The effect of live streaming commerce quality on customers purchase intention extending the elaboration likelihood model with herd behaviour

Full Terms & Conditions of access and use can be found atBehaviour & Information TechnologyISSN: Print Online Journal homepage: www.tandfonline.com/journals/tbit20The effect of live stre

Full Terms & Conditions of access and use can be found at

Behaviour & Information Technology

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/tbit20

The effect of live streaming commerce qualityon customers’ purchase intention: extending theelaboration likelihood model with herd behaviourQin Yang & Young-Chan Lee

To cite this article: Qin Yang & Young-Chan Lee (2024) The effect of live streaming

commerce quality on customers’ purchase intention: extending the elaboration likelihood model with herd behaviour, Behaviour & Information Technology, 43:5, 907-928, DOI: 10.1080/0144929X.2023.2196355

To link to this article: https://doi.org/10.1080/0144929X.2023.2196355

Published online: 31 Mar 2023.

Submit your article to this journal

Article views: 1262

View related articles

View Crossmark data

Citing articles: 2 View citing articles

The effect of live streaming commerce quality on customers’ purchase intention: extending the elaboration likelihood model with herd behaviour

Qin Yang aand Young-Chan Lee b

School of Economics, Jiaxing University, Jiaxing, People’s Republic of China;b

Department of Business Administration, College ofManagement and Economics, Dongguk University, Gyeongju, Korea

This study examines how technology quality, experience quality, and herd behaviour in livestreaming commerce affect customers’ purchase intention We proposed an integrated researchmodel based on the elaboration likelihood model (ELM) and herd behaviour This study usedcovariance-based structural equation modelling (CB-SEM) to analyse data and assess theresearch model and hypotheses We surveyed 872 Chinese customers who have experience inlive streaming commerce, from which the data of 845 were used to test the hypotheses Ourfindings show that good technology and experience quality lead customers to discount theirown information and imitate their peers Customers’ herd behaviour positively affects theirpurchase intention Further, discounting own information positively mediates the indirect linkbetween live streaming commerce quality (technology quality and experience quality),imitation, and customers’ purchase intention This study is the first to combine live streamingcommerce quality and herd behaviour to investigate customers’ purchase intention in livestreaming commerce It highlights the value of incorporating herd behaviour into the ELM andadds to the body of knowledge by providing a deeper insight into customers’ purchaseintention in live streaming shopping It also has managerial implications for live streamingcommerce practitioners to sever the sustainable growth of e-commerce.

Live streaming commerce is a business model in which a streamer sells goods and services directly to customers via online video streams in which the former exhibits, discusses, and responds to audience inquiries in real time A live streaming session may occur on an e-commerce website or social media platform It can be store-specific or brand-specific, and the streamer can also hold live streaming events where they promote products from various suppliers Live streaming commerce combines instant product purchases with customer interaction via chatting or reaction buttons In the last few years, live streaming commerce has greatly impacted the retail industry in China and has become a major sales channel Many traditional e-commerce vendors are transforming their business and products on famous live stream-ing platforms such as Taobao Live, Douyin, Kuaishou, etc According to research data from China Internet Watch, the gross merchandise value (GMV) of e-commerce live streaming is estimated to account for over 20% of China’s online shopping

GMV by 2022 (Source: www.forbes.com; www statista.com).

Research data from Statista.com and iResearchChi-na.com shows that in 2020, e-commerce live streaming in China represented 10.6% of the total online shopping gross merchandise value (GMV) The proportion was estimated to grow to 24.3% in 2023 According to a 2022 survey of Chinese consumers, around 43% of the respondents aged between 25 and 30 confirmed that they watched shopping-related live streaming The sur-vey also showed that 39% of all age groups of respon-dents used live commerce Live streaming is vital for brands to increase sales and for small-scale operators such as farmers to reach consumers more effectively It grew exponentially during the pandemic, encouraging people to shop online and gain interactive and immer-sive experiences amid lockdowns The e-commerce giant Alibaba’s Taobao Live has captured the lion’s share of live streaming consumers, accounting for 68.5%, followed by Douyin and Kuaishou Other major Chinese internet players like JD.com and Baidu are also trying to grow their presence in the market.

CONTACTYoung-Chan Leechanlee@dongguk.ac.krDepartment of Business Administration, College of Management and Economics, DonggukUniversity, 38066, Gyeongju, Korea

2024, VOL 43, NO 5, 907–928

https://doi.org/10.1080/0144929X.2023.2196355

In addition, the‘2022 China Retail Digitalization White Paper,’ jointly published by McKinsey and the China Chain Store & Franchise Association (CCFA), suggests that the onset of COVID-19 has accelerated the adop-tion of new retail technologies and that online channels have flourished and fragmented Retailers need to not only manage online communities (e.g Weibo or Wechat official accounts) but also create strategies for social platforms(e.g Xiaohongshu), and short video live streaming platforms(e.g Douyin).

Compared to traditional e-commerce, live stream-ing commerce offers significant benefits and helps cus-tomers make more informed purchasing decisions On a live streaming commerce platform, a streamer com-municates precise and comprehensive information about products by combining real-time visual rep-resentations of the products with their own move-ments (Gao et al 2021) Further, customers can assess the streamer’s trustworthiness and attractiveness by examining simple clues or instructive inferences in the live streaming environment based on the strea-mer’s appearance, tailored services, and guidance (Wongkitrungrueng and Assarut 2020) Customers can also ask questions spontaneously on the bullet screen and receive live responses from the streamer, increasing the currency of the information available to them (Gao et al.2021) From observing their fellow co-viewers, customers can acquire valuable knowledge and use the bullet-screen remarks and responses of others to make more informed purchasing decisions (Wongkitrungrueng and Assarut 2020; Yu et al.

2018) In sum, live streaming commerce allows custo-mers to better evaluate products and make educated consumption decisions, as it provides them with infor-mation signals and persuasive messages (Zhou, Chen, and Su 2019; Chen, Zhao, and Wang 2020) It is notable that products on live streaming commerce platforms have a much higher purchase rate than those on traditional e-commerce platforms (Lu and Chen2021).

Past studies have explored customer behaviour in live streaming shopping from various aspects such as continuance intention (Zhang et al 2022), stickiness (Li, Li, and Cai 2021), purchase intention (Gao et al.

2021; Lu and Chen 2021; Meng et al 2021; Doong

2021; Park and Lin 2020; Zhang et al 2020; Sun et al.2019), response intention (Gao et al 2021), cus-tomer engagement behaviour (Kang et al 2021; Guo et al 2021; Wongkitrungrueng and Assarut 2020), and urge to buy impulsively (Zuo and Xiao 2021) Those studies have found a number of factors that explain why people use live streaming shopping and buy products from live streaming shopping The

critical factors include trust-related variables (e.g trust in streamers or products), psychology-related variables (e.g attachment to streamers or platforms, psychological distance, etc.), and relationship variables (e.g swift guanxi, tie strength, etc.) However, one crucial phenomenon (i.e herd behaviour) has been overlooked Herd behaviour is frequently a beneficial tool in marketing and, when employed correctly, can result in increased sales and changes to the social structure It has been identified as particularly influen-tial for user behaviour in various information systems (IS) fields For example, Shen, Zhang, and Zhao (2016) extended past research by incorporating the influence of herd behaviour to investigate the adop-tion of online reviews in online communities Their findings showed that herd factors positively influenced user information adoption Vedadi, Warkentin, and Dennis (2021) identified that herd factors substantially affected information security decisions Herd behav-iour was also discussed in the social media area, Mattke et al (2020) analysed click-through and view-through intention based on herd theory In e-commerce, research on herd behaviour is still very rare One recent study by Erjavec and Manfreda (2022) identified herd behaviour as a critical determi-nant of online shopping adoption After a comprehen-sive review of the previous literature, to the best of the authors’ knowledge, a few studies investigated the herd behaviour behind customers’ purchase intention in the context of live streaming shopping As a result, this study aims to fill in the gaps in the research on live streaming shopping.

Furthermore, previous research in the field of live streaming commerce primarily used four frameworks: 1 Technical characteristics-customer behaviour frame-work, a framework that focuses on the influences of a live streaming commerce platform’s technology charac-teristics such as visibility, metavoicing, and guidance shopping affordance (Sun et al.2019) 2 Social charac-teristics-customer behaviour framework, a framework that emphasises the social stimulus generated from the external environment, such as responsiveness and per-sonalisation (Kang et al 2021) 3 Technical & social characteristics-customer behaviour framework, the cooperation between the technology and interpersonal subsystem (Li, Li, and Cai 2021; Zhang et al 2022) 4 Streamer characteristics-customer behaviour frame-work, a framework that integrates the characteristics of the streamer (e.g streamer trustworthiness and strea-mer attractiveness) and custostrea-mers’ behavioural inten-tions (Zuo and Xiao 2021; Park and Lin 2020) For instance, Li, Li, and Cai (2021) explored the effects of social system factors and technical system factors on

customers’ attachment to streamers and platforms and mediating influence of users’ attachment on stickiness Gao et al (2021) analysed how the viewers processed the persuasive message in live streaming commerce from the perspectives of information quality, streamer quality, and system quality Zhang et al (2022) exam-ined customers’ intention to continue using live stream-ing commerce from the point of view of social enablers and technical enablers Additionally, Guo et al (2021) and Wongkitrungrueng and Assarut (2020) also investi-gated the relationship between trust and consumer engagement From a social point of view, Kang et al (2021) looked into how customer engagement changes over time based on the strength of ties in live streaming commerce However, an integrated framework that examines key technological (i.e technology quality), experiential (i.e experience quality), and herd behav-ioural (i.e discounting own information and imitation) determinants of purchase intention in live streaming commerce is lacking Therefore, it is necessary to fill the existing gaps by examining technological, experien-tial, and herding factors to understand how and when purchase intention emerges in live streaming shopping Live streaming commerce represents a connection not only to information technology but also to streamers and peers Therefore, there is a need to analyse the role of technology quality, experience quality, and herd behaviour in the live streaming shopping context to provide researchers and practitioners with a full understanding of how customers develop purchase intention and their border requirements The current literature demonstrates a lack of research in this area, especially concerning the Chinese live streaming indus-try context.

The findings of this study add to the existing body of knowledge in several aspects First, we applied the ELM to investigate the quality-related factors that led customers to form herd behaviours This broad-ened the scope of the study beyond the social facilita-tors and technological enablers that had been discussed in previous research Second, this study extended the ELM to include herd behaviour in a novel research context, demonstrating that herd behaviour influences purchase intention Third, we assessed quality-related factors (technology and experience quality) as second-order constructs, which provided insights into how to interpret customer behaviour on a quality basis Finally, the findings aid live streaming commerce vendors in cultivating their customers’ herd behaviour, which is critical for devel-oping purchase intention, and provide valuable impli-cations for practitioners developing sustainable live streaming commerce marketing strategies.

2 Theoretical background and hypothesis development

2.1 Live streaming commerce

Live streaming commerce offers the convenience of real-time shopping, gives customers comprehensive infor-mation from multiple perspectives, and allows them to make informed choices (Cai et al 2018; Wongkitrun-grueng, Dehouche, and Assarut2020) It has received adequate academic attention in the last few years because of its growing popularity (seeTable 1) Many recent studies investigated customers’ behaviour inten-tions in live streaming shopping scenarios, such as pur-chase intention, continuance intention, and response intention The potential influencing factors include technology-enabled functions or mechanisms, such as interactivity (Kang et al 2021), information quality (Gao et al 2021), personalisation, visibility (Zhang et al.2022; Sun et al.2019), metavoicing, and guidance shopping (Sun et al.2019) Product-related factors are also found to be crucial determinants of behavioural intentions, such as product popularity (Kang et al.

2021), product fit uncertainty, product quality uncer-tainty (Guo et al.2021), and perceived product quality (Chen, Zhao, and Wang2020) Mistrusted live stream-ing platforms or streamers exist since China’s live streaming commerce market is still in the development stage, which is one of the critical reasons customers are reluctant to buy products from live streaming shopping Several scholars mentioned this issues and discussed the factors that influence customer trust Chen, Zhao, and Wang (2020) examined the mechanism behind consu-mers’ trust building and purchasing intention in live streaming commerce They also discovered that in live streaming commerce, there is a trust transfer effect from streamer to product Similarly, Zhang et al (2022) investigated the influence of social and techno-logical enablers on trust and how trust influences consu-mers’ intention to continue using a service in the context of live streaming commerce Wongkitrungrueng and Assarut (2020) developed a comprehensive framework and examined the relationships between customers’ per-ceived value of live streaming and customer trust Their findings showed that trust in products positively affected trust in sellers, which provided additional insights into the phenomena that drive customers to participate in live streaming commerce In addition, because live streaming shopping platforms offer social networking capabilities, social factors like tie strength and swift guanxi were found to positively influence customers’ purchase intentions and engagement in live streaming commerce (Kang et al 2021; Guo et al 2021; Chen, Tsai, and Tang2021).

In summary, most of the research on customer behaviour in live streaming commerce reveals a ten-dency to examine technological or social factors How-ever, live streaming commerce provides a technological platform for customers to buy products and enhances customers’ shopping experience Studies examining key technological, experiential, and herding determinants of customers’ purchase intention are lack-ing This study aims to analyse the quality of live streaming commerce from technical aspects, such as vicarious expression and visibility, and experiential elements, like immersion, emotion, symbolic value, and admiration Additionally, it will investigate the impact of customers’ herd behaviour, such as discount-ing personal information and imitation, on their pur-chase intentions in live streaming commerce.

2.2 Elaboration likelihood model (ELM)

The ELM is based on a dual-process theory and explains how persuasion influences individuals’ attitudes in two ways: using central and peripheral information proces-sing Before making a decision, the central route involves contemplating messages and examining the quality of their arguments, such as evaluating their rela-tive merits and relevance (Bhattacherjee and Sanford

2006) As customers take a peripheral route, they spend less effort judging information related to products

and pay more attention to the opinions of other people, such as comments left by their previous customers and the star rating of the products (Filieri and McLeay2014; Shihab and Putri2019; Tang, Jang, and Morrison2012) Throughout the literature, the ELM has often been used to explain persuasive information behaviour; therefore, this framework provides a strong theoretical and empirical basis for studying behaviour in live streaming commerce This way, the central and peripheral routes would be demonstrated to have a significant influence on the customers’ herd behaviour to purchase from live streaming commerce, and this herd behaviour would further drive them to make a purchase decision The ELM has been used extensively to explain changes in attitudes toward, or perceptions of, specific topics in a wide range of disciplines, including market-ing, psychology, health, education, tourism, human resources, information technology, environment, and entrepreneurship In the field of information systems (IS), the ELM has been looked at from several perspec-tives In Meng and Choi (2019), the ELM concept is integrated into the theory of planned behaviour to investigate tourists’ intentions to use location-based ser-vices Using a combination of the ELM, commitment– trust theory, and social presence theory, Bao and Wang (2021) identified the mechanisms of consumer participation in brand microblogs Based on the ELM, Shi et al (2018) proposed a theoretical framework for

Table 1.Review of the most recent studies on live streaming commerce.

Gao et al (2021)Information completeness, Information accuracy,Information currency, Streamer trustworthiness,Streamer attractiveness, Bullet-screenconsistency, Co-view involvement

Perceived persuasivenessMindfulnessPurchase intention,

Productfit uncertainty, Trust,Product quality uncertainty

n/aPurchase intentionMeng et al (2021)Pleasure emotion, Arousal emotion, Emotional

trust, Admiration

Doong (2021)Reputation, InteractivityTrust, Perceived valuen/aPurchase intentionPark and Lin (2020)Wanghong-productfit, Live content-product fitWanghong trustworthiness,

the systematic investigation of the factors determining individual dissemination behaviours on social network-ing services (SNSs) Zhou (2021) examined users’ inten-tion to adopt informainten-tion in online health communities using the ELM Chen, Tsai, and Tang (2021) investi-gated the influence of informational-based readiness, customer readiness, and social influence on the inten-tion to utilise self-service stores and the moderating effects of social influence The ELM model has also been applied in several studies in the live streaming commercefield For example, using the ELM and trust transfer theory, Chen, Zhao, and Wang (2020) explored how live streaming affects consumers’ trust and pur-chase intention By integrating mindfulness as a mod-erator, Gao et al (2021) constructed an ELM to assess the influences of central and peripheral factors on per-ceived persuasiveness in live streaming commerce Nonetheless, there has been little analysis of customers’ purchase intention in live streaming commerce inte-grating the ELM and herd behaviour Therefore, this study examines customers’ herd behaviour and pur-chase intention by considering technology quality-related factors as central route factors and experience quality-related factors as peripheral route factors.

2.2.1 Central route factors

The live streaming commerce platform comprises exter-nal technological characteristics and features as part of its status as a unique type of information system (IS) As part of the current study, a measurable technology quality of live streaming commerce is proposed as the combination of the technology characteristics of the live streaming commerce platform The technological system’s processes, tasks, and technologies are essential in transforming inputs into outputs (Li, Li, and Cai

2021; Bostrom and Heinen1977) Vicarious expression and visibility are typical representatives of the central route factors as they reveal the quality of technology embedded in the live streaming commerce platform With the development of collaborative properties of social media, advanced technology can facilitate live streaming commerce with excellent customer relation-ship capabilities Customers can communicate with ser-vice providers efficiently and conveniently using the intermediary technological device.

In recent years, some studies have examined the tech-nological characteristics of live streaming commerce For example, Zhang et al (2022) developed a theoretical model to examine how social and technological enablers might influence the development of trust and whether trust affects customers’ continuance intentions Simi-larly, Li, Li, and Cai (2021) proposed a socio-technologi-cal model to explain how live streaming services affect

user stickiness Platform-based services allow customers to participate in exchange-related activities and experi-ences Using conceptual frameworks derived from IS usage and service marketing literature, this study exam-ines the technology quality of live streaming commerce focusing on vicarious expression and visibility Vicar-ious expression refers to the display of a streamer’s con-sumption of products to enable viewers to experience what it is like to consume these products (Yi and Davis 2003; Li, Li, and Cai 2021) Further, visibility refers to the ability to present products visually to the user, which is crucial to the interaction between custo-mers and streacusto-mers in live streaming shopping (Zhang et al.2022).

Chen et al (2019) identified four functions of vicar-ious expression: analogy mapping, trialability, vicarvicar-ious experience demonstrability, and transferring First, ana-logous mapping describes some characteristics that are similar between a viewer and a streamer, which can enrich viewers’ experiences (Yi and Davis2003; Li, Li, and Cai2021) Live streaming can trigger different feel-ings in individuals (Huang, Lurie, and Mitra2009) For instance, the same coat displayed by streamers may evoke different emotions in different customers Through analogy, viewers can see the streamer as a sub-stitute for themselves, thus solving the problem of indi-vidual differences This can effectively compensate for the lack of an individual experience in traditional online shopping Second, trialability refers to the streamers’ ability to test products in response to viewer requests (Moore and Benbasat1991) The third function is vicar-ious experience demonstrability, which means a strea-mer can demonstrate the products’ physical appearance and express their psychological impressions of products Customers can get to know the product information better with the help of a streamer by show-ing the products based on their needs (Li, Li, and Cai

2021) Lastly, transferring entails a streamer taking on the customer’s role For example, streamers can be asked to perform certain actions (e.g trying on a coat) Viewers can imagine what it would be like to use a product through streamers’ displays.

Besides, visibility is another critical technological fac-tor of live streaming commerce Traditionally, e-com-merce makes sellers invisible to their customers (Treem and Leonardi2013), making it difficult for the customers to develop trust relationships and acquire clues to interpersonal communication (Bai, Yao, and Dou2015; Zhang et al.2022) By communicating visu-ally in live streaming, customers can see streamers’ true reflection, reducing their perceived distance (Lv, Jin, and Huang 2018) and enabling social connection (O’Riordan, Feller, and Nagle2016).

2.2.2 Peripheral route factors

In addition to central routes, customers are also influenced by peripheral cues in processing shopping information and developing their evaluation of pro-ducts This study proposed experienced quality as the peripheral route of live streaming shopping, which cus-tomers use to assess product recommendations The experience quality conception considers how customers react to the psychological benefits they expect from live streaming shopping (Hu, Dai, and Salam 2019) Research on consumer behaviour has become increas-ingly focused on customers’ experience (Novak, Hoffman, and Yung 2000; Rose et al 2012; Kumar and Anjaly 2017) The technical-experiential approach of this study provides a better and deeper understanding of how live streaming commerce affects customers’ pur-chasing intentions Watching and participating in live streaming commerce can be emotional and experiential in nature (Hu, Dai, and Salam2019) With live stream-ing commerce, customers can connect and share infor-mation with streamers and other customers As a result, customers develop emotions and sensations that are live, exciting, and simultaneous, and feel fully immersed The creation and delivery of services are dri-ven by the customers’ experiences (Heinonen and Strandvik 2015; Strandvik, Holmlund, and Edvardsson

2012) Accordingly, experience quality is proposed in this study as customers’ experiential feelings, emotions, and sensations associated with live streaming com-merce Based on a literature review of service marketing and IS usage, this study defines experience quality as a multi-dimensional construct, identifying four dimen-sions: immersion, emotion, symbolic value, and admiration.

Live streaming shopping immersion is defined as the customers’ feeling when they become engrossed in, involved with, and absorbed by their experience (Yim, Chu, and Sauer 2017) Customers may be able to identify the values and benefits they can obtain from an activity based on their experience with per-ceived immersion (Fang et al.2018) Emotion describes consumer experiences associated with exchange-based activities in live streaming commerce (Hu, Dai, and Salam 2019) An emotional connection occurs when a viewer is positioned in a fast-paced, interactive chat environment, and those emotions are subsequently expressed in their responses to the live-streamer or others in the chat (Lim et al 2015) An important source of competitive advantage is the customer experience (Schmitt 1999) The customer’s individual evaluation of an experience results in the creation of an individual experience Experience is characterised by ‘uniqueness that sets an activity apart from others’

(Palmer 2010) Moreover, customer experiences that are memorable, unique, and sustainable over time are successful (Pine and Gilmore 1998) When someone shops via live streaming, he/she creates symbolic meanings, social codes, relationships, and identities about him/herself (Firat and Venkatesh 1993) Thus, for customers, an experience that can enhance their identity and reflect their personality is important (Wongkitrungrueng and Assarut 2020) Furthermore, as an aspect of positive psychology, admiration refers to high levels of esteem or praise for notable individ-uals or role models and occurs when people observe someone with exceptional abilities or virtuous behav-iour (Becker and Luthar 2007; Immordino-Yang et al.2009).

2.3 The effect of technology quality on experience quality

Vicarious expression and visibility enable streamers’ charms and emotions to be conveyed directly through the screen, which improves customers’ sense of identity and engagement, resulting in a pleasant shopping experience Live streaming shopping environments with higher-quality technology allow customers to feel more at ease and enjoy a more immersive shop-ping experience (Yim, Chu, and Sauer2017) Feelings and thoughts connected to service are a reflection of how people experience the service This way, vivid rep-resentations of products can evoke emotional responses and assist customers with their decision-making (Li et al 2013) Furthermore, through live streaming shopping, consumers can establish their own identities, symbolic values, social codes, and relationships In light of this, customers value shopping experiences that can enhance their identities through a shopping mode with high technology quality (Wongki-trungrueng and Assarut 2020) Lastly, admiration is another aspect that contributes to attracting customers Customers’ obsession with a streamer can be largely attributed to their admiration for them (Wohlfeil and Whelan2012) This admiration motivates them to pur-chase products displayed by a streamer In this study, we aim to explore how consumers’ perception of tech-nology quality affects their admiration The audience’s admiration for the streamer will be expressed through bullets on the screen after watching live streaming with vivid product demonstration and professional expla-nations Based on the above discussion, we propose the following hypothesis:

H1: In live streaming commerce, technology quality is positively associated with experience quality.

2.4 Herd behaviour

There is considerable discussion regarding herd behav-iour in various fields, including finance, consumer behaviour, and organisational decision-making (Erjavec and Manfreda2022) An example of this would be when decision-makers use information about what other people do even though their private data suggest doing something quite different (Banerjee1992) There-fore, herd behaviour can be described as a form of heur-istic where individuals conform to the preferences of the majority in their environment by choosing similar beha-viours (Çelen and Kariv2004; Antony and Joseph2017) Banerjee (1992) stated that herd behaviour is character-ised by two factors, namely discounting own infor-mation and imitation Discounting their own information refers to the extent to which people do not consider their information and beliefs when making decisions The imitation of herding behaviours involves individuals following other people’s decisions.

Research on herd behaviour focuses mostly on dis-crete decisions across different disciplines, such as decisions on online reviews adoption (Shen, Zhang, and Zhao 2016), online shopping adoption (Erjavec and Manfreda 2022), information security (Vedadi, Warkentin, and Dennis2021), social media herd behav-iour (Mattke et al 2020), and information security message persuasiveness (Xu and Warkentin2020) Cus-tomers can purchase based on their assessment of the situation in live streaming commerce It often involves searching for information and evaluating the results The nature of live streaming commerce is relatively complex; therefore, a detailed assessment requires specific knowledge and a considerable investment of time and energy (Duan, Gu, and Whinston2009; Jasper-son, Carter, and Zmud 2005) A lack of information regarding the live streaming business can raise uncer-tainty over its appropriateness (Sun2013).

It is more effective to follow the herd than to invest one’s own time and efforts This approach assumes that the herd has completed the investigation and assessment process and determined that purchasing from a live streaming commerce platform is reasonable While watching live streaming, customers are more likely to trust the streamer and other customers as they perceive good technology quality and have a plea-sant experience They tend to disregard their beliefs and copy other customers’ actions (Sun2013) The customer ultimately incorporates other people’s behaviour into their preconceived notions and makes a herd-based decision based on group behaviour (Vedadi, Warkentin, and Dennis2021) In summary, the fact that many cus-tomers purchase from live streaming commerce may

signify its popularity Thus, the streamer’s performance and the number of current customers in live streaming commerce may send signals, such as purchasing signals, that influence a customer to purchase from live stream-ing commerce.

2.5 The effect of live streaming commerce quality on discounting own information

Discounting own information in live streaming com-merce involves the degree to which a customer disre-gards their own information when making a purchase decision and consults with other customers instead Different customers feel differently when watching live streaming shopping (Huang, Lurie, and Mitra 2009) Vicarious expression enables customers to see the anchor as a substitute for themselves, and this solves the problem of individual differences This, in turn, can effectively compensate for the lack of personalisa-tion in tradipersonalisa-tional e-commerce By enabling visibility, customers can see the streamers in live shopping, creat-ing psychological closeness and fostercreat-ing social connec-tions (Lv, Jin, and Huang 2018) Thus, live streaming leads to a deeper relationship between customers and a greater level of trust in streamers as they spend more time watching it According to prior research on herding, it can be viewed as a form of social inter-action-based behaviour (Chen, Wang, and Xie 2011) Herding describes how individuals learn from and fol-low other persons’ behaviours Live streaming shopping with high technology quality enables streamers, custo-mers, and other peers to communicate and exchange ideas efficiently In light of this, we anticipate a strong relationship between technology quality and discount-ing our own information Therefore, this study postu-lates the following hypothesis:

H2: In live streaming commerce, technology quality is positively associated with discounting own information.

In addition to advanced technological features, live streaming shopping promises a rich user experience as well An emotional connection is an assessment of an individual’s perceptions and judgments about the characteristics of their partners (Hansen, Morrow, and Batista2002) and associates Through emotional bond-ing, this study forms an emotional association between the streamer, customers, and peers (Kim and Park

2013) Live streaming commerce can enhance the custo-mer’s shopping experience and make it more exciting by enhancing their hedonic experience These positive experiences can help establish an emotional connection among the streamer, customers, and other peers.

Customers who obtain a pleasant emotional state while watching live streaming shopping will be more inclined to participate actively in shopping-related activities and exhibit a more positive attitude toward the products presented In this case, customers are more likely to trust the streamer and other peers when they have a positive mental state and experience, thus relying on the herd and being less attentive to their own information.

While watching live streaming, customers have per-ceptions of themselves, the streamer, and other custo-mers (Wongkitrungrueng and Assarut 2020) People usually tend to shop at places where they can associate with others like themselves In live streaming shopping, streamers and viewers share the same platform Conse-quently, social identification—a self-defining process that indicates belongingness to a group of people (Ash-forth and Mael1989) and establishes self-efficacy (Bhat-tacharya and Sen 2003)—can occur in relation to an individual (streamer) and a larger group (other custo-mers) (Vukadin, Wongkitrungrueng, and Assarut

2018) In this regard, customers will tend to trust the streamer and other customers with the same identifi-cation They may be more inclined to rely on infor-mation recommended by others than their own information Moreover, the admiration for streamers also positively impacts customers’ attitudes, partici-pation, and loyalty (You and Robert2018; Trivedi and Sama 2020) Additionally, customers watching the live streaming will unconsciously show their admiration emotion via the screen bullets; this causes them to res-onate emotionally and further increases their admira-tion Hence, admiration emotion can contribute to the psychological bond between the streamer and viewers (Meng et al.2021) Customers are more likely to build positive relationships with the streamer they admire and have positive attitudes toward the products the streamer recommends Therefore, they are less likely to rely on their own information but trust the streamer instead Hence, the following hypothesis is proposed:

H3: In live streaming commerce, experience quality is positively associated with discounting own information.

2.6 The effect of discounting own information on imitation and purchase intention

A customer’s imitation in a live streaming shopping scenario is defined as the measure of how much the cus-tomer is influenced by other customers’ choices or decisions while watching Customers who discount their own information are more likely to imitate the

actions of others rather than make a decision entirely based on their own information and assessment When a consumer has dismissed their own thoughts, it is appropriate for the customer to replicate other people’s actions in similar circumstances (Au and Kauffman 2003; Thies, Wessel, and Benlian 2016) Therefore, we argue that when customers discount their own information, imitation becomes an acceptable alternative strategy, as customers might conclude that the streamer and other peers provide better and more complete information to them on everything related to live streaming shopping Hence, we propose the follow-ing hypothesis:

H4: In live streaming commerce, discounting own information is positively associated with customers’ tendency to imitate the behavior of others.

Additionally, we propose that customers’ decision to purchase products and accept services from live streaming commerce is positively associated with herd factors This is consistent with the notion that herding may have an effect on how individuals make decisions (Duan, Gu, and Whinston 2009) Accord-ingly, we suggest that discounting one’s information may positively affect purchase intention in live stream-ing commerce When customers discount their own information, they need to seek external information to increase the likelihood of making the right choice (Chen and Chaiken 1999) When it comes to live streaming commerce, this means that customers are likely to refer to and even rely on online recommen-dations from other audiences (Zhang et al 2014) As a result, external information’s influence may outweigh their own evaluation’s influence in this scenario In turn, consumers will be more inclined to accept sug-gestions from other audiences and become intend to buy from live streaming commerce Thus, we propose the following hypothesis.

H5: In live streaming commerce, discounting own information is positively associated with customers’ purchase intention.

2.7 The effect of imitation on purchase intention

Banerjee (1992) described a logical strategy for imitation as‘adopting the behavior of others.’ In the context of this study, it means choosing to acquire things based on the suggestions of other customers on a live stream-ing commerce platform Usually, to avoid bestream-ing per-ceived as incompetent if investments generate low returns, somefinance investors replicate the investment strategies of professional investment managers (Scharf-stein and Stein1990) A person might prefer the chances

of being mistaken with everyone else over the risk of making an uncommon prediction that turns out to be the only correct one (Graham1999) and may consider imitation to be an effective approach for dealing with uncertainty (Field et al 2006) Similarly, a variety of informational cues, including online reviews and evalu-ations, are frequently provided in live streaming shop-ping context The use of these cues may have an influence on herd behaviour and lead to a strong ten-dency to imitate others’ behaviours Customers who have a strong proclivity to imitate others’ actions nor-mally use online informational cues to determine the quality and popularity of products (Park, Lee, and Han 2007) By evaluating how other audiences have assessed and reviewed products, customers can learn about and follow the purchasing decisions of others As Zhang and Liu (2012) suggested, people tend to jus-tify their imitating behaviour by learning deliberately from others Research on the adoption of an IS also pro-vides empirical evidence of the relationship between imitation and behavioural intention (Sun 2013) In light of these considerations, we suggest that imitation may influence customers’ purchase intention in live streaming commerce Consequently, we propose the fol-lowing hypothesis:

H6: In live streaming commerce, imitation is positively associated with customers’ purchase intention.

3 Methodology and research design

3.1 Measurement development

We have nine variables in the questionnaire, namely, vicarious expression (VE), visibility (VIS), immersion (IMME), emotion (EMO), symbolic value (SV), admira-tion (ADM), discounting own informaadmira-tion (DOI), imi-tation (IMI), and purchase intention (PI) All measurement items were developed from the estab-lished scale and revised appropriately according to the current study context We adopted measurement items of vicarious expression and visibility for technol-ogy quality from the studies of Li, Li, and Cai (2021) and Zhang et al (2022) Further, we developed the measure-ment items of immersion, emotion, symbolic value, and admiration for experience quality by reviewing the works of Hu, Dai, and Salam (2019), Wongkitrungrueng and Assarut (2020), and Meng et al (2021) The items assessed herd behaviour (i.e discounting own infor-mation and imitation) by referring to the works of Erja-vec and Manfreda (2022) and Mattke et al (2020) Lastly, we measured and developed the items of custo-mers’ purchase intention according to the study of Gao et al (2021) A Likert five-point scale (from 1 =

strongly disagree to 5 = strongly agree) was applied to measure these items.

To verify these hypotheses, we conducted a survey to evaluate our research model As we collected the survey data in China, wefirst translated the original items into Chinese and then adopted the back-translation to ensure the accuracy of the items’ intended meaning between English and Chinese Moreover, to ensure the survey’s validity, we invited two professors in the e-commerce major and two live streaming e-commerce platform managers to review the questionnaire After the review, one item measuring admiration and one item measuring emotion were removed The valid measurement items are presented in Appendix A Then, the formal questionnaires were distributed to the potential respondents.

3.2 Sampling and data collection

Customers who have had live streaming shopping experiences are the target respondents of this study, regardless of the categories of the products In 2022, the total revenue of China’s e-commerce live streaming sector is projected to reach RMB 1.2 trillion (US$180 billion) with a total of 660 million viewers Thisfigure is expected to grow to RMB 4.9 trillion (US$720 billion) in 2023, according to a 2021 iResearch report This will account for 11.7 percent of total e-commerce sales in the country, injecting new impetus into the economy Because the live streaming commerce market in China is rapidly developing and China has a large population, this study focused on investigating Chinese customers and decided to collect data in China.

The survey was conducted via a famous and reliable online service, Wenjuanxing (www.wjx.cn) To main-tain the quality and precision of the questionnaires, we used the premium service provided by the survey provider In the questionnaire, wefirst ask the respon-dents whether they have experience shopping on live streaming commerce platforms If their answer is‘yes,’ they can continue the survey; however, if their answer is ‘no,’ the survey will stop automatically Random sampling was applied for this study to target all the potential live streaming commerce experienced custo-mers The surveying period lasted from early February to late March 2022 Overall, 872 participants took the survey After collecting the data, we checked the survey sources and found that our collected questionnaires covered most of the provinces of China To keep the val-idity of the questionnaires, we checked the IP address and time spent by each respondent in answering the survey Finally, 845 valid questionnaires remained The effective recovery rate was 96.9% The sample size

for the data was determined based on the guidelines pre-sented by Ahmed et al (2021), which state that a sample size of 50 is regarded as inadequate, 300 as good, 500 as very good, and 1000 as an excellent sample for factor analysis Therefore, the sample size of this study is sat-isfactory for performing estimations.

As shown inTable 2, 381 are male (45.1%) among all valid samples, and 464 are female (54.9%) Most of the respondents are aged between 20 and 29 years (n = 345, 40.8%), and most of them have an undergraduate education background (n = 437, 51.7%) As to their occupations, 198 of them are students (23.4%), 174 are self-employed (20.6%), and 338 are working in private or state-owned companies (40%) Further, 498 respon-dents have a monthly income between 3000–9000 RMB (59%) As for the use period of live streaming shopping, most have used it for more than one year (n = 502, 59.4%) Detailed information on demographic characteristics can be found inTable 2.

3.3 Preliminary analysis

Podsakoff and Organ (1986) suggested that common method variance may exist in single-source data We performed Harman’s single-factor test to check com-mon method bias (CMB) in the collected data All the

measurement items were loaded into a principal com-ponent without rotation It is suggested that the pro-blem with CMB exists if the total variance of one factor exceeds 50% The first factor accounts for 33.14% of this study’s variance (less than 50%) Thus, the data do not have a problem with CMB.

In addition to CMB, we checked the normality of the distributions of all the major research variables Hair et al (2010) and Byrne (2013) argued that data is con-sidered to be normal if skewness is between−2 to +2 and kurtosis is between−7 to +7 The values of skew-ness in this study range from −1.436 to −1.026, and the values of kurtosis range from 1.206 to 3.12, indicat-ing an acceptable level Therefore, we continue to adopt the variables in the following analyses A multicollinear-ity test is the next step to determine whether there are similarities between the variables in the proposed model Multicollinearity refers to a situation in which two or more predictor variables strongly correlate and do not contribute distinct information to the regression model When the degree of correlation between vari-ables is sufficiently strong, it can create difficulties in both the fitting process and the interpretation of the model One method for detecting multicollinearity is to use the variance inflation factor (VIF), a statistic that evaluates the correlation between predictor vari-ables in a regression model By reviewing the works of Liang et al (2012) and Leong, Jaafar, and Ainin (2018), the existence of multicollinearity was evaluated via the VIF and collinearity tolerance Hair et al (2016) and Gao et al (2021) suggested that when the VIF was below 10, and the tolerance was above 0.10, this indicated that there was no multicollinearity pro-blem The VIFs of the items measured in the current study range from 1.376 to 1.726, which means multicol-linearity is unlikely to be a problem.

4 Data analysis and results

Considering the research model, research purpose, and data characteristics, we adopted covariance-based struc-tural equation modelling (CB-SEM) for assessing the research model (Gefen, Rigdon, and Straub 2011) since it can provide us with complete resources for ver-ifying the hypotheses CB-SEM is used mostly when we have an existing theory to test, whereas PLS-SEM (par-tial least squares structural equation modelling) is appropriate for theory building and prediction in the exploratory stage We proposed an integrated research model based on the ELM and herd theory, which are established theories Therefore, CB-SEM is used in our study It is a preferred method when the goal is theory testing or theory confirmation Second, in the case of

Table 2.Respondents’ demographics.