With the rise of live-streaming shopping, this research examines how vicarious learning experiences influence consumers' impulse buying behaviors.. Adopting vicarious learning theory, th
INTRODUCTION
Research background and statement of the problem
The modern economic landscape is undergoing a significant transformation, characterized by the rise of online platforms and innovative business models (Bui et al., 2023; Swchee et al., 201) TikTok has become a global sensation, boasting around 1 billion users and ranking among the most popular applications worldwide A TikTok report on shopping behavior in Vietnam during the 2023 holiday season revealed that 69% of users prefer short-form videos to discover products, while 84% are influenced to purchase brands showcased in these videos (Meltwater, 202) This shift is revolutionizing traditional e-commerce by enabling real-time interactions between sellers and consumers (Aug et al., 2018; Forrester, 2021).
Livestream commerce has become a vital marketing channel for companies globally, leveraging the impulsive psychology of consumers during their buying journey While earlier research focused on customer satisfaction and purchase intent, recent studies highlight how livestream content on social media significantly influences spontaneous purchasing decisions This interactive platform not only promotes products but also creates an engaging environment that excites viewers Impulsive purchases often arise from sudden buying urges, and the ease of shopping through live streams fosters unplanned shopping behaviors, leading consumers to indulge in online shopping The rise of livestream technologies has revolutionized consumer engagement, reshaping how individuals discover and interact with products Both direct and independent vicarious learning through live streams play crucial roles in shaping consumer behavior, especially regarding impulse buying, as more consumers seek real-time demonstrations and peer interactions.
Exploring the intricate dynamics of interactive engagement is essential, as it allows viewers to fulfill their entertainment needs while also gaining indirect knowledge Research indicates that audiences can vicariously learn about various subjects, including health (Song et al., 2022), business (Park et al., 2022), and travel experiences This dual benefit enhances the overall viewing experience and promotes educational growth through entertainment.
Vicarious learning, defined as acquiring knowledge or skills by observing others, plays a significant role in consumer behavior during live streams on platforms like TikTok (Gioia and Manz, 1985) Through coactive vicarious learning, viewers gain direct insights into product usage from live streamers, enhancing their understanding and aiding informed purchasing decisions (Ying Hua et al., 2023; Myers et al., 2018) In contrast, independent vicarious learning allows consumers to absorb experiences and opinions from fellow viewers, contributing to a collective knowledge base that influences their choices (Gioia and Manz, 1985) This blend of learning methods transforms the live streaming environment into a hub of shared knowledge, encouraging immediate and informed consumer decisions (Men et al., 2023; Myers et al., 2018) The real-time interaction with streamers fosters a relationship between vicarious learning and impulsive buying, highlighting the dynamic nature of this interactive marketplace (Sun et al., 2019) As consumers engage with live streamers, the likelihood of spontaneous purchasing decisions increases, illustrating the synergy between vicarious learning and impulsive buying behaviors (Clemen Addo et al., 2021).
This study explores how coactive and independent vicarious learning through live streaming influences impulse buying behaviors among consumers By examining the psychological factors, social influences, and real-time interactions involved in these learning experiences, the research aims to offer valuable insights into the changing nature of consumer decision-making in the digital era.
Research objectives
To contextualize the study’s purpose, we propose two questions that this study aim to answer:
7 Are CVL and IVL from the vicarious learning experience critical components needed for consumers to understand the live streamer?
2 How CVL and IVL influence consumer perceived value and impulse buying behavior in Iivest ream ing sh opping ?
This study aims to explore the role of coactive vicarious learning (CVL) in enhancing consumers' understanding of live streamers within live-streaming shopping environments We will investigate how active engagement with content creators during live streams shapes consumers' perceptions of the personalities behind the content Additionally, we will analyze the effects of independent vicarious learning (IVL) on consumers' attitudes and opinions, focusing on how passive observation of live streams and the behaviors of content creators influence these perceptions.
Our research focuses on the impact of coactive vicarious learning (CVL) and independent vicarious learning (IVL) on consumer perceived value in live-streaming shopping We aim to understand how these learning experiences affect consumers' perceptions of value when engaging with live streamers and participating in shopping activities Additionally, we will analyze the relationship between CVL, IVL, and impulse buying behavior, investigating how these learning experiences drive consumers' tendencies to make impulsive purchases during live streams.
Research contribution
This research enhances the understanding of vicarious learning theory in live-streaming shopping, focusing on how coactive vicarious learning (CVL) and independent vicarious learning (IVL) affect impulse buying behavior By analyzing the interplay between CVL and IVL, the study reveals the intricate dynamics of vicarious learning experiences in e-commerce Furthermore, it examines how these learning mechanisms shape consumers' perceptions of social cues and influence their impulse purchases, highlighting the cognitive and behavioral outcomes tied to vicarious learning in online shopping environments.
This research aims to provide valuable insights for live streaming platforms and merchants, focusing on enhancing customer engagement and encouraging impulse buying The findings on interactivity and EULS1T will guide the design and development of platform features, as well as improve streamer communication strategies through virtual presence and psychological proximity.
LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT 5 2.1 Definitions and theoretical backgrounds
Vicarious learning theory
Vicarious learning is a cognitive process where individuals acquire knowledge through observation rather than direct experience, as highlighted by Bandura (1977) This learning method has been widely studied across various fields, demonstrating its significance in skill and knowledge acquisition (Wedgewood, 2006; Zhou & Hu, 2019) In education, vicarious learning is examined in both classroom and online environments (Schunk & Hanson, 1985; Williams et al., 2022), while in sports, athletes utilize observational practices and video modeling to improve performance (Law & Stc-Maric, 2005; Morgan et al., 2013) The medical sector employs vicarious learning through simulation training and expert procedure observation (Van de Ridder et al., 2008; Burke & Mancuso, 2012), and in the military, it enhances tactical instruction and learning models (Kemple et al., 2000; Litz et al., 2007) Organizational research also highlights the impact of vicarious learning on workplace outcomes, including safety practices and skills development (Tullett et al., 2018; Turner et al., 2022) Recently, the live streaming industry has introduced new interactive vicarious learning experiences (Weibel et al., 2008; He et al., 2022), while marketing contexts explore its influence on consumer behavior through vicarious goal theory and comparative advertising (Lockwood & Kunda, 1997; Kärkkiäinen & Vincent-Lanchor, 2010) Overall, vicarious learning, initially rooted in education, has evolved into a vital area of research for understanding complex, social, and experiential learning across diverse domains (Bandura & Walters, 1963; Greve & Seijts, 2006).
Vicarious learning in live streaming involves acquiring knowledge through the observation of streamers' actions and their consequences, rather than through personal experience Live streaming platforms facilitate this learning process by allowing viewers to watch streamers showcase skills, engage in activities, and interact through chat features Research has shown that viewers can learn about new hobbies, creative skills, and gaming strategies by passively observing live streams (Zagal et al., 2020) As they watch these live demonstrations, viewers develop mental models that inform their understanding of effective behaviors, strategies, and best practices.
Live streaming offers two distinct forms of vicarious learning: Coactive Vicarious Learning (CVL) and Independent Vicarious Learning (IVL) CVL involves active viewer engagement through comments, questions, and multiplayer interactions, while IVL consists of passive observation without direct interaction Both methods facilitate the acquisition of knowledge aligned with viewers' interests and goals The unique interactive and social aspects of live streaming enhance vicarious learning, allowing viewers to quickly grasp concepts through streamers' live trial-and-error experiences, all within a vibrant virtual community This dynamic environment makes vicarious learning in live streaming both engaging and impactful.
Coactive vicarious learning (CVL) is a learning process that occurs through active interaction and observation of others' behaviors (Zhou & Hu, 2019) In the realm of live streaming, the interactive nature of the platform creates ample opportunities for CVL to thrive This interactivity facilitates two-way communication between streamers and their audience, allowing viewers to engage actively by commenting, asking real-time questions, and collaborating during gameplay sessions (Zhou & Hu).
Active observation of streamers' demonstrations, combined with direct feedback through live chat, fosters a collaborative learning environment where viewers can discuss strategies, share experiences, and clarify approaches This interactive engagement enhances social learning by facilitating knowledge exchange and discussion (Sun et al., 2019) Unlike passive observational learning, this immersive experience allows learners to ask questions and co-construct understanding with peers, making it more engaging (Jee & Lee, 2018) Viewers can immediately apply concepts demonstrated by streamers and receive feedback for improvement Research indicates that interactivity boosts sense of presence, relationship intimacy, and learning performance (Wongkil Rungrueng & Assarut, 2020; Zhou & Hu, 2019) Consequently, synchronous observation and guidance through interactivity promote more effective vicarious learning than passive exposure alone.
Various methods have been utilized to measure interactivity, with Steuer (1992) creating scales that assess speed, range, and mapping to evaluate user-content interactions Subsequent research employed questionnaires to analyze factors such as active control and two-way communication (Liu & Shrum, 2002; McMillan & Hwang, 2002) McMillan and Hwang (2002) focused on communication direction, information exchange, and sensory stimuli to gauge user perceptions of interactivity, influencing later conceptual frameworks Additional studies have explored interactivity through constructs related to active control, two-way communication, and synchronicity among participants (Hairong et al., 2009) In this research, we adopt these established frameworks to further investigate interactivity.
Ou ct al (2014) measurement scale through capturing active control, two-way communication and synchronicity between participants.
Independent vicarious learning (IVL) is a form of passive observational learning where the observer does not interact directly with the individual being observed (Bandura, 1977) A key attribute of IVL is EULSIT, which allows consumers to learn independently by reviewing documentation of streamers' past performances without direct engagement (Myers, 2018) EULSIT is defined as the belief that using an information presentation tool aids in evaluating livestreamers by observing their broadcasts and past consumer comments (Ou et al., 2014) This aligns with IVL's one-way learning mechanism, facilitating asynchronous learning from archived livestream performances (He et al., 2022; Williams et al., 2019) These materials preserve the authenticity of informal problem-solving observed through IVL (Zhou & Hu, 2019) Research indicates that EULSIT enhances purchasing decisions by providing access to transaction records and reviews, proving more effective than interactive discussions (Jin et al., 2020) Additionally, exposure to streamers’ unedited shopping experiences has been shown to improve cognitive abilities beyond direct engagement (Cho et al., 2021) Overall, IVL from observation fosters superior long-term learning compared to no modeling, as EULSIT enables exposure to unrehearsed experiences, thereby enhancing learning outcomes (Schunk & Hanson, 1985; Jang et al., 2023).
Virtual presence
Virtual presence is the subjective experience of "being there" in a digital environment with virtual entities (Xu et al., 2021) This concept has gained traction in various research areas, particularly in live streaming commerce (Sun et al., 2019; Ma, 2021; Xu et al., 2021; Chen and Liao, 2022) Despite the importance of different aspects of virtual presence in live streaming, research on their combined effects remains limited (Gao et al., 2018) Some studies have evaluated overall perceived virtual presence using single-item or multi-item scales to measure the intensity of one's virtual experience (Draper et al., 1998; Kim et al.).
Research has explored various behavioral indicators, such as task performance and body movements, to assess levels of virtual presence (Sanchez-Vives & Slater, 2005; Jung & Park, 2017) Furthermore, the use of physiological sensors to measure immersion depth, arousal, and emotional contagion offers additional objective metrics for evaluating presence (Wadlinger & Isaacowitz, 2011; Jin et al., 2022).
Virtual presence has two dimensions, social presence and telepresence, which have been widely adopted in past research (Ou et al, 2014; Gao et al., 2018; Ye et al., 2020;
In this study, we utilize the conceptual framework of virtual presence proposed by Ou et al (2014), which encompasses both social presence and telepresence dimensions This approach enables us to effectively capture and validate perceptual experiences related to virtual interactions, as informed by previous research (Xu et al., 2021).
Psychological proximity
Psychological proximity is defined as the perceived closeness between an individual and another person or object, as noted by Trope and Liberman (2010) Research indicates that this sense of proximity significantly impacts various cognitive and behavioral outcomes, as highlighted by Liberman and others.
Our research conceptualizes psychological proximity in livestreaming through three key dimensions: spatial, temporal, and social proximity Spatial proximity refers to the perceived distances between live streamers and viewers during broadcasts, while temporal proximity evaluates whether live streams feel immediate or distant in real-time Social proximity measures the emotional closeness viewers feel towards streamers, including the strength of parasocial bonds Previous studies have highlighted these dimensions, with Chen et al (2021) emphasizing spatial proximity based on physical distances mentioned in streams, and Jang and Shin (2022) investigating temporal proximity by contrasting real-time streams with pre-recorded videos By exploring these dimensions, our study sheds light on how perceived nearness affects vicarious learning and consumption behaviors in the context of live streaming.
Means-end chain framework
The MEC framework effectively analyzes vicarious learning in live streaming, indicating that product and service attributes can yield beneficial outcomes that align with consumer values (Huang and Chang, 2020) Key attributes facilitating vicarious learning in this context are interactivity and exposure to influencers' trial-and-error experiences, which enhance active engagement and passive observation, respectively (Zhou & Hu, 2019) These elements foster perceived benefits such as virtual presence and psychological proximity, creating a sense of connection with both the live streamer and the community (Wongkitrungrueng & Assarut, 2020) Enhanced immersion and social intimacy through interactivity and exposure influence consumers' intention to make impulsive purchases, a vital aspect of live streaming shopping (Sun et al., 2019) Research underscores the role of learning experiences in improving socialization and presence, with live streaming amplifying these effects through real-time audio/visual content and interaction (Edirisingha et al., 2009; Kehrwald, 2008; Lin et al., 2022) Thus, the MEC framework provides a solid theoretical foundation for understanding how interactivity and exposure to influencers drive vicarious learning, leading to positive consumer outcomes in live streaming shopping environments.
Literature review
No Journal title Research objectives Research methods
Empirical results Limitations and future research
1 The role of vicarious learning strategies in shaping consumers' uncertainly: the case of live- streaming shopping (Jinqi
The study aims to understand how live- streaming technology (interactivity and effective use of live- streaming shopping information presentation tools) impacts consumers' credibility perception of live streamers.
Questionn a ire survey with data collected from 405 consumers who engage in
Psycho logical proXi mity (Supported).
This study has three main limitations First, it only looked at general live streaming shopping perceptions, not impacts of different product types
Second, it only examined two vicarious learning factors - interactivity and
EULSIT Third, the data came solely from Chinese sites, reducing cross- cultural applicability.
2023). live- streaming shopping via an online survey.
H3a Interactivity and EƯLSIT (+) -> Virtual presence (Supported).
To overcome existing limitations, future research should focus on three key areas: testing the model across various product categories, integrating additional vicarious learning elements such as streamer voice interactivity, and gathering data from multiple countries to identify cultural differences in buying behavior These steps will enhance the international applicability and generalization of the findings.
118: Purchase intention (+) —* Purchase behavior (Supported).
2 A moderated mediation model for e-impulse
Examine the mediating role of e impulse buying in
H1 e-ỈB mediates the relationship between e- IBT and CS.
Despite the valuable insights gained, this research faces certain constraints. buying tendency, customer the relationship between c-impulsc ve study through H2 c-IBT is positively
The article highlights the challenge of social desirability bias in survey research related to customer satisfaction and buying tendencies Respondents may provide socially acceptable answers rather than truthful ones, despite efforts to ensure anonymity and reduce bias This limitation affects the reliability of findings, particularly in the context of e-IBT's interaction with various shopping website characteristics and promotional measures.
The study by Parayitam and Anuj examines the effectiveness of websites in influencing e-impulse buying tendencies among respondents in the Union Territory of Delhi It highlights a relationship between website features and customer satisfaction, though the generalizability of these findings may be limited, particularly in rural areas where shopping behaviors could differ significantly The research, conducted by Sharma, Nripendra P Rana, and Yogesh K Dwivedi in 2022, suggests that varying levels of stimulation on effective websites can impact consumer purchasing decisions.
The study evaluates the impact of the ent model and promotions on impulse buying behavior, highlighting the moderating role of regression analysis in this relationship It finds that e-impulse buying is positively correlated with customer satisfaction, emphasizing the significance of online hedonic motives Future research should explore additional dimensions of this relationship to enhance understanding of consumer behavior in the digital marketplace.
The study explores the moderating role of personality traits on the relationship between customer satisfaction and e-impulse buying (e-IB), revealing that hedonic motives significantly influence this dynamic It highlights that higher levels of hedonic motives strengthen the association between e-IB and satisfaction, while lower levels weaken it Additionally, the research emphasizes the importance of social media and networking in shaping e-impulse buying behaviors, particularly in the post-pandemic era, offering insights into shifting consumer preferences in India Furthermore, it underscores the relevance of website trustworthiness and the effects of electronic word-of-mouth (eWOM) on consumers' intentions to continue engaging in e-shopping, suggesting areas for future research.
The analysis of impulsive buying behavior highlights the significant role of social interaction and the urgency created by scarcity during live streaming events Despite the valuable insights gained from quantitative research, limitations exist, such as the social desirability bias in surveys, where respondents may provide socially acceptable answers rather than truthful ones Efforts to mitigate this bias, including ensuring anonymity, have been implemented, but it remains a potential constraint Additionally, the study examines the mediating role of affective reactions in the relationship between impulsive buying urges and consumer behavior, employing dual-stage structural equation modeling (SEM) and artificial neural network (ANN) analysis to explore these dynamics further.
Dwivedi Garr}' cognitive and collected study's focus on respondents from the
Impulsive buying behaviors in the Union Territory of Delhi are influenced by affective reactions, as studied by Wei-1 Ian Tan Additionally, Keng-Boon Ooi highlights that the tendency to shop at malls serves as a moderating factor, which may affect the generalizability of these findings to other regions.
Eugene Cheng- effect of impulsive visiting correlation between regions, particularly rural areas where e-
Xi Aw, buying tendency customers impulsive buying urge shopping behaviors may differ
Bhimaraya Mctri, in India and impulsive buying significantly, with a preference for in
2022) through questionna ires. behaviour (Unsupported).
H3: Affective reactions positively correlate with impulsive buying urges.
H4: Cognitive reactions positively correlate with affective reactions.
H5: Parasocial interaction positively correlates with affective reactions. store impulse buying over online impulse buying.
Future research should focus on the moderating effects of personality traits on the connection between e-impulse buying tendencies and actual e-impulse buying behavior, as this could yield valuable insights.
Additionally, comparative studies between rural and urban customers regarding their e-impulse buying
H6: Vicarious experience positively correlates with affective reactions.
117: Vicarious experience positively correlates with cognitive reactions.
H8: Scarcity persuasion positively correlates with affective reactions.
H9: Scarcity- persuasion positively correlates with cognitive reactions.
Price perception significantly influences consumer behaviors, highlighting regional differences Additionally, examining the role of social media and networking on e-impulse buying in the post-pandemic context can provide insights into changing consumer preferences The trustworthiness of websites and the effects of electronic word-of-mouth (eWOM) on e-impulse purchases are also crucial areas for future research Lastly, exploring the differences between compulsive and impulsive buying behaviors, particularly in relation to the pandemic's effects on consumer habits, offers a compelling direction for further investigation.
Hl 1: Social contagion positive correlates with affective reactions.
Hl2: Susceptibility to social influence moderates the relationship between social contagion and affective reactions. investigation.
4 Boosting customers* impulsive buying tendency in live-
The research aims to explore the influences of parasocial
H1 Product information quality (+) —> Customer engagement (Supported).
This study, while contributing valuable insights, has limitations that open avenues for future research The data on streamer interaction was gathered in China, which may restrict its generalizability due to the unique cultural context Future studies could explore the role of customer persuasion and price engagement across different cultures, as well as the impact of streamer credibility on impulsive buying behaviors This could involve deploying the proposed model to better understand engagement and perception in diverse settings.
(Xi Luo behavior in Utilizing (+) —> Customer settings Given that our study employed , Jun-Hwa livestreaming Structural engagement (Supported) a purposive sampling method, it's
Cheah, Linda D commerce Equation important to acknowledge the potential
Hollcbcck Xin- Additionally, it Modelling H4 Review consistency limitation of generalizability in our
Jean Lim (2024) investigates the structural equation modeling (SEM) for customer findings, addressing limitations through the analysis of mediating roles and engagement Supported by data from 508 participants in India, the study reveals a significant relationship between customer engagement and impulsive buying tendencies Future research should explore cognitive and affective reactions, utilizing resonant methods and diverse sampling approaches to enhance understanding of customer behavior This comparative analysis aims to deepen insights into the moderating effects of engagement on impulsive buying tendencies, ultimately contributing to a comprehensive understanding of customer dynamics.
H7a : IFQ -> CE — IBT Customer engagement mediates the relationship between product information quality and impulse buying tendency (Supported).
H7b ITỌ -> CE -> IBT Customer engagement mediates the relationship between streamer topic.
Future research in Livestream Commerce (LSC) presents exciting opportunities to investigate key factors affecting impulse buying, such as streamer credibility, various streamer types, and resonant contagion By examining these elements, researchers can enhance our comprehension of consumer behavior within LSC environments This understanding will aid in creating more effective models and strategies for marketers and platform operators, ultimately improving interaction quality and impulse buying tendencies.
H7c SC — CE —IBT Customer engagement mediates the relationship between streamer credibility and impulse buying tendency
H7d RCN — CE — IBT Customer engagement mediates the relationship between review consistency and impulse buying tendency
Customer engagement mediates the relationship between resonant contagion and impulse buying tendency
When customers are more receptive to deal proneness, the relationship between customer engagement and impulsive buying tendency is stronger (Supported).
5 How do product recommendations affect impulse buying? An empirical study on WcChat social commerce
Examine how product recommendations on social media affect a user's urge to buy impulsively, by proposing a model based on signaling theory
VC study using regression analysis
Utilization of regression analysis to examine the
HI: Information quality (+) —♦ Cognitive trust in recommender
Cognitive trust in recommender (Supported).
While valuable, this study is limited by its focus on data from China, potentially restricting its generalizability.
Additionally, the use of purposive sampling may limit broader applicability.
Future research could explore additional factors influencing impulse buying in
Li vestream Commerce (LSC), such as brand knowledge, platform credibility, relations!!! p between
Investigating the impact of various product types and categories on vicarious expression streamers can enhance our understanding of product affection and the factors influencing streamer credibility Additionally, examining the role of aesthetic appeal and resonant contagion in media can reveal insights into consumers' urges to buy and their potential for group identification based on shared values This exploration opens promising avenues for future investigations into cognitive and affective trust among consumers.
Data collected H6: Affective trust in through recommender (-) —♦ surveys Product affection from an unspecific d number of social media users.
Urge to buy impulsively (Supported).
—* Urge to buy impulsively (Supported).
Commerce Based on the Stimulus
The study aim to propose a research model based on the stimulus organism-response (S-O-R) framework
Firstly, the questionnaires did not specify a particular live streaming commerce platform, potentially overlooking variations in consumer experiences across different platforms or
Response Io understand (Partial (Unsupported) product categories Future research
Chao-Hsing consumers exhibit notable reactions and behaviors in relation to least squares analysis, suggesting that expertise positively influences their responses To gain deeper insights, comparative studies across different platforms could be conducted Additionally, Lee and Chien-Wen highlight the importance of perceived enjoyment and its impact on various product categories, emphasizing the role of consumer engagement in shaping purchasing decisions.
Chen 2021) factors in live equation (Supported) demographics Additionally, the cross- streaming commerce
And explore the impulsive buying behavior of consumers in live streaming commerce in China. modeling) statistical analysis used for empirical evaluation.
Survey conducted in the Chinese context.
H4: Product usefulness (+) —♦ Perceived usefulness (Supported).
Products (Unsupported). sectional methodology used in this study may not capture dynamic changes over time.
Future research opportunities abound in exploring impulse buying behavior in live streaming commerce Comparative studies across different platforms, product categories, and consumer demographics could offer deeper insights Moreover, investigating
Data collected via 433 valid questionna ires from consumers with shopping experience on live streaming platforms.
H7: Perceived usefulness (-) —ằ The urge to buy impulsively
H8: Perceived usefulness (+) —♦ Perceived enjoyment (Supported).
Perceived enjoyment significantly influences the urge to buy impulsively, as supported by research Exploring impulse buying behavior across different cultures can offer valuable cross-cultural insights Utilizing experimental designs, econometric models, and advanced technologies like big data and AI can reveal new findings and deepen our understanding of impulse buying in live streaming commerce.
To examine the influence of social relationship factors
HI Information tit-to- task ( ! ) —>
Research framew ork and hypothesis development
2.3.1 Attributes of vicarious learning and perceived benefits by customer
Prior research emphasizes the importance of interactive behaviors in live streaming, such as liking, recommending content, and giving virtual gifts, which enhance social engagement and reduce perceived risks (Hamari and Sjoblom, 2017; Yu et al., 2018) These interactions deepen viewers' connection to the community and enhance their sense of social presence (Yu et al., 2018; Xue et al., 2020) The theory of social presence suggests that a platform's social presence influences users' understanding of content and promotes social interaction (Chang and Hsu, 2016; McLean and Osei-Frimpong, 2017; Bonner, 2010) In live streaming commerce, dynamic interactions create diverse experiences, fostering emotional connections and increasing immersion during the shopping process (Van Noort et al., 2012) As interactivity rises, users experience a greater sense of virtual presence, leading to enhanced engagement within the online environment.
Research by Zhang et al (2023) and Luo et al (2022) highlights the significant role of interactivity in enhancing psychological proximity The interactive features of livestreaming foster stronger social connections between sellers and viewers, creating a sense of societal closeness This direct and synchronous interaction diminishes perceived spatial distance (Wu, 1999) and enhances the feeling of temporal proximity (Song and Zinkhan, 2008), ultimately leading to a greater sense of psychological proximity Consequently, our study posits that interactivity has a positive impact on both virtual presence and psychological proximity in the realm of digital commerce.
HI a: interactivity positively affects on virtual presence
Hlh: Interactivity positively affects on psychological proximity^
The effective use of information presentation tools in livestreaming commerce enhances users' perceptions of virtual presence, as highlighted by Kim and Park (2020), who emphasize the role of product visualization in boosting telepresence This is supported by Marsh and Pei (2017), who note that user proficiency with technology enhances the sense of virtual presence through 3D rendering and live demonstrations Additionally, Lee (2021) and Wen (2019) point out that EULSIT is essential for facilitating dynamic user interactions via chat and comments, which strengthens social presence (O'Cinneide and Anderson, 2021; Ye and Kwan, 2021) Furthermore, research by Tang & Shen (2020) and Shi & Choi (2020) highlights the importance of virtual assistants and augmented reality features as key elements in improving both telepresence and social presence in the livestreaming commerce environment.
H2a: EƯLSIT has a positive effect Oil virtual presence online shopping
Recent studies emphasize the significant impact of EULSIT on psychological proximity in online shopping, particularly through social, spatial, and temporal dimensions EULSIT effectively reduces social distances by enhancing product information presentation, overcoming language barriers, and promoting social interactions via features like chat and comments, which fosters trust and encourages customer engagement (Leckie et al., 2020; Han & Jung, 2019) Additionally, research by Park and Kim (2018) indicates that accessible presentation tools facilitate real-time interactions through quick feedback mechanisms Similarly, Xu and Nam (2019) highlight that user-friendly livestream tools reduce temporal distances by optimizing viewer-broadcaster interactions during live events Furthermore, EULSIT enhances spatial proximity by creating a sense of virtual space through visualization features (Jung and Park, 2019), with Zheng and Wang (2020) noting that multimedia effects further shorten spatial distances.
H2b: EULSIT has a positive effect on psychological proximity
Kim and Biocca (1997) demonstrated that enhancing virtual presence in a virtual reality system increases feelings of social richness and intimacy, reducing psychological distance This suggests a connection between virtual presence and basic human needs for belongingness and connection (Nowak, 2001), which further diminishes perceived distance in relationships Gao et al (2022) found that higher levels of virtual presence in virtual reality foster empathy and psychological proximity among users Jin et al (2021) also showed that virtual presence enhances perceptions of relatedness and belonging within virtual communities, supporting the cognitive aspects of psychological proximity (Bakhtar et al., 2018) These findings collectively affirm the hypothesis and contribute to a refined conceptual model proposed by Kociatkiewicz and Kostulski.
In 2021, Alqahtani et al explored the distinct dimensions of virtual, social, and self-presence, highlighting how these elements recursively influence an individual's mental representation of others during interactions This mental modeling plays a crucial role in minimizing psychological distance by establishing shared frames of reference As immersive technologies continue to gain prevalence, their impact on interpersonal dynamics becomes increasingly significant.
(2021) found that virtual presence has an even stronger hand in collapsing physical boundaries in e-commerce context They argue presence will increasingly mediate interpersonal affinity and togetherness online.
H3: Virtual presence positively affects psychological proximity
2.3.2 Perceived benefits and customers* perception of value
Virtual presence enhances engagement by enabling real-time interaction through comments and shares, fostering a sense of connection that increases viewer involvement and perceived value (Hu et al., 2020) This immersive experience boosts credibility and trustworthiness (Arnold and Reynolds, 2003), as consumers feel heard and valued by the brand Consequently, when consumers believe their opinions matter, they are more likely to develop a positive attitude towards the product, resulting in greater satisfaction and brand loyalty.
Virtual presence also improves perceived usefulness by allowing customers to experience how a product or service can meet their needs According to Fang et al
User immersion in activities significantly enhances the ability to recognize values and benefits This immersive experience aligns with user motivations and needs, influencing perceptions of the advantages gained from interacting with various objects Ultimately, virtual presence can positively impact customers' perceptions of credibility, product attachment, and perceived usefulness.
H4: Virtual presence has a positive effect on customers' perception of credibility (4a) and perceived usefulness (4b)
Psychological proximity refers to the perceived closeness or similarity between individuals and entities, influencing customer perceptions of brand credibility (Y Liu et al., 2013) Interactive features in virtual environments, such as live comments and Q&A sessions, enable real-time engagement, allowing customers to actively participate with live streamers (Li et al., 2015) This active involvement creates a two-way communication channel, making customers feel heard and valued (Rice and Love, 1987).
Research by Ah Ram Lee et al (2018) highlights a positive link between psychological proximity and perceived usefulness, significantly impacting customer satisfaction and repeat purchases Consumers engage with live-streamed content, gaining indirect knowledge through vicarious learning about various topics, including business and health (Short et al., 1976; Song et al., 2022; Park et al., 2022; Lu, 2019) Emotional connections between streamers and viewers enhance the perceived usefulness of live streams, fostering positive emotions and reducing feelings of loneliness Additionally, the interactivity of live streaming platforms provides engaging, personalized experiences, further boosting perceived usefulness Ultimately, psychological proximity enhances customer perceptions and engagement with businesses and products.
H5: Psychological proximity has a positive effect on customers' perception of credibility (5a) and perceived usefulness (5b)
2.3.3 Value perception and customer intention to buy impulsively
A positive experience with an e-store significantly increases the likelihood of online purchases (Wells et al., 2011; Zheng et al., 2019) The credibility of a streamer is influenced by their expertise, reputation, and authenticity, leading viewers to trust the promoted products or services as reliable and high-quality (Zeng et al., 2023; Racherla et al., 2012) Furthermore, the authenticity and transparency of the livestreaming environment enhance the streamer's credibility, which can further stimulate viewers' impulse buying behavior (Wells et al., 2011).
Psychological proximity significantly influences impulse buying behavior, as the emotional and cognitive closeness to a product or brand enhances the likelihood of spontaneous purchases (Lee et al., 2018) Live streaming amplifies this effect by fostering a sense of urgency and immediacy (Wu et al., 2023; Racherla et al., 2012) Factors such as emotional attachment, perceived relevance, and situational context contribute to this psychological proximity, ultimately stimulating impulsive decision-making (Singh et al., 2021).
H6: Credibility has a direct and positive impact on impulse buying behavior
H7: Psychological proximity has a direct and positive impact on impulse buying behavior
When customers view a product or brand as credible and trustworthy, it builds a foundation of confidence that can trigger impulsive purchases Research by Fang et al (2018) highlights the significant influence of credibility on consumer behavior, emphasizing its role in facilitating impulsive buying When a brand is seen as credible, consumers are more likely to trust the information provided, leading them to perceive the product as high quality and capable of meeting their needs This perceived credibility fosters an environment that encourages impulsive buying by instilling trust, minimizing perceived risks, and shaping consumer attitudes and perceptions (Yang et al., 2024).
The perceived usefulness of a product significantly influences customers' impulsive buying intentions, particularly in experiential marketing contexts (Racherla et al., 2012) When items are showcased through live, interactive experiences, their perceived value increases, prompting impulsive purchases driven by a desire for engagement (Vonkcman et al., 2017) Customers are more likely to make spontaneous purchases when they view a product as functional or capable of solving a specific problem, as this practical value intensifies their buying intention (Singh et al., 2021; Li et al., 2015).
H8a: Credibility has a positive effect on customers' intention to buy impulsively
H8b: Perceived usefulness has a positive effect on customers' intention to buy impulsively
2.3.4 Customer buying intention and impulse buying behavior.
Furthermore, studies have consistently demonstrated a direct correlation between the strength of buying intention and the likelihood of impulsive purchasing (Sawar et al.,
In 2023, research by Lo et al highlights that a strong intent to purchase significantly increases the likelihood of impulsive buying behaviors (Herzallah et al., 2022) This relationship underscores the psychological complexities of consumer decision-making and emphasizes the importance of strategic interventions aimed at influencing buying intentions to shape impulsive buying patterns (Rook, D W., 1987; Loewenstein and Lerner, 2003) For businesses, understanding the intricate connection between buying intentions and impulsive tendencies is essential for crafting effective marketing strategies.
H8: Urge to buy intention has a positive effect on impulse buying behavior
The proposed research model
Based on the hypotheses discussed above, we propose the following research model:
RESEARCH METHOD
Research process
The cuiTcnt thesis adopts a mixed-methods approach, combining qualitative and quantitative methodologies to ensure a comprehensive research framework Qualitative methods are instrumental in refining the questionnaire's validity and clarity, with a thorough literature review informing the adoption of measurement scales for key constructs These scales were then adapted to fit the research context, and all construct items were translated into Vietnamese to facilitate accurate data collection.
Following the development of the measurement scale, a questionnaire was designed and pre-tested with 20 participants Iterative modifications were made to enhance the questionnaire's clarity before its finalization and distribution.
The second phase of the study included paper-based interviews with participants aged 18-34 who had engaged in livestream shopping across various platforms The survey was conducted on multiple commercial platforms, such as TikTok Shop, Shopee, and Facebook, all of which feature a livestreaming function for sellers The gathered data was analyzed using SmartPLS 3.2.9.
The analysis began with a two-stage approach to evaluate the measurement model, employing reliability indicators such as Cronbach's alpha and composite reliability Additionally, convergent validity was assessed using measures like indicator reliability and average variance extracted (AVE) Discriminant validity was examined through cross loadings, the Fornell-Larcker criterion, and the Heterotrait-Monotrait Ratio (HTMT) Furthermore, common method bias (CMB) was analyzed to ensure the robustness of the research findings.
The evaluation of the structural model involved several key criteria, including the Variance Inflation Factor (VIF) to assess collinearity, the Standardized Root Mean Residual (SRMR) for model fit, and R² and Ọ² to determine predictive power and relevance Hypotheses were tested through a bootstrapping procedure with 5,000 samples to analyze direct, mediating, and moderating effects Additionally, the FIMIX-PLS approach was utilized to investigate the potential influence of unobserved heterogeneity on the research outcomes, as illustrated in Figure 3.1.
- Reliability (cronbach's alpha and composite reliability)
- Convergent validity (indicator reliability and average variance extracted - AVE)
Fornell-Larcker criterion, and the
- The collinearity issues (VIF value)
- The model fit (SRMR value)
- The predictive power (R ‘) and predictive relevance (Q 2 )- Hypothesis testing (bootstrapping 5.000): direct effects
Measurement scale
/Ml items were measured with a live -point Likert scale (I = strongly disagree, 5 = strongly agree).
1 I felt that I had a lot of control over my experience at this live streamer's live-streaming shopping site (ACT1) Ou et al (2014)
2 While I was watching this live streamer's live- streaming shopping site 1 could freely choose what I wanted to see (ACT2)
1 During the live-streaming shopping, this live streamer facilitates two-way communication between him/herself and consumers (TCI)
2 During the live-streaming shopping, this live streamer gives consumers the opportunity to talk to him/her (TC2)
1 During the live-streaming shopping, this live streamer responded to my questions very quickly (SYN1)
2 During the live-streaming shopping, 1 was able to get information from this live streamer very rapidly (SYN2)
1 1 feel confident that the live-streaming shopping’s information presentation tool provides accurate information about this live streamer’s reputation (EUL1)
2 A considerable amount of useful feedback information about the transaction history of this live streamer is available through live-streaming shopping's information presentation tool (EUL2)
3 I believe that the live-streaming shopping's information presentation tool is effective for consumers to know about this live streamer (EUL3)
4 I believe that the live-streaming shopping's information presentation tool is reliable and dependable so as to help me evaluate this live streamer (EƯL4)
1 There is a sense of human contact in this livestreamer's live-streaming shopping site (SP1)
2 There is a sense of personalness in this livestreamer’s live-streaming shopping site (SP2)
3 There is human warmth in this livcstrcamcr's live- streaming shopping site (SP3)
4 There is a sense of human sensitivity in this livestreamer's live-streaming shopping site (SP4)
1 While watching this livcstrcamcr’s live-streaming shopping site, my body was in the room, but 1 felt my mind was inside the world created by this live streamer.
2 While watching this live streamer's live-streaming shopping site, 1 felt that 1 was immersed in the world this livestreamer had created (TP2)
3 This live streamer-generated world seemed to me to be
"somewhere 1 visited" rather than "something I saw"
4 I felt 1 was more in the "real world" than the "computer world" while I was watching this livcstrcamcr’s live streaming shopping site (TP4)
1 While watching live-streaming shopping I felt I was in the same place as the lives!reamer (SPP1)
2 While watching live-streaming shopping 1 felt 1 was spatially close with the livestreamer (SPP2)
3 While watching live-streaming shopping, I fell the livestreamer responded to me closely (SPP3)
4 While watching live-streaming shopping, I felt the live streamer interacted in the same place as I was (SPP4)
1 While watching live-streaming shopping 1 felt I was interacting simultaneously with the livcstreamcr (TEP1)
2 While watching live-streaming shopping I felt the livestreamer was not temporally distant (TEP2)
3 While watching live-streaming shopping, the live streamer gave quick responses to my actions (TEP3)
1 While watching live-streaming shopping, it was easy to become friends with the livestreamer (SOP1)
2 While watching live-streaming shopping, I fell I became more intimate with the livestreamer (SOP2) Lim ct al (2014)
3 While watching live-streaming shopping, I felt the livcstrcamcr held a socially important meaning to me (SOP3)
1 I believe this livestreamer will deliver to me that matches the posted description (CRE1) a product
2 I believe this livestreamer will deliver to me a product according to the posted delivery terms and conditions
3 This livestreamer is likely to be honest (CRE3)
4 This live streamer is likely to be reliable (CRE4)
5 This live streamer is likely to be credible (CRE5)
1.1 find it useful (PU1) Joo, E., & Yang,
3 It can help me find (PU3)
4 1 find it is convenient and high - quality products (PƯ4)
Urge to buy impulsively 1 As I read the product recommendations in this official Wells et al (2011)
(UTBI) account 1 had the urge to purchase items other than in addition to my specific shopping goal (L I BI 1)
2 As I read the product recommendations in this official account, 1 had a desire to buy items that did not pertain to my specific shopping goal (UBTI2)
3 As 1 read the product recommendations in this official account I had the inclination to purchase items outside of my specific shopping goal (UBTI3)
1 When 1 bought (the item), I felt unprompted urge to buy Jeon (1990); it(IBl) Badgaiyan and
2 I couldn't help myself when 1 saw (the item) (IB2)
3 Without intended to I ended up purchasing the thing (IB3)
4 1 bought the item on the heat of the moment (IB4)
Sample and data collection
After consulting with the instructor, we developed a draft survey and conducted a preliminary survey with 20 university students in Ho Chi Minh City to tailor it for the Vietnamese market Once the questionnaire was finalized, our team carried out an online survey using Google Forms to gather data from December 2023 to January 2024 We employed a random sampling method, targeting customers aged 18 to over 65 in Ho Chi Minh City who have actively engaged with livestreams on platforms like Shopee, TikTok, or Facebook for purchasing goods in the past year.
To gather authentic and reliable data on livestream shopping experiences, we initiated our survey by asking participants a preliminary question: "Have you made purchases while watching livestreams on any platforms in the last 12 months?" Only individuals who responded with a "Yes" were eligible to proceed with the main survey.
Our survey focuses on livestream buyers, requiring respondents to identify their preferred livestream platform and their purchasing frequency after watching Only participants with a clear understanding of livestream shopping will complete the questionnaire The findings will be published, and screening questions regarding age, income, and monthly spending on livestream purchases will be utilized to ensure the results accurately reflect relevant customer segments.
CHAPTER ĨV: DATA ANALYSIS AND RESULTS
DATA ANALYSIS AND RESULTS
Sample characteristics
The sampling method employed in this study is simple random sampling, utilizing a Google Form survey questionnaire to gather data from both students and professionals in Ho Chi Minh City A total of 300 samples were collected from participants in the area, with general information about these samples detailed in Table 4.1.
This study presents a comprehensive analysis of a sample demographic comprising
300 individuals, shedding light on their gender distribution, educational backgrounds, employment status, and monthly income levels.
The gender distribution in the sample shows a slight female majority at 58.3%, slightly below the targeted 60% A notable 88.7% of participants have attained a college or university education, while only 9.3% hold a high school diploma Employment data indicates that 83.7% of the respondents are students, with the remaining 16.3% working in office roles, and there are no participants identified as housewives or from other employment sectors.
Regarding the monthly income, most respondents, comprising 58.3%, fall into the
The income distribution among participants reveals that a significant number earn less than 5,000,000 VND monthly, indicating modest earnings In the range of 5,000,000 to less than 9,000,000 VND, 26.0% of participants fall, showcasing a moderate-income level A smaller segment, 10.0%, earns between 9,000,000 and less than 15,000,000 VND, reflecting a higher financial standing Additionally, 5.3% of respondents report incomes above 15,000,000 VND, identifying them as part of a more affluent demographic.
A significant 61.3% of consumers engage in transactions with an average order value below VND 1,000,000, while 16.0% fall within the VND 1,000,000 to < VND 3,000,000 range Additionally, 9.7% of participants are willing to invest in higher-value items, spending between VND 3,000,000 and < VND 5,000,000 Notably, 13.0% make purchases exceeding VND 5,000,000 In terms of shopping frequency, 19.3% participate in livestreaming shopping a few times a week, and 37.7% engage a few times a month, indicating moderate regularity in this shopping trend Conversely, 29.7% prefer to shop via livestreaming only a few times a year, while 13.3% are particularly selective, participating just 1-2 times.
Employment Freq % Monthly Income (*) Freq %
This study utilized SmartPLS 3.2.8 (Ringle et al., 2015) to implement the partial least squares-structural equation model (PLS-SEM) for evaluating the precision of measurement scales and the structural model The findings of the analysis are presented below.
Assessment of measurement scale
The research framework included unidimensional, multidimensional, and single-item constructs, following the two-stage approach recommended by Becker et al (2012) In Stage I, the repeated indicators method was utilized to derive latent variable scores, which were then saved for subsequent analysis In Stage II, these scores served as indicators for their respective constructs The findings regarding the reliability and validity of the studied constructs are detailed in Tables 4.1 and 4.2.
In Stage II of the research model, the reliability of constructs was evaluated using Cronbach's alpha and composite reliability, both meeting the 0.7 threshold (Hair et al., 2017) Table 4.1 demonstrates satisfactory scale reliability, while convergent validity was confirmed with average variance extracted (AVE) values exceeding the 0.5 cutoff, including Active Control (0.833), Two-way Communication (0.836), EULS1T (0.648), Social Presence (0.629), and Telepresence (0.632) However, Synchronicity was excluded from the model due to its AVE being below 0.5 Discriminant validity was assessed through cross-loadings and the Heterotrait-Monotrait Ratio (HTMT), with each indicator loading higher on its respective construct than on others As shown in Table 4.2, the square root of each construct's AVE surpassed its highest correlations with other constructs, and all HTMT values remained below the conservative maximum of 0.737, confirming the overall reliability and validity of the measurement model.
Hierarchical measurement model No of scale items
Alpha CR AVE Factor Loadings/
Alpha CR AVE Dimension loading/Highest cross loading.
Table 4.2 Scale accuracy analysis: Discriminant validity assessment
ACT CRE EUL IB PU SP SOP SPP TP TEP TC UTBI
Urge to buy impulsively (ƯTBI) 0.080 0.133 0.105 0.894 W2 0.124 0.049 0.072 0.079 0.050 0.101
Assessment Stage II t CRE EƯL IB PƯ UTBI
Urge to buy impulsively (UTBI) 0.174 0.172 0.188 0.157
Assessment of structural model
Following the procedure to evaluate the structural model as proposed by Hair et al
(2017), the collinearity issues among each set of predictor variables were firstly checked; all VIF values (see Table 4.3) of less than 5.0 demonstrated that collinearity was unlikely to be a concern.
CRE EUL PU pp VP UTBĨ IB
To assess the quality of the structural model, the SRMR value of 0.076 - asserted a good fit of the model for theory testing (see Figure 4.3) Furthermore, the R2 and Ọ2
In the evaluation of the proposed research model, 70 endogenous constructs were analyzed for predictive relevance and power The Q2 values indicated predictive relevance, as all were greater than zero However, the R2 values demonstrated varying degrees of predictive accuracy: 0.26 (large), 0.13 (moderate), and 0.02 (weak), with specific R2 values for constructs such as Perceived Utility (0.164), Psychological Proximity (0.183), and others, all indicating weak predictive accuracy Despite this, the exogenous constructs displayed substantial explanatory capability, confirming the overall quality of the structural model The subsequent section presents the results of the hypothesis testing.
A t-test utilizing a bootstrapping procedure with 5,000 samples was conducted to assess the direct effects within the research model The results indicated that the direct hypotheses were supported at a minimum confidence level of 95%, while five hypotheses met a confidence level of 90% However, hypothesis H8b was excluded due to a p-value of 0.213, which exceeds the acceptable threshold of 0.05.
Table 4.4 Significance testing results of the structural model path coefficients
Credibility -> Urge to buy impulsively (H8a) 0.108 0.091 0.245 SUPPORTED
Virtual Presence -> Psychological proximity (H3) 0.131 0.083 0.220 SUPPORTED
Perceived Usefulness -> Impulse buying (H7) 0.112 0.053 0.435 SUPPORTED
Urge to buy impulsively "> Impulse buying (H9) 0.178 0.013 0.223 SUPPORTED
Interactivity -> Virtual Presence (Hla) 0.235 0.001 0.308 SUPPORTED
Interactivity -> Psychological proximity (Hlb) 0.285 0.000 0.302 SUPPORTED
Virtual Presence -> Perceived Usefulness (H4b) 0.321 0.000 0.447 SUPPORTED
Perceived Usefulness -> Urge to buy impulsively
Discussion of Result
The current thesis delves into the influence of vicarious learning on impulse buying behavior in the context of live-streaming shopping.
Utilizing the Means-end chain framework, this article suggests that vicarious learning attributes, particularly interactivity and EULSIT, can enhance perceived benefits and value in livestreaming settings The interplay of credibility and perceived usefulness, along with effective information technology utilization, significantly influences impulsive buying urges and behaviors Findings indicate a positive correlation between interactivity and both virtual presence and psychological proximity, supporting earlier research by Hamari and Sjoblom (2017) and Yu et al.
(2018) consider interactive behaviors as crucial elements in enhancing users' perceived benefits.
Virtual presence and psychological proximity, as opposed to interactivity alone, have a stronger impact on credibility and perceived usefulness (Hamari and Sjoblom, 2017;
Yu ct al., 2018) This finding implies that customers strongly demand an immersive and close psychological distance during their online shopping journey (Zhang et al., 2023; Luo et al., 2022).
The study revealed a direct link between the strength of impulsive buying intentions and actual impulsive purchase behaviors This finding aligns with previous research, indicating that customers who are more determined in their purchase intentions are more likely to engage in impulsive buying behaviors (Herzallah et al., 2022).
The perception of value, particularly in terms of credibility, significantly influences impulsive buying behavior by fostering trust and reducing perceived risks (Yang et al., 2024) While credibility enhances consumers' attitudes and perceptions, perceived usefulness does not demonstrate a significant impact on the urge to buy impulsively, as indicated by a p-value greater than 0.1 This suggests a lack of strong evidence for a direct relationship between perceived usefulness and impulse buying in the studied population.
A sample of 75 participants was analyzed, revealing results that contradict previous studies which indicated that perceived usefulness positively impacts purchase intention Research has shown that customers are more likely to make impulsive purchases when they perceive a product as practical or when it meets an immediate need (Li et al.).
The research conducted in Vietnam highlights the significant impact of geographical and cultural contexts on consumer behavior, particularly impulsive buying tendencies Cultural norms, economic factors, and individual values shape these behaviors, suggesting that the findings may not be applicable to other regions or countries Additionally, the study did not account for other intervening variables, such as personal self-control, mood, or promotional cues, which could also affect the relationship between perceived usefulness and impulsive buying behaviors.