JAPR_3-1_73.pdf477.1KB

 Online streaming services allow users to access content such as tele-vision programs or movies without temporal constraints. Not only have online-based video websites such as hulu.com and youtube.com become extremely popular, most traditional broadcasting television networks have also expanded their services into the online environment. According to the Harris Poll, over half of Americans (53%) watch digitally streamed television programs, and the number of users for digitally streaming tele-vision continues to grow yearly (Shannon-Missal, 2012). Besides con-venience, the crucial contributing factor to the growth of digital streaming video consumption is the fact that it is generally free. It is a great benefit for audiences to watch television programs without paying for cable TV subscriptions. Free service is an important factor in attracting users to services and brands because it creates emotional arousal in the consum-er’s mind. In return, users experience appreciation and gratitude toward the brand. Several studies have examined the effects of free services on relationship building between brand and consumers (Raghubir, 2004; Wu, Ray, Geng, & Whinston, 2004); however, no studies have investigated the effect of free services on relationship building in online streaming video services.

 This study emphasizes two original ideas that differentiate it from pre-vious relationship building studies. First, this study examines brand- con-sumer relationship building in the context of online streaming video serv-ices, which has not been explored before. Second, this study investigates the role of the emotional component - gratitude - in building relationships as suggested in Palmatier, Jarvis, Bechkoff, and Kardes’s (2009) study. Palmatier et al. (2009) developed the processing model to highlight the role of gratitude in relationship building in a traditional marketing environment. In addition to the role of gratitude as a mediating varia-ble-as suggested by Palmatier et al. (2009), this study tests the concept of gratitude as the output of relationship building with online video services.

 The purpose of this study is to propose and test a conceptual model that shows associations among variables affecting relationship building in online television websites, based on Palmatier et al.’s (2009) relationship marketing model. This study makes an academic contribution to the field of marketing research by confirming the role of gratitude in online rela-tionship building between consumers and free online video websites. This study also has practical implications for online marketing practitioners by emphasizing the effects of “free” marketing.

Theoretical Background: Reciprocity

 Reciprocity, a key concept explaining the durability and stability of exchange relationships (Larson, 1992), is regarded as a useful theoretical framework in relationship research (De Wulf et al., 2001; Huppertz, Arenson, & Evans, 1978; Palmatier et al., 2009). The principle of reci-procity holds that people feel psychological obligations to return what they receive from others (Bagozzi, 1995; Gouldner, 1960), as well as psychological guilt whenever they violate the reciprocal norm (Li & Dant, 1997). The concept of reciprocity has been used in a number of studies to un-derstand the psychological processes involved in building relationships between customers and companies (Bagozzi, 1995; De Wulf et al., 2001; Kang & Ridgeway, 1996). Bagozzi (1995) argued that a consumer displays loyalty to a company in reciprocation of its investment in the relationship. Kang and Ridgeway (1996) argued that a company’s friendliness evokes the payback obligation toward the company from consumers. Additionally, De Wulf et al.’s (2001) model of customer relationship building showed statistically significant links between relationship marketing tactics, per-ceived relationship investment, relationship quality, and behavioral loyalty.

 Moon (2000) has questioned whether the principle of reciprocity is compatible with the realities of consumer research, because a reciprocal interaction between a consumer and a company would require a one-to-one interaction with every consumer. However, the current study considers customer relationships in the context of the online environment. Unlike in traditional offline environments, online environments allow for one-to-one interaction between a customer and a company. The current study focuses on how users of online streaming video perceive the web-sites that offer the service for free and how users reciprocate the websites’ investment in building relationships with them. The users’ reciprocal pay-back may be the feeling of gratitude toward the websites and may even be linked to the feeling of gratitude toward the commercials on the videos provided by the websites.

Gratitude in Relationship Marketing

 Many earlier sociologists have explained the concept of gratitude as a social element of human interactions that offers an emotional foundation for reciprocal behaviors (Gouldner, 1960; Schwartz, 1967; Simmel, 1950; Recited from Palmatier et al., 2009). Gouldner (1960) regarded the concept of gratitude as a force that facilitates people’s reciprocal obligations. Schwartz (1967) argued that gratitude would cause relationships to maintain people’s prosocial orientation. Simmel (1950) viewed gratitude as an es-sential component that supports the chain of reciprocity. Based on mutual feelings of gratitude, people tend to feel better when they reciprocate and guilty when they fail to reciprocate (Becker, 1986; Trivers, 1985).

 From the marketing perspective, gratitude refers to consumers’ emo-tional appreciation for benefits received from product or service providers and their desire to reciprocate the benefits (Emmons & McCullough, 2004; Morales, 2005). Trust and commitment are traditional variables in the extant model of relationship marketing, but Palmatier et al.’s (2006) meta-analysis of relationship marketing literature suggested the possibility of other mediating variables. Palmatier et al. (2009) tested the role of gratitude and found that consumers would feel appreciative toward the providers’ extra efforts and favors to build relationships with them and would reciprocate such providers’ relationship investment by showing trust and then commitment, increasing consumers’ purchase intentions. Feelings of gratitude could also have a positive effect on consumers’ purchase intentions without the mediation of trust and commitment.

 This study developed a new relationship building model with gratitude as a mediator between relationship investment and relationship quality on free streaming video websites, based on Palmatier et al.’s (2009) approach.

Proposed Model and Hypotheses

 Palmatier et al. (2009) proposed their relationship building model by adding the variable of gratitude as the mediator of relationship investment on trust and commitment, and purchase intention. The current study adopts Palmatier et al.’s (2009) model and revises it to fit the context of online relationship building, specifically websites that offer free streaming video services. Therefore, a dependent variable for this study is the users’ intention to visit the website. As suggested in Palmatier et al.’s (2009) study, this study adopts a second-order factor model for relationship quality due to high correlation between trust and commitment (r = .692, p < .001) that causes discriminant validity issues with the two variables. Observed variables for trust and commitment significantly loaded on their respective first-order factors. Accordingly, the average scores for the first-order latent variables were obtained based on the three items of each factor, and used as indicators of the higher-order construct of relationship quality. Figure 1 hypothesizes how a dependent variable, intention to visit the website, is associated with relationship investment, gratitude toward the website, and relationship quality.

Figure 1. The Hypothesized Model

Relationship Investment

 Many studies on relationship marketing highlight the importance of product or service providers’ investment in relationship building with au-diences (Baker, Simpson, & Signaw, 1999; Bennet, 1996; De Wulf et al., 2001; Ganesan, 1994; Hart & Johnson, 1999; Palmatier et al., 2009). According to De Wulf et al. (2001), perceived relationship investment as an antecedent of relationship quality refers to “a consumer’s perception of the extent to which a retailer devotes resources, efforts, and attention aimed at maintaining or enhancing relationships with regular customers that do not have outside value and cannot be recovered if these relationships are terminated” (p. 35). Thus, when customers perceive that a retailer is working to build a favorable relationship with them, the customers should be positively impressed (Hart & Johnson, 1999).

 According to the theoretical explications provided by Palmatier et al. (2009), relationship investment triggers the relationship building process, which is associated with trust and commitment (relationship quality) that subsequently affects purchase intention (visit intention in the current study).

Relationship Quality (Trust and Commitment)

 Trust is an indication of successful relationship marketing (Garbarino & Johnson, 1999; Gruen, Summers, & Acito, 2000; Hunt & Morgan, 1995). Trust is defined as “a consumer’s confidence in a retailer’s reliability and integrity” (De Wulf et al., 2001, p. 36). The development of trust is re-garded as a crucial result of investing in dyadic buyer-seller relationships (Gundlach, Achrol, & Mentzer, 1995). Molm (2010) studied the differences between levels of trust in reciprocal versus negotiation-based relationships. According to Molm (2010), trust is greater in reciprocal relationships since each party feels more united and harmonious.

 The concept of commitment is defined as “a consumer’s enduring desire to continue a relationship with a retailer accompanied by this consumer’s willingness to make efforts at maintaining it” (De Wulf et al., 2001, p. 37). Relationship commitment strongly reflects consumers’ perceptions of relationship quality (De Wulf et al., 2001; Dorsch et al., 1998; Hennig-Thurau & Klee, 1997). Several studies deal with the variable of commitment as one of three subdimensions―satisfaction, trust, and com-mitment―to represent the construct of relationship quality (De Wulf et al., 2001; Yoon et al., 2008; Yoon et al., 2013). However, other studies based on commitment-trust theory recognize the linear relationship between trust and commitment―trust as an antecedent of commitment (Palmatier et al., 2006; Palmatier et al., 2009; Morgan & Hunt, 1994). This study adopts the latter approach.

 Relationship management studies have found a significant positive ef-fect of relationship investment on trust (Ganesan, 1994) and commitment (Bennet, 1996). In this study, users of online streaming video services who perceive relationship investment from the website would have trust in the website, leading to greater commitment to the website.

 H1. Relationship investment leads free online streaming video users to feel higher levels of relationship quality-trust and commitment-toward the website.

Gratitude toward the Website

 Previous studies on gratitude identified two aspects of grati-tude-affective and behavioral (Palmatier et al., 2009). The affective aspect of gratitude is the internal cognitive and affective process of individuals as “the recipient of an intentionally rendered benefit” (Emmons, 2004, p. 9; Recited from Palmatier et al., 2009, p. 2). Based on the theory of reci-procity, individuals who are well treated, or obtain significant benefits from others, feel an obligation to repay the other party (Becker, 1986; Komter, 2004). As individuals’ motivation to repay the other party grows, individuals engage in behavior to reciprocate the generosity of the other party. Thus, the behavioral aspect of gratitude originates from the affective aspect of gratitude. The behavioral aspect of gratitude is the “act of giving in return” (Bartlett & DeSteno, 2006; Palmatier et al., 2009, p. 3). This re-turn behavior is triggered by feeling gratitude toward the other party, which creates the cycle of reciprocity (Becker, 1986; Schwartz, 1967). Palmatier et al.’s (2009) study proposed the mediating role of gratitude between relation-ship investment and trust/commitment, and between relationship investment and purchase intention, as well. Following the model explications by Palmatier et al. (2009), this study assumes the mediating effects of website gratitude on relationship quality―trust and commitment―and on website visit intention.

 H2. Relationship investment leads free online streaming video users to feel higher levels of gratitude toward the online video website.

 H3. Free online streaming video users’ feelings of gratitude lead to higher perceptions of relationship quality―trust and commitment―toward the website.

Visit Intention

 As a dependent variable of this study, visit intention refers to how like-ly users are to keep vising an online video website in the future. Palmatier et al.’s (2009) model suggested two antecedents of purchase intention―feelings of gratitude and relationship commitment. Thus, the current study hypothesizes that gratitude toward the website would positively associate with visit intention and that relationship quality―trust and commitment―would positively associate with visit intention.

 H4. Free online streaming video users’ feelings of gratitude lead to their intention to visit the website.

 H5. Free online streaming video users’ perceptions of relationship quality―trust and commitment―lead to their intention to visit to the website.

Methods

Samples and Procedures

 A survey with 205 college students was conducted to test the proposed model. Survey participants were recruited on a volunteer basis from the student population of a large Southwestern public university. Recruitment was conducted in two ways―classroom advertisement and mass e-mail via a student enrollment list. College students are deemed appropriate for the current study because they tend to be familiar with new media (Yang, Zhou, & Liu, 2010) and represent a significant proportion of online video users (Shannon-Missal, 2012). According to the Harris Poll, 68% of U.S. adults who watch online streaming video are between the ages of 18 and 35 (Shannon-Missal, 2012).

 Respondents who agreed to participate in the study were given the web-site address for an online survey (https://www.surveymonkey.com/s/onli-netvs). Respondents were directed to log onto the first page of the study site where they read general instructions. Participants were then asked to indicate the name or URL of one website they visit most regularly for watching online streaming video. Respondents were given a specific defi-nition and an example of an online video website that offers no offline service, such as hulu.com, and that offers both online and offline service, such as nbc.com and espn.com. Respondents were instructed to exclude YouTube because it contains mostly user-created content. Respondents were then instructed to fill out the questionnaire regarding their responses to the website they had listed. Most of the survey items were adapted from previous studies (De Wulf et al., 2001; Palmatier et al., 2009).

Measurement

 Except relationship quality, all other latent constructs―relationship in-vestment, visit intention, and gratitude toward the website―had three items measured with 7-point Likert scales, ranging from 1 (strongly dis-agree) to 7 (strongly agree). As stated, due to discriminant validity issues, the trust and commitment variables were composted and then combined into one factor, relationship quality.

 Measurement reliability was tested using factor analyses and Cronbach’s alpha. All scales were found to be internally consistent. One component was extracted from the three survey items for each variable. Survey items and Cronbach’s alphas are displayed in Table 1.

Table 1. Summary of Measures

Analysis

 Structural Equation Modeling (SEM) with AMOS 18 was employed to test the hypotheses in the proposed model. To assess the fit of the pro-posed model, this study used (1) the chi square statistic, specifically the ratio of chi square value to degree of freedom, (2) the goodness-of-fit in-dex (GFI), (3) the normed fit index (NFI), (4) the Tucker-Lewis Index (TLI), (5) the comparative fit index (CFI), and (6) the root mean square error of approximation (RMSEA).

Results

Descriptive Statistics

 A total of 205 responses were valid for the data. Of the 205 partic-ipants, 159 (77.6%) were female and 46 (22.4%) were male. The average age of the participants was 20.5 (SD = 1.874), ranging from 18 to 28. With respect to the frequency of watching television online, the partic-ipants watched television online for an average of 3.29 hours per week (SD = 4.04), with a range of 1 to 30 hours per week.

Analysis of Structural Equation Modeling

 Prior to model testing, analyses assessed if the scales achieved sat-isfactory levels of reliability, validity, and whether factor loadings sig-nificantly related to their corresponding constructs. The measurement model, including the latent constructs and their respective observed varia-bles was analyzed first. Then, the structural model with the hypothesized relationships was tested and refined through a series of tests in an at-tempt to better explain the data. A final, modified model was ultimately developed.

Measurement Model Evaluation

 A confirmatory factor analysis of the full measurement model showed that all of the indicators significantly loaded on their corresponding latent construct (p < .001). The fit of the measurement model was acceptable (Chi Square = 79.662, df = 38, p < .001, CMIN/DF = 2.096, GFI = .932, NFI = .948, TLI = .959, CFI = .972, and RMSEA = .073).

 The convergent and discriminant validity of the latent constructs were examined. Convergent validity was checked in two ways: (1) checking factor loading of each item to the latent constructs, and (2) checking AVE (Average Variance Extracted) value, which should be greater than .05. As stated, all items significantly loaded onto the correspondent latent con-structs (p < .001), and all latent constructs’ AVEs were greater than .05.

 Discriminant validity was checked by comparing AVE estimates with MSV (Maximum Shared Variance) and ASV (Average Shared Variance) as suggested by Fornell and Larcker (1981). All AVE estimates of latent constructs were greater than the MSV and ASV estimates as shown in Table 2. Thus, both convergent and discriminant validity of latent con-structs were obtained.

Table 2. Test of Convergent and Discriminant Validity of Latent Variables

 Table 3 presents correlations, variance, and covariance of the latent con-structs in the measurement model. All correlation coefficients, except that of gratitude toward website and relationship quality (.727), were below the recommended threshold of .70 (Tabachnik & Fidell, 2001), indicating no significant multicollinearity issues. Overall, the results indicated that the scales assessed what they were intended to measure and were reliable in the sample.

Table 3. Correlation, Covariance, and Variance for Latent Constructs

Structural Model Evaluation

 The structural model was assessed using maximum likelihood. All five hypothesized relationships among four latent variables were statistically significant in the directions hypothesized. Relationship investment (RV) from the website was positively associated with relationship quality, lend-ing support for H1. RV from the website also was positively associated with gratitude toward the website (GRAT_website), which led to a positive association with relationship quality, lending support for H2 and H3. There were positive significant associations between GRAT website and intention to visit the website, and between relationship quality and in-tention to visit the website. Thus, H4 and H5 were confirmed. The good-ness-of-fit indices suggested that the model fit the data well; x² = 79.886, df = 39, p < .001, x²/df ratio = 2.048; GFI = .932; NFI = .948; TLI = .961, CFI = .972; RMSEA = .072. The parameter estimates for the pro-posed structural model are reported in Table 4.

Table 4. Analysis of Structural Models

Discussion

 This study examined the mediating effect of gratitude toward online video websites between relationship investment and viewers’ intention to visit the websites. A modified model from Palmatier et al.’s (2009) relationship marketing model was proposed and tested. All paths in the model were found to be significant. The results of this study suggest two implications for free online streaming video research: (1) theoretical support for the tra-ditional relationship marketing model in free online streaming video web-sites, and (2) the importance of viewers’ gratitude-based emotion.

Relationship Building with Viewers in Free Streaming Video Websites

 The assumptions of the traditional relationship marketing model were supported in the online environment, specifically in the context of free online streaming video. Perceived relationship investment positively affected online video viewers’ perception of relationship quality (trust and commit-ment) toward websites that offer streaming video services. Relationship quality ultimately had a positive association with viewers’ intention to visit the website. When online video viewers perceive that the website makes regular efforts to maintain or improve a tie with them, they are more likely to have trust in the website, to make extra efforts to remain a user of the website, and continue to visit the website in the future.

 In particular, even though it is not tested in this study, websites’ efforts to build relationships with users would be more readily accessible to users because of their inherently interactive nature. Several studies have found interactivity to be an antecedent of perceived relationship invest-ment (Yoon et al., 2008). Online video users obtain information (e.g., newly updated or coming television shows) from the website regularly and quickly via two-way communication tools such as email. Moreover, users can watch the television program simultaneously via streaming. This technological convenience should lead users to perceive that websites care about their regular customers.

The Role of Gratitude-Based Emotion

 Consistent with the assumption of Palmatier et al.’s (2009) model, the current study showed that perceived relationship investment had a positive influence on free online video users’ feelings of gratitude toward the website, which led to their intention to visit the website. In addition to a direct effect of gratitude on visit intention, gratitude had an indirect effect on visit intention through relationship quality (trust and commitment).

 Several previous studies have demonstrated the effects of feelings of gratitude on customers’ behavioral change. Based on the theory of reci-procity, customers feel obligated to pay back a company when they get benefits from that company. By changing their purchase behavior, cus-tomers act upon their feelings of gratitude and satisfy their obligation to repay (Dahl, Honea, & Manchanda, 2005; Recited from Palmatier et al., 2009). Thus, free online streaming video users who have feelings of grati-tude toward the website may keep visiting as their payback to the website. As expected, the current study found that these feelings of gratitude are positively associated with the relationship quality.

 In the context of online video, therefore, users who perceive relation-ship investment by the website may feel gratitude toward the website and satisfaction with the service, may build a relationship with it, and may become regular customers. In this reciprocal process, users’ gratitude may play a role as a motivating force for both users and the website―producing a higher quality of relationship with regular users and a greater loyalty toward the website. This assumption is well explained by Komter (2004)’s definition of gratitude―as part of the chain of reciprocity, an imperative force that compels people to return the benefits they have received.

Conclusion

 The number of people who use digital streaming video continues to grow; thus, its advertising market is also growing quickly (Shannon-Missal, 2012). This study proposed and tested a conceptual model of relationship building in free online streaming video. Recently, consumers’ feelings of gratitude have been regarded an additional important variable for relation-ship marketing research (Dewani & Sinha, 2012; Kolyesnikova, Sullivan, & Callison, 2011; Palmatier et al., 2009; Soscia, 2007).

 Gratitude seems to enhance the effectiveness of relationship building in free online video websites in two ways. First, the users’ gratitude-based emotion, induced by the website’s relationship investment, has a direct effect on users’ intention to visit the website regularly. Second, the users’ relationship with the website formed by the feeling of gratitude may also lead to increased visit intention. A study conducted by Whatley, Webster, Smith, and Rhodes (1999) reported that individuals tend to internalize the importance of gratitude and reciprocity, which suggests the possibility of strong and long-term relationships. Thus, once individuals develop feelings of gratitude toward the brand and services, it may accelerate and strengthen the relationship building process.

 This study has practical implications for online marketing practitioners, including digital media planners and advertisers, who need to consider persuasive appeals to elicit reciprocal emotions such as gratitude. More benefits may lead to consumers’ feelings of gratitude. For example, de-spite free service, if online streaming video viewers can experience pre-mium services such as high-definition (HD) video and non-buffering, which are usually offered for paid-membership users, they may have more feelings of gratitude. Possible financial loss that may occur due to fewer paid-memberships might be counteracted by an increase in advertising. Advertisers with exclusive contracts for video-embedded com-mercials could sponsor free streaming services for popular television series. Advertisers of product placement (PPL) in the video content might also be interested in streaming commercials because of the interactive nature of online ads. When a brand is specified as the sponsor of free online service for the television series, the viewers may feel gratitude toward the brand. Considering the rapid growth of the online video service market, online marketing practitioners may have to continue to develop various marketing strategies that can build consumers’ feelings of gratitude. Over the long term, online marketers may need a deeper understanding of the social psychology of reciprocal tendencies.

 This study also makes an academic contribution to the field of market-ing research by suggesting a conceptual model for relationship building in the new market of free online video service. Future research may be inter-ested in how various functions of interactivity in the online environment affect the exchange of emotions with consumers, leading to relationship building. Additional emotion-based factors on online relationship marketing may be an interesting topic as well.