Abstract

Electronic word of mouth (eWOM) on social networking sites (SNS) has become crucial in
shaping online consumers’ purchase intentions, reflecting the growing reliance on rich nature
of user-generated content and recommendations shared on these platforms. This study aims to
investigate the intricate relationships among different factors of eWOM information,
including information credibility, information usefulness, information adoption, attitude
towards information, and their impacts on online purchase intention. This study draws upon
the Information Adoption Model (IAM) and the Information Acceptance Model (IACM)
foundation theory which reveal the steps individuals take in adopting new information, and
focuses on how perceptions and attitudes towards information shape acceptance. The study
used a quantitative approach, utilizing a structured questionnaire to collect data from 337
Vietnamese individuals actively engaged in eWOM activities on various SNS platforms.
Statistical analysis was conducted using partial least squares structural equation modeling
(PLS-SEM). The findings show that information credibility significantly influences both
information usefulness and adoption. Interestingly, while information usefulness shapes user
attitudes, it does not directly impact online purchase intention. Similarly, a positive attitude
toward eWOM does not necessarily lead to information adoption. This suggests that
consumers may hold favorable views of eWOM content without actively adopting it.
Information adoption emerges as a critical determinant of online purchase intention,
emphasizing the importance of promoting user engagement and trust in eWOM information.
These findings contribute to a deeper understanding of online consumer behavior and provide
valuable guidance for stakeholders seeking to leverage eWOM effectively in digital
marketing.
Keywords: Electronic word-of-mouth (eWOM), Social networking site (SNS), Information
credibility, Information usefulness, Information adoption, Online consumer behavior
1. Introduction

The bidirectional communication inherent in Social Networking Sites (SNS) creates a

dynamic social ecosystem that facilitates extensive interactions among users and a thriving
sea of information where ideas, opinions and conversations converge [1]. Within this realm,
users navigate multifaceted dimensions where online experiences are shaped, and decisions
are influenced by the exchange of ideas and opinions [2].

Social networking sites transcend mere platforms for social engagement, evolving into arenas
where users engage in shopping activities [3]. Delving into a wealth of information allows
users to extract the necessary information, gaining profound insights into their purchasing
journey from vast sources of information [4]. However, navigating this digital abundance
poses challenges, particularly in distinguishing between authentic information and
misinformation. In this virtual complexity, electronic word-of-mouth (eWOM) emerges as a
guiding beacon, empowering users to make informed online purchasing decisions [5]. Unlike
traditional word-of-mouth confined by geographical limitations and social circles, eWOM
penetrates the boundless digital terrain, reaching a wider audience with extraordinary speed
and scale [6]. Moreover, the richness of eWOM content fosters a robust information
landscape. Consumers can gauge real-world experiences and compare offerings across the
marketplace. This empowers consumers to move beyond superficial attraction and engage in
conscious, informed purchasing behaviors [7].
However, this empowerment is complicated by the rise of misinformation and manipulative
online practices because not all instances of eWOM are equivalent [8]. In the era of digital
omnipresence, companies often construct elaborate schemes, disseminating exaggerated or
false product information, exploiting customers for financial gain [9]. Amidst these changing
circumstances, the credibility of eWOM emerges as pivotal a factor that significantly
influences consumer decisions. The inherent authenticity, expertise, and trustworthiness
characteristics of credible eWOM, contribute to establishing the foundation for disseminating
reliable information [7]. This credibility-driven information ecosystem empowers consumers
to make conscious choices, elevating their decision-making beyond mere superficial appeal.
Additionally, the positive impact of credible eWOM extends to fostering a sense of
communal experience and trust, where individuals find resonance in the recommendations
and testimonials of others. This communal validation acts as a potent catalyst, significantly
influencing online purchasing intentions [10].

Various research studies have explored the roles of credible eWOM on diverse online
platforms. Matute [11] proposes a model highlighting the influence of perceived credibility
and usefulness on purchase intention. Verma and Dewani [12] provide a comprehensive
framework and literature review on eWOM credibility, revealing its significant influences on
the decision-making process of consumers. Nofal [13] shed light on the role of weakly
attached eWOM (W-tie) and strongly attached eWOM (S-tie) in shaping consumer behavior.
Notably, the Information Adoption Model (IAM) by Sussman and Siegal [14] and the
Information Acceptance Model (IACM) proposed by Van Der Heijden [15], which have
played crucial roles in elucidating the complex interconnections among user perceptions,
attitudes, and intentions concerning information technologies, have gained widespread
acceptance within the context of eWOM information analysis. Recently, Ngo [16] integrated
these models and introduced the factor of information quantity to analyze the impacts of
information quality, quantity, credibility, need of information and attitude towards
information on purchase intention, elucidating the mediating roles of eWOM information
usefulness and adoption. However, the complex relationships among these factors and the
mechanisms guiding consumers’ evaluation of eWOM credibility and subsequent decision-
making processes have not been thoroughly investigated. Therefore, this study aims to bridge
this gap by providing a nuanced examination of the cognitive processes that individuals
engage in when evaluating eWOM credibility and deciding whether to adopt it. More
specifically, the research examines the interrelationships among information credibility,
attitude toward information, information usefulness, information adoption, and the purchase
intention of online consumers on SNS platforms.

By integrating insights from the Information Adoption Model (IAM) and the Information
Acceptance Model (IACM), this study aims to enhance our understanding of how credibility
assessments influence other elements of eWOM evaluation. Moreover, the research seeks to
distinguish between factors like attitude and information adoption, which are often conflated
in existing literature. This distinction is expected to offer a more comprehensive
understanding of how favorable perceptions of eWOM lead to tangible consumer behaviors.
Ultimately, the study aims to provide practical insights for businesses and marketers
interested in harnessing the potential of credible and influential eWOM to enhance
engagement and drive online sales.
This paper is structured into five main sections as follows. First, the literature review delves
into previous research pertaining to eWOM, with a focus on seminal frameworks such as the
Information Adoption Model (IAM) and the Information Acceptance Model (IACM), along
with key factors influencing eWOM dynamics. This section lays the groundwork for
developing the research framework and hypotheses. Second, the methodology section
outlines the approach taken in this study, detailing participant selection, instrument
development, data collection methods, and analytical techniques employed. Third, the results
section presents the main findings of the study, focusing on examining the proposed
hypotheses and evaluating the research model. Fourth, the discussion section interprets the
study's results, comparing them to existing findings, highlighting novel insights, and
elucidating theoretical and practical implications. Additionally, it addresses any discrepancies
between our findings and previous research, while also acknowledging the study's limitations
and offering recommendations for future research endeavors. Finally, the conclusion
succinctly summarizes the study's objectives, key findings, and contributions to the field.
2. Literature review
2.1. Electronic word-of-mouth (eWOM)

Electronic Word-of-Mouth (eWOM) within the context of Social Networking Sites (SNS)
refers to the digital sharing of user-generated content, encompassing information, opinions,
recommendations, and evaluations concerning products, services, or experiences. This
dissemination involves the voluntary and typically informal exchange of information among
members of online communities, taking advantage of the interconnected and interactive
features of SNS platforms [17,18]. In the contemporary era, an increasing number of
consumers turn to SNS platforms to seek information. According to Erkan and Evans [19],
users can easily express their viewpoints through various mediums like images, text, or
videos, using mobile apps or reposting functionalities. This facilitates the rapid spread of
information through eWOM within SNS. Within these platforms, eWOM plays a crucial role
in shaping consumer perceptions, influencing decision-making processes, and fostering
virtual communities where individuals actively engage in information exchange, thereby
nurturing a dynamic and influential online environment conducive to consumer-driven
discourse [20,21]. Recognizing its impact, businesses widely embrace SNS as an efficient
platform for eWOM to distribute information about their products or services [20].
2.1.1. Information Adoption Model (IAM)

The Information Adoption Model (IAM), as introduced by Sussman and Siegal [14], is a
derivative of the Technology Acceptance Model (TAM). It provides a structured approach to
understanding how individuals integrate new information technologies into their decision-
making processes. The IAM is particularly relevant in the context of online shopping within
Social Networking Sites (SNS), where the exchange of information and social interaction are
integral to the user experience [22]. It elucidates the cognitive evaluation of electronic Word-
of-Mouth (eWOM) credibility and its impact on consumer purchasing decisions. This focus is
well-suited to the nuanced nature of eWOM in online environments, where users are often
encountering a multitude of information [23]. The model suggests that users engage in a
mental filtering process, assessing eWOM credibility against personal needs and social
interaction contexts. Empirical studies applying the IAM framework have explored the
influence of eWOM credibility on purchase intentions. For example, Rahaman [3], examining
consumer perceptions of online review or platform trustworthiness, provides insights by
applying IAM and TAM theories, highlighting factors like source, message, and media
credibility shaping trust. Considerations of social capital in eWOM, where reviews from
known or respected individuals carry greater influence, are crucial. Similarly, Bueno and
Gallego [23] identified service quality, e-commerce satisfaction, argument quality, and
source credibility as key determinants in consumer-to-consumer (C2C) domain. These factors
underscore the critical role of trust in building successful consumer relationships on C2C
platforms. The tourism industry presents a unique context for the application of IAM, where
eWOM significantly influences consumer attitudes toward travel destinations. For instance, a
study by Tapanainen [24] explored eWOM's impact within the tourism sector, revealing
positive effects of information quality, credibility, and usefulness on consumer attitudes
toward travel destinations. Collectively, these IAM-based studies underscore the critical role
of credibility assessments in shaping eWOM adoption and subsequent attitudes toward
products, services, and destinations. They highlight IAM's utility in understanding the
complex processes of information evaluation and decision-making within the SNS context,
setting the stage for nuanced investigations on eWOM.
2.1.2. Information Acceptance Model (IACM)

The Information Acceptance Model (IACM), an extension of the Information Adoption

Model (IAM) developed by Van Der Heijden [15], provides deeper insights into the complex
dynamics of user engagement with information technologies. It facilitates a thorough
examination of user perceptions, attitudes, and intentions, crucial for understanding
information system acceptance. Particularly within the context of electronic word-of-mouth
(eWOM) on Social Networking Sites (SNS), the IACM proves invaluable for exploring how
factors like information credibility, interpersonal trust, and social influence collectively shape
user acceptance and drive online transaction behaviors [25]. Empirical research utilizing the
IACM extensively explored the impact of eWOM on consumer purchasing intentions.
Findings consistently highlight the significant role of information credibility, perceived
quality, necessity, and overall attitude toward information in influencing consumers'
decisions to purchase goods or services [26,27]. This body of evidence firmly supports the
idea that eWOM, when perceived as credible and high-quality, substantially influences
consumer behavior in the digital marketplace. Further studies, such as Park's [28]
investigation, delve into how eWOM influences customer loyalty through the IACM
framework, examining the connections between user attitudes towards information, its
credibility, quality, and usefulness with loyalty outcomes. Similarly, Tjongirin [29] research
applies the IACM to assess the relationship between eWOM and purchase intentions on
social media, emphasizing the critical role of information usefulness in consumer behavior.
The IACM's proven utility in eWOM research underscores the importance of examining the
multifaceted ways users perceive and respond to different dimensions of eWOM. This study
builds upon this foundation by applying the IACM framework, and aims to contribute to a
more nuanced understanding of how eWOM truly influences consumer behavior on SNS
platforms.
2.2. Theoretical framework and hypotheses
development

In this section, the hypotheses are formulated based on an extensive review of literature and
prior studies examining the relationships among the relevant variables, leading to the
development of the theoretical framework. Drawing on the fundamental theories of the
Information Adoption Model (IAM) and the Information Acceptance Model (IACM), this
research investigates the pivotal role of information credibility (IFC) within the context of
electronic word-of-mouth (eWOM) on social networking sites (SNS). Specifically, the study
rigorously examines a series of hypotheses, providing a comprehensive analysis of how IFC
impacts key factors, including information usefulness (IFU), information adoption (IFA), and
attitude towards information (ATI), all of which are integral to shaping online purchase
intentions (PI).
2.2.1. Information credibility (IFC)

Information credibility (IFC) is conceptualized as the audience's perception of the

trustworthiness and expertise of the information source [12], is a critical factor influencing
how eWOM is processed within online environments. Through the lens of the Information
Adoption Model (IAM), information credibility serves as a cognitive gatekeeper, impacting
how users evaluate the suitability and trustworthiness of eWOM messages. Individuals
engage in a critical assessment process, seeking source credibility cues to determine whether
the information merits their trust and subsequent acceptance into their decision-making [30].

This perception is particularly salient in the context of electronic word-of-mouth (eWOM),

where messages are often regarded as emanating from users who are both knowledgeable and
experienced, thus enhancing their perceived helpfulness and informativeness [31,32]. The
trustworthiness of the eWOM source is pivotal; elements such as the sender's online
reputation, a track record of consistent reviews, and the openness about any potential biases
significantly influence the perceived usefulness of the information [33]. Keshavarz [34]
proposed a framework for evaluating the credibility of social media information, highlighting
the importance of source-related dimensions. Building on this foundation, Aghakhani [31]
provides a comprehensive analysis of how review quality defined as a composite of review
comprehensiveness and review topic consistency and source credibility interact to affect
review usefulness. Therefore, the following hypothesis was developed.
H1

Information credibility positively influences information usefulness.

Electronic word-of-mouth (eWOM) credibility is increasingly acknowledged as a reliable

form of social validation that significantly impacts online decision-making processes [35,36].
When consumers perceive both the source and content of eWOM as trustworthy, it fosters a
heightened sense of confidence and security, positively influencing their propensity to engage
in online purchases [37]. Research by Amarullah [38] reinforces this idea, demonstrating
that credible eWOM can notably reduce perceived risks associated with online transactions.
By effectively addressing potential concerns, eWOM credibility not only builds trust but also
alleviates consumer hesitancy, thereby facilitating the completion of online purchases. For
instance, in the domain of health-related information, cognitive trust and perceived value are
important. Users actively seek information they perceive as accurate and valuable, shaping
their health decisions and behaviors. Motivations for seeking such information vary from
personal health improvement to caregiving and social interaction needs [39]. Therefore, the
following hypothesis was developed.
H2
Information credibility positively influences online purchase intention.

The credibility of eWOM is a pivotal factor in shaping consumer behavior, acting as a

catalyst for the adoption of information [40]. Positive eWOM information about a product,
especially when originating from trusted sources like family and friends, can prompt
individuals to consider trying the product themselves [31,41]. This aligns with the
fundamental premise of the IACM, which posits that credible information reduces cognitive
barriers associated with decision-making [42]. Credible eWOM also can simplify the
decision-making process by providing easily accessible and dependable information, thereby
reducing the cognitive effort required for research or evaluation, thus increasing the
likelihood of adoption [43,44]. Furthermore, the persuasive power of eWOM is amplified
when it corresponds with the consumer's existing beliefs and aligns with other
recommendations, further enhancing its credibility and the subsequent adoption of
information [42]. On the other hand, the technological aspects of AI-driven websites, such as
visibility and interactivity, contribute to the creation of an engaging online environment
conducive to a flow state. This state is facilitated by the seamless integration of visibility and
interactivity features, which are designed to captivate users and encourage prolonged
engagement [45]. Therefore, the following hypothesis was developed.
H3

Information credibility positively influences information adoption.

2.2.2. Information usefulness (IFU)

The usefulness of eWOM content is a critical determinant in shaping the attitudes of

recipients towards the information conveyed. Information usefulness (IFU) refers to the
perceived degree to which eWOM content assists recipients in decision-making or problem-
solving [21]. When eWOM information successfully caters to the recipient's requirements by
offering comprehensive, relevant, and reliable information, it fosters a positive perception of
the content [[46], [47], [48]]. eWOM information that resolves uncertainties, alleviates
decision-making dilemmas, and evokes positive emotions tends to be valued and appreciated
by recipients [49]. Valuable eWOM can strengthen the relational bond between the recipient
and the peers providing the information and build trust in the received messages and
opinions. This bond relies on the credibility of the message and the trustworthiness of the
peer delivering the eWOM, which subsequently enhances the recipients' trust in the
information received. This social connection amplifies attitudes toward the information
[12,50]. Moreover, Pang and Zhang [51] discovered that the frequency of mobile social
media usage can positively impact social trust, which, in turn, can predict users' engagement
with eWOM. Attitude towards eWOM can partially mediate the effect of social capital on
eWOM engagement, indicating that the usefulness of eWOM content is likely to enhance
users' attitudes and engagement with the information. Therefore, the following hypothesis
was developed.
H4

Information usefulness positively influences attitude towards information.

The perceived usefulness of eWOM information has been demonstrated to significantly
reduce the perceived risk and uncertainty associated with online purchases, thereby
encouraging a higher intention to purchase [11]. This outcome is attributed to the specific
benefits and features of products that eWOM information provides, facilitating comparisons
between products and aiding decision-making processes, ultimately leading to actionable
intentions towards online purchases [52]. Additionally, Ismagilova [37] conducted a meta-
analysis to reconcile conflicting evidence and offer a comprehensive understanding of
eWOM which identified argument quality, sentiment, eWOM usefulness, and trust in the
message as the most reliable predictors of purchase intention. Furthermore, Chen [53]
forecasted the impact of eWOM on purchase intention based on factors such as helpfulness,
credibility, information quality, and professionalism, employing feature filtering algorithms
and an ensemble model for precise predictions. Therefore, the following hypothesis was
developed.
H5

Information usefulness positively influences online purchase intention.

Useful eWOM information can simplify complex decisions and reduce the need for recipients
to invest additional cognitive effort in searching for more information to reinforce their
confidence in the provided information regarding a product or service. The perceived ease of
processing and low perceived risk encourages information adoption [[54], [55]]. Useful
eWOM information can provide actionable insights and concrete recommendations, serving
as a trigger for online users to progress towards purchase or engagement. This enhances the
probability of information adoption among online users and influences their decision-making
processes based on this information [56]. Previous research has demonstrated that the
usefulness of eWOM information has a positive influence on individuals’ inclination to adopt
it [[57], [58], [59]]. This indicates that individuals are more inclined to adopt information
when they perceive it as relevant and valuable in addressing their specific needs and
objectives. Therefore, the following hypothesis was developed.
H6

Information usefulness positively influences information adoption.

2.2.3. Attitude towards information (ATI)

Attitude towards information (ATI) refers to the overall disposition or feeling an individual
holds towards eWOM content. It encompasses various cognitive, affective, and behavioral
dimensions [60]. When eWOM messages elicit positive emotions such as excitement, trust,
or enthusiasm, individuals tend to be more persuaded by the information and feel a
heightened inclination to act upon it [61,62]. A positive attitude towards eWOM reflects
positively on the brand or product under discussion, building a more positive image and
strengthening associations with positive emotions. This improved perception can contribute
to higher purchase intention [13]. Pang and Zhang [63] suggest that the quality of service on
mobile social media platforms influences user evaluations, including satisfaction,
identification, and belongingness. These factors contribute to emotional attachment, which is
a strong predictor of service attachment and lead to online intention. Therefore, the following
hypothesis was developed.
H7

Attitude towards information positively influences online purchase intention.

The cognitive dimension of ATI concerns consumers’ beliefs and thoughts about eWOM,
including its credibility, relevance, and usefulness. When eWOM is seen as credible and
useful, it positively affects consumer attitudes and purchase intentions [60]. The affective
dimension involves the emotions and affective responses elicited by eWOM messages.
Positive emotions like excitement, trust, and enthusiasm can enhance the persuasiveness of
eWOM, leading to a stronger inclination to act upon the information [37,61]. The emotional
engagement with eWOM content also can create a favorable brand image and associate the
product with positive emotions, consequently increasing the likelihood of purchase [56]. A
positive attitude towards eWOM reflects positively on the brand or product being discussed,
building a more favorable image and strengthening associations with positive emotions. This
enhanced perception can lead to a higher intention to purchase [64]. Therefore, the
following hypothesis was developed.
H8

Attitude towards information positively influences information adoption.

2.2.4. Information adoption (IFA)

Information Adoption (IFA) is defined as the integration of eWOM into consumers' decision-
making processes [65]. This integration involves incorporating valuable insights from eWOM
to streamline choices, thereby reducing the necessity for extensive research and facilitating a
clearer path towards purchase intent [3,66]. The process of IFA is not merely about the
accumulation of information but involves a critical evaluation and assimilation of eWOM
into one's personal knowledge base. This integration is crucial as it aligns with personal
values and fosters a commitment that increases the likelihood of making a purchase [67]. The
theoretical basis of IFA can be traced back to the IAM, which posits that consumers' behavior
is influenced by their attitudes towards the eWOM content and the credibility of its source
[[42], [68]]. The antecedents of eWOM, such as quality, consumer attitude, credibility,
usefulness, and the needs of the consumer, have been identified as significant predictors of
IFA. These factors collectively contribute to the adoption of eWOM, which subsequently
mediates the influence of these antecedents on consumers' purchase intentions [56].
Therefore, the following hypothesis is proposed.
H9

Information adoption positively influences online purchase intention.

The research framework used in this study and its hypotheses are presented in Fig. 1.
Fig. 1.
 Open in a new tab
Research framework.
3. Methodology
3.1. Participants

This study targeted individuals actively engaged in eWOM activities on SNS and possessing
experience in purchasing goods through these online platforms. It exclusively targeted
individuals aged 18 and above, deliberately refraining from collecting data from minors. The
sample consisted of 337 survey participants residing in various regions across Vietnam, with
their demographic information presented in Table 1. In terms of gender distribution,
approximately half of the participants identified as male (49.8 %), while 48.1 % identified as
female, and a smaller proportion (2.1 %) identified as ‘Other.’ Regarding age distribution, the
majority of respondents fell within the 22–25 age range, comprising 29.3 % of the sample.
The participants represented a diverse age range, with individuals aged 18–21 constituting
33.8 % of the sample, and those over 25 accounting for 6.9 %, ensuring a broad spectrum of
perspectives on the impact of eWOM. Among the SNS utilized by participants, Facebook
emerged as the most frequently used platform, with 28.8 % of respondents, followed by
YouTube with 26.6 %. In contrast, Instagram and TikTok exhibited smaller user bases, with
22.9 % and 21.7 % of participants, respectively.
Table 1.
Respondents’ information and SNSs usage.
Respondents' Information (N = 337) Frequency Percentage
Gender Male 168 49.8
Female 162 48.1
Other 7 2.1
 Respondents' Information (N = 337) Frequency Percentage
Age 18–21 114 33.8
22–25 200 59.3
Over 25 23 6.9
SNS usage Facebook 335 28.8
Instagram 226 22.9
YouTube 310 26.6
TikTok 253 21.7
Open in a new tab
3.2. Instrument

The primary tool utilized for quantitative data collection in this study is a structured
questionnaire. The questionnaire was initially developed in English and subsequently
translated into Vietnamese using Brislin's back-translation method, to mitigate the potential
influence of linguistic and cultural disparities. The questionnaire was meticulously designed
with two distinct sections to facilitate comprehensive data collection and enable a systematic
analysis of the impact of eWOM.

The questionnaire comprised two main sections: the first section collected demographic
information from participants, including gender, age, and SNS usage patterns. The second
section evaluated five constructs of the study model, consisting of 17 items measured on a 5-
point Likert scale (ranging from 1-strongly disagree to 5-strongly agree). The items were
adapted from validated studies to ensure appropriateness and measurement robustness.
Specifically, information credibility was assessed using four items adapted from validated
studies by Refs. [19,65,69], attitude toward information with three items from Ref. [70],
information usefulness with three items from Refs. [19,[71], [72], [73]], information adoption
from Ref. [74], and online purchase intention with four items from Ref. [75] (see Table 2).
These adapted items ensured alignment with the context of eWOM on the internet and
measurement robustness. Moreover, previous research on eWOM's influence on online
purchase intention supported the consistency of these factors [16]. This structured
questionnaire approach facilitated comprehensive data collection conducive to robust analysis
of eWOM's impact.
Table 2.
Convergent validity.
Constructs' Indicators Loadings Cronbach's Composite Average vari
alpha reliability (CR) extracted (A
Information Credibility (IFC) 0.835 0.89 0.672
IFC1: “I think eWOM information in social 0.758
networking sites is convincing.”
IFC2: “I think eWOM information in social 0.914
networking sites is credible.”
IFC3: “I think eWOM information in social 0.713
 Constructs' Indicators Loadings Cronbach's Composite Average vari
alpha reliability (CR) extracted (A
networking sites is believable.”
IFC4: “I think eWOM information in social 0.877
networking sites is trustworthy.”
Information Usefulness (IFU) 0.896 0.935 0.827
IFU1: “eWOM information in social 0.913
networking sites is generally useful for me to
evaluate the product.”
IFU2: “eWOM information in social 0.912
networking sites is generally helpful for me to
evaluate the product.”
IFU3: “eWOM information in social 0.904
networking sites is generally informative for me
to evaluate the product.”
Attitude Towards Information (ATI) 0.877 0.924 0.803
ATI1: “I always review eWOM information in 0.916
social networking sites when I want to buy a
product.”
ATI2: “I find eWOM information in social 0.862
networking sites helpful for me to make
decisions when buying a product.”
ATI3: “I feel eWOM information in social 0.908
networking sites makes me more confident in
purchasing products.”
Information Adoption (IFA) 0.936 0.959 0.886
IFA1: “I learn something new about brands 0.936
through eWOM information in social
networking sites.”
IFA2: “I accept the eWOM information of 0.957
brands in social networking sites.”
IFA3: “I accept the eWOM recommendation of 0.931
brands in social networking sites.”
Online Purchase Intention (OPI) 0.821 0.879 0.646
After reviewing the eWOM information spread 0.839
in social networking sites about the products:
OPI1: “It is very likely that I will buy the
product.”
OPI2: “I will buy the product next time I need a 0.744
product.”
OPI3: “I will definitely try the product.” 0.809
OPI4: “I will recommend the product to my 0.819
 Constructs' Indicators Loadings Cronbach's Composite Average vari
alpha reliability (CR) extracted (A
friends.”
Open in a new tab
3.3. Data collection and data analysis

This empirical study employed a quantitative methodology, utilizing online surveys

administered through Google Forms to collect data. A convenience sample technique was
chosen for its ease of contact, time efficiency, and cost-effectiveness in data collection [57].
The questionnaire was disseminated via SNS to reach a diverse pool of Vietnamese citizens
across the country, seeking their voluntary participation in the study. Prior to engagement,
explicit consent was obtained from each participant, acknowledging their involvement and
the utilization of their data for research purposes. The questionnaire provided a
comprehensive delineation of the study's purposes and assured participants of the
confidentiality measures safeguarding their privacy. It is essential to emphasize that this
research is situated within the field of business and management, with a primary focus on
conducting surveys to elucidate individuals' perceptions. Importantly, no experiments
involving human, or animal subjects were conducted as part of this study. Over a three-month
period, from September to December 2023, the research collected responses from 337
eligible participants who actively engaged in eWOM activities on diverse SNS such as
TikTok, Facebook, YouTube, Instagram, and Zalo. These participants had also conducted at
least one online purchase or transaction through these platforms. Responses with incomplete
answers or uniform selections across all questions were excluded to enhance the objectivity,
suitability, and accuracy of the dataset.

This study employed partial least squares structural equation modeling (PLS-SEM) as a
statistical analysis technique to explore complex models and conduct exploratory research.
PLS-SEM offers several advantages, including flexibility and the ability to handle small
sample sizes [[58], [76]]. To evaluate the convergent validity, internal consistency, and
reliability of the measurement model for each latent construct, the Average Variance
Extracted (AVE) and Composite Reliability (CR) analyses were performed. Additionally, the
study utilized the Heterotrait-Monotrait Ratio of Correlations (HTMT) to assess discriminant
validity. These analytical approaches collectively contribute to ensuring the robustness and
validity of the measurement model, thereby enhancing the reliability of the study's findings
[77].
4. Results
4.1. Measurement model analysis

In order to determine the reliability and validity of the measurement model in the study
context, Hair [78] suggests two main ways to assess how reliable a measurement model is:
Cronbach's alpha and Composite reliability rho_c. However, Cronbach's alpha might not
reflect the true reliability of the model, while Composite reliability might overestimate it.
Therefore, they recommend presenting both measures, with a reasonable reliability value
typically falling between these two measures.
In evaluating the measurement model, this study followed the guidelines outlined by Hair
[78] to examine outer loadings, average variance extracted (AVE) for convergent validity,
Cronbach's alpha, composite reliability (CR), and discriminant validity. As shown in Table 2,
all the outer loadings for items were between 0.692 and 0.910, nearly meeting the minimum
of 0.70. Additionally, AVE values (reflecting convergent validity) exceeded the required
minimum level of 0.50, confirming convergent validity for all measures [79]. Furthermore,
internal consistency reliability was high, with Cronbach's alpha coefficients for all constructs
exceeding the acceptable threshold of 0.70. These findings confirm the reliability and validity
of the research model in this study.

To evaluate discriminant validity, the study employed two approaches: the Heterotrait-
Monotrait ratio of correlations (HTMT) and the Fornell-Lacker criterion. Although the
Fornell-Lacker criterion along with cross-loadings examination is frequently used for this
assessment, it has been suggested that these methods might not always effectively identify a
lack of discriminant validity in common research scenarios [50]. The HTMT method offers a
preferable and more effective alternative for assessing discriminant validity and was applied
in this research [78]. A HTMT value less than 0.90 is indicative of adequate discriminant
validity between two reflective constructs [80]. According to the data presented in Table 3,
all HTMT values for the reflective constructs were below 0.90, offering strong evidence of
discriminant validity.
Table 3.
Discriminant validity.
ATI IFA IFC IFU OPI
ATI
IFA 0.079
IFC 0.385 0.16
IFU 0.143 0.162 0.275
OPI 0.208 0.269 0.245 0.136
Open in a new tab
4.2. Model structure
4.2.1. Assessment of common method bias

The study employed a self-response survey methodology for data collection, which may
introduce the possibility of common method bias (CMB). CMB has been extensively studied
in research, with scholars exploring its potential impact on empirical analyses involving
respondents as data sources, including the consideration of multicollinearity among variables.
To address this concern, collinearity statistics, particularly the Variance Inflation Factor
(VIF), a widely accepted metric for detecting multicollinearity, were utilized to evaluate
multicollinearity between independent latent variables. The VIF serves as a crucial index for
assessing both multicollinearity and common method bias within datasets. VIF values
exceeding 5 indicate significant collinearity issues among the indicators of constructs
measured in a formative manner. Additionally, a VIF threshold of 2 is sometimes used for
quantitative variables. As proposed by Ref. [81], a VIF of 2.5 indicates considerable
collinearity. In this study, reflective variables such as ATI, IFA, IFC, IFU, and OPI naturally
exhibit correlations among themselves, serving as important indicators for identifying
multicollinearity within regression models. The study's results revealed VIF values ranging
from 1 to 1.202, as detailed in Table 4. These findings suggest the absence of
multicollinearity or common method bias concerns, as the values are within the preferred
range, indicating no problematic collinearity among the indicators.
Table 4.
Results of testing hypotheses in the theoretical framework.
Original Sample Standard Deviation T Statistics P Result V
Sample (O) Mean (M) (STDEV) (|O/STDEV|) Values
IFC - > 0.245 0.248 0.054 4.898 0.000 Accepted 1
IFU
IFC - > 0.142 0.145 0.060 2.441 0.019 Accepted 1
OPI
IFC - > 0.114 0.113 0.066 1.738 0.085 Accepted 1
IFA
IFU - > 0.128 0.132 0.055 2.338 0.020 Accepted 1
ATI
IFU - > 0.039 0.040 0.052 0.746 0.468 Rejected 1
OPI
IFU - > 0.12 0.124 0.053 2.238 0.023 Accepted 1
IFA
ATI - > 0.109 0.114 0.057 1.869 0.055 Accepted 1
OPI
ATI - > 0.019 0.022 0.061 0.310 0.759 Rejected 1
IFA
IFA - > 0.213 0.216 0.049 4.333 0.000 Accepted 1
OPI
Open in a new tab
4.2.2. Examination of research hypotheses

To assess the structural model and examine the hypothesis, bootstrapping analysis was
conducted, with the findings displayed in Table 4 and Fig. 2. This analysis reveals the
intricate relationships among eWOM information credibility, information usefulness, attitude
towards information, information adoption, and the online purchase intention. Specifically,
Hypothesis 1 (H1) was supported, demonstrating a positive impact of IFC on IFU (O = 0.245;
T-statistic = 4.898; p-value = 0.000). This suggests that when users encounter eWOM
information perceived as credible, they are more likely to perceive it as valuable and useful.
Hypothesis (H2) was also supported, indicating a positive influence of IFC on OPI
(O = 0.142; T-statistic = 2.441; p-value = 0.019). This implies that users are inclined to make
online purchases when they trust the eWOM information they receive from other users.
Hence, ensuring information credibility is a key factor for enhancing customers' purchasing
intentions. Furthermore, Hypothesis 3 (H3) was supported (O = 0.114, T-statistic = 1.738; p-
value = 0.085), indicating that IFC positively influences IFA. This suggests that when users
perceive eWOM information as credible, they are more likely to adopt it, reducing skepticism
and increasing adoption rates. Hypothesis 4 (H4) found significant support, demonstrating a
positive relationship between IFU and ATI(O = 0.128; T-statistic = 2.338; p-value = 0.02).
This means that users tend to develop a positive attitude towards eWOM information when
they perceive it as useful and valuable to them. However, Hypothesis 5 (H5), positing a
positive impact of IFU on OPI, was rejected (O = 0.039; T-statistic = 0.746; p-value = 0.468).
This suggests that while users may find eWOM information useful, other factors likely
influence their online purchase intentions. Therefore, they may feel the information they
receive is useful but not enough encouraging them to have intention buying that product or
service. Hypothesis H6 was also supported (O = 0.12; T-statistic = 2.238; p-value = 0.023),
indicating a positive impact of IFU on IFA. This suggests that when a user feels the
information provided by others is useful, they exhibit higher probability to adopt and
integrate such information into their decision-making process for buying a product or service.
In addition, Hypothesis 7 (H7) was supported, showing a positive effect of ATI on OPI
(O = 0.109; T-statistic = 1.869; p-value = 0.055). This implies that when peer users have a
positive attitude towards eWOM information, their intention to make online purchases also
increases significantly. In the study, Hypothesis 8 (H8) was found to be not significant,
suggesting that ATI alone does not strongly impact IFA (O = 0.019; T-statistic = 0.310; p-
value = 759). This indicates that while a user's attitude may change upon receiving eWOM
information, this change alone may not lead to full adoption of such information. Finally,
Hypothesis 9 (H9) was accepted, demonstrating a positive impact of IFA on OPI (O = 0.213;
T-statistic = 4.333; p-value = 0). This suggests that once users accept and absorb the
information, they are more likely to use it in their decision-making processes regarding
product or service purchases.
Fig. 2.

Open in a new tab

PLS-SEM structural model.
4.2.3. Evaluation of explanation and prediction power of
research model

The coefficient of determination, also known as R-squared (R2), is a crucial statistic in

statistical modeling that explains the proportion of variance in dependent variables that can
be predicted from independent variables. Essentially, R2 measures the strength of a model's
explanatory power. In this research, the analysis identified the highest R2 value for the
variable OPI, which was found to be 0.108, or 10.8 %. This suggests that about 10.8 % of the
variation in OPI is explained by the interrelations among other variables in the study namely,
IFC, IFU, ATI, and IFA. In the realm of social sciences, R2 values from 0.10 (10 %) to 0.50
(50 %) are typically deemed to be within an acceptable range. Ozili [82] points out that the
acceptability of these figures largely depends on the statistical significance of the variables
that explain the model, underscoring the importance of thorough evaluation.

Additionally, the Q-squared (Q2), a metric indicative of predictive goodness of fit, measures
the model's performance in forecasting outcomes. A Q2 value exceeding 0 signifies that the
reconstructed values align closely with actual observations, thus confirming the predictive
relevance of the model [83]. In this study, all indicators point towards positive predictive
performance: the Q2 index surpasses 0 for specific variables ATI (0.011), IFA (0.026), IFU
(0.047), and OPI (0.059). These values underscore the model's effectiveness in anticipating
outcomes, indicating a strong alignment between the model's predictions and the observed
data, thereby validating its utility for predictive purposes.
5. Discussion

The empirical findings of this study underscore a substantial role of eWOM information
credibility in the context of online shopping on SNS. The identified significance of
information credibility in influencing information usefulness supports the notion that when
users encounter trustworthy and reliable information, it fosters confidence and perceived
value, thereby significantly contributing to its perceived usefulness of the information in
shaping individuals’ decision-making processes. This finding aligns with existing literature,
highlighting the enduring importance of information credibility as a key determinant in
perceiving information as useful in online purchase processes [12,31]. Moreover, the
observed impact of information credibility on online purchase intention emphasizes its
enduring influence beyond the assessment of information, extending its sway into the
consequential domain of online purchase intentions. This suggests that consumers, influenced
by perceived information credibility, are more likely to exhibit a positive inclination toward
making online purchases, consistent with the findings of [84]. This alignment with the
broader landscape of digital consumer behavior seamlessly resonates with the IACM model,
positioning credibility as a catalyst for shaping consumer attitudes and subsequent behavioral
intentions. An intriguing aspect of the findings is the effect of information credibility on
information adoption. The exploration reveals that information credibility exerts a strong
influence on information adoption in the research context, implying that users prioritize
information credibility as a decisive factor in accepting and incorporating information into
their knowledge base, aligning well with existing literature [40,44]. This phenomenon implies
that users are more inclined to adopt and integrate information from sources deemed credible
in the online decision-making process, even when its utility may not be immediately
apparent. This heightened concern about information credibility may be attributed to the
inherent risks and uncertainties associated with online shopping, where users lack the ability
to physically inspect product quality, assess delivery timelines, or evaluate security and
privacy measures [85]. The significance of this finding lies in the understanding that, in the
online shopping environment, users prioritize credibility as a primary criterion for evaluating
eWOM information. The more emphasis on credibility as a risk-mitigation strategy indicates
a strategic approach by consumers to navigate the uncertainties inherent in online
transactions. Practically, enhancing the information credibility can yield tangible benefits in
influencing consumer behavior. Businesses can leverage this insight by prioritizing trusted
sources and actively involving the community in encouraging authentic user-generated
content that contributes to the genuine and unbiased eWOM. By resolving negative reviews
and concerns promptly, while maintaining transparent communication and responsiveness,
businesses can demonstrate a strong commitment to customer satisfaction and significantly
enhances the credibility of the eWOM information. However, building a credible eWOM
environment presents challenges. It requires continuous investment in cultivating
relationships with established experts or professionals in the field, moderating user-generated
content, and promptly addressing customer concerns, which can be costly. Therefore, it is
essential to balance these expenses with the long-term benefits of increased trust, purchase
intention, and brand loyalty. Moreover, the findings offer valuable guidance to policymakers,
who can enact regulations to protect consumers from deceptive practices and promote a more
trustworthy online marketplace, while educating consumers about evaluating information
credibility to enable them to make rational choices.

In line with the foundational principles of the IACM model, the results reveal a notable and
positive correlation between information usefulness and the attitude towards eWOM
information. This finding underscores the fundamental principle that individuals tend to
develop a favorable attitude towards information perceived as utilitarian [86]. In this research
context, it emphasizes the critical role of providing valuable information to users. The
positive attitude generated in response to useful information becomes a crucial determinant,
enhancing the overall efficacy of online purchasing processes. The study's results also
indicate the crucial role of information usefulness in its correlation with information
adoption. This finding is consistent with existing literature on customer behavior, as
evidenced by studies of Hussain [65] and Ngo [16]. This reflects the tendency of individuals
to actively integrate and consider information they perceive as valuable and contributory to
their understanding. In the context of online purchasing, where users heavily rely on the
experiences and opinions of others, the positive correlation between information usefulness
and information adoption gains heightened significance. When users encounter information
perceived as useful, it not only addresses their immediate informational needs but also
enhances their trust in the credibility of the eWOM sources, thereby increasing the likelihood
of adopting such information. According to Filieri [33], positive experiences with valuable
content establish a positive reinforcement loop, where users are more likely to actively seek
out and adopt similar content in subsequent decision-making scenarios. Contrary to
expectations, however, the study discovered a surprising result that information usefulness
alone does not exert a significant direct impact on purchase intention. This nuanced finding
prompts a reevaluation of the factors influencing online purchase decisions. While
information usefulness is acknowledged as a crucial factor, the study suggests that other
variables, including product quality, pricing, trust, and perceived risk, also hold significant
roles in shaping consumers' decisions to make online purchases [87]. This finding
underscores the complexity of consumer decision-making in the online environment,
highlighting the need for a comprehensive and holistic approach to understand the
multifaceted determinants of online purchase intentions. This nuanced understanding of
consumer behavior has significant implications for businesses across various sectors. E-
commerce platforms can utilize these insights to optimize their user review systems, ensuring
that the most useful and credible information is prominently displayed. This approach can
enhance user engagement and ultimately drive sales.

Moreover, the study's findings revealed a significant positive correlation between attitude
towards eWOM information and online purchase intention. A positive attitude signifies a
favorable evaluation of the credibility, relevance, and utility of information propagated
through eWOM channels on SNS [88]. Consumers who maintain positive attitudes towards
eWOM information are more likely to convert these attitudes into intentions for online
purchases, which is consistent with established literature indicating attitudes as key predictors
of behavioral intentions [89]. In the situation of information overload, a positive attitude
towards eWOM content becomes a valuable asset for driving consumers towards actual
behavioral conversions. However, contrary to the logical expectation that a positive attitude
should directly lead to information adoption, the results introduce a nuanced and intriguing
dimension to the discussion. The observed disconnect suggests that expressing a positive
attitude towards eWOM information does not necessarily ensure active integration into the
decision-making process. This potential gap between emotional responses and conscious
processing is influenced by various factors such as information overload and the cognitive
demands associated with processing extensive information. In the fast-paced online
environment, consumers often resort to mental shortcuts, such as brand loyalty or simplified
product features, which may overshadow active adoption of all encountered information.
Additionally, confirmation bias and selective attention play a role, with consumers selectively
focusing on information that reinforces existing beliefs or purchase intentions, potentially
limiting the impact of attitude on actual information adoption. This finding is consistent with
the Theory of Planned Behavior (TPB) [90], which posits that any behavior requiring
reflection and/or planning, such as completing an online purchase, is directly influenced by
intention. Understanding these dynamics is crucial for marketers to craft strategies that not
only foster positive attitudes towards eWOM information but also address cognitive barriers,
streamline decision-making processes, and align with consumers' mental shortcuts, thereby
bridging the gap between attitude and action.

Importantly, the results underscored the critical role of information adoption in shaping
online purchase intention. The significant and positive impact observed in the study
underscores the idea that consumers who actively integrate eWOM information into their
decision-making processes are more likely to exhibit intention towards making online
purchases, consistent with previous research findings of Sardar [56] and Khoa [57]. This
revelation indicates the importance of cultivating an environment that promotes the active
adoption of such information by consumers, aligning with the theoretical foundations of
models like IACM, which emphasize the significance of user engagement and active
involvement in shaping behavioral outcomes. In the realm of eWOM on SNS, where
information is rapidly and extensively disseminated, users actively engaging with and
incorporating eWOM information into their decision-making process signifies a level of trust
and reliance on the information source. This trust, in turn, contributes to the effectiveness of
eWOM in influencing consumer behavior. This novel insight has profound implications for
both businesses and platform designers. To fully leverage the power of eWOM, strategies
must go beyond fostering positive sentiment to actively integrating eWOM information into
consumers' decision-making processes. This may involve designing eWOM campaigns that
are not only emotionally appealing but also cognitively engaging, providing easily digestible
information that aligns with consumers’ existing beliefs and habits. Moreover, platforms can
play a crucial role by incorporating features that help users navigate information overload and
mitigate cognitive biases, thereby empowering them to make more informed purchase
decisions based on eWOM information.
5.1. Limitations and future directions

This study has provided valuable insights, but certain limitations require further investigation.
Relying solely on self-reported survey data introduces biases such as social desirability and
recall errors. Future studies could mitigate these biases by integrating additional
methodologies such as behavioral observations, in-depth interviews, or focus groups. An
acknowledged constraint is the inability to control for all external variables influencing online
purchase intention, such as individual personality traits, brand loyalty, or cultural influences.
To address this limitation, future studies could incorporate a broader range of control
variables or utilize advanced statistical methods to isolate the specific impact of eWOM from
these external factors.

To advance research in this field, future directions should include implementing a more
diverse sampling strategy and exploring different geographical and cultural contexts.
Furthermore, considering the evolving landscape of new technologies, particularly the rise of
artificial intelligence (AI), presents a compelling avenue for investigation within the eWOM
domain. Subsequent research could delve into the implications of AI-generated eWOM,
investigating how individuals distinguish between AI-generated and user-generated content
as credible. Additionally, broadening the scope to unravel the influence of cultural factors on
eWOM effectiveness by exploring the impact of cultural values, norms, and online behavior
patterns across diverse contexts can provide valuable insights for future research in this area.
5.2. Implications

The research, from a theoretical standpoint, illuminated intricate connections among various
eWOM information factors, encompassing elements such as information credibility,
usefulness, attitude towards information, and information adoption, in relation to online
purchase intention. Notably, the study underscored the pivotal role of information credibility
in shaping online consumer behavior. It emerged as a potent influencer, impacting perceived
value, information adoption, and purchase intention. These findings are consistent with
established theoretical models, particularly the Information Acceptance Model (IACM).
Furthermore, the study reinforced the importance of information adoption within the
theoretical framework of the Information Adoption Model (IAM). By integrating these
models, the research provides a comprehensive understanding of the complex relationships
among eWOM information factors, highlighting user engagement and information adoption
as key drivers of behavior.

The empirical findings offer practical insights for various stakeholders navigating the
dynamic landscape of electronic word-of-mouth (eWOM). Businesses can foster customer
loyalty by establishing trust through transparent practices, collaborating with credible
influencers, and promoting authentic user-generated content. By prioritizing the creation of
high-quality eWOM that addresses consumer needs and showcases genuine user experiences,
businesses can significantly enhance the impact of eWOM on purchase decisions.
Additionally, marketers can leverage insights from information adoption psychology to
design campaigns that extend beyond mere positive sentiment generation. By crafting
messaging that resonates with consumers and facilitates the active integration of eWOM into
decision-making processes, marketers can cultivate a more engaged and loyal customer base.
Policymakers play a pivotal role in creating a trustworthy online environment by enforcing
regulations that curb deceptive eWOM practices. Moreover, by educating consumers about
evaluating information credibility equips them to make rational choices in the complex digital
landscape. Ultimately, empowered consumers contribute to a transparent and reliable eWOM
ecosystem by developing critical thinking skills, seeking diverse perspectives, and actively
participating in online communities. Platforms and businesses can further promote
information adoption by facilitating user engagement through discussions and user-friendly
tools for information filtering and retrieval.
6. Conclusion

This study investigated the intricate relationships among information credibility, information
usefulness, attitude towards information, information adoption, and online purchase intention
in the context of eWOM on SNS. The findings provided valuable insights into the
multifaceted nature of consumer behavior in online shopping contexts. Specifically, the study
emphasized the significant influence of eWOM information credibility on shaping user
perceptions and subsequent actions. Credible eWOM not only enhances the perceived
usefulness of information but also directly impacts online purchase intention and information
adoption. This underscores the importance of fostering trust and authenticity within online
shopping ecosystems. Moreover, information usefulness emerges as a crucial factor shaping
user attitudes towards eWOM information.

The study also confirms that information usefulness contributes to positive attitudes towards
eWOM. However, an interesting finding is that usefulness alone doesn't guarantee a
purchase. This suggests that other factors, such as product quality, pricing, and perceived
risk, also play crucial roles in influencing purchasing decisions. Therefore, future studies
could delve deeper into how eWOM interacts with these traditional and more rational
decision-making considerations.

Furthermore, the research revealed a disconnect between attitude and information adoption,
indicating a significant gap in our understanding of consumer behavior. While a positive
attitude towards information translates into higher purchase intentions, it does not necessarily
lead to active information adoption. This discrepancy underscores the potential influence of
cognitive constraints, information overload, and confirmation bias, hindering users from
integrating eWOM insights into their purchase decisions. Addressing these challenges
presents an opportunity for researchers to develop more user-friendly decision-support tools
that assist consumers in navigating abundant eWOM information effectively.

Finally, this research highlights the critical role of information adoption in driving online
purchase intentions. This reinforces the need to cultivate online environments that encourage
user engagement, active evaluation of eWOM, and a sense of confidence in the information
shared.
Ethics declarations

All participants provided informed consent to participate in the study. The questionnaires
were anonymized, and respondents were free to opt out of participation in the study whenever
they were uncomfortable. The study procedure received ethical approval from FPT Can Tho
University, Vietnam (Approval No. 20230911.02).