Exploring antecedents of online group-buying: Social commerce perspective

2.1 Online group-buying

Group-buying or known as “Tuangou”, roots in the Chinese social shopping convention that emphasis its people-powered nature, has become hot and popular group commerce worldwide [64]. The rise of group-buying websites in recent years has changed the traditional model in which the price of a product is set by its seller. Today, online shoppers gather together to negotiate prices with sellers; in simpler terms, group-buying is a pricing strategy in which consumers with common needs combine to lower the price of a product by placing a large order to gain the economies of scale [18, 26]; the more group-buying participants, the higher the price-negotiation capacity [92]. Lai and Zhuang [57] believe that through the transaction model of group-buying, both buyers and sellers also may effectively reduce their transaction costs. Through the mechanism of group-buying, therefore, a seller may reduce the cost of attracting customers while a buyer may get a desired product at a lower price [49], creating a win-win situation for both parties [59, 76].

Recently, online group-buying has been recognized as a kind of social commerce integration [82], specifically Sterling et al. [81] points out that local social commerce bring the explosion of group-buying. It is thus posited in this present study that online group-buying is possible since its members have a common need, and if an order placed through group-buying is large enough, these buyers may also get certain price-related or other economic benefits from the seller. Based on the above, in this study the behavior of online group-buying is defined as a kind of community commerce for purpose of shopping and consumption, whereby netizens are collaborated to increase consumers’ power, with the same desire to reach the threshold of a discounted mass order and enjoy price-related or other perceived benefits.

2.2 Theory foundation of online group-buying: TPB

The TPB, which is an extension of the well-known theory of reasoned action (TRA, see Fishbein and Ajzen [29]), is the base theory of this study. TRA asserts that behavior is a direct function of behavioral intention. Attitude is treated in several theories as having certain influences over behavior intention, including TRA [29], TPB [3], and the technology acceptance model [23, 24], which have also been tested under different environmental backgrounds. Taylor and Todd [85] state that attitude is the belief that a certain behavior would lead to a certain outcome.

Ajzen and Fishbein [5] believe that a behavior is sometimes influenced more by social-environmental pressure than it is by personal attitude. Previous studies believe subjective norm increases an individual tendency of systematically using behavior intention [47, 87, 47, 87]. Hansen et al. [38] believe an online consumer is influenced by support from family members, friends, and acquaintances regarding the online purchase. Thus, an online consumer who faces more social pressure or stronger stimuli to comply would have stronger subjective norm and behavior intention. Thus, in this study this variable is defined as the “peer norm”, the reference group for online group-buying consists of friends and peers.

Further the TPB model extends from the TRA model by incorporating an additional construct – perceived behavioral control (PBC) – to account for situations in which an individual lacks substantial control over the targeted behavior [3]. Ajzen [1] defines perceived behavioral control refers to individuals’ belief in their own capability to execute a behavior. TPB shows that perceived behavioral control does influence behavior intention and also directly affects actual behaviors [1, 2]. Ajzen [4] believes the perceived behavioral control in TPB is not a unitary concept but a two-tier concept consisting of controllability and self-efficacy. The same two-tier concept has been adopted by numerous researchers in their discussions on perceived behavioral control. Examples include “the application to e-services,” “satisfaction with digitalization and sustainable intention” [42], and ‘information searching and the behavior of product purchase” [71].

In previous online shopping research, the TPB is often used as the theoretical background, and past empirical findings often yield good interpretive results [45, 90]. However, while online shopping and/or auction behavior is different from that of group-buying auction [17]; therefore, this paper specific focuses on online group-buying to study consumers’ intentions and behaviors, and posits that a consumer with a stronger intention of online group-buying is more likely to engage in the behavior of online group-buying.

2.3 Exogenous variables of online group-buying

To complete a group-buying transaction, certain factors influence the activity of group-buying for the initiator and the group buyer. Of those factors, trust in the initiator plays a critical role [53]. In addition, the benefits perceived by a consumer and the inter-consumer interactions in the process of group-buying all influence the completion of group-buying transactions [50, 52]. Therefore, our study proposes the three exogenous factors of trust in the initiator, perceived benefits, and social interaction to understand their influence on the behavior of online group-buying from social commerce perspective.

Trust in the initiator. Trust refers to a person’s (believer’s) feeling that the trusted person (trustee) will meet the expectation and ensure the security of the believer [34]. Besides, trust is treated as a resource that reduces uncertainty and allows the believer to face the uncertainties in society [71, 78]. In online shopping, trust is always the major concerns of consumers, and more specific, for online group-buying auctions, trust in the initiator is a concern that influences consumer bidding [21, 86].

Perceived benefits. When conducting a group-buying purchase, consumers must perceive several benefits from group-buying as an incentive so as to support their buying behavior [52]. Group-buying therefore provides several benefits for consumers. For example, convenience is a form of psychological cost expressed in nonmonetary forms such as time, effort, and stress [11, 93]. Reducing cost for transaction cost, such as time and effort saving in searching for products and placing orders [7, 92], also, driving down the price for buyers [60].

Social interaction. Group-buying is a kind of group commerce model [64], in such collective shopping behavior, both the initiator and group buyers must have interpersonal interactions to complete the transaction. Yet, the lack of social interaction makes online shopping unattractive to consumers who enjoy socializing or those who treat buying online as a form of social shopping [46, 84]. Recently, Animesh et al. [8] and Wei et al. [89] in their study posited that online users who get more involved in their activities are likely to focus entirely on their virtual interactions, therefore, generating their purchasing intention. Internet plays a pivotal role in communication between the online group-buying initiator and the group buyers, yet the quality of the social interaction depends on both online group-buying parties.

3.1 Research model and hypotheses development

Numerous researchers have applied TPB to the study of online consumer behavior; for example Hansen et al. [39], Pavlou and Fygenson [71]; Herrero Crespo and Rodríguez del Bosque [41]; Lin [62]; Cheng and Huang [19], Wu et al. [90]. TPB is also utilised in this study to examine consumers’ online group-buying behavior. In this paper, the three social commerce antecedents are explored to facilitate understanding of online group-buying. As Kim et al. [52] provides empirical evidence that trust and perceived benefit are strong determinants of a consumer’s e-commerce transaction decision. Most recently Kim and Park [53] has confirmed trust is still play an significant role in the context of social commerce. Moreover, the social interaction is treated as a requisite in social network environment [50]. Therefore, the three variables of trust in the initiator, perceived benefits, and social interaction are included as the social commerce antecedents in this study. This research model shown in Fig. 1.

Pavlou and Fygenson [71] used TPB to investigate information searching and product purchasing behavior. Wu et al. [90] also adopted TPB to examine consumers’ online keyword searching behavior. Hansen et al. [38] was to test the ability of the TRA and TPB in predicting consumer online grocery behavior intention. Cheng and Huang [19] was combined TPB with characteristics of online group-buying of customers-initiated transactions to examine antecedents that influence online group-buying intentions. TPB is also utilised in this study to examine consumers’ online group-buying behavior. Therefore we propose the hypothesis 1 to hypothesis 7:

A user’s intention to participate in an online group-buying transaction is positively associated with the user’s online group-buying behavior.

In online shopping, trust is always the major concerns of consumers, and more specific, for online group-buying auctions, trust in the initiator is a concern that influences consumer bidding [48, 86]. Sun et al. [83] indicated the degree of trust consumers had on initiators influences online group-buying. Besides, Luhmann [63] also proposes that the relationship between trust and perceived behavioral control is that trust helps to reduce uncertainties in a transaction. According to Ku [55] and Tsai et al. [86], online trust affects users’ group-buying intention, and Kim and Park [53] has confirmed that trust is an indispensable factor in the social commerce context. Therefore, trust facilitates the behavior of online group-buying by overcoming the worries in engaging in online group-buying instead of controlling the behaviors of the initiator of online group-buying. In addition, when the group-buying members have much trust in the initiator, they are more likely to discuss group-buying with their friends or to persuade them to join online group-buying under the same initiator. Hence, we hypothesize that:

A user’s trust in the initiator is positively associated with attitude toward online group-buying.

In addition, the flow theory proposed by Csikszentmihalyi [22] claims that perceived enjoyment and perceived playfulness are important drives leading individuals into the flow state, and those who are in the flow state are more likely to voluntarily interact with others. Previous researchers believe that to most consumers, the enjoyment of shopping plays a critical role in motivating them to continue purchasing online [6, 30]. Altogether, the variable of perceived benefits includes convenience, saving time, reducing cost, enjoyment, and playfulness of online group-buying.

The premise of gaining these benefits, however, is that the initiator must be trusted by all other group-buying members. In other words, the consumers participating in online group-buying must believe that they will not be harmed by the initiator. As a result, the benefits gained from online group-buying also increase the trust group buyers have in the initiator. Hence, we hypothesize that:

A user’s perceived benefits are positively associated with trust in the initiator.

The study of Huang [43] called for the importance of social focus than the shopping focus, and found that interactive among online users influence their intention to purchase. As the nature of social commerce, social interaction should be the fundamental requirement in the group transactions [50]. In this study, we posit that social interaction is an important antecedent in the online group-buying decision. Hence, we hypothesize that:

A user’s perceived social interaction is positively associated with trust in the initiator.

3.2 Subjects

A survey study was conducted to examine the proposed hypotheses. Subjects of this research were individuals all of whom had practical experience of online group-buying during the past six months [19, 88]. An announcement was posted on a group-buying based discussion forum with PTT (telnet://ptt.cc) to recruit participants. PTT is the largest and most popular BBS (Bulletin Board System) in Taiwan with approximately 100,000 visitors per day. The announcement stated the purpose of this study and specified that only those with group-buying experiences would be qualified to participate in the survey. Sample information is gathered by online surveys conducted at my3q.com. To preserve confidentiality, all subjects were informed that their responses would remain anonymous and would be used for academic purposes only. A total of 268 surveys were returned. The exclusion of incomplete questionnaires resulted in a total of 202 usable responses. Given that approximately 485 people logged into the system, the presumed response rate was 41.6% . Among the respondents, 6.4% were males and 93.6% were females. This sample structure is consistent with the past finding that the percentage of female group buyers is higher than their male counterparts by 90% [20, 67].

In addition, we compared early and late respondents on all variables (with late respondents being assumed to be similar to non-respondents) in order to assess non-response bias [10]. No significant differences were found between these two groups in all constructs, so non-response bias was not expected to have an effect on the results of our study. Also, common method bias was evaluated by using Harman’s one-factor test to ensure that there was no significant method effect on the predefined causal relationship [74]. Given the small magnitude and insignificance of the method variance, we contend that the method is unlikely to be a serious concern for this study. Therefore, the representativeness of our sample is assured. Table 1 details the respondents’ characteristics.

3.3 Measurement development

Totally ten constructs are employed in our study: trust, perceived benefits, social interaction, group-buying attitude, group-buying peer norm, group-buying perceived behavioral control, self-efficacy, controllability, group-buying intention, and group-buying behavior. These constructs were measured via a multiple-item scale drawn from prevalidated measurement items that were re-worded for the study of group-buying behaviors. The applicability of the modified items was enhanced by literature reviews, using two professors and three master’s students as expert judges, and by pilot testing relevant samples. During this process, scale items were trimmed and refined and dimensions were modified to hold the content validity as understanding of the constructs improved. All items used seven-point Likert scales anchored from “strongly disagree (= 1)” to “strongly agree (= 7)” (see Appendix A). Table 2 provides operational definitions for these constructs.

3.4 Reliability and validity

The partial least squares model was employed to test our measurements and proposed hypotheses. We have chosen Smart PLS 2.0.M3 because both reflective and formative constructs are involved in our model that it was ability to model latent constructs under conditions of nonnormality with small to medium sample sizes [37]. Perceived benefit is a latent formative construct that is formed by the first-order factor of convenience, cost, search time, order time, shopping fun, and playfulness. The SmartPLS 2.0 M3 [77], which has also been frequently adopted in many information systems journals such as MIS Quarterly [15, 80], was employed for this analysis.

Reliability, convergent validity, and discriminant validity tests are often used to evaluate the measurement model in PLS. Reliability can be assured through composite reliability, Cronbach’s alpha, and factor loading. Factor loadings higher than 0.7 may be viewed as highly reliable, whereas factor loadings lower than 0.5 should be dropped. Convergent validity should be assured when multiple indicators are used to measure one construct. This can be examined by composite reliability (CR), and averaged variance extracted (AVE) by constructs [77]. For convergent validity to be necessary, CR should be higher than 0.7, and AVE should be higher than 0.5. For discriminant validity to be required, the correlation between construct pairs should be lower than 0.90 and the square root of AVE should be higher than the interconstruct correlation coefficients [12]. Data shown in Tables 3 and 4 indicate that all minimum requirements were met; therefore, the results exhibit strong construct reliability and validity. Tables 3 and 4 also show the descriptive statistics and correlation matrix of aggregated data. For each variable, the mean, standard deviation (SD), skewness (M3), and kurtosis (M4) are provided [31].

3.5 Perceived benefits as a second-order formative construct

The multidimensional nature of perceived benefits makes it a formative second-order construct [36]. To examine the appropriateness of treating perceived benefits as a second-order formative construct, a series of activities adopted from Petter et al. [72] was used to verify the existence of a second-order formative construct.

As shown in Fig. 2, the relative weights of the first-order constructs are all significant (p <  0.001). A strong and significant mediating effect of second-order constructs between first-order constructs and dependent variables shows that perceived benefits may represent six first-order constructs with similar predictability but with more compactness. Finally, low variance inflation factor (VIF) values (convenience = 1.692, cost = 1.240, search time = 2.446, order time = 2.289, shopping fun = 1.790, and playfulness = 1.268), all of which are far below the common cut-off threshold of 10 [25], exclude the potential for a collinearity problem.

5.1 Implications to academics

The Theory of Planned Behavior has been widely adopted in past studies of information systems [12, 85] and the behavior of online shopping [40, 71]. When using TPB to discuss the behavior of online shopping, however, past researchers often have overlooked the influence of norm on the behavior of online shopping [71]. In this study, peer norm has a significant impact on both personal attitude and intention. This finding is interesting because online group-buying is a kind of collective buying behavior; therefore, it is believed that peer norm plays an imperative role in this kind of buying behavior.

In addition, whereas past researchers discussed the direct influence of perceived benefits of online shopping on the behavior of online shopping [27], this study simultaneously incorporates the exogenous beliefs that influence online consumer behaviors, such as perceived benefits, trust in the initiator, and social interaction, in online group-buying. Besides, this study adopts the TPB adjusted by Ajzen [4] to discuss the important exogenous variables that influence the behavior of online group-buying as a way to improve the insufficiency of TPB regarding research on online shopping.

5.2 Implications to practitioners

Online group-buying continues to gain momentum in Taiwan. After its initial appearance in BBS, major online shopping websites also introduced this novel shopping mechanism as a way to accommodate consumers. A group-buying website allows consumers with similar needs to gather together, making group-buying a new way to shop online. Sellers who wish to adopt group-buying should first understand the factors behind consumers’ participation in online group-buying and their attraction to it.

In this study, we have determined that social interaction, trust in the initiator, and perceived benefits are important exogenous social commerce factors behind consumers’ participation in online group-buying. For the importance of perceived benefits, our findings suggest that time- and money-saving and convenience are valued benefits of group-buying for consumers. Sellers who are launching group-buying interactions are therefore advised to satisfy these consumer needs and to make consumers feel their purchases are more rewarding than expected. In addition, enjoyment and playfulness in online group-buying are also benefits valued by consumers, indicating that a sense of novelty and mystery is aroused in consumers when they discover a rare product. George [35] also believe the behavior of seeking diversity is one of the important motivations behind consumers’ participation in online shopping. Sellers are therefore recommended to provide unique and diverse products that are rare or novel, as this would make shopping more refreshing to consumers. A vast selection of products on a group-buying website also allows consumers more choices and satisfies their various needs, while ensuring consumers that they are only a few clicks away from getting their desired products. A seller who can make consumers feel that participating in group-buying is enjoyable and fun will then put them in a state of flow where they enhance their attitudes toward group-buying and their beliefs in being able or having the resources to engage in group-buying. In turn, as an added benefit, these consumers may even discuss online group-buying with their friends.

In addition to perceived benefit, it is also important for a seller to consider whether an initiator is trustworthy since trust is difficult to establish in online shopping. If the initiator is not trustworthy, consumers may not wish to join group-buying out of fear of being tricked, and all the group-buying-related benefits provided by the seller would not be experienced by the consumers. Therefore, sellers are advised to establish a “trust mechanism” for online group-buying as a way to improve the trust of the group-buying members in the group-buying website and to reduce uncertainties in the process of online group-buying.

The need for interpersonal interactions is also valued by group buyers, especially in the social commerce context. If consumers feel there is a lack of interpersonal interaction among themselves, their attitudes toward participation in online group-buying may be affected, as they may not believe they have the ability or confidence to engage in online group-buying. In addition, perceived social interaction may also improve the trust of group buyers in the initiator. Sellers are thus recommended to establish a family community in which interactions between the initiator and group buyers and/or among group buyers are facilitated as a way to satisfy the need for interpersonal relationships. This would prompt the initiator to quickly answer questions from the members, making them believe the initiator is active and capable of serving them and, in turn, placing more trust in the initiator.

Our findings also indicate that members’ perceived benefits and social interaction significantly influence their trust in the initiator. Sellers thus need to establish an evaluation mechanism of the initiator as a way to reassure consumers who wish to join group-buying and to reduce any relevant uncertainties. In other words, members’ attitudes toward online group-buying can be improved only when they believe their initiator can meet their expectations, and only then would they discuss group-buying with their friends or persuade them to join group-buying under the same initiator. Moreover, members are also affected by their peers’ opinions. In other words, a member’s attitude and intention to join online group-buying is affected when that member’s friends discuss group-buying. Sellers thus must consider how to establish a good reputation among consumers and increase the number of group buyers through word-of-mouth advertising.

In conclusion, while group-buying has been recognized as a kind of social commerce integration, the three antecedents are considered from this kind perspective. The results also demonstrate that the three antecedents play important role in influencing group buyers’ decision making.

5.3 Limitations and future research

The sample in this study was limited to group buyers in Taiwan, and therefore the generalizability of the findings to other countries may be limited; therefore, this provides an avenue for future research to collect data from the diverse countries. Group-buying is different from conventional electronic commerce in that group-buying, or named group commerce, is kind of social commerce integration, therefore future research is suggested to explore the online group-buying issue from the social commerce perspective. In addition, High involvement purchases would involve differing perception on a variety of the research constructs as opposed to low involvement purchases that these needs to be controlled for or treated as moderating variables in future.