- Email: [email protected]
Why do you return the favor in online knowledge communities? A study of the motivations of reciprocity
Computers in Human Behavior 63 (2016) 342e349
Contents lists available at ScienceDirect
Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh
Full length article
Why do you return the favor in online knowledge communities? A study of the motivations of reciprocity Hua (Jonathan) Ye b, *, 1, Yuanyue Feng a, 1 a b
Department of Management Science, College of Management, Shenzhen University, No. 3688, Nanhai Ave, Shenzhen, Guangdong, 518060, China School of Management, Harbin Institute of Technology, 13 Fayuan Street, Nangang Dist., Harbin, 150001, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 June 2015 Accepted 3 May 2016 Available online 26 May 2016
Online knowledge community administrators are attempting to encourage their users to contribute knowledge in order to provide value to members and maintain sustainability. A large number of online knowledge communities fail mainly due to the reluctance of users to return the favor and share knowledge. Many studies on this topic have highlighted the importance of reciprocity for knowledge contribution which forms a virtuous feedback loop for the community sustainability. However, it is unclear how reciprocity is developed and what influences its development. Motivated by this, this study focuses on investigating the antecedents of knowledge receivers’ reciprocity in online knowledge communities. It formulates and tests a theoretical model to explain reciprocity behavior of community members based on equity theory and Social Identity explanation of De-individuation Effects (SIDE) model. Our proposed model is validated through a large-scale survey in an online forum for English learning. Results reveal that indebtedness and community norm not only are key antecedents of intention to reciprocate but are also positively related to each other. The perceived anonymity of the online knowledge community not only has a positive effect on intention to reciprocate, but also has an interactive effect with community norm on intention to reciprocate. Theoretical and practical implications of this study are discussed. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Online knowledge community Reciprocity Indebtedness Community norm Perceived anonymity
1. Introduction Online knowledge communities comprise of individuals with common interests, goals, or practices, who share and combine knowledge for their own benefits and engage in social or personal interactions (Ye, Feng, & Choi, 2015). They serve not only as sources of information, social support, and recreation, but also as a platform for knowledge exchange (Armstrong & Hagel, 1996; Phang, Kankanhalli, & Sabherwal, 2009). Interactions and knowledge embedded in online knowledge communities are the key factors to allow them to survive and thrive (Wasko & Faraj, 2005). Online knowledge community administrators are hence attempting to encourage their users to contribute knowledge and resources in order to provide value to members and ensure sustainability (Phang et al., 2009). Motivated by this, researchers have been
* Corresponding author. E-mail addresses: [email protected] (H. Ye), [email protected] (Y. Feng). 1 Both authors equally contribute to this paper and are listed in an alphabetical order. http://dx.doi.org/10.1016/j.chb.2016.05.007 0747-5632/© 2016 Elsevier Ltd. All rights reserved.
investigating the drivers of online community members’ knowledge contribution behaviors. They conclude that various factors motivate the behavior of online knowledge contribution, such as the anticipation of extrinsic benefits (organizational rewards, reputation) (e.g., Kankanhalli, Tan, & Wei, 2005; Wasko & Faraj, 2005), intrinsic benefits (sense of self worth, sense of belongingness, and social affiliation) (e.g., Bock, Zmud, Kim, & Lee, 2005), and social capital (social interaction ties, trust, norm of reciprocity, perceived identity, and shared language) (e.g., Chiu, Hsu, & Wang, 2006; Wasko & Faraj, 2005). Even in the absence of organizational rewards, a key motivator found in these studies is the idea of reciprocity (e.g., Ardichvili, Page, & Wentiling, 2003; Chiu et al., 2006; Wasko & Faraj, 2000, 2005). This is because reciprocity enables the formation of a virtuous feedback loop to knowledge contribution and thus the community sustainability. Constructs like subjective norms or pro-sharing norms have been identified to explain how receivers’ reciprocity behavior is regulated (e.g., Bock et al., 2005). In order to understand reciprocity behaviors, researchers (e.g., Mathews & Green, 2009; Nowak & Sigmund, 2005) propose that
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
besides being driven by incentives, reciprocity may also derive from a desire to repay the favor or knowledge received from the community before as explained by equity theory. Such a desire to repay tends to exist in those who frequently obtained necessary information from the communities and learned skills for their tasks (Wasko & Faraj, 2000). This desire may derive from the feeling of indebtedness (Kolm, 2008), or from community norm (Wasko & Faraj, 2005). However, no empirical research has systematically examined the effects of indebtedness and community norm on reciprocity behaviors. Further, online reciprocity behaviors may be driven by certain unique factors since online knowledge communities differ notably from conventional organizations (Chiu et al., 2006). One of the salient differences is the anonymity of online knowledge communities (Wasko, Faraj, & Teigland, 2004). In online knowledge communities, users can interact anonymously and indirectly (Fehr and €chter 2000). They can easily remain anonymous or change their Ga identities (Ba & Pavlou, 2002), since most forum websites identify users by e-mail addresses, which can be readily obtained from multiple sources. The anonymity of members in the online community makes it more likely that individual interactions will go unnoticed by other network members (Wasko et al., 2004). This may lead individuals to have different psychological responses to social interactions (Faraj, Jarvenpaa, & Majchrzak, 2011; Pinsonneault & Heppel, 1998). Such differences require researchers to investigate the influence of anonymity on online reciprocity. Although prior studies have provided clues about the possibility of anonymity’s influence on reciprocity (e.g., Alpizar, Carlsson, & Johansson-Stenman, 2008; Kolm, 2008) in the offline context, there is no empirical study to test the relationship in the online context. Without knowing the effects of anonymity, there will be a gap in our understanding of what affects reciprocity behaviors in online knowledge communities. With the above practical and theoretical motives, we are interested to study the antecedents of reciprocity in online knowledge communities from knowledge receivers’ perspective. Based on the equity theory and Social Identity explanation of De-individuation Effects (SIDE) model, this study develops a model to explain the effects of perceived anonymity, community norm, and indebtedness on knowledge receivers’ intention to reciprocate in online knowledge communities. A survey was conducted in an English learning forum to test the model. The study expects to contribute to the existing literature in following ways. First, it helps improve our understanding of online reciprocity behaviors by adopting new theoretical lenses, i.e., equity theory and SIDE model. Second, it models and tests the antecedents of online reciprocity through a large-scale survey. Third, it generates new insights about online knowledge contribution from the perspective of knowledge receivers’ reciprocity behaviors.
343
used to explain social interaction via computer-mediated communication, which fits well with the context of our study. 2.1. Equity theory Equity theory focuses upon an individual’s perception and request of fairness or equity with respect to a relationship (Cohen & Greenberg, 1982). During a social exchange, an individual assesses the ratio of what is output from the relationship to what is input in the relationship, and also the ratio of what the other person in the relationship output from the relationship to what is input into the relationship. Equity theory posits that an equitable relationship exists when individuals perceive that they are receiving equal relative outcomes from the exchange compared with their inputs (Adams, 1965; Watkins, Scheer, Ovnicek, & Kolts, 2006). That is, whether they are receiving a fair return for the efforts or resources that they put into the exchange (Glass & Wood, 1996). A perception of inequity in an exchange results in the feeling of indebtedness (Gouldner, 1960). Such a feeling motivates individuals to commit to a reciprocal behavior in order to avoid being perceived as socially insensitive (Mathews & Green, 2009). For example, in Glass and Wood (1996)’s software piracy study, the debt perceived to be owed to others from a prior exchange are identified as the main factor for an individual’s intention to provide an illegal software copy to others. Therefore, the more inequitable the relationship, the more indebted the participants will feel and the greater they will be motivated to reduce the inequity (Greenberg, 1986). There are two ways that an individual can restore equity in an inequitable relationship. First, the individual can restore “actual equity” by appropriately altering his own outputs or inputs in the exchange. Second, the individual can restore psychological equity by appropriately distorting perceptions of his or her own outputs and inputs compared with other participants’ (Walster, Berscheid, & Walster, 1973), or by reducing the importance of the inequity (Watkins et al., 2006). In the context of online knowledge contribution, the input and output of knowledge exchange in online communities can be considered as the knowledge contributed and the benefits received from the exchange (Chiu et al., 2006). When the benefits received from the exchange (outputs) exceed the perceived value of knowledge contributed (input), knowledge receivers will perceive inequity and feel indebted to the community (i.e., indebtedness). They will generate a desire to reciprocate the community by contributing knowledge with higher quality and quantity, so as to restore the inequity inside (Wiertz & Ruyter, 2007). Therefore, indebtedness is expected to affect individuals’ reciprocity intention and behaviors in online knowledge communities. Accordingly, we include indebtedness into our model. 2.2. Social identity model of de-individuation effects (SIDE Model)
2. Theoretical foundations Previous theories that have been used to study knowledge contribution motives include motivation theory (Bock et al., 2005; Chiu, Wang, Shih, & Fan, 2011), social exchange theory (Kankanhalli et al., 2005; Wasko & Faraj, 2005; Watson & Hewett, 2006), social capital and social cognitive theories (Chiu et al., 2006; Hsu, Ju, Yen, & Chang, 2007; Lin, Hung, & Chen, 2009), and public goods theory (Hollingshead, Fulk, & Monge, 2002; Wasko & Faraj, 2000). These theories highlight reciprocity as a motive for knowledge contribution but do not investigate the antecedents of reciprocity. For this reason, equity theory has been adopted in this study as the theoretical foundation to investigate the antecedents of reciprocity. Another theoretical foundation for this study is Social Identity explanation of De-individuation Effects (SIDE) model. SIDE model is
SIDE model suggests that anonymity changes the relative salience of personal and social identity, and thereby having a profound effect on group behavior (Spears & Lea, 1994). Anonymity in an online community obscures individual features and interpersonal differences, and hence enhances the salience of social identity. It thereby depersonalizes social perceptions of others and the self (Postmes, Spears, Sakhel, & De Groot, 2001). This decreased visibility of the individual within anonymous groups results in the accentuation of the depersonalization process and amplification of cognitive efforts to perceive the group as an entity (Postmes, Spears, & Lea, 1998). In online knowledge communities, the de-individuating features (e.g., perceived anonymity, physical isolation, and selective self-presentation) decrease perceptions of individual differences.
344
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
This decreased perception of individual differences leads individuals to focus less on their self-conception and hence to increase their adherence to community norms and form more positive impression about the online knowledge communities (Walther, 1996). If paired with high group norm salience, these results become more salient. A strong community norm indicates strong expectations from peer members in the community. It includes the expectation of reciprocity, equity, and mutual support between members in the exchange (Wasko & Faraj, 2005). Behaviors conflicting with equity expectation are considered illegitimate and are not encouraged in the community (Yee, Bailenson, Urbanek, Chang, & Merget, 2007). In online knowledge communities, in order to avoid being perceived alien and keep a positive self-concept, community members tend to conform to the norms and hence conduct reciprocity behaviors. Therefore, a strong community norm will encourage members to conform to the expectation of their peers and hence to reciprocate. Based on the above reasoning, we include perceived anonymity and community norm in our model. 3. Research model and hypotheses Reciprocity is defined as the propensity among at least two cooperative partners to repay the favor that the partner has received before (Kolm, 2008). In online settings, reciprocity tends to be generalized (Wasko & Faraj, 2000). Generalized reciprocity refers to the propensity to reciprocate another’s action, not by directly rewarding the benefactor, but by benefiting another actor implicated with the benefactors in a social exchange (Westphal & Zajac, 1997). Following Kolm (2008) and Westphal and Zajac (1997), we define our dependent variable, i.e., intention to reciprocate, as the intention of beneficiary to return help to the benefactors or those who are in the benefactors’ group. Based on the two theoretical perspectives discussed, we propose the research model for this study (see Fig. 1). Our dependent variable intention to reciprocate is expected to be influenced by community norm, indebtedness, and perceived anonymity. Furthermore, community norm is expected to influence indebtedness. Perceived anonymity is also proposed to moderate the relationship between community norm and intention to reciprocate. 3.1. Community norm Community norm is defined as a behavioral regularity (Yee et al., 2007), based on a socially shared belief of how one ought to behave. It refers to a rule for what should be done, accepted, and internalized by the group members (Scott, 1971). It triggers the enforcement of the prescribed behavior by informal social sanctions (Fehr and G€ achter, 2000) Previous literature proposes that through social influence
processes (Fulk, 1993; Schmitz & Fulk, 1991), community norms can have an important influence on attitude and intention (Ajzen, 1991). According to the SIDE model, the stronger the community norms are, the more individuals are motivated to conform to them, and the more their attitudes and behaviors tend to be communitydetermined than individual-determined (Lee, 1990). When reciprocity behaviors are regarded as a salient community norm in a community, individuals in this community are more likely to be aware of this community norm and the expectations of significant others for reciprocity behaviors. Such awareness prompts them to comply with these expectations (Bock et al., 2005) because conformity to the community norm renders them legitimacy in the community. Thus, a higher level of community norm of reciprocity can lead to a higher intention to reciprocate towards the community. In online knowledge communities, similarly, when an individual perceives a high level of community norm of reciprocal knowledge sharing behaviors, the individual may unconsciously or consciously feel that reciprocal knowledge sharing are expected by others in the online knowledge community. In view of that, the individual may perceive a need to follow such community norm so as to gain legitimacy in the community. Thus, the individual would intend to engage in reciprocal knowledge contribution to the community. H1. Community norm is positively related to the intention to reciprocate in online knowledge communities.
3.2. Indebtedness As a psychological construct, indebtedness has been defined as “a state of obligation to repay another” in the context of receiving a benefit from another (Greenberg, 1980, p. 4). The state of indebtedness constitutes a violation of the sense of “ought” and therefore involves attendant feelings of guilt analogous to the feelings presumed to accompany distributive injustice or inequity (Cohen & Greenberg, 1982). It is an emotional state of “arousal and discomfort”, where the individual is alert to opportunities to reduce this discomfort (Greenberg & Shapiro, 1971). In online knowledge communities, when members receive knowledge or favor from other members, they compare it with their own input into the online knowledge community. After comparison, if they find an inequity in the exchange, they will feel indebtedness because this is in conflict with their own positive impression and self-perception (Constant, Kiesler, & Sproull, 1994). Equity theory suggests that a desire to reduce such a feeling of indebtedness will motivate individuals to reciprocate other members in the community with their knowledge in order to avoid being perceived as socially insensitive (Mathews & Green, 2009) and to maintain their self perception. Hence we hypothesize: H2. Indebtedness is positively related to intention to reciprocate in online knowledge communities.
Community Norm H3
H1 H2
Intention to Reciprocate
Indebtedness H5
H4
Perceived Anonymity Fig. 1. Research model.
Based on the SIDE model, a strong community norm of reciprocal knowledge sharing implies a strong expectation for members to return the help or knowledge from the community and a high pressure to get legitimized. With increasing expectations from the community, the feeling of indebtedness will be enhanced (Watkins et al., 2006). Under the condition of salient community norm, individuals tend not to distort perceptions of benefits and appreciate the value of knowledge contributed in the exchange comparing with other participants’ (Walster et al., 1973), such as by reducing the importance of inequity (Cooper, 2007). Therefore, this strong community norm enhances individuals’ feeling of indebtedness. H3. Community norm is positively related to indebtedness in online
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
knowledge communities.
3.3. Perceived anonymity Anonymity refers to the extent to which community members cannot identify other members or the source of a particular contribution to the group (Pinsonneault & Heppel, 1998). In online knowledge communities, the richness of normal face-to-face interaction and self-regulation (Kiesler, Siegel, & Mcguire, 1984) is absent. Anonymity of others leads to heightened self-awareness, and thus to greater adherence to group norms when a social identity is salient (Joinson, 2001). The heightened self-awareness leads members to more focus on their own feelings and behaviors in the community. This could lead to pro-social behaviors online to self-fulfill their own perception about what they want to be (Ferraro, Pfeffer, & Sutton, 2005). Moreover, anonymity enhances members’ participation in the discussion (Sia, Tan, & Wei, 2002). For example, anonymity has been found to increase equity and participation rates in e-learning (Chester & Gwynne, 1998). This positive effect of anonymity is explained by Walther (1996). Drawing on the SIDE model developed by Lea and Spears (1992), Walther proposed that the de-individuating features of online communities (visual anonymity, physical isolation, and selective self-presentation) lead to decreased perceptions of individual differences, increased adherence to group norms, and more positive impression formation. In order to sustain positive impression formation, people tend to be more pro-social and helpful to others (Leary & Kowalski, 1990). Thus, during the knowledge exchange in online knowledge communities, members who perceive people as anonymous tend to be active in reciprocating the community to maintain their self perception. H4. Perceived anonymity is positively related to the intention to reciprocate in online knowledge communities. The perception of anonymity may have a moderating effect on the relationship between community norm and reciprocity in online knowledge communities (Ba & Pavlou, 2002; Spears & Lea, 1994). According to the SIDE model (Lea and Spears 1992; Spears & Lea, 1994), with anonymity inherent in most online knowledge communities, social identity becomes salient. This serves to strengthen the impact of social norms, and hence normative influence. When a social identity is salient, through a heightened selfawareness, anonymity increases members’ intent to maintain a good self concept, thereby leading to their greater adherence to group norms (Joinson, 2001). After receiving favors or benefits, anonymous members want to reciprocate especially when the community norm is strong. Therefore, we hypothesize that: H5. The positive relationship between community norm and the intention to reciprocate in online knowledge communities will be stronger when perceived anonymity is high. We also include age, gender, forum tenure, education level, previous experience of being helped, and posting frequency in our model as control variables that might affect reciprocity in online knowledge communities. 4. Research methodology The research model was empirically tested by the data collected from an English learning forum (bbs.gter.net) through survey methodology. Most of the constructs in our theoretical model are latent variables, which are best studied using the survey approach (Nunnally & Bernstein, 1994). Besides, the survey methodology provides a basis for establishing generalizability, allows
345
replicability, and has statistical power (Teo, Wei, & Benbasat, 2003). 4.1. Survey administration We collect our survey data from bbs.gter.net, an online student discussion forum for language learning, particularly English learning. Individuals post their questions related to English learning or literature translation. There were an estimated 5 million registered users and 4 thousand online users per day with about 2 thousand posting per day in this online knowledge community.2 We posted the questionnaire in bbs.gter.net for one week with the help of forum administrators to highlight the survey invitation. An invitation letter is also delivered through a well known Southeast Asian university to the international students who have registered an account in this forum. In appreciation of the respondents’ effort, we promised a lucky draw with 2 prizes of $100 each for the respondents. Overall, we received 180 responses from registered users of this online forum. Due to attrition and missing values, the resulting sample size used in the analyses is 169. Of the 169 respondents, the proportion of males exceeded female respondents (57.9% are males and 42.1% are females), with the majority of the respondents aged between 20 and 26 years old (86.3%). On average, 91.8% respondents had internet experience of more than 4 years. The respondents had an average of 1.94 years of experience using the forum for knowledge sharing. 4.2. Operationalization of constructs The survey items were generated based on a review of the relevant information systems (IS), knowledge management, and reciprocity literature. Where previously tested measures were not available, we developed items based on the construct definition and description. Table 1 provides formal definitions of the constructs. Where available, the constructs are measured using instruments adapted from past studies to enhance validity (Stone, 1978). Otherwise, new instruments are developed based on a review of the knowledge management, information systems and reciprocity literature. All the items are measured by using seven-point scales anchored from “strongly disagree” to “strongly agree” (see Table 2 for the survey items for each construct), unless otherwise indicated. Given that the items for measuring the constructs were adapted from various sources or developed for this study, all of the items were subjected to a two-stage conceptual validation exercise based on procedures prescribed by Moore and Benbasat (1991). All items have an acceptable level of agreement among sorters (Cohen’s Kappa > 0.9, hit rate > 0.95). We also develop items for all the control variables (see Table 3). 5. Data analysis For this study, structural equation modelling (SEM) analysis was chosen over regression analysis, because SEM can simultaneously analyze all of the paths in one analysis (Chin, 1998). Within SEM, PLS was chosen because the research is exploratory in nature and the model proposed is multi-stage (Ma & Agarwal, 2007). This study aims to explore the antecedents of reciprocity intention in online knowledge communities. Therefore, we used Partial Least Squares (PLS) as the main statistical technique to analyze the data. Constructs were modelled using reflective indicators according to Chin (1998). SmartPLS 2.0 (M3 Release) was used to run our PLS
2
http://bbs.gter.net/bbs/.
346
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
Table 1 Formal definitions of constructs. Construct (abbreviation)
Definition
Source
Intention to Reciprocate (RECP) Indebtedness (INDEB) Community Norm (NORM) Perceived Anonymity (ANY)
The intention of beneficiary to return help to the benefactors or those who are in the benefactors’ group
Kolm (2008)
A state of obligation to repay another in the context of receiving a benefit from another A behavioral regularity based on a socially shared belief of how one ought to behave The extent to which community members cannot identify other members or the source of a particular contribution to the group
Greenberg (1980) Yee et al. (2007) Pinsonneault and Heppel (1998)
Table 2 Operationalization of constructs. Construct
Items
Intention to Reciprocate RECP1 (RECP) RECP2 RECP3
Indebtedness (INDEB)
Community Norm (NORM) Perceived Anonymity (ANY)
RECP4 INDEB1 INDEB2 INDEB3 NORM1 NORM2 NORM3 ANY1 ANY2 ANY3
Source When receiving help from this community, I will return the help to the community When obtaining experience from this community, I will share my own experiences later with this community When receiving suggestions about my situation from this community, I will contribute my suggestions to others When receiving support from this community, I will support others in this community When I receive help from the community, I feel that I owe something to the community When I receive help from the community, I feel guilty if I cannot repay the community When I receive help from the community, I feel indebted to the community When receiving help from the community, others expect receivers to return the help There is an atmosphere in this community to return others’ help Receivers do not live up to others’ expectation if they do not return the help Others can identify my topics and comments (Reverse) The community is large enough that nobody can trace topics and comments back to their authors Others can recognize my topics and comments (Reverse)
Self-developed
Greenberg and Shapiro (1971); Watkins et al. (2006) Bock et al. (2005)
Pinsonneault and Heppel (1998)
Table 3 Control variables. Control Variable (abbreviation)
Survey items
Age (AGE) Gender (GENDER) Forum Tenure (TENURE) Education Level (EDU) Previously being helped (PREHELPED) User Posting (POSTING)
Please provide your age (in years). Male/Female How long have you been using this forum (in years)? (1: Bachelor and below; 2: Master; 3: PhD) (0: Previously not being helped; 1: Previously being helped) Please indicate your number of posting per week in this forum
analysis. 5.1. Measurement model To validate the measurement model, reliability (Nunnally, 1978), discriminant validity (Chin, 1998) and convergent validity (Fornell & Larcker, 1981) were assessed for the reflective indicators (Hair, Anderson, Tatham, & Black, 1998). All reflective constructs in our model exhibited acceptable levels of reliability, convergent validity (see Table 5), and discriminant validity (see Tables 4 and 5). The single method test of Harman (1976) was used to test for common method variance. The factor analyses produced neither a single factor nor one general factor that accounted for the majority of the variance (>50%). Each factor accounted for more than the viable cut-off of 5% (see Table 3). Therefore, this method indicated that common method bias was not a problem. 5.2. Structural model results The results of the data analysis are shown in Fig. 2 and Table 6. As seen from Fig. 2 below, our structural model could explain 42% of the total variance in intention to reciprocate and 43% of indebtedness which is greater than the acceptable threshold of 10 percent (Falk & Miller, 1992). Fig. 2 also summarizes the path coefficients and their significance from bootstrapping resampling.
Table 4 Item loadings. Items
RECP1 RECP2 RECP3 RECP4 INDEB1 INDEB2 INDEB3 ANY1 ANY2 ANY3 NORM1 NORM2 NORM3 Eigen value % of variance Cumulative %
Component 1
2
3
4
0.92 0.79 0.84 0.84 0.12 0.01 0.17 0.14 0.12 0.12 0.05 0.23 0.19 5.51 39.37 39.37
0.12 0.14 0.09 0.12 0.87 0.84 0.83 0.08 0.17 0.09 0.40 0.22 0.17 2.15 15.37 54.74
0.06 0.19 0.09 0.10 0.14 0.17 0.05 0.78 0.86 0.94 0.16 0.00 0.08 1.45 10.37 65.11
0.08 0.13 0.14 0.18 0.16 0.23 0.16 0.34 0.07 0.03 0.68 0.78 0.83 1.20 8.62 73.74
Demographic variables (i.e., age, gender, forum tenure, education level, previously being helped, and user posting) are included in the analysis as controls for reciprocity. Among all the controls, forum tenure is found to positively affect intention to reciprocate. We found adequate support for our research model in this study.
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
347
Table 5 Descriptive statistics, construct correlation matrix. Construct
Mean
RECP INDEB NORM ANY
Cronbach’s alpha
5.79 4.32 4.81 3.22
Composite reliability
0.90 0.87 0.78 0.85
0.93 0.91 0.87 0.91
SD
1.20 1.50 1.26 1.35
Correlations 1
2
3
4
0.87 0.40 0.38 0.29
0.84 0.56 0.26
0.83 0.29
0.88
Note: Diagonal elements are the square root of AVE.
Community Norm H3 0.65**
Indebtedness
H1 0.28** H2 0.11*
Intention to Reciprocate
H5 -0.32***
R2=.42
0.22** H4
R2=.43
Perceived Anonymity Fig. 2. Graphical display of results (*p < 0.05,
**
p < 0.01,
***
p < 0.001).
All paths were significant at the 0.05 level with three paths significant at the 0.01 level. As predicted, perceived anonymity, community norm, and indebtedness were positively related to members’ intention to reciprocate in online knowledge communities (H1, H2, and H4 were supported). Community norm were significantly positively related to indebtedness (H3 was supported). Contrary to our prediction, there was a negative interaction effect of perceived anonymity and community norm on the intention to reciprocate (H5 was not supported). 6. Discussion and implications In online knowledge communities, individuals usually search for knowledge to fulfill their needs (Ye et al., 2015). After they fulfilled their knowledge needs, it was thought that individuals will leaf until they have new needs. Surprisingly, studies found that individuals return the favor to the community by helping others (Wasko & Faraj, 2005; Ye et al., 2015). This reciprocal behavior has been believed to be critical to the survival and sustainability of
online knowledge communities. Researchers and practitioners are interested in how to promote such a behavior. Considering the importance of reciprocity in engaging community members in knowledge contribution and hence the sustainability of online knowledge communities, we study the antecedents of reciprocity in online knowledge communities based on equity theory and the SIDE model. In line with our hypotheses, results show that community norm, indebtedness, and perceived anonymity affect members’ intention to reciprocate. Also, community norm positively influence members’ feeling of indebtedness. In contrast to our hypothesis, perceived anonymity has a negative interactive effect with community norm on members’ intention to reciprocate (H5 not supported). Under a condition of high perceived anonymity, the positive effect of community norm on members’ intention to reciprocate will be weakened, rather than being strengthened. A plausible explanation for this unexpected result may be that, as members perceive their real identities to be unseen by others in the online knowledge community, they might be less concerned with or care about their legitimacies in the forum, thereby being less motivated to conform to the community norm of reciprocity. This could be because individuals’ self awareness was not heightened and individuals are more likely to free ride (Spears & Lea, 1994). Hence, their intention to comply with the community norm would be undermined in this situation. Another possible reason could be due to the difference of online communities from physical organizations where online communities have no formal regulations and reward mechanisms (Faraj et al., 2011). The reciprocal behavior in online communities is purely voluntary and may not be constrained by such regulations and reward mechanisms. 6.1. Theoretical implications This study contributes to the literature in several ways. First, our research is novel in examining the antecedents of reciprocity behavior in the online context. Reciprocal contribution of knowledge by members has been emphasized as a critical factor for the
Table 6 Hypotheses testing. DV ¼ Intention to reciprocate
Age Gender Forum tenure Education level Previously being helped User posting Community norm Indebtedness Perceived anonymity Community norm * Perceived anonymity R2 *
p < 0.05;
**
p < 0.01;
***
p < 0.001.
DV ¼ Indebtedness
1
2
Result
0.07 0.01 0.32* 0.10 0.01 0.09
0.03 0.02 0.21* 0.07 0.01 0.08 0.28* 0.11* 0.22* 0.32*** 0.42
N.S. N.S. Sig. N.S. N.S. N.S. H1 Supported H2 Supported H4 Supported H5 not Supported
0.09
3
Result
0.65***
H3 Supported
0.43
348
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
sustainability and growth of online knowledge communities (Wasko & Faraj, 2005; Watson & Hewett, 2006; Ye et al., 2015). Through the exploration and examination of the motives for reciprocal knowledge contribution, this study advances theoretical development in the area of knowledge contribution in general and reciprocity behavior in online knowledge communities in particular. Second, unlike most of the prior knowledge contribution studies, our research adopts new theoretical lenses to examine the antecedents of online reciprocal knowledge contribution. These new theoretical lenses, i.e., equity theory and SIDE model, provide additional insights into reciprocity behaviors in the online context. On one hand, this study echoes to and advances the equity theory by discovering the positive effect of indebtedness on members’ intention to reciprocate in online knowledge communities. Consistent with prior literature on reciprocity in offline context (Glass & Wood, 1996; Mathews & Green, 2009), we find significant relationship between the intrinsic factor, i.e., feeling of indebtedness, and reciprocity intention in online context. This finding suggests that in the online knowledge communities, knowledge receivers are likely to believe that they are receiving help from the community as a whole rather than from a specific individual. Thereby they tend to reciprocate by contributing their knowledge to the online knowledge community so as to repay the favors received from the community. On the other hand, this study enriches the SIDE model by examining its applicability to the context of online knowledge sharing. In accord with the theoretical essence of the SIDE model, community norm of reciprocity fosters the intention to reciprocate by forum members. A strong community norm indicates a strong peer expectation and individuals tend to conform to this expectation in order to be legitimate and be liked by others. In addition, perceived anonymity by members increases their self-awareness and hence their intention to maintain a good self concept, which in turn motivates them to engage in reciprocity. 6.2. Practical implications Collectively, results of this study indicate the circumstances under which mechanisms to promote reciprocity intention and behavior in online knowledge communities may be more effective. First, online knowledge communities can raise the community norm of reciprocity among members by rewarding the reciprocity behaviors and announcing this to the whole community. This can be done by highlighting reciprocal postings or adding reputation scores for those who reciprocate. Enhancing communication channels between community members can also strengthen the community’s reciprocity norms. Second, community norm of reciprocity appears to be particularly effective under conditions of low online anonymity. While there is a strong norm of reciprocity in the online community, community designers should design artifacts that reduce the anonymity of members. Making the identity of community members visible to their peers can work as a self-monitoring mechanism which prevents members from free-riding behaviors that could jeopardize the online knowledge communities. Thus, online knowledge communities should be wary of the values and hazards of anonymity and strive to moderate the anonymity level of the online knowledge community in different situations. 6.3. Limitation and future work Results of this study must be interpreted in the context of its limitations. First, the use of cross-sectional data and SEM do not allow the possibility of bidirectional (feedback) effects to be explored. For instance, the effects of reciprocity on subsequent
perceptions of community norm by knowledge receivers has been recognized but cannot be examined. Future studies can collect longitudinal data to assess such bidirectional (feedback) effects. Second, our research model was empirically tested based on the responses of knowledge receivers from one online community in Asia, which belongs to the knowledge sharing type of community. Caution must be exercised when attempting to generalize the results across different types of online communities in varied contexts. Future studies can replicate this study in different type of online communities, e.g., transactional or fantasy communities to test the validity of the findings. Third, we found a negative interaction between perceived anonymity and community norm contrary to the prediction by the SIDE model. Perception of anonymity, while enhancing individuals’ self identity, may unexpectedly reduce individuals’ need for legitimacies, thereby reducing their proactiveness to conform to the reciprocity norm. Future study may need to further investigate the mechanisms behind this relationship. 7. Conclusion We developed a theoretical model based on equity theory and SIDE model to explain what factors affect the development of intention to reciprocate, and the influence of anonymity perception on online knowledge community members’ online reciprocity behaviors. It was tested through a survey of members of an online community for English learning. Overall, our survey data provides strong empirical support for the proposed relationships. An important finding is that in online knowledge communities perceived anonymity has important consequences with regard to members’ reciprocity intention. When members perceive themselves to be anonymous, they are more willing to reciprocate the favors received from the online community. However, when the community norm of reciprocity is salient, anonymity will counterintuitively lead to more free-riding behaviors. Therefore, anonymity and community norm not only have a direct effect, but also have an interactive effect on the reciprocity behaviors. Future studies of this direction can improve our understanding of reciprocity and knowledge contribution in online knowledge communities. Acknowledgements The work described in this paper was partially supported by a grant from the National Natural Science Foundation of China (Grant No.: 71532004) and the Natural Science Foundation of Guangdong Province (Grant No.: 2014A030310314). References Adams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental psychology (pp. 267e299). New York, NY: Academic Press. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179e211. Alpizar, A., Carlsson, F., & Johansson-Stenman, O. (2008). Anonymity, reciprocity, and conformity: evidence from voluntary contributions to a National Park in Costa Rica. Journal of Public Economic, 92, 1047e1060. Ardichvili, A., Page, V., & Wentiling, T. (2003). Motivation and barriers to participation in virtual knowledge-sharing communities of practice. Journal of Knowledge Management, 7(1), 64e77. Armstrong, A., & Hagel, J. (1996). The real value of online communities. Harvard Business Review, 74(3), 134e141. Ba, S., & Pavlou, P. A. (2002). Evidence of the effect of trust building technology in electronic markets: price premiums and buyer behavior. MIS Quarterly, 26(3), 243e268. Bock, G. W., Zmud, R. W., Kim, Y. G., & Lee, J. N. (2005). Behavioral intention formation in knowledge sharing: examining the roles of extrinsic motivators, social-psychological forces, and organizational climate. MIS Quarterly, 29(1), 87e111. Chester, A., & Gwynne, G. (1998). Online teaching: encouraging collaboration through anonymity. Journal of Computer-Mediated Communication, 4(2). To be
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349 found on-line at http://www.ascusc.org/jcmc/. Chin, W. W. (1998). The partial least squares approach to structural equation modeling. In G. A. Marcoulides (Ed.), Modern methods for business research (pp. 295e336). Mahwah, NJ: Lawrence Erlbaum Associates. Chiu, C.-M., Hsu, M.-H., & Wang, E. T. G. (2006). Understanding knowledge sharing in virtual communities: an integration of social capital and social cognitive theories. Decision Support Systems, 42(3), 1872e1888. Chiu, C.-M., Wang, E. T. G., Shih, F. J., & Fan, Y. W. (2011). Understanding knowledge sharing in virtual communities: an integration of expectancy disconfirmation and justice theories. Online Information Review, 35(1), 134e153. Cohen, R. L., & Greenberg, J. (1982). The justice concept in social psychology. In J. Greenberg, & R. L. Cohen (Eds.), Equity and justice in social behavior (pp. 2e42). New York, NY: Academic Press. Constant, D., Kiesler, S., & Sproull, L. (1994). What’s mine is ours or is it? a study of attitudes about information sharing. Information Systems Research, 5(4), 400e421. Cooper, J. (2007). Cognitive dissonance: Fifty years of a classic theory. Thousand Oaks, CA: Sage Publications. Falk, R. F., & Miller, N. B. (1992). A Primer for soft modeling. Akron,OH.,: University of Akron Press. Faraj, S., Jarvenpaa, S. L., & Majchrzak, A. (2011). Knowledge collaboration in online communities. Organization Science, 22(5), 1224e1239. €chter, S. (2000). Cooperation and punishment in public goods experFehr, E., & Ga iments. American Economic Review, 90, 980e994. Ferraro, F., Pfeffer, J., & Sutton, R. I. (2005). Economics language and assumptions: how theories can become self-fulfilling. Academy of Management Review, 30(1), 8e24. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39e50. Fulk, J. (1993). Social construction of communication technology. Academy of Management Journal, 36(5), 921e950. Glass, R. S., & Wood, W. A. (1996). Situational determinants of software piracy: an equity theory perspective. Journal of Business Ethics, 15, 1189e1198. Gouldner, A. W. (1960). The norm of reciprocity. American Sociological Review, 25, 165e178. Greenberg, M. S. (1980). A theory of indebtedness. In K. J. Gergen, M. S. Greenberg, & R. H. Wills (Eds.), Social exchange: Advances in theory and research. New York: Plenum. Greenberg, M. S. (1986). A preliminary statement on a theory of indebtedness. In Justice in social exchange (Symposium presented at the western psychological association, San Diego). Greenberg, M. S., & Shapiro, S. P. (1971). Indebtedness: an adverse aspect of asking for and receiving help. Sociometry, 34, 290e301. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5th ed.). Prentice-Hall. Harman, H. H. (1976). Modern factor analysis. Chicago, IL: University of Chicago Press. Hollingshead, A. B., Fulk, J., & Monge, P. (2002). Fostering intranet knowledge sharing: an integration of transactive memory and public goods approaches. Distributed Work, 335e355. Hsu, M. H., Ju, T. L., Yen, C. H., & Chang, C. M. (2007). Knowledge sharing behavior in virtual communities: the relationship between trust, self-efficacy, and outcome expectations. International Journal of Human Computer Studies, 65, 153e169. Joinson, A. (2001). Self-disclosure in computer-mediated communication: the role of self-awareness and visual anonymity. European Journal of Social Psychology, 31, 177e192. Kankanhalli, A., Tan, B. C. Y., & Wei, K. K. (2005). Contributing knowledge to electronic knowledge repositories: an empirical investigation. MIS Quarterly, 29(1), 113e143. Kiesler, S., Siegel, J., & Mcguire, J. W. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39, 1123e1134. Kolm, S. C. (2008). Reciprocity: An economics of social relations. Cambridge; New York: Cambridge University Press. Leary, M. R., & Kowalski, R. M. (1990). Impression management: a literature review and two-component model. Psychological Bulletin, 107(1), 34e47. Lea, M., & Spears, R. (1992). Paralanguage and social perception in computermediated communication. Journal of Organizational Computing, 2, 321e341. Lee, C. (1990). Modifying an American consumer behavior model for consumers in confucian culture: the case of the Fishbein behavioral intention model. Journal of International Consumer Marketing, 3(1), 27e50. Lin, M. J. J., Hung, S. W., & Chen, C. J. (2009). Fostering the determinants of
349
knowledge sharing in professional virtual communities. Computers in Human Behavior, 25, 929e939. Ma, M., & Agarwal, R. (2007). Through a glass darkly: information technology design, identity verification, and knowledge contribution in online communities. Information Systems Research, 18(1), 42e67. Mathews, M. A., & Green, J. D. (2009). Looking at me, appreciating you: self-focused attention distinguishes between gratitude and indebtedness. Cognition & Emotion, 24(4), 710e718. published on: 28 January 2009 (iFirst). Moore, G. C., & Benbasat, I. (1991). Development of an instrument to measure perceptions of adopting an information technology innovation. Information Systems Research, 2(3), 192e222. Nowak, M. A., & Sigmund, K. (2005). Evolution of indirectly reciprocity. Nature, 437(27), 1291e1298. Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill. Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory. New York: McGrawHill. Phang, C. W., Kankanhalli, A., & Sabherwal, R. (2009). Usability and Sociability in Online Communities: a comparative study of knowledge seeking and contribution. Journal of AIS, 10(10), 721e747. Pinsonneault, A., & Heppel, N. (1998). Anonymity in group support systems research: new conceptualization, measure, and contingency framework. Journal of Management Information Systems, 14(3), 89e108. Postmes, T., Spears, R., & Lea, M. (1998). Breaching or building social boundaries? SIDE-effects of computer-mediated communication. Communication Research, 25, 689e715. Postmes, T., Spears, R., Sakhel, K., & De Groot, D. (2001). Social influence in computer-mediated communication: the effects of anonymity on group behavior. Personality and Social Psychology Bulletin, 27, 1242e1254. Schmitz, J., & Fulk, J. (1991). Organizational colleagues, media richness, and electronic mail: a test of the social influence model of technology use. Communication Research, 18(4), 487e523. Scott, J. F. (1971). Internalization of norms: A sociological theory of moral commitment. Prentice-Hall. Sia, C. L., Tan, B. C. Y., & Wei, K. K. (2002). Group polarization and computermediated communication: effects of communication cues, social presence, and anonymity. Information Systems Research, 13(1), 70e90. Spears, R., & Lea, M. (1994). Panacea or panopticon? the hidden power in computermediated communication. Communication Research, 21, 427e459. Stone, E. F. (1978). Research methods in organizational behavior. Santa Monica, CA: Goodyear. Teo, H. H., Wei, K. K., & Benbasat, I. (2003). Predicting intention to adopt interorganizational linkages: an institutional perspective. MIS Quarterly, 27(1), 19e49. Walster, E., Berscheid, E., & Walster, G. W. (1973). New directions in equity research. Journal of Personality and Social Psychology, 151e176. Walther, J. B. (1996). Computer-mediated communication: impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23(1), 3e43. Wasko, M., Faraj, S., & Teigland, R. (2004). Collective action and knowledge contribution in electronic networks of practice. Journal of the Association for Information Systems, 5(December), 493e513. Wasko, M. M., & Faraj, S. (2000). ‘It is what one does’: why people participate and help others in electronic communities of practice. The Journal of Strategic Information Systems, 9(2e3), 55e173. Wasko, M. M., & Faraj, S. (2005). Why should i share? examining social capital and knowledge contribution in electronic networks of practice. MIS Quarterly, 29(1), 35e56. Watkins, P., Scheer, J., Ovnicek, M., & Kolts, R. (2006). The debt of gratitude: dissociating gratitude and indebtedness. Cognition and Emotion, 20(2), 217e241. Watson, S., & Hewett, K. (2006). A multi-theoretical model of knowledge transfer in organizations: determinants of knowledge contribution and knowledge reuse. Journal of Management Studies, 43(2), 141e173. Westphal, J. D., & Zajac, E. J. (1997). Defections from the inner circle: social exchange, reciprocity, and the diffusion of board independence in U.S. corporations. Administrative Science Quarterly, 42, 161e183. Wiertz, C., & Ruyter, K. (2007). Beyond the call of duty: why customers contribute to firm-hosted commercial online communities. Organization Studies, 28, 347e376. Yee, N., Bailenson, J. N., Urbanek, M., Chang, F., & Merget, D. (2007). The unbearable likeness of being digital: the persistence of nonverbal social norms in online virtual environments. Cyberpsychology & Behavior, 10, 115e121. Ye, H. J., Feng, Y., & Choi, B. C. F. (2015). Understanding knowledge contribution in online knowledge communities: a model of community support and forum leader support. Electronic Commerce Research and Applications, 14, 34e45.
Contents lists available at ScienceDirect
Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh
Full length article
Why do you return the favor in online knowledge communities? A study of the motivations of reciprocity Hua (Jonathan) Ye b, *, 1, Yuanyue Feng a, 1 a b
Department of Management Science, College of Management, Shenzhen University, No. 3688, Nanhai Ave, Shenzhen, Guangdong, 518060, China School of Management, Harbin Institute of Technology, 13 Fayuan Street, Nangang Dist., Harbin, 150001, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 June 2015 Accepted 3 May 2016 Available online 26 May 2016
Online knowledge community administrators are attempting to encourage their users to contribute knowledge in order to provide value to members and maintain sustainability. A large number of online knowledge communities fail mainly due to the reluctance of users to return the favor and share knowledge. Many studies on this topic have highlighted the importance of reciprocity for knowledge contribution which forms a virtuous feedback loop for the community sustainability. However, it is unclear how reciprocity is developed and what influences its development. Motivated by this, this study focuses on investigating the antecedents of knowledge receivers’ reciprocity in online knowledge communities. It formulates and tests a theoretical model to explain reciprocity behavior of community members based on equity theory and Social Identity explanation of De-individuation Effects (SIDE) model. Our proposed model is validated through a large-scale survey in an online forum for English learning. Results reveal that indebtedness and community norm not only are key antecedents of intention to reciprocate but are also positively related to each other. The perceived anonymity of the online knowledge community not only has a positive effect on intention to reciprocate, but also has an interactive effect with community norm on intention to reciprocate. Theoretical and practical implications of this study are discussed. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Online knowledge community Reciprocity Indebtedness Community norm Perceived anonymity
1. Introduction Online knowledge communities comprise of individuals with common interests, goals, or practices, who share and combine knowledge for their own benefits and engage in social or personal interactions (Ye, Feng, & Choi, 2015). They serve not only as sources of information, social support, and recreation, but also as a platform for knowledge exchange (Armstrong & Hagel, 1996; Phang, Kankanhalli, & Sabherwal, 2009). Interactions and knowledge embedded in online knowledge communities are the key factors to allow them to survive and thrive (Wasko & Faraj, 2005). Online knowledge community administrators are hence attempting to encourage their users to contribute knowledge and resources in order to provide value to members and ensure sustainability (Phang et al., 2009). Motivated by this, researchers have been
* Corresponding author. E-mail addresses: [email protected] (H. Ye), [email protected] (Y. Feng). 1 Both authors equally contribute to this paper and are listed in an alphabetical order. http://dx.doi.org/10.1016/j.chb.2016.05.007 0747-5632/© 2016 Elsevier Ltd. All rights reserved.
investigating the drivers of online community members’ knowledge contribution behaviors. They conclude that various factors motivate the behavior of online knowledge contribution, such as the anticipation of extrinsic benefits (organizational rewards, reputation) (e.g., Kankanhalli, Tan, & Wei, 2005; Wasko & Faraj, 2005), intrinsic benefits (sense of self worth, sense of belongingness, and social affiliation) (e.g., Bock, Zmud, Kim, & Lee, 2005), and social capital (social interaction ties, trust, norm of reciprocity, perceived identity, and shared language) (e.g., Chiu, Hsu, & Wang, 2006; Wasko & Faraj, 2005). Even in the absence of organizational rewards, a key motivator found in these studies is the idea of reciprocity (e.g., Ardichvili, Page, & Wentiling, 2003; Chiu et al., 2006; Wasko & Faraj, 2000, 2005). This is because reciprocity enables the formation of a virtuous feedback loop to knowledge contribution and thus the community sustainability. Constructs like subjective norms or pro-sharing norms have been identified to explain how receivers’ reciprocity behavior is regulated (e.g., Bock et al., 2005). In order to understand reciprocity behaviors, researchers (e.g., Mathews & Green, 2009; Nowak & Sigmund, 2005) propose that
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
besides being driven by incentives, reciprocity may also derive from a desire to repay the favor or knowledge received from the community before as explained by equity theory. Such a desire to repay tends to exist in those who frequently obtained necessary information from the communities and learned skills for their tasks (Wasko & Faraj, 2000). This desire may derive from the feeling of indebtedness (Kolm, 2008), or from community norm (Wasko & Faraj, 2005). However, no empirical research has systematically examined the effects of indebtedness and community norm on reciprocity behaviors. Further, online reciprocity behaviors may be driven by certain unique factors since online knowledge communities differ notably from conventional organizations (Chiu et al., 2006). One of the salient differences is the anonymity of online knowledge communities (Wasko, Faraj, & Teigland, 2004). In online knowledge communities, users can interact anonymously and indirectly (Fehr and €chter 2000). They can easily remain anonymous or change their Ga identities (Ba & Pavlou, 2002), since most forum websites identify users by e-mail addresses, which can be readily obtained from multiple sources. The anonymity of members in the online community makes it more likely that individual interactions will go unnoticed by other network members (Wasko et al., 2004). This may lead individuals to have different psychological responses to social interactions (Faraj, Jarvenpaa, & Majchrzak, 2011; Pinsonneault & Heppel, 1998). Such differences require researchers to investigate the influence of anonymity on online reciprocity. Although prior studies have provided clues about the possibility of anonymity’s influence on reciprocity (e.g., Alpizar, Carlsson, & Johansson-Stenman, 2008; Kolm, 2008) in the offline context, there is no empirical study to test the relationship in the online context. Without knowing the effects of anonymity, there will be a gap in our understanding of what affects reciprocity behaviors in online knowledge communities. With the above practical and theoretical motives, we are interested to study the antecedents of reciprocity in online knowledge communities from knowledge receivers’ perspective. Based on the equity theory and Social Identity explanation of De-individuation Effects (SIDE) model, this study develops a model to explain the effects of perceived anonymity, community norm, and indebtedness on knowledge receivers’ intention to reciprocate in online knowledge communities. A survey was conducted in an English learning forum to test the model. The study expects to contribute to the existing literature in following ways. First, it helps improve our understanding of online reciprocity behaviors by adopting new theoretical lenses, i.e., equity theory and SIDE model. Second, it models and tests the antecedents of online reciprocity through a large-scale survey. Third, it generates new insights about online knowledge contribution from the perspective of knowledge receivers’ reciprocity behaviors.
343
used to explain social interaction via computer-mediated communication, which fits well with the context of our study. 2.1. Equity theory Equity theory focuses upon an individual’s perception and request of fairness or equity with respect to a relationship (Cohen & Greenberg, 1982). During a social exchange, an individual assesses the ratio of what is output from the relationship to what is input in the relationship, and also the ratio of what the other person in the relationship output from the relationship to what is input into the relationship. Equity theory posits that an equitable relationship exists when individuals perceive that they are receiving equal relative outcomes from the exchange compared with their inputs (Adams, 1965; Watkins, Scheer, Ovnicek, & Kolts, 2006). That is, whether they are receiving a fair return for the efforts or resources that they put into the exchange (Glass & Wood, 1996). A perception of inequity in an exchange results in the feeling of indebtedness (Gouldner, 1960). Such a feeling motivates individuals to commit to a reciprocal behavior in order to avoid being perceived as socially insensitive (Mathews & Green, 2009). For example, in Glass and Wood (1996)’s software piracy study, the debt perceived to be owed to others from a prior exchange are identified as the main factor for an individual’s intention to provide an illegal software copy to others. Therefore, the more inequitable the relationship, the more indebted the participants will feel and the greater they will be motivated to reduce the inequity (Greenberg, 1986). There are two ways that an individual can restore equity in an inequitable relationship. First, the individual can restore “actual equity” by appropriately altering his own outputs or inputs in the exchange. Second, the individual can restore psychological equity by appropriately distorting perceptions of his or her own outputs and inputs compared with other participants’ (Walster, Berscheid, & Walster, 1973), or by reducing the importance of the inequity (Watkins et al., 2006). In the context of online knowledge contribution, the input and output of knowledge exchange in online communities can be considered as the knowledge contributed and the benefits received from the exchange (Chiu et al., 2006). When the benefits received from the exchange (outputs) exceed the perceived value of knowledge contributed (input), knowledge receivers will perceive inequity and feel indebted to the community (i.e., indebtedness). They will generate a desire to reciprocate the community by contributing knowledge with higher quality and quantity, so as to restore the inequity inside (Wiertz & Ruyter, 2007). Therefore, indebtedness is expected to affect individuals’ reciprocity intention and behaviors in online knowledge communities. Accordingly, we include indebtedness into our model. 2.2. Social identity model of de-individuation effects (SIDE Model)
2. Theoretical foundations Previous theories that have been used to study knowledge contribution motives include motivation theory (Bock et al., 2005; Chiu, Wang, Shih, & Fan, 2011), social exchange theory (Kankanhalli et al., 2005; Wasko & Faraj, 2005; Watson & Hewett, 2006), social capital and social cognitive theories (Chiu et al., 2006; Hsu, Ju, Yen, & Chang, 2007; Lin, Hung, & Chen, 2009), and public goods theory (Hollingshead, Fulk, & Monge, 2002; Wasko & Faraj, 2000). These theories highlight reciprocity as a motive for knowledge contribution but do not investigate the antecedents of reciprocity. For this reason, equity theory has been adopted in this study as the theoretical foundation to investigate the antecedents of reciprocity. Another theoretical foundation for this study is Social Identity explanation of De-individuation Effects (SIDE) model. SIDE model is
SIDE model suggests that anonymity changes the relative salience of personal and social identity, and thereby having a profound effect on group behavior (Spears & Lea, 1994). Anonymity in an online community obscures individual features and interpersonal differences, and hence enhances the salience of social identity. It thereby depersonalizes social perceptions of others and the self (Postmes, Spears, Sakhel, & De Groot, 2001). This decreased visibility of the individual within anonymous groups results in the accentuation of the depersonalization process and amplification of cognitive efforts to perceive the group as an entity (Postmes, Spears, & Lea, 1998). In online knowledge communities, the de-individuating features (e.g., perceived anonymity, physical isolation, and selective self-presentation) decrease perceptions of individual differences.
344
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
This decreased perception of individual differences leads individuals to focus less on their self-conception and hence to increase their adherence to community norms and form more positive impression about the online knowledge communities (Walther, 1996). If paired with high group norm salience, these results become more salient. A strong community norm indicates strong expectations from peer members in the community. It includes the expectation of reciprocity, equity, and mutual support between members in the exchange (Wasko & Faraj, 2005). Behaviors conflicting with equity expectation are considered illegitimate and are not encouraged in the community (Yee, Bailenson, Urbanek, Chang, & Merget, 2007). In online knowledge communities, in order to avoid being perceived alien and keep a positive self-concept, community members tend to conform to the norms and hence conduct reciprocity behaviors. Therefore, a strong community norm will encourage members to conform to the expectation of their peers and hence to reciprocate. Based on the above reasoning, we include perceived anonymity and community norm in our model. 3. Research model and hypotheses Reciprocity is defined as the propensity among at least two cooperative partners to repay the favor that the partner has received before (Kolm, 2008). In online settings, reciprocity tends to be generalized (Wasko & Faraj, 2000). Generalized reciprocity refers to the propensity to reciprocate another’s action, not by directly rewarding the benefactor, but by benefiting another actor implicated with the benefactors in a social exchange (Westphal & Zajac, 1997). Following Kolm (2008) and Westphal and Zajac (1997), we define our dependent variable, i.e., intention to reciprocate, as the intention of beneficiary to return help to the benefactors or those who are in the benefactors’ group. Based on the two theoretical perspectives discussed, we propose the research model for this study (see Fig. 1). Our dependent variable intention to reciprocate is expected to be influenced by community norm, indebtedness, and perceived anonymity. Furthermore, community norm is expected to influence indebtedness. Perceived anonymity is also proposed to moderate the relationship between community norm and intention to reciprocate. 3.1. Community norm Community norm is defined as a behavioral regularity (Yee et al., 2007), based on a socially shared belief of how one ought to behave. It refers to a rule for what should be done, accepted, and internalized by the group members (Scott, 1971). It triggers the enforcement of the prescribed behavior by informal social sanctions (Fehr and G€ achter, 2000) Previous literature proposes that through social influence
processes (Fulk, 1993; Schmitz & Fulk, 1991), community norms can have an important influence on attitude and intention (Ajzen, 1991). According to the SIDE model, the stronger the community norms are, the more individuals are motivated to conform to them, and the more their attitudes and behaviors tend to be communitydetermined than individual-determined (Lee, 1990). When reciprocity behaviors are regarded as a salient community norm in a community, individuals in this community are more likely to be aware of this community norm and the expectations of significant others for reciprocity behaviors. Such awareness prompts them to comply with these expectations (Bock et al., 2005) because conformity to the community norm renders them legitimacy in the community. Thus, a higher level of community norm of reciprocity can lead to a higher intention to reciprocate towards the community. In online knowledge communities, similarly, when an individual perceives a high level of community norm of reciprocal knowledge sharing behaviors, the individual may unconsciously or consciously feel that reciprocal knowledge sharing are expected by others in the online knowledge community. In view of that, the individual may perceive a need to follow such community norm so as to gain legitimacy in the community. Thus, the individual would intend to engage in reciprocal knowledge contribution to the community. H1. Community norm is positively related to the intention to reciprocate in online knowledge communities.
3.2. Indebtedness As a psychological construct, indebtedness has been defined as “a state of obligation to repay another” in the context of receiving a benefit from another (Greenberg, 1980, p. 4). The state of indebtedness constitutes a violation of the sense of “ought” and therefore involves attendant feelings of guilt analogous to the feelings presumed to accompany distributive injustice or inequity (Cohen & Greenberg, 1982). It is an emotional state of “arousal and discomfort”, where the individual is alert to opportunities to reduce this discomfort (Greenberg & Shapiro, 1971). In online knowledge communities, when members receive knowledge or favor from other members, they compare it with their own input into the online knowledge community. After comparison, if they find an inequity in the exchange, they will feel indebtedness because this is in conflict with their own positive impression and self-perception (Constant, Kiesler, & Sproull, 1994). Equity theory suggests that a desire to reduce such a feeling of indebtedness will motivate individuals to reciprocate other members in the community with their knowledge in order to avoid being perceived as socially insensitive (Mathews & Green, 2009) and to maintain their self perception. Hence we hypothesize: H2. Indebtedness is positively related to intention to reciprocate in online knowledge communities.
Community Norm H3
H1 H2
Intention to Reciprocate
Indebtedness H5
H4
Perceived Anonymity Fig. 1. Research model.
Based on the SIDE model, a strong community norm of reciprocal knowledge sharing implies a strong expectation for members to return the help or knowledge from the community and a high pressure to get legitimized. With increasing expectations from the community, the feeling of indebtedness will be enhanced (Watkins et al., 2006). Under the condition of salient community norm, individuals tend not to distort perceptions of benefits and appreciate the value of knowledge contributed in the exchange comparing with other participants’ (Walster et al., 1973), such as by reducing the importance of inequity (Cooper, 2007). Therefore, this strong community norm enhances individuals’ feeling of indebtedness. H3. Community norm is positively related to indebtedness in online
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
knowledge communities.
3.3. Perceived anonymity Anonymity refers to the extent to which community members cannot identify other members or the source of a particular contribution to the group (Pinsonneault & Heppel, 1998). In online knowledge communities, the richness of normal face-to-face interaction and self-regulation (Kiesler, Siegel, & Mcguire, 1984) is absent. Anonymity of others leads to heightened self-awareness, and thus to greater adherence to group norms when a social identity is salient (Joinson, 2001). The heightened self-awareness leads members to more focus on their own feelings and behaviors in the community. This could lead to pro-social behaviors online to self-fulfill their own perception about what they want to be (Ferraro, Pfeffer, & Sutton, 2005). Moreover, anonymity enhances members’ participation in the discussion (Sia, Tan, & Wei, 2002). For example, anonymity has been found to increase equity and participation rates in e-learning (Chester & Gwynne, 1998). This positive effect of anonymity is explained by Walther (1996). Drawing on the SIDE model developed by Lea and Spears (1992), Walther proposed that the de-individuating features of online communities (visual anonymity, physical isolation, and selective self-presentation) lead to decreased perceptions of individual differences, increased adherence to group norms, and more positive impression formation. In order to sustain positive impression formation, people tend to be more pro-social and helpful to others (Leary & Kowalski, 1990). Thus, during the knowledge exchange in online knowledge communities, members who perceive people as anonymous tend to be active in reciprocating the community to maintain their self perception. H4. Perceived anonymity is positively related to the intention to reciprocate in online knowledge communities. The perception of anonymity may have a moderating effect on the relationship between community norm and reciprocity in online knowledge communities (Ba & Pavlou, 2002; Spears & Lea, 1994). According to the SIDE model (Lea and Spears 1992; Spears & Lea, 1994), with anonymity inherent in most online knowledge communities, social identity becomes salient. This serves to strengthen the impact of social norms, and hence normative influence. When a social identity is salient, through a heightened selfawareness, anonymity increases members’ intent to maintain a good self concept, thereby leading to their greater adherence to group norms (Joinson, 2001). After receiving favors or benefits, anonymous members want to reciprocate especially when the community norm is strong. Therefore, we hypothesize that: H5. The positive relationship between community norm and the intention to reciprocate in online knowledge communities will be stronger when perceived anonymity is high. We also include age, gender, forum tenure, education level, previous experience of being helped, and posting frequency in our model as control variables that might affect reciprocity in online knowledge communities. 4. Research methodology The research model was empirically tested by the data collected from an English learning forum (bbs.gter.net) through survey methodology. Most of the constructs in our theoretical model are latent variables, which are best studied using the survey approach (Nunnally & Bernstein, 1994). Besides, the survey methodology provides a basis for establishing generalizability, allows
345
replicability, and has statistical power (Teo, Wei, & Benbasat, 2003). 4.1. Survey administration We collect our survey data from bbs.gter.net, an online student discussion forum for language learning, particularly English learning. Individuals post their questions related to English learning or literature translation. There were an estimated 5 million registered users and 4 thousand online users per day with about 2 thousand posting per day in this online knowledge community.2 We posted the questionnaire in bbs.gter.net for one week with the help of forum administrators to highlight the survey invitation. An invitation letter is also delivered through a well known Southeast Asian university to the international students who have registered an account in this forum. In appreciation of the respondents’ effort, we promised a lucky draw with 2 prizes of $100 each for the respondents. Overall, we received 180 responses from registered users of this online forum. Due to attrition and missing values, the resulting sample size used in the analyses is 169. Of the 169 respondents, the proportion of males exceeded female respondents (57.9% are males and 42.1% are females), with the majority of the respondents aged between 20 and 26 years old (86.3%). On average, 91.8% respondents had internet experience of more than 4 years. The respondents had an average of 1.94 years of experience using the forum for knowledge sharing. 4.2. Operationalization of constructs The survey items were generated based on a review of the relevant information systems (IS), knowledge management, and reciprocity literature. Where previously tested measures were not available, we developed items based on the construct definition and description. Table 1 provides formal definitions of the constructs. Where available, the constructs are measured using instruments adapted from past studies to enhance validity (Stone, 1978). Otherwise, new instruments are developed based on a review of the knowledge management, information systems and reciprocity literature. All the items are measured by using seven-point scales anchored from “strongly disagree” to “strongly agree” (see Table 2 for the survey items for each construct), unless otherwise indicated. Given that the items for measuring the constructs were adapted from various sources or developed for this study, all of the items were subjected to a two-stage conceptual validation exercise based on procedures prescribed by Moore and Benbasat (1991). All items have an acceptable level of agreement among sorters (Cohen’s Kappa > 0.9, hit rate > 0.95). We also develop items for all the control variables (see Table 3). 5. Data analysis For this study, structural equation modelling (SEM) analysis was chosen over regression analysis, because SEM can simultaneously analyze all of the paths in one analysis (Chin, 1998). Within SEM, PLS was chosen because the research is exploratory in nature and the model proposed is multi-stage (Ma & Agarwal, 2007). This study aims to explore the antecedents of reciprocity intention in online knowledge communities. Therefore, we used Partial Least Squares (PLS) as the main statistical technique to analyze the data. Constructs were modelled using reflective indicators according to Chin (1998). SmartPLS 2.0 (M3 Release) was used to run our PLS
2
http://bbs.gter.net/bbs/.
346
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
Table 1 Formal definitions of constructs. Construct (abbreviation)
Definition
Source
Intention to Reciprocate (RECP) Indebtedness (INDEB) Community Norm (NORM) Perceived Anonymity (ANY)
The intention of beneficiary to return help to the benefactors or those who are in the benefactors’ group
Kolm (2008)
A state of obligation to repay another in the context of receiving a benefit from another A behavioral regularity based on a socially shared belief of how one ought to behave The extent to which community members cannot identify other members or the source of a particular contribution to the group
Greenberg (1980) Yee et al. (2007) Pinsonneault and Heppel (1998)
Table 2 Operationalization of constructs. Construct
Items
Intention to Reciprocate RECP1 (RECP) RECP2 RECP3
Indebtedness (INDEB)
Community Norm (NORM) Perceived Anonymity (ANY)
RECP4 INDEB1 INDEB2 INDEB3 NORM1 NORM2 NORM3 ANY1 ANY2 ANY3
Source When receiving help from this community, I will return the help to the community When obtaining experience from this community, I will share my own experiences later with this community When receiving suggestions about my situation from this community, I will contribute my suggestions to others When receiving support from this community, I will support others in this community When I receive help from the community, I feel that I owe something to the community When I receive help from the community, I feel guilty if I cannot repay the community When I receive help from the community, I feel indebted to the community When receiving help from the community, others expect receivers to return the help There is an atmosphere in this community to return others’ help Receivers do not live up to others’ expectation if they do not return the help Others can identify my topics and comments (Reverse) The community is large enough that nobody can trace topics and comments back to their authors Others can recognize my topics and comments (Reverse)
Self-developed
Greenberg and Shapiro (1971); Watkins et al. (2006) Bock et al. (2005)
Pinsonneault and Heppel (1998)
Table 3 Control variables. Control Variable (abbreviation)
Survey items
Age (AGE) Gender (GENDER) Forum Tenure (TENURE) Education Level (EDU) Previously being helped (PREHELPED) User Posting (POSTING)
Please provide your age (in years). Male/Female How long have you been using this forum (in years)? (1: Bachelor and below; 2: Master; 3: PhD) (0: Previously not being helped; 1: Previously being helped) Please indicate your number of posting per week in this forum
analysis. 5.1. Measurement model To validate the measurement model, reliability (Nunnally, 1978), discriminant validity (Chin, 1998) and convergent validity (Fornell & Larcker, 1981) were assessed for the reflective indicators (Hair, Anderson, Tatham, & Black, 1998). All reflective constructs in our model exhibited acceptable levels of reliability, convergent validity (see Table 5), and discriminant validity (see Tables 4 and 5). The single method test of Harman (1976) was used to test for common method variance. The factor analyses produced neither a single factor nor one general factor that accounted for the majority of the variance (>50%). Each factor accounted for more than the viable cut-off of 5% (see Table 3). Therefore, this method indicated that common method bias was not a problem. 5.2. Structural model results The results of the data analysis are shown in Fig. 2 and Table 6. As seen from Fig. 2 below, our structural model could explain 42% of the total variance in intention to reciprocate and 43% of indebtedness which is greater than the acceptable threshold of 10 percent (Falk & Miller, 1992). Fig. 2 also summarizes the path coefficients and their significance from bootstrapping resampling.
Table 4 Item loadings. Items
RECP1 RECP2 RECP3 RECP4 INDEB1 INDEB2 INDEB3 ANY1 ANY2 ANY3 NORM1 NORM2 NORM3 Eigen value % of variance Cumulative %
Component 1
2
3
4
0.92 0.79 0.84 0.84 0.12 0.01 0.17 0.14 0.12 0.12 0.05 0.23 0.19 5.51 39.37 39.37
0.12 0.14 0.09 0.12 0.87 0.84 0.83 0.08 0.17 0.09 0.40 0.22 0.17 2.15 15.37 54.74
0.06 0.19 0.09 0.10 0.14 0.17 0.05 0.78 0.86 0.94 0.16 0.00 0.08 1.45 10.37 65.11
0.08 0.13 0.14 0.18 0.16 0.23 0.16 0.34 0.07 0.03 0.68 0.78 0.83 1.20 8.62 73.74
Demographic variables (i.e., age, gender, forum tenure, education level, previously being helped, and user posting) are included in the analysis as controls for reciprocity. Among all the controls, forum tenure is found to positively affect intention to reciprocate. We found adequate support for our research model in this study.
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
347
Table 5 Descriptive statistics, construct correlation matrix. Construct
Mean
RECP INDEB NORM ANY
Cronbach’s alpha
5.79 4.32 4.81 3.22
Composite reliability
0.90 0.87 0.78 0.85
0.93 0.91 0.87 0.91
SD
1.20 1.50 1.26 1.35
Correlations 1
2
3
4
0.87 0.40 0.38 0.29
0.84 0.56 0.26
0.83 0.29
0.88
Note: Diagonal elements are the square root of AVE.
Community Norm H3 0.65**
Indebtedness
H1 0.28** H2 0.11*
Intention to Reciprocate
H5 -0.32***
R2=.42
0.22** H4
R2=.43
Perceived Anonymity Fig. 2. Graphical display of results (*p < 0.05,
**
p < 0.01,
***
p < 0.001).
All paths were significant at the 0.05 level with three paths significant at the 0.01 level. As predicted, perceived anonymity, community norm, and indebtedness were positively related to members’ intention to reciprocate in online knowledge communities (H1, H2, and H4 were supported). Community norm were significantly positively related to indebtedness (H3 was supported). Contrary to our prediction, there was a negative interaction effect of perceived anonymity and community norm on the intention to reciprocate (H5 was not supported). 6. Discussion and implications In online knowledge communities, individuals usually search for knowledge to fulfill their needs (Ye et al., 2015). After they fulfilled their knowledge needs, it was thought that individuals will leaf until they have new needs. Surprisingly, studies found that individuals return the favor to the community by helping others (Wasko & Faraj, 2005; Ye et al., 2015). This reciprocal behavior has been believed to be critical to the survival and sustainability of
online knowledge communities. Researchers and practitioners are interested in how to promote such a behavior. Considering the importance of reciprocity in engaging community members in knowledge contribution and hence the sustainability of online knowledge communities, we study the antecedents of reciprocity in online knowledge communities based on equity theory and the SIDE model. In line with our hypotheses, results show that community norm, indebtedness, and perceived anonymity affect members’ intention to reciprocate. Also, community norm positively influence members’ feeling of indebtedness. In contrast to our hypothesis, perceived anonymity has a negative interactive effect with community norm on members’ intention to reciprocate (H5 not supported). Under a condition of high perceived anonymity, the positive effect of community norm on members’ intention to reciprocate will be weakened, rather than being strengthened. A plausible explanation for this unexpected result may be that, as members perceive their real identities to be unseen by others in the online knowledge community, they might be less concerned with or care about their legitimacies in the forum, thereby being less motivated to conform to the community norm of reciprocity. This could be because individuals’ self awareness was not heightened and individuals are more likely to free ride (Spears & Lea, 1994). Hence, their intention to comply with the community norm would be undermined in this situation. Another possible reason could be due to the difference of online communities from physical organizations where online communities have no formal regulations and reward mechanisms (Faraj et al., 2011). The reciprocal behavior in online communities is purely voluntary and may not be constrained by such regulations and reward mechanisms. 6.1. Theoretical implications This study contributes to the literature in several ways. First, our research is novel in examining the antecedents of reciprocity behavior in the online context. Reciprocal contribution of knowledge by members has been emphasized as a critical factor for the
Table 6 Hypotheses testing. DV ¼ Intention to reciprocate
Age Gender Forum tenure Education level Previously being helped User posting Community norm Indebtedness Perceived anonymity Community norm * Perceived anonymity R2 *
p < 0.05;
**
p < 0.01;
***
p < 0.001.
DV ¼ Indebtedness
1
2
Result
0.07 0.01 0.32* 0.10 0.01 0.09
0.03 0.02 0.21* 0.07 0.01 0.08 0.28* 0.11* 0.22* 0.32*** 0.42
N.S. N.S. Sig. N.S. N.S. N.S. H1 Supported H2 Supported H4 Supported H5 not Supported
0.09
3
Result
0.65***
H3 Supported
0.43
348
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349
sustainability and growth of online knowledge communities (Wasko & Faraj, 2005; Watson & Hewett, 2006; Ye et al., 2015). Through the exploration and examination of the motives for reciprocal knowledge contribution, this study advances theoretical development in the area of knowledge contribution in general and reciprocity behavior in online knowledge communities in particular. Second, unlike most of the prior knowledge contribution studies, our research adopts new theoretical lenses to examine the antecedents of online reciprocal knowledge contribution. These new theoretical lenses, i.e., equity theory and SIDE model, provide additional insights into reciprocity behaviors in the online context. On one hand, this study echoes to and advances the equity theory by discovering the positive effect of indebtedness on members’ intention to reciprocate in online knowledge communities. Consistent with prior literature on reciprocity in offline context (Glass & Wood, 1996; Mathews & Green, 2009), we find significant relationship between the intrinsic factor, i.e., feeling of indebtedness, and reciprocity intention in online context. This finding suggests that in the online knowledge communities, knowledge receivers are likely to believe that they are receiving help from the community as a whole rather than from a specific individual. Thereby they tend to reciprocate by contributing their knowledge to the online knowledge community so as to repay the favors received from the community. On the other hand, this study enriches the SIDE model by examining its applicability to the context of online knowledge sharing. In accord with the theoretical essence of the SIDE model, community norm of reciprocity fosters the intention to reciprocate by forum members. A strong community norm indicates a strong peer expectation and individuals tend to conform to this expectation in order to be legitimate and be liked by others. In addition, perceived anonymity by members increases their self-awareness and hence their intention to maintain a good self concept, which in turn motivates them to engage in reciprocity. 6.2. Practical implications Collectively, results of this study indicate the circumstances under which mechanisms to promote reciprocity intention and behavior in online knowledge communities may be more effective. First, online knowledge communities can raise the community norm of reciprocity among members by rewarding the reciprocity behaviors and announcing this to the whole community. This can be done by highlighting reciprocal postings or adding reputation scores for those who reciprocate. Enhancing communication channels between community members can also strengthen the community’s reciprocity norms. Second, community norm of reciprocity appears to be particularly effective under conditions of low online anonymity. While there is a strong norm of reciprocity in the online community, community designers should design artifacts that reduce the anonymity of members. Making the identity of community members visible to their peers can work as a self-monitoring mechanism which prevents members from free-riding behaviors that could jeopardize the online knowledge communities. Thus, online knowledge communities should be wary of the values and hazards of anonymity and strive to moderate the anonymity level of the online knowledge community in different situations. 6.3. Limitation and future work Results of this study must be interpreted in the context of its limitations. First, the use of cross-sectional data and SEM do not allow the possibility of bidirectional (feedback) effects to be explored. For instance, the effects of reciprocity on subsequent
perceptions of community norm by knowledge receivers has been recognized but cannot be examined. Future studies can collect longitudinal data to assess such bidirectional (feedback) effects. Second, our research model was empirically tested based on the responses of knowledge receivers from one online community in Asia, which belongs to the knowledge sharing type of community. Caution must be exercised when attempting to generalize the results across different types of online communities in varied contexts. Future studies can replicate this study in different type of online communities, e.g., transactional or fantasy communities to test the validity of the findings. Third, we found a negative interaction between perceived anonymity and community norm contrary to the prediction by the SIDE model. Perception of anonymity, while enhancing individuals’ self identity, may unexpectedly reduce individuals’ need for legitimacies, thereby reducing their proactiveness to conform to the reciprocity norm. Future study may need to further investigate the mechanisms behind this relationship. 7. Conclusion We developed a theoretical model based on equity theory and SIDE model to explain what factors affect the development of intention to reciprocate, and the influence of anonymity perception on online knowledge community members’ online reciprocity behaviors. It was tested through a survey of members of an online community for English learning. Overall, our survey data provides strong empirical support for the proposed relationships. An important finding is that in online knowledge communities perceived anonymity has important consequences with regard to members’ reciprocity intention. When members perceive themselves to be anonymous, they are more willing to reciprocate the favors received from the online community. However, when the community norm of reciprocity is salient, anonymity will counterintuitively lead to more free-riding behaviors. Therefore, anonymity and community norm not only have a direct effect, but also have an interactive effect on the reciprocity behaviors. Future studies of this direction can improve our understanding of reciprocity and knowledge contribution in online knowledge communities. Acknowledgements The work described in this paper was partially supported by a grant from the National Natural Science Foundation of China (Grant No.: 71532004) and the Natural Science Foundation of Guangdong Province (Grant No.: 2014A030310314). References Adams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental psychology (pp. 267e299). New York, NY: Academic Press. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179e211. Alpizar, A., Carlsson, F., & Johansson-Stenman, O. (2008). Anonymity, reciprocity, and conformity: evidence from voluntary contributions to a National Park in Costa Rica. Journal of Public Economic, 92, 1047e1060. Ardichvili, A., Page, V., & Wentiling, T. (2003). Motivation and barriers to participation in virtual knowledge-sharing communities of practice. Journal of Knowledge Management, 7(1), 64e77. Armstrong, A., & Hagel, J. (1996). The real value of online communities. Harvard Business Review, 74(3), 134e141. Ba, S., & Pavlou, P. A. (2002). Evidence of the effect of trust building technology in electronic markets: price premiums and buyer behavior. MIS Quarterly, 26(3), 243e268. Bock, G. W., Zmud, R. W., Kim, Y. G., & Lee, J. N. (2005). Behavioral intention formation in knowledge sharing: examining the roles of extrinsic motivators, social-psychological forces, and organizational climate. MIS Quarterly, 29(1), 87e111. Chester, A., & Gwynne, G. (1998). Online teaching: encouraging collaboration through anonymity. Journal of Computer-Mediated Communication, 4(2). To be
H. Ye, Y. Feng / Computers in Human Behavior 63 (2016) 342e349 found on-line at http://www.ascusc.org/jcmc/. Chin, W. W. (1998). The partial least squares approach to structural equation modeling. In G. A. Marcoulides (Ed.), Modern methods for business research (pp. 295e336). Mahwah, NJ: Lawrence Erlbaum Associates. Chiu, C.-M., Hsu, M.-H., & Wang, E. T. G. (2006). Understanding knowledge sharing in virtual communities: an integration of social capital and social cognitive theories. Decision Support Systems, 42(3), 1872e1888. Chiu, C.-M., Wang, E. T. G., Shih, F. J., & Fan, Y. W. (2011). Understanding knowledge sharing in virtual communities: an integration of expectancy disconfirmation and justice theories. Online Information Review, 35(1), 134e153. Cohen, R. L., & Greenberg, J. (1982). The justice concept in social psychology. In J. Greenberg, & R. L. Cohen (Eds.), Equity and justice in social behavior (pp. 2e42). New York, NY: Academic Press. Constant, D., Kiesler, S., & Sproull, L. (1994). What’s mine is ours or is it? a study of attitudes about information sharing. Information Systems Research, 5(4), 400e421. Cooper, J. (2007). Cognitive dissonance: Fifty years of a classic theory. Thousand Oaks, CA: Sage Publications. Falk, R. F., & Miller, N. B. (1992). A Primer for soft modeling. Akron,OH.,: University of Akron Press. Faraj, S., Jarvenpaa, S. L., & Majchrzak, A. (2011). Knowledge collaboration in online communities. Organization Science, 22(5), 1224e1239. €chter, S. (2000). Cooperation and punishment in public goods experFehr, E., & Ga iments. American Economic Review, 90, 980e994. Ferraro, F., Pfeffer, J., & Sutton, R. I. (2005). Economics language and assumptions: how theories can become self-fulfilling. Academy of Management Review, 30(1), 8e24. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39e50. Fulk, J. (1993). Social construction of communication technology. Academy of Management Journal, 36(5), 921e950. Glass, R. S., & Wood, W. A. (1996). Situational determinants of software piracy: an equity theory perspective. Journal of Business Ethics, 15, 1189e1198. Gouldner, A. W. (1960). The norm of reciprocity. American Sociological Review, 25, 165e178. Greenberg, M. S. (1980). A theory of indebtedness. In K. J. Gergen, M. S. Greenberg, & R. H. Wills (Eds.), Social exchange: Advances in theory and research. New York: Plenum. Greenberg, M. S. (1986). A preliminary statement on a theory of indebtedness. In Justice in social exchange (Symposium presented at the western psychological association, San Diego). Greenberg, M. S., & Shapiro, S. P. (1971). Indebtedness: an adverse aspect of asking for and receiving help. Sociometry, 34, 290e301. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5th ed.). Prentice-Hall. Harman, H. H. (1976). Modern factor analysis. Chicago, IL: University of Chicago Press. Hollingshead, A. B., Fulk, J., & Monge, P. (2002). Fostering intranet knowledge sharing: an integration of transactive memory and public goods approaches. Distributed Work, 335e355. Hsu, M. H., Ju, T. L., Yen, C. H., & Chang, C. M. (2007). Knowledge sharing behavior in virtual communities: the relationship between trust, self-efficacy, and outcome expectations. International Journal of Human Computer Studies, 65, 153e169. Joinson, A. (2001). Self-disclosure in computer-mediated communication: the role of self-awareness and visual anonymity. European Journal of Social Psychology, 31, 177e192. Kankanhalli, A., Tan, B. C. Y., & Wei, K. K. (2005). Contributing knowledge to electronic knowledge repositories: an empirical investigation. MIS Quarterly, 29(1), 113e143. Kiesler, S., Siegel, J., & Mcguire, J. W. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39, 1123e1134. Kolm, S. C. (2008). Reciprocity: An economics of social relations. Cambridge; New York: Cambridge University Press. Leary, M. R., & Kowalski, R. M. (1990). Impression management: a literature review and two-component model. Psychological Bulletin, 107(1), 34e47. Lea, M., & Spears, R. (1992). Paralanguage and social perception in computermediated communication. Journal of Organizational Computing, 2, 321e341. Lee, C. (1990). Modifying an American consumer behavior model for consumers in confucian culture: the case of the Fishbein behavioral intention model. Journal of International Consumer Marketing, 3(1), 27e50. Lin, M. J. J., Hung, S. W., & Chen, C. J. (2009). Fostering the determinants of
349
knowledge sharing in professional virtual communities. Computers in Human Behavior, 25, 929e939. Ma, M., & Agarwal, R. (2007). Through a glass darkly: information technology design, identity verification, and knowledge contribution in online communities. Information Systems Research, 18(1), 42e67. Mathews, M. A., & Green, J. D. (2009). Looking at me, appreciating you: self-focused attention distinguishes between gratitude and indebtedness. Cognition & Emotion, 24(4), 710e718. published on: 28 January 2009 (iFirst). Moore, G. C., & Benbasat, I. (1991). Development of an instrument to measure perceptions of adopting an information technology innovation. Information Systems Research, 2(3), 192e222. Nowak, M. A., & Sigmund, K. (2005). Evolution of indirectly reciprocity. Nature, 437(27), 1291e1298. Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill. Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory. New York: McGrawHill. Phang, C. W., Kankanhalli, A., & Sabherwal, R. (2009). Usability and Sociability in Online Communities: a comparative study of knowledge seeking and contribution. Journal of AIS, 10(10), 721e747. Pinsonneault, A., & Heppel, N. (1998). Anonymity in group support systems research: new conceptualization, measure, and contingency framework. Journal of Management Information Systems, 14(3), 89e108. Postmes, T., Spears, R., & Lea, M. (1998). Breaching or building social boundaries? SIDE-effects of computer-mediated communication. Communication Research, 25, 689e715. Postmes, T., Spears, R., Sakhel, K., & De Groot, D. (2001). Social influence in computer-mediated communication: the effects of anonymity on group behavior. Personality and Social Psychology Bulletin, 27, 1242e1254. Schmitz, J., & Fulk, J. (1991). Organizational colleagues, media richness, and electronic mail: a test of the social influence model of technology use. Communication Research, 18(4), 487e523. Scott, J. F. (1971). Internalization of norms: A sociological theory of moral commitment. Prentice-Hall. Sia, C. L., Tan, B. C. Y., & Wei, K. K. (2002). Group polarization and computermediated communication: effects of communication cues, social presence, and anonymity. Information Systems Research, 13(1), 70e90. Spears, R., & Lea, M. (1994). Panacea or panopticon? the hidden power in computermediated communication. Communication Research, 21, 427e459. Stone, E. F. (1978). Research methods in organizational behavior. Santa Monica, CA: Goodyear. Teo, H. H., Wei, K. K., & Benbasat, I. (2003). Predicting intention to adopt interorganizational linkages: an institutional perspective. MIS Quarterly, 27(1), 19e49. Walster, E., Berscheid, E., & Walster, G. W. (1973). New directions in equity research. Journal of Personality and Social Psychology, 151e176. Walther, J. B. (1996). Computer-mediated communication: impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23(1), 3e43. Wasko, M., Faraj, S., & Teigland, R. (2004). Collective action and knowledge contribution in electronic networks of practice. Journal of the Association for Information Systems, 5(December), 493e513. Wasko, M. M., & Faraj, S. (2000). ‘It is what one does’: why people participate and help others in electronic communities of practice. The Journal of Strategic Information Systems, 9(2e3), 55e173. Wasko, M. M., & Faraj, S. (2005). Why should i share? examining social capital and knowledge contribution in electronic networks of practice. MIS Quarterly, 29(1), 35e56. Watkins, P., Scheer, J., Ovnicek, M., & Kolts, R. (2006). The debt of gratitude: dissociating gratitude and indebtedness. Cognition and Emotion, 20(2), 217e241. Watson, S., & Hewett, K. (2006). A multi-theoretical model of knowledge transfer in organizations: determinants of knowledge contribution and knowledge reuse. Journal of Management Studies, 43(2), 141e173. Westphal, J. D., & Zajac, E. J. (1997). Defections from the inner circle: social exchange, reciprocity, and the diffusion of board independence in U.S. corporations. Administrative Science Quarterly, 42, 161e183. Wiertz, C., & Ruyter, K. (2007). Beyond the call of duty: why customers contribute to firm-hosted commercial online communities. Organization Studies, 28, 347e376. Yee, N., Bailenson, J. N., Urbanek, M., Chang, F., & Merget, D. (2007). The unbearable likeness of being digital: the persistence of nonverbal social norms in online virtual environments. Cyberpsychology & Behavior, 10, 115e121. Ye, H. J., Feng, Y., & Choi, B. C. F. (2015). Understanding knowledge contribution in online knowledge communities: a model of community support and forum leader support. Electronic Commerce Research and Applications, 14, 34e45.