need; and (4) informational support, such as offering advice, factual input, and feedback to help network members eval-uate actions and make decisions (Lin and Bhattacherjee 2009). Due to a potential overlap among these four typologies mentioned above, this study categorizes social support into two types: (1) expressive support that contains the first two typologies above and (2) instrumental support that includes the last two typologies above. The dimension reduction for four typologies is essential, be-cause these four typologies that are highly correlated with each other may be confusing and detrimental for researchers when try-ing to provide clear and specific managerial recommendations.
Expressive support with four items and instrumental support with four items are drawn and refined fromEastin and LaRose
(2005), who modify the items fromCohen et al. (1984).
Commit-ment with three items and reciprocity with three items are modi-fied from Wasko and Faraj (2005). However, as there are no suitable scale items for shared codes, language, and narratives in the existing research, this study develops three items for shared codes and language and another three items for shared narratives based on the definition of cognitive capital in previous literature
(Nahapiet and Ghoshal 1998, Wasko and Faraj 2005) through the
use of a focus group. Centrality with three items and network ties with four items were modified fromObst and White (2005). IM usage with four items was modified fromCheung et al. (2000). These items are confirmed to be effective, because they accurately reflect the users’ IM usage in terms of duration, intensity, and fre-quency (Cheung et al. 2000). Collectively, these constructs were modified from previous literature by being embedded with the fea-tures related to IM usage.Appendix A lists all the measurement items in detail.
4. Data analysis
4.1. Measurement model testing
The final survey data with a sample size of 364 responses were analyzed via a two-step structural equation modeling (SEM) ap-proach proposed byAnderson and Gerbing (1988)using SAS soft-ware. The first stage performed a confirmatory factor analysis (CFA) on all data collected to assess scale reliability and validity. The second stage examined the structural model for testing the hypotheses. The test results from the two stages are described as follows.
CFA analysis was done on all items corresponding to the nine constructs. The goodness-of-fit of the hypothesized CFA model was assessed applying a variety of fit metrics, as shown in
Table 1. The normed fit index (NFI) and the adjusted
goodness-of-fit index (AGFI) were both slightly lower than the recommended value of 0.9, while the normalized chi-square (chi-square/degrees of freedom) of the CFA model was smaller than the recommended value of 3.0. The root mean square residual (RMR) was smaller than 0.05, and the root mean square error of approximation (RMSEA) was smaller than 0.08. At the same time, the comparative fit index (CFI), the non-normed fit index (NNFI), and the goodness-of-fit index (GFI) all exceeded 0.90. These figures reveal that the CFA model hypothesized in this study fits the empirical data well (Bentler and Bonnett 1980).
Convergent validity was assessed applying three criteria suggested byFornell and Larcker (1981). First, all factor loadings inTable 1were statistically significant at p < 0.001 to assure con-vergent validity of construct (Anderson and Gerbing 1988). Second, the reliabilities for each construct exceeded 0.70 inTable 1, satis-fying the general requirement of reliability for research instru-ments. The average variance extracted (AVE) for all constructs exceeded 0.50 except for two constructs (reciprocity and central-ity) being slightly lower than 0.5, indicating that the measurement items overall captured sufficient variance in the underlying con-struct than those attributable to measurement error (Fornell and Larcker 1981). Some AVEs of social capital constructs were slightly lower than 0.5 and may have been caused by three potential rea-sons related to our discordant sample subjects between males (62.6%) and females (37.4%). First, the previous literature indicates that females are hindered in their efforts to achieve career advancement and its associated benefits owing to their inability to access social capital (Timberlake 2005). Second, prior research
(Darley and Smith 1995) indicates that females are comprehensive
information processors who consider both subjective and objective question attributes and respond to subtle cues (during a question-naire survey). Conversely, males are selective information proces-sors who tend to use heuristics processing and miss subtle cues
(Darley and Smith 1995). Such phenomenon could slightly
dimin-ish the variance extracted estimates of our factors. Third, the sig-nificant gender-related differences in exposure to and use of computers (e.g., IM usage) have been confirmed (Bannert and Arbinger 1996, Celik and Ipcioglu 2007, Liu and Chang 2010, Park
2009, Sánchez-Franco 2007, Wang 2010). These differences may
be another source that lessens the variance extracted estimates of our online social capital.
Due to the three possible reasons described above, the amount of variance that is captured by an underlying factor in relation to the amount of variance due to measurement error is thus slightly lower in this study. To further detect the cause of low AVEs in our constructs (i.e., reciprocity and centrality), we conducted post hoc analyses by splitting the data into two different gender sub-groups (males vs. females) based on our CFA model and then calcu-lated separate AVEs for each gender. The empirical results summarized inAppendix Bindicate that the AVEs for females are
Table 1
Standardized loadings and reliabilities. Construct Indicatorsa
Standardized loading
AVE Cronbach’s
a
Expressive support ES1 0.93 (t = 21.94) 0.73 0.88 ES2 0.95 (t = 22.70)
ES3 0.66 (t = 13.84)
Instrumental support IS1 0.83 (t = 16.74) 0.53 0.76 IS2 0.66 (t = 12.59) IS3 0.68 (t = 13.08) Commitment CO1 0.80 (t = 15.68) 0.50 0.73 CO2 0.65 (t = 12.29) CO3 0.66 (t = 12.60) Reciprocity RE1 0.76 (t = 14.33) 0.49 0.74 RE2 0.72 (t = 13.55) RE3 0.62 (t = 11.54) Shared codes and
language SC1 0.75 (t = 14.62) 0.51 0.76 SC2 0.71 (t = 13.67) SC3 0.69 (t = 13.13) Shared narratives SN1 0.93 (t = 20.52) 0.64 0.81 SN2 0.86 (t = 18.36) SN3 0.56 (t = 11.02) Centrality CE1 0.72 (t = 14.01) 0.49 0.74 CE2 0.69 (t = 13.28) CE3 0.69 (t = 13.43) Network ties NT1 0.79 (t = 15.78) 0.53 0.77 NT2 0.72 (t = 13.98) NT3 0.66 (t = 12.76) IM usage IM1 0.81 (t = 16.12) 0.50 0.73 IM2 0.65 (t = 12.40) IM3 0.64 (t = 12.15) Goodness-of-fit indices (N = 364): v2 288¼ 493:77 (p-value < 0.001); NNFI = 0.93;
NFI = 0.88; CFI = 0.95; GFI = 0.91; AGFI = 0.88; RMR = 0.02; RMSEA = 0.04.
a
Indicators remaining after CFA purification. A few indicators are excluded from this measurement model due to their insignificance.
Table 2
Chi-square difference tests for examining discriminant validity.
Construct pair v2 288¼ 493:77 (unconstrained model) v2 169(constrained model) v2 difference (Expressive support, Instrumental
support)
768.12 274.35***
(Expressive support, Commitment) 731.11 237.34***
(Expressive support, Reciprocity) 734.14 240.37***
(Expressive support, Shared codes and language)
753.24 259.47***
(Expressive support, Shared narratives) 942.44 448.67***
(Expressive support, Centrality) 715.13 221.36***
(Expressive support, Network ties) 765.77 272.00***
(Expressive support, IM usage) 725.40 231.63***
(Instrumental support, Commitment) 671.66 177.89***
(Instrumental support, Reciprocity) 696.96 203.19***
(Instrumental support, Shared codes and language)
742.30 248.53***
(Instrumental support, Shared narratives) 745.40 251.63***
(Instrumental support, Centrality) 594.99 101.22***
(Instrumental support, Network ties) 643.45 149.68***
(Instrumental support, IM usage) 631.61 137.84***
(Commitment, Reciprocity) 631.71 137.94***
(Commitment, Shared codes and language) 626.56 132.79***
(Commitment, Shared narratives) 713.84 220.07***
(Commitment, Centrality) 600.02 106.25***
(Commitment, Network ties) 653.82 160.05***
(Commitment, IM usage) 619.39 125.62***
(Reciprocity, Shared codes and language) 655.91 162.14***
(Reciprocity, Shared narratives) 723.93 230.16***
(Reciprocity, Centrality) 635.74 141.97***
(Reciprocity, Network ties) 697.35 203.58***
(Reciprocity, IM usage) 669.19 175.42***
(Shared codes and language, Shared narratives)
727.31 233.54***
(Shared codes and language, Centrality) 633.94 140.17***
(Shared codes and language, Network ties) 725.81 232.04***
(Shared codes and language, IM usage) 648.10 154.33***
(Shared narratives, Centrality) 692.89 199.12***
(Shared narratives, Network ties) 726.04 232.27***
(Shared narratives, IM usage) 710.23 216.46***
(Centrality, Network ties) 601.70 107.93***
(Centrality, IM usage) 566.00 72.23***
(Network ties, IM usage) 639.76 145.99***
*** Significance at the 0.001 overall significance level by using the Bonferroni
higher than those for males, providing some support for our argu-ment above. Nevertheless, the previous literature states that the test of AVEs (Fornell and Larcker 1981) is quite conservative and very often variance extracted estimates will be below 0.50, even when reliabilities are acceptable (Hatcher 1994, p. 331).
This study applied the chi-square difference tests for assessing discriminant validity. The advantage of using chi-square difference tests is that their simultaneous pair-wise comparisons for the con-structs were based on the Bonferroni method. Controlling for the experiment-wise error rate yielded an overall significance level of 0.001, and the Bonferroni method suggests that the critical value of the chi-square difference should be 17.56. As the chi-square dif-ference statistics for all pairs of constructs inTable 2exceeded the critical value of 17.56, this study achieved the discriminant validity for the data sample. The test results suggest that the instruments used for measuring the constructs of interest in this study are sta-tistically adequate.
4.2. Structural model testing
After the above CFA was completed, this study performed struc-tural model testing that reflected the hypothesized associations for the purpose of hypotheses’ testing. To avoid unpredictable biases caused by individuals’ age and their industry, this study includes both of them as control variables in our structural modeling.
Fig. 1andTable 3present the test results of this analysis. The test results for the six hypotheses of this study are ex-plained as follows. First, instrumental support rather than expres-sive support has an indirect influence on IM usage via the mediation of commitment, suggesting thatH1is only partially ported. Second, neither expressive support nor instrumental sup-port has an influence on IM usage via reciprocity, suggesting that
H2 is not supported. Third, both expressive support and
instru-mental support have an indirect influence on IM usage via the mediation of shared codes and language, supportingH3. Fourth, both expressive support and instrumental support have no influ-ence on IM usage via shared narratives, suggesting thatH4is not supported. Fifth, both expressive support and instrumental support have an indirect influence on IM usage through the mediation of centrality, supporting H5. Lastly, both expressive support and instrumental support have no influence on IM usage via network ties, suggesting thatH6is not supported.
To further confirm the mediation effects of our six mediators (e.g., commitment, reciprocity, etc.) on IM usage, we conducted a post hoc analysis with four steps based on the recommendations
fromBaron and Kenney (1986)(seeAppendix C). The test results
presented inAppendix Cconfirm that when both the independent variables (i.e., expressive and instrumental support) and mediators (i.e., six social capital constructs) are used together to explain the outcome (i.e., IM usage) in step 4, the mediators which were signif-icant in step 2 remain significant and the independent variables which were significant in step 3 are no longer significant. Such test results support the full mediation effects for some of our social capital dimensions (i.e., commitment, shared codes and language, and centrality). Overall, it would be assertive to conclude that so-cial capital fully mediates the relationship between soso-cial support and IM usage. These empirical results suggest the mediating effects of social capital, fully or partially, and sometimes depend heavily on issues in which social capital is taken into account. For example, whileAquino and Serva (2005) find that social capital partially mediates the relationship between regular communication and perceived in-role performance, andLin and Huang (2005) empiri-cally show that the effects of human capital on developmental po-tential are fully mediated by social capital.
Based on the test results above, a further analysis for indirect ef-fects is also performed as shown inTable 4. The analysis indicates
Commitment Shared codes and language Reciprocity Shared narratives Network ties Centrality IM usage Expressive support Instrumental support 0.03 0.09 0.13* 0.16** 0.12* 0.12* 0.60*** 0.49*** 0.45*** 0.36*** 0.78*** 0.65*** 0.14* 0.04 0.14* 0.02 0.43*** 0.13 Industry Age -0.05 -0.04
that social support, expressively and instrumentally, affects IM usage indirectly through three mediators in which centrality is the most influential one among the three mediators.
The unsupportedH2, H4 and H6based onFig. 1suggest that not all online social capital elements significantly influence IM usage. Because reciprocity, shared narratives, and network ties are all important and impact different aspects of individuals’ behavior in the real world, they may be less influential to IM usage in the settings of a virtual world. Of course, this phenomenon does not suggest that the influence of social capital on IM usage should be completely ignored. Given that some social capital elements are still significantly influential to IM usage, the unexpected results for the unsupported hypotheses warrant further study.
The true reasons behind the unsupported hypotheses are not misinterpreted. To further explore our model, this study performed a post hoc analysis by removing three significant paths out of six model paths between social capital and IM usage. Structural mod-eling is then conducted and the test results show that the influence of both reciprocity and network ties on IM usage becomes signifi-cant with standardized coefficients of 0.24 (p < 0.01) and 0.41 (p < 0.01), respectively (whereas the influence of shared narratives on IM usage remains insignificant). The phenomenon may suggest future researchers to more carefully look into social capital from different dimensions.
5. Discussions
This study evaluates a formation of IM usage by considering so-cial capital as an important theoretical mediator. It is argued that online communities, which provide the contingency for enhancing
social support both expressively and instrumentally, are likely to foster greater IM usage. Our empirical research (from the aspects of social support and social capital) emphasizes the social utility of developing and maintaining social relationships. This is an important but often ignored perspective of IM applications (Li et al. 2005).
The findings of this study shed light on several areas that could benefit marketers, service providers, or corporate managers. For example, employees who frequently provide social support to oth-ers via IM are likely to have intensive social capital and frequent IM usage. As a result, these employees may be effectively appointed as key coordinators who require high levels of responsiveness to-wards their co-workers (or customers) via IM. For instance, if the task of HRM service to employees in an organization is assigned to staff who reveal a low inclination of providing social support, then they are unlikely to establish social capital and become un-able to solve problems (e.g., problems of training and education) with IM in a timely manner.
This study complements previous studies that have demon-strated the power of social influence in explaining virtual commu-nity participation (e.g.,Bagozzi and Dholakia 2002,Dholakia et al. 2004). Particularly, the findings of this study help to distinguish different weights between six dimensions of social capital regard-ing their influence on IM usage. Marketers (or corporate managers) leveraging critical mediators (i.e., commitment, shared codes, and language and centrality) into position can be rewarding for the vendors and marketers in order to boost users’ IM usage. The test results of this study suggest that IM usage is indirectly affected by social support via the mediation of different social capital dimensions – that is, low IM usage is likely attributed to a lack of commitment, shared codes and language, and centrality, which may be prompted by weak social support. This finding implies that vendors or marketers who want to promote their online product or service should establish supportive communities in which online individuals can easily obtain online social support. It is important that online promotion or incentives, such as virtual time dollars or gift vouchers that positively catalyze individuals’ social support for online others, may be provided. In the long run, both expressive and instrumental support can become reciprocal and expanded, eventually boosting IM usage.
Given that centrality is one of the most influential mediators for motivating users’ IM usage, marketers or corporate managers can make good use of this factor as a checkpoint to periodically assess both users’ social support and their IM usage. The tests re-sults of this study reveal that centrality can be a powerful role that significantly overwhelms other social capital dimensions. This finding is understandable, because most users in a modern society become heavily dependent on IM in their daily life, result-ing in a significant role of centrality between social support and IM usage. Marketers should invent daily life activities (e.g., e-charity, e-group improvisation, or e-group creation) to enhance the opportunity for users to perform social support, strengthening their centrality and subsequent IM usage. Marketers can also in-vent online solutions that link people’s daily ein-vents to online sys-tems. For instance, IM users may be regularly reminded of providing social support to their friends (e.g., sending
encourage-Table 3
Path coefficients and t values.
Hypothesis Standardized
coefficient
t value H1: Expressive support ? Commitment 0.03 0.46
H2: Expressive support ? Reciprocity 0.09 1.58
H3: Expressive support ? Shared codes and
language
0.13* 2.27
H4: Expressive support ? Shared narratives 0.16** 2.98
H5: Expressive support ? Centrality 0.12* 2.14
H6: Expressive support ? Network ties 0.12* 2.19
H7: Instrumental support ? Commitment 0.60*** 8.20
H8: Instrumental support ? Reciprocity 0.49*** 6.81
H9: Instrumental support ? Shared codes and
language
0.45*** 6.37
H10: Instrumental support ? Shared narratives 0.36*** 5.80
H11: Instrumental support ? Centrality 0.78*** 9.44
H12: Instrumental support ? Network ties 0.65*** 8.82
H13: Commitment ? IM usage 0.15* 2.10
H14: Reciprocity ? IM usage 0.05 0.72
H15: Shared codes and language ? IM usage 0.14* 2.11
H16: Shared narratives ? IM usage 0.02 0.30
H17: Centrality ? IM usage 0.43*** 4.72
H18: Network ties ? IM usage 0.13 1.74
*p < 0.05. **p < 0.01. ***p < 0.001.
Table 4
Analysis of indirect effects.
Path Indirect effects through Total effects
Commitment Shared codes and language Centrality
Expressive support ? IM usage 0.000 (0%) 0.018 (26%) 0.052 (74%) 0.070
ment e-cards) so that their perceived centrality and IM usage are strengthened.
This study collectively recommends that those marketers or managers who want to aggrandize users’ IM usage can place emphasis on social support and the significant social capital dimensions empirically confirmed herein. The marketers or man-agers should learn that IM usage can be substantially discouraged if social support is ignored.
A major limitation of this study is the possibility of a common method bias due to a single questionnaire applied to measure all constructs, and this possibility could inflate the strength of the hypothesized relationships among these constructs. Researchers may apply different instruments rather than only a single set of questionnaires to obtain their data in the future. Another limita-tion is the cross-seclimita-tional survey employed in this study. The development of social capital leading to its outcomes, such as IM usage and knowledge sharing, is an ongoing phenomenon (e.g.,
Chiu et al. 2006), and IM usage may generate some feedback to
so-cial capital in the long run.
This study’s social capital constructs were measured at a static point rather than as they were developing, hence losing some time richness of explanation (Chiu et al. 2006). Thus, future longitudinal research may complementarily support the findings of this study. Collectively, the model proposed and validated herein could bene-fit by further testing on the basis of a longitudinal survey. Future researchers can also improve the above shortcomings by observing the subjects’ IM usage behavior over time so that genuine associa-tions among research constructs in a virtual world can be transpar-ently revealed. This study also suggests that an issue which future research should address is whether access to and participation in online communities perpetuate gender, age, and racial inequalities. Lastly, given that using IM is much cheaper than using a traditional phone or a mobile phone, the costs may be an important driver for users to employ such Internet-based communication tools (e.g., mobile instant messaging), which could be taken into account in future research.
Acknowledgment
The author wishes to thank Professor Chou-Kang Chiu and Miss Siou-Jyuan Guo who helped assist in data collection.
Appendix A. Measurement items
Expressive support (1: Strongly Disagree; 5: Strongly Agree) ES1. Over the last one month, I received adequate emotional concern from people using IM.
ES2. Over the last one month, I felt relieved by getting sympathy from online people using IM.
ES3. Over the last one month, I met many people on IM whose company I really enjoy.
ES4. Over the last one month, I have been encouraged to make some choices related to my career by online people using IM.
Instrumental support (1: Strongly Disagree; 5: Strongly Agree) IS1. Over the last one month, I received numerous personal advice from online people using IM.
IS2. Over the last one month, I acquired a variety of information from online people using IM.
IS3. Over the last one month, I obtained sufficient assistance from online people using IM.
IS4. Over the last one month, I consulted online people using IM for practical issues and matters.
Commitment (1: Strongly Disagree; 5: Strongly Agree)
CO1. I would feel at a loss if IM were no longer available for my interaction with my friends (and/or relatives).
CO2. I care about the fate of IM regarding its future advance-ment on my communication efficiency with my friends (and/ or relatives).
CO3. I feel a great deal of loyalty to IM due to its importance in my interaction with my friends (and/or relatives).
Reciprocity (1: Strongly Disagree; 5: Strongly Agree)
RE1. When using IM, I think that my friends (and/or relatives) and I should trust each other.
RE2. When using IM, I think that my friends (and/or relatives) and I should maintain a relationship with each other.
RE3. When using IM, I think that my friends (and/or relatives) and I need identification with each other.
RE4. I never think that I have obligations to improve the rela-tionship with others on IM.
Shared codes and language (1: Strongly Disagree; 5: Strongly Agree)
SC1. When using IM, my friends (and/or relatives) and I under-stand each other with online jargon.
SC2. When using IM, my friends (and/or relatives) and I follow similar codes or rules.
SC3. When using IM, my friends (and/or relatives) and I easily obtain a consensus after discussion.
Shared narratives (1: Strongly Disagree; 5: Strongly Agree) SN1. My friends (and/or relatives) and I share interesting narra-tives through our IM usage.
SN2. My friends (and/or relatives) and I enjoy pleasant dialogue through our IM usage.
SN3. My friends (and/or relatives) and I share life events in detail through our IM usage.
Centrality (1: Strongly Disagree; 5: Strongly Agree)
CE1. I often think of talking with my friends (and/or relatives) via IM.
CE2. I think that using IM to contact my friends (and/or rela-tives) reflects a part of centrality in my life.
CE3. I count on IM for interacting with my friends (and/or relatives).
Network ties (1: Strongly Disagree; 5: Strongly Agree)
NT1. I have a lot of interests in common with my friends (and/or relatives) who use IM.
NT2. I feel strong interpersonal ties with my friends (and/or relatives) who use IM.
NT3. I find it difficult to form a bond with my friends (and/or relatives) who use IM.
NT4. I feel a sense of being strongly connected to my friends (and/or relatives) who use IM.
IM usage (1: Strongly Disagree; 5: Strongly Agree) IM1. I use IM very intensively.
IM2. I use IM very frequently.
IM3. I use IM for a variety of applications. IM4. Overall, I spend a lot of time using IM.
References
Anderson, J. C., and Gerbing, D. W. Structural equation modeling in practice: a review and recommended two-step approach. Psychological Bulletin, 103, 3, 1988, 411–423.
Aquino, K., and Serva, S. Using a dual role assessment to improve group dynamics and performance. Journal of Management Education, 29, 1, 2005, 17–38. Bagozzi, R. P., and Dholakia, U. M. Intentional social action in virtual communities.
Journal of Interactive Marketing, 16, 2, 2002, 2–21.
Bannert, M., and Arbinger, P. R. Gender-related differences in exposure to and use of computers: results of a survey of secondary school students. European Journal of Psychology of Education, 11, 3, 1996, 269–282.
Baron, R. M., and Kenny, D. A. The moderator–mediator variable distinction in social psychological research: conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 6, 1986, 1173–1182.
Bentler, P. M., and Bonett, D. G. Significance tests and goodness-of-fit in the analysis of covariance structures. Psychological Bulletin, 88, 3, 1980, 588–606. Boisot, M. Information Space: A Framework for Learning in Organisations, Institutions
and Culture. Routledge, 1995.
Boneva, B., Quinn, A., Kraut, R., Kiesler, S., and Shklovski, I. Teenage communication in the instant messaging era. In R. Kraut, M. Brynin, and S. Kiesler (eds.), Computers, Phones and the Internet, Oxford University Press, Oxford, 2006.
Boland, R. J., and Tenaski, R. V. Perspective making and perspective taking in communities of knowing. Organizational Science, 6, 4, 1995, 350–372. Bottrell, D. Dealing with disadvantage: resilience and the social capital of young
people’s networks. Youth and Society, 40, 4, 2009, 476–501.
Brown, J. S., and Duguid, P. Organizational learning and communities-of-practice: toward a unified view of working, learning, and innovation. Organization Science, 2, 1, 1991, 40–57.
Bryant, J. A., Sanders-Jackson, A., and Smallwood, A. M. IMing, text messaging, and adolescent social networks. Journal of Computer Mediated Communication, 11, 2, 2006. Available at:http://jcmc.indiana.edu/vol11/issue2/bryant.html. Retrieved November 20, 2006.
Buller, D. B., and Aune, R. K. The effects of speech rate similarity on compliance; application of communication accommodation theory. Western Journal of Communication, 56, 1, 1992, 37–53.
Celik, H., and Ipcioglu, I. Gender differences in the acceptance of information and communication technologies: the case of internet usage. International Journal of Knowledge and Learning, 3, 6, 2007, 576–591.
Chau, P. Y. K., Hu, P. J. H., Lee, B. L. P., and Au, A. K. K. Examining customers’ trust in online vendors and their dropout decisions: an empirical study. Electronic Commerce Research and Applications, 6, 2, 2007, 171–182.
Cheung, W., Chang, M. K., and Lai, V. S. Prediction of Internet and World Wide Web usage at work: a test of an extended Triandis model. Decision Support Systems, 30, 1, 2000, 83–100.
Cheung, C. M. K., and Lee, M. K. O. A theoretical model of intentional social action in online social networks. Decision Support Systems, 49, 1, 2010, 24–30. Chiu, C. M., Hsu, M. H., and Wang, E. T. G. Understanding knowledge sharing in
virtual communities: an integration of social capital and social cognitive theories. Decision Support Systems, 42, 3, 2006, 1872–1888.
Cohen, S., Mermelstein, R., Kamarck, T., and Hoberman, H. Measuring the functional components of social support. In I. G. Sarason and B. R. Sarason (eds.), Social Support: Theory, Research and Applications, Martines Niijhoff, The Hague, Holland, 1984, 73–94.
Coleman, J. S. Foundations of Social Theory. Belknap Press, Cambridge, MA, 1990. Cummings, J., Lee, J., and Kraut, R. Communication technology and friendship during
the transition from high school to college. In R. E. Kraut, M. Brynin, and S. Kiesler (eds.), Computers, Phones, and the Internet: Domesticating Information Technology, Oxford University Press, New York, 2006, 265–278.
Cutrona, C. E., and Suhr, J. A. Controllability of stressful events and satisfaction with spouse support behaviors. Communication Research, 19, 2, 1992, 154–174. Darley, W. K., and Smith, R. E. Gender differences in information processing
strategies: an empirical test of the selectivity model in advertising response. Journal of Advertising, 24, 1, 1995, 41–56.
Dholakia, U. M., Bagozzi, R. P., and Pearo, L. K. A social influence model of consumer participation in network- and small-group-based virtual communities. International Journal of Research in Marketing, 21, 3, 2004, 241–263.
Appendix B. AVE values of reciprocity and centrality
Construct Indicators Males Females
Standardized loading AVE Standardized loading AVE Reciprocity RE1 0.77 (t = 11.28) 0.47 0.77 (t = 9.25) 0.55 RE2 0.68 (t = 9.83) 0.77 (t = 9.20) RE3 0.59 (t = 8.51) 0.69 (t = 7.98) Centrality CE1 0.71 (t = 10.90) 0.47 0.67 (t = 7.85) 0.51 CE2 0.65 (t = 9.82) 0.75 (t = 8.90) CE3 0.69 (t = 10.56) 0.71 (t = 8.33)
Appendix C. Empirical tests of mediation effects and their regression coefficients
Step 1 Step 2 Step 3 Step 4
Commitment Reciprocity Shared codes and language Shared narratives Centrality Network ties IM usage IM usage IM usage Independent variables Expressive support 0.07 0.09 0.13** 0.16** 0.14** 0.15** 0.13** 0.04 Instrumental support 0.28** 0.21** 0.13** 0.23** 0.44** 0.37** 0.36** 0.06 Mediators Commitment 0.15** 0.15** Reciprocity 0.05 0.05
Shared codes and language 0.15** 0.15**
Shared narratives 0.03 0.03
Centrality 0.28** 0.26**
Network ties 0.15** 0.13**
*p < 0.05.
Notes: Step 1 presents that the mediator variables are explained by the independent variables (i.e., expressive and instrumental support). Step 2 presents that the outcome (i.e., IM usage) is explained by the mediator variables.
Step 3 presents that the outcome (i.e., IM usage) is explained by the independent variables.
Step 4 presents that both the independent variables and mediators are used together to explain the outcome (i.e., IM usage).
Drentea, P., and Moren-Cross, J. L. Social capital and social support on the web: the case of an Internet mother site. Sociology of Health & Illness, 27, 7, 2005, 920– 943.
Eastin, M. S., and LaRose, R. Alt. support: modeling social support online. Computers in Human Behavior, 21, 6, 2005, 977–992.
Fornell, C., and Larcker, D. F. Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 1, 1981, 39–50.
Granovetter, M. The strength of weak ties: A network theory revisited. In P. V. Marsden and N. Lin (eds.), Social Structure and Network Analysis, Sage, Beverly Hills, CA, 1982, 105–130.
Hatcher, L. A Step-by-Step Approach to Using SAS for Factor Analysis and Structural Equation Modeling. SAS Institute, Inc, Cary, NC, 1994.
He, W., Qiao, Q., and Wei, K. K. Social relationship and its role in knowledge management systems usage. Information & Management, 46, 3, 2009, 175–180.
Hirsch, B. J. Natural support systems and coping with major life changes. American Journal of Community Psychology, 8, 2, 1980, 159–172.
Horrigan, J. B. Online Communities: Networks that Nurture Long-distance Relationships and Local Ties. Pew Internet and American Life Study, Washington, DC, 2002. Hsu, C. L., and Lu, H. P. Why do people play on-line games? An extended TAM with
social influences and flow experience. Information & Management, 41, 7, 2004, 853–868.
Isaacs, E., Walendowski, A., Whittaker, S., Schiano, D. J., and Kamm, C. The character, functions, and styles of instant messaging in the workplace. In Proceedings of the 2002 ACM Conference on Computer Supported Cooperative Work, New Orleans, LA, USA, 2002, 11–20.
Kalichman, S. C., Benotsch, E. G., Weinhardt, L., Austin, J., Luke, W., and Cherry, C. Health-related Internet use, coping, social support, and health indicators in people living with HIV/AIDS: preliminary results from a community survey. Health Psychology, 22, 1, 2003, 111–116.
Kankanhalli, A., Tan, B. C. Y., and Wei, K. K. Contributing knowledge to electronic knowledge repositories: an empirical investigation. MIS Quarterly, 29, 1, 2005, 113–143.
Kim, C., Tao, W., Shin, N., and Kim, K. S. An empirical study of customers’ perceptions of security and trust in e-payment systems. Electronic Commerce Research and Applications, 9, 1, 2010, 84–95.
Lee, M. C. Factors influencing the adoption of internet banking: an integration of TAM and TPB with perceived risk and perceived benefit. Electronic Commerce Research and Applications, 8, 3, 2009, 130–141.
Lee, M. K. O., Cheung, C. M. K., and Chen, Z. Understanding user acceptance of multimedia messaging services: an empirical study. Journal of the American Society for Information Science and Technology, 58, 13, 2007, 2066–2077. Li, D., Chau, P. Y. K., and Lou, H. Understanding individual adoption of instant
messaging: an empirical investigation. Journal of the Association for Information Systems, 6, 4, 2005, 102–129.
Lin, N. Social Capital. Cambridge University Press, Cambridge, UK, 2001.
Lin, C. P., and Bhattacherjee, A. Understanding online social support and its antecedents: a socio-cognitive model. Social Science Journal, 46, 4, 2009, 724– 737.
Lin, C. P., and Bhattacherjee, A. Learning online social support: an investigation of network information technology based on UTAUT. CyberPsychology & Behavior, 11, 3, 2008, 268–272.
Lin, J., Chan, H. C., and Wei, K. K. Competing application usage with theory of planned behavior. Journal of the American Society for Information Science and Technology, 57, 10, 2006, 1338–1349.
Lin, S., and Huang, Y. The role of social capital in the relationship between human capital and career mobility: moderator or mediator? Journal of Intellectual Capital, 6, 2, 2005, 191–205.
Lin, T. C., and Huang, C. C. Understanding the determinants of EKR usage from social, technological and personal perspectives. Journal of Information Science, 35, 2, 2009, 165–179.
Litt, J. S. Medicalized Motherhood: Perspectives from the Lives of African-American and Jewish Women. Rutgers University Press, New Brunswick, NJ, 2000.
Liu, E. Z. F., and Chang, Y. F. Gender differences in usage, satisfaction, self-efficacy, and performance of blogging. British Journal of Educational Technology, 41, 3, 2010, E39–E43.
Lu, Y., Zhou, T., and Wang, B. Exploring Chinese users’ acceptance of instant messaging using the theory of planned behavior, the technology acceptance model, and the flow theory. Computers in Human Behavior, 25, 1, 2009, 29–39. Magni, M., and Pennarola, F. Intra-organizational relationships and technology acceptance. International Journal of Information Management, 28, 6, 2008, 517– 523.
Miyata, K. Social support for Japanese mothers online and offline. In B. Wellman and C. Haythornthwaite (eds.), The Internet in Everyday Life, Blackwell Publishers, Malden, MA, 2002.
Nahapiet, J., and Ghoshal, S. Social capital, intellectual capital, and the organizational advantage. Academy of Management Review, 23, 2, 1998, 242–266.
Nardi, B., Whittaker, S., and Bradner, E. Interaction and outeraction: Instant messaging in action. In Proceedings of CSCW 2000, Philadelphia, 2000, 79–88. Obst, P. L., and White, K. M. Three-dimensional strength of identification across
group memberships: a confirmatory factor analysis. Self and Identity, 4, 1, 2005, 69–80.
Oliva, R. A. Instant messaging comes of age. Marketing Management, 12, 3, 2003, 49– 52.
Park, S. Concentration of internet usage and its relation to exposure to negative content: does the gender gap differ among adults and adolescents? Women’s Studies International Forum, 32, 2, 2009, 98–107.
Park, J., Yang, S. J., and Lehto, X. Adoption of mobile technologies for Chinese consumers. Journal of Electronic Commerce Research, 8, 3, 2007, 196–206. Putnam, R. Tuning in, tuning out: the strange disappearance of social capital in
America. Political Science and Politics, 28, 4, 1995, 664–683.
Quan-Haase, A., Wellman, B., Witte, J., and Hampton, K. Capitalizing on the Internet: Social contact, civic engagement, and sense of community. In B. Wellman and C. Haythornthwaite (eds.), The Internet and Everyday Life, Blackwell, Oxford, UK, 2002.
Quan-Haase, A., and Wellman, B. How does the internet affect social capital? In M. Huysman and V. Wulf (eds.), Social Capital and Information Technology, MIT Press, Cambridge, MA, 2004, 113–135.
Radin, P. To me, it’s my life: medical communication, trust, and activism in cyberspace. Social Science & Medicine, 62, 3, 2006, 591–601.
Reagans, R., and Zuckerman, E. W. Networks, diversity, and productivity: the social capital of corporate R&D teams. Organization Science, 12, 4, 2001, 502– 517.
Recchiuti, J. K. College Student’s Uses and Motives for E-mail, Instant Messaging, and Online Chat Rooms. Unpublished master’s thesis, University of Delaware, Newark, Delaware, 2003.
Reynolds, N., Diamantopoulos, A., and Schlegelmilch, B. B. Pretesting in questionnaire design: a review of the literature and suggestions for further research. Journal of the Market Research Society, 35, 2, 1993, 171–182. Rouibah, K., and Hamdy, H. Factors affecting information communication
technologies usage and satisfaction: perspective from instant messaging in Kuwait. Journal of Global Information Management, 17, 2, 2009, 1–29. Rouibah, K. Social usage of instant messaging by individuals outside the workplace
in Kuwait; a structural equation model. Information Technology & People, 21, 1, 2008, 34–68.
Snchez-Franco, M. J. Exploring the influence of gender on Web usage via partial least squares. Behavior and Information Technology, 25, 1, 2007, 19–36. Seibert, S. E., Kraimer, M. L., and Liden, R. C. A social capital theory of career success.
Academy of Management Journal, 44, 2, 2001, 219–237.
Shiu, E., and Lenhart, A. How Americans use instant messaging? Pew Internet & American Life Project (September 1, 2004). Available athttp://www.pewinternet. org.
Shumaker, S., and Brownell, A. Toward a theory of social support: closing conceptual gaps. Journal of Social Issues, 40, 4, 1984, 11–36.
Song, L., and Lin, N. Social capital and health inequality: evidence from Taiwan. Journal of Health and Social Behavior, 50, 2, 2009, 149–163.
Swarbrick, A. Visualising knowledge networks: Leveraging the potential of work based communities. Project report submitted as part of a BSc (Hons) in Information Technology, Business Management and Language Degree in the Department of Computer Science at the University of York, 2002.
Timberlake, S. Social capital and gender in the workplace. Journal of Management Development, 24, 1, 2005, 34–44.
Vaux, A. Social Support: Theory, Research, and Intervention. Praeger, New York, 1988. Walther, J. B., and Boyd, S. Attraction to computer-mediated social support. In C. A. Lin and D. J. Atkin (eds.), Communication Technology and Society: Audience Adoption and Uses, Hampton Press, Cresskill, NJ, 2002, 153–188.
Wang, C., Hsu, Y., and Fang, W. Acceptance of technology with network externalities: an empirical study of Internet instant messaging services. Journal of Information Technology Theory and Application, 6, 4, 2004, 15–28. Wang, E. S. T. Internet usage purposes and gender differences in the effects of
perceived utilitarian and hedonic value. CyberPsychology, Behavior, and Social Networking, 13, 2, 2010, 179–183.
Wasko, M. M., and Faraj, S. Why should I share? Examining social capital and knowledge contribution in electronic networks of practice. MIS Quarterly, 29, 1, 2005, 35–57.
Wellman, B., and Frank, K. Network capital in a multi-level world: Getting support from personal communities. In N. Lin, R. Burt, and K. Cook (eds.), Social Capital: Theory and Research, Hawythorne, Aldine de Gruyter, NY, 2001.
Willemyns, M., Gallois, C., Callan, V., and Pittam, J. Accent accommodation in the job interview: impact of interviewer, accent and gender. Journal of Language and Social Psychology, 16, 1, 1997, 3–22.
Yoon, C., and Kim, S. Convenience and TAM in a ubiquitous computing environment: the case of wireless LAN. Electronic Commerce Research and Applications, 6, 1, 2007, 102–112.
Zaman, M., Anandarajan, M., and Dai, Q. Experiencing flow with instant messaging and its facilitating role on creative behaviors. Computers in Human Behavior, 2010. doi:10.1016/j.chb.2010.03.001.
Zhang, Z. J. Feeling the sense of community in social networking usage. IEEE Transactions on Engineering Management, 57, 2, 2009, 225–239.
Zhou, L. Application of TPB to punctuation usage in instant messaging. Behaviour & Information Technology, 26, 5, 2007, 399–407.
