The impact of online micro-influencers on behavioural intentions: The case of mobile phone customers in Ho Chi Minh city
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
47
The impact of online micro-influencers on behavioural
intentions: The case of mobile phone customers
in Ho Chi Minh City
Kieu Anh Tai1*, Le Minh Duy1, Nguyen The Khai1
1Ho Chi Minh City Open University, Vietnam
ARTICLE INFO
ABSTRACT
DOI:10.46223/HCMCOUJS.
econ.en.11.1.1309.2021
The advances of the Internet and social media offer customers
nowadays more opportunities to inform their decisions with product
information - not only that promoted by the company but also
content shared by micro-influencers. This study aims to investigate
the impact of online micro-influencers on the customer’s
behavioural intentions. A survey was conducted using convenience
sampling with customers in Ho Chi Minh City who have bought or
intended to buy smartphones upon viewing online product reviews.
Data, upon having been collected and screened, included 371 usable
responses and were analysed using PLS-SEM. Results show that the
customer’s perceptions of opinion leadership of micro-influencers
and their parasocial interactions with micro-influencers positively
influence the customer’s behavioural intentions (intention to follow
the advice and intention to recommend) through message quality.
Managerial implications are also discussed in the paper.
Received: November 12th, 2020
Revised: January 23rd, 2021
Accepted: January 26th, 2021
Keywords:
online opinion leaders, micro-
influencers, parasocial
interactions, behavioural
intentions, smartphone
1. Introduction
In traditional marketing communications, information source could be either the businesses
called as advertisers disseminating messages about their products or services to the customers or
famous individuals called as celebrities or key opinion leaders (KOL), who are able to influence
the public’s opinions and behaviour due to their expertise or attractiveness appeal (Stokburger-
Sauer & Hoyer, 2009). Indeed, much research has been done on the latter type of information
source, having focused on celebrities as influencers or opinion leaders. Those studies investigated
either the influence of celebrities' personal traits, the congruence of their image that brands want
to leverage on secondary brand associations (Gentina, Butori, & Heath, 2014; Min, Chang, Jai, &
Ziegler, 2019); or the impact of celebrities' endorsement thanks to their image or reputation or the
customer's engagement with the celebrity on the consumer's attitude and buying behaviour (Ilicic
& Webster, 2014).
With the advances of the Internet and social network sites, consumers are increasingly
given the opportunity to access more information channels and sources of information to evaluate
alternatives and make purchase decisions. On online forums and social network sites (social
media), there has emerged a number of individuals, through their social media accounts to create,
contribute and share useful content for another internet/ social media users, gradually forming a
community of people that “follow”, interact and take into account their opinions. Those
individuals, commonly known in practice as bloggers and vloggers, are also referred to as micro-
48
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
influencers who influence others’ opinions in the internet environment. This type of social
influencers, though having less number of followers than traditional celebrities, have more positive
followers’ interactions (e.g., viewing or re-sharing), thereby generating higher communication
effectiveness and efficiency (Abidin, 2016).
Businesses and brands are increasingly interested in finding ways to leverage online “brand
ambassadors” who are social media (e.g., YouTube or Facebook) bloggers presenting and
reviewing on products or services that they tried out and promoted the product to other online users
(Sokolova & Kefi, 2020). The information generated and spread by those social influencers is
usually in the form of an image or a short video clip embedded with product message or textual
description. The popularity of the blogger/vlogger phenomenon and user-generated content (UGC)
has led to the rise of a new form of marketing “influencer marketing”, in which brands or
businesses collaborate with online opinion leaders to promote their products or services (De
Veirman, Cauberghe, & Hudders, 2017).
Some research has focused on user-generated content (UGC), also called consumer-
generated content (CGC), in terms of electronic word-of-mouth (EWOM) (Chatterjee, 2011;
Trusov, Bucklin, & Pauwels, 2009). However, some researchers noticed the ephemeral nature of
attention and inequal popularity of UGC (Susarla, Oh, & Tan, 2016). Recently, some researchers
have paid attention to online opinion leaders, who could be either celebrity on social network
platforms (Li, Lee, & Lien, 2012) or just individuals who have a lot of followers, share or interact
(e.g., tweet or retweet) on online social platforms (De Veirman et al., 2017; Park & Kaye, 2017).
Researchers suggested that online opinion leaders on online social platforms are
necessarily not traditional celebrities (famous athletes or actors) and that the sharing content is not
advertising, and even that the quality of image or video content may not be professional; so, it is
needed to the persuasive cues about those online opinion leaders (Sokolova & Kefi, 2020). Prior
research revealed that online opinion leaders were often seen as a reliable source of information,
having a more positive impact on brand trust and purchase intent than the influence of conventional
celebrities (Bijen, 2017; Djafarova & Trofimenko, 2019). Research also found that online opinion
leaders, compared to traditional celebrities, were more likeable because of more frequent
interactions and more likeliness of the promoting content to be positively rated by consumers
(Sokolova & Kefi, 2020). Nonetheless, there has been scant research on the specific cues of
persuasion concerning online opinion leaders on the online social platforms such as the consumer’s
perception of their parasocial interactions with online opinion leaders (Sokolova & Kefi, 2020) or
about the extent of the customer’s perceptions about an individual who can influence others’
opinion online (Casaló, Flavián, & Ibáñez-Sánchez, 2020).
This study aims to investigate the influence of the specific persuasive cues of online
opinion leaders (parasocial interaction and online opinion leadership) on the consumer’s perceived
message quality, which in turn affects the consumer’s behavioural intentions both relating to
consumer behaviour (intention to follow the advice of online opinion leaders) and online opinion
leader (intention to recommend the online opinion leaders to others). This research is theoretically
significant as it contributes to bridging the above-described gap in the literature. In addition, this
research also responds to the call for research in emerging market contexts like Vietnam, where
marketing research is still limited with mixed evidence (Sheth, 2011), let alone research into online
opinion leaders and their cues of persuasion in marketing communications. Furthermore, a report
published by the research agency Indochina Research has shown Vietnam has high internet and
social media penetration, with 66% and 64% of the population, respectively (Depouilly, 2019).
That report also found 88% of their survey respondents indicated product information search as
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
49
one of their digital activities, implying the context of online product reviews is of practical
relevance and significance. Another synthetic statistic by We Are Social and Hootsuite (2019) also
indicated a high penetration of smartphones, with also 96% of internet users accessing via mobile
devices.
2. Theoretical foundation and hypothesis development
2.1. Elaboration Likelihood Method (ELM) model
In communication studies, researchers often draw on ELM theory to explain how receivers
(consumers) process information or cues exposed to them (Sokolova & Kefi, 2020). According to
ELM model, there are two routes of information processing: The central route in which the
receivers carefully and thoughtfully process the received information; and the peripheral route in
which receivers rely on cues used by the communicator (such as the attractiveness of the
information source), rather than to evaluate the merits of information or argument quality (Petty
& Cacioppo, 1986). Also, Petty, Cacioppo, and Schumann (1983) proposed that information
quality has a strong influence on attitude change only when receivers have the high motivation or
high ability to evaluate information; otherwise, whether and the extent that receivers have
attitudinal shift is strongly influenced by peripheral persuasive cues such as information sources.
Drawing on ELM theory, previous studies often considered message/information quality (the
quality of arguments) taking the central route and information source characteristics (e.g., source
credibility) taking the peripheral route to influence consumers’ attitudes and behaviours (Le,
Robinson, & Dobele, 2020; Mahapatra & Mishra, 2017). There is also research examining the
impact of both source and message (Cheung & Thadani, 2012; Mahapatra & Mishra, 2017), as the
two drivers of communication effectiveness in accordance with Hovland (1948)’s social
communication theory.
The current study also uses ELM theory as the theoretical foundation to devise hypotheses
and research model to address the research objectives. Specifically, the study examines the
relationship between the information source characteristics (namely, online opinion leadership and
parasocial interaction) and the message (message quality) in influencing behavioural intentions in
the online social platforms.
2.2. Online opinion leadership
In the Internet environment (including social media), due to the characteristics of short-
lived attention to and disproportionate popularity of the information (Susarla et al., 2016), the role
of opinion leaders (information source) is very important, especially with a huge amount of
information available on the Internet, it is challenging for consumers to be able to process all
information. Several studies into information source often incorporated opinion leadership – a
personal source characteristic (Casaló et al., 2020; Le et al., 2020). Opinion leadership is defined
as the extent to which an individual is perceived as a role model to others and the extent to which
the information provided by that person is considered to be interesting and to be his or her
persuasiveness (Casaló et al., 2020). Opinion leaders are those who have real expertise and
experience with products or services, are searched for advice, or are able to disseminate
information about products or services to others (Stokburger-Sauer & Hoyer, 2009).
Thus, it can be argued in this research that the concept of opinion leadership can be
extended to the online environment as online opinion leadership as the perceptions of a social
platform-mediated individual as opinion leader that his or her information is seen interesting and
has persuasive power. On social platforms, the concept of online opinion leaders can refer to either
individual who has a broad knowledge or experience of a particular topic (expertise) or have many
50
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
connections with others on social platforms; these individuals can disseminate information about
products or services to others in the online environment (Goldenberg, Lehmann, Shidlovski, &
Barak, 2006). Individuals are perceived by social network sites’ “friends” as the online opinion
leaders can exert a certain influence on those “social friends” (Iyengar, Van den Bulte, & Valente,
2010). Indeed, some research examined opinion leadership impacting behavioural intentions,
independently of the message factor (Casaló et al., 2020).
2.3. Parasocial interaction
Parasocial interaction emerged from communication theory to explain the development of
the relationship between personas on the mass media and receivers (Horton & Wohl, 1956).
Labrecque (2014) defined parasocial interaction as the illusionary experience in which the
consumer interacts with personas (mediated representations of celebrities or characters) as if they
engage directly in a two-way relationship with that celebrity or character. This is a persuasive cue
of information source (Sokolova & Kefi, 2020). In the context of the Internet and social media,
the relationship between online opinion leaders and followers can sometimes be two-way because
users can add comments or discuss content or receive feedback from mediated representatives
(website/blog administrator) (Labrecque, 2014). However, the real opinion leader cannot respond
to all requests or participate in real discussions with followers due to a large number of followers,
hence the relationship between online influencers and followers are indeed parasocial interactions
(Sokolova & Kefi, 2020). In this way, social platform users can establish relationships with online
opinion leaders (bloggers/vloggers) by following or subscribing to their social channels/accounts
and following the content that online influencers share. Thus, unlike Le et al. (2020)’s research
including source intimacy that could be more of reciprocal relationship, this study considers
parasocial interaction instead as the interpersonal source characteristic regarding online opinion
leaders.
2.4. Message quality
Communication research has also paid considerable attention to the type of information
about products or services that are informally spread between people and termed as word-of-mouth
(WOM) (Anderson, 1998). The Internet and social media have led to a new form of WOM -
electronic WOM (EWOM), which referred to information about products or services that is shared
by online users to a large number of other online users (Hennig-Thurau, Gwinner, Walsh, &
Gremler, 2004) and is conceptually similar to UGC or CGC terms often used in influencer
marketing practice. Message quality is defined as the compelling strength of the arguments
embedded in the content of the information (Bhattacherjee & Sanford, 2006)
2.5. Behavioural intentions
Prior research has examined the consequences of opinion leadership (Casaló et al., 2020)
or final consequences of the causal path within the ELM (Le et al., 2020) that include behavioural
intentions such as the intention to follow the advice of the opinion leader or the intention to
recommend the message source. In the context of product reviews on social platforms, the current
research investigated two behavioural intentions that are impacted by micro-influencership: The
intention to follow the advice of the online opinion leader and the intention to recommend the
social platform account/channel of the online opinion leader to others. The former captures the
outcome favouring the firm and product embedded in the review, and the latter concerns the result
in favour of the micro-influencer themselves. Intent to follow the advice is defined as the degree
to which an individual thinks and follows the advice or reviews of online opinion leaders; and
intention to recommend is the extent to which an individual is willing to refer (the social platform
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
51
account/channel of) an online opinion leader to others with the same need or interest (Casaló et
al., 2020).
2.6. Development of hypothesis and research model
Some prior EWOM research examined only the direct effects of both source and message
factors on behavioural intentions (Cheung & Thadani, 2012) or only source characteristics (Casaló
et al., 2020; Sokolova & Kefi, 2020). Those studies did not consider a basic tenet of ELM model
regarding the causal relationship of two routes of information processing. When the motivation or
ability to process information is low, the receiver follows a peripheral route to process cues, but
gradually develops the ability to process the merits of information as in the central route (Petty &
Cacioppo, 1986). When consumers start to search for product reviews from online opinion leaders
or to connect (subscribe/follow) to online opinion leaders' channels/accounts, consumers may not
be able to process tons of information (posts or reviews by online opinion leaders, comments and
replies by them and other followers…). Therefore, consumers may first form perceptions of source
cues, i.e. characteristics of the online opinion leader, then gradually with acquired knowledge and
interactive experience with the source they develop the ability and motivation to process the
information offered by the online opinion leader (Petty & Cacioppo, 1986).
Le et al. (2020)’s research has shown both characteristics of the source - personal
characteristics and the interpersonal characteristics, have significant effects on the consumer
perceptions of the message quality. In a similar vein, in the online environment, online opinion
leadership and parasocial interaction, which are two persuasive cues of the information source,
can be argued to relate to the perceived quality of the message, because the appreciation of the
source can lead to thorough processing of information and positive message evaluation (Petty &
Cacioppo, 1986). The online opinion leader can also at times through their mediated
representations interact with other users, thereby making others feel more attached to the source
and trust the source’s posts or reviews (Casaló et al., 2020). Thus, it can be hypothesised:
H1: Online opinion leadership has a positive influence on message quality
H2: Parasocial interaction has a positive influence on message quality
According to the ELM theory, the quality of the message (information or arguments) is a
concept that is mostly processed in the central route, leading to the change in attitudinal evaluations
and finally consumer behaviours (Petty & Cacioppo, 1986). Indeed, WOM and EWOM research
have demonstrated information characteristics such as information/message quality can influence
intentions or actual behaviours (Cheung & Thadani, 2012; Sweeney, Geoffrey, & Mazzarol, 2012).
In the current study, behavioural intentions include both the intention relating to what is embedded
in the message spread by the one opinion leader (intention to follow advice) and intention relating
to the online opinion leader themselves for their having good taste regarding specific buying
circumstances. If UGC/CGC is assessed positively, consumers will be more likely to have
behavioural intentions relating to products or services imbedded in the message that online opinion
leaders create and share. Thus, it can be hypothesised that:
H3: The message quality has a positive influence on the intention to follow the advice of
the online opinion leader
H4: The message quality has a positive influence on the intention to recommend the social
platform account/channel of the online opinion leader to others
Figure 1. presents a visual representation of the hypothesised relationships discussed above
52
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
Figure 1. The proposed research model
3. Methods
The study adopted scales with good psychometric properties from previous studies to
measure the concepts in the research model. Specifically, the scale of online opinion leadership (6
items) is adopted from (Casaló et al., 2020); parasocial interaction (6 items) from Labrecque
(2014); message quality (5 items) from Le et al. (2020); intention to recommend (3 items) from
Al-Ansi, Olya, and Han (2019); and intention to follow advice (4 items) from Casaló et al. (2020).
Respondents will rate the degree of agreement of the statements (observed variable) using the 5-
point Likert scale.
The questionnaire begins with screening questions to ensure that the criteria are met for
selecting the survey respondents, who are consumers who view reviews about mobile phone
products on social networking sites / online forums in the past 6 months in making a purchase or
in planning to purchase. The scales are translated from English to Vietnamese and back-translated,
then the original version and the back-translated version are compared, and adjustments are made
to ensure the scale content equivalence. After being designed, the survey questionnaire was pilot-
tested with 2 experts and 10 consumers to ensure that the questionnaire content is easy to
understand, unbiased, and free of grammatical and format errors.
Data was collected using the convenience sampling method, by sending survey
questionnaires to consumers who have previously purchased or planned to buy a mobile phone
and viewed reviews about mobile phones before buying or to plan for the purchase. Given the high
penetration of mobile devices in Vietnam, this sampling method deems appropriate to reach the
above described target population for this study. The mobile phone product was chosen because it
was a high-involvement product and information search is often conducted before making a
purchase. The questionnaire is developed on the Google Form and the link is emailed or posted to
online social platforms such as Facebook or Zalo and online forum in August 2020. The
respondents first indicated the social accounts (that is the micro-influencers) they have followed
and watched, then rated the scale items in relation to their mentioned social accounts.
4. Results and discussions
4.1. Sample characteristics
There were 453 returned responses. After 82 responses from those who did not view
product reviews about mobile phones on social platforms were excluded, a data of 371 responses
were usable for analysis using PLS-SEM. The demographic information of the sample is shown
in Table 1 below.
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
53
Table 1
Sample descriptive statistics
Demographic variables
Frequency (n=371)
Percentages (%)
Gender
Male
149
222
40.2%
59.8%
Female
Age
18-25
108
186
61
29.1%
50.1%
16.4%
4.3%
26-35
36-45
Above 45
Education
Up to high school
University
61
266
44
16.4%
71.7%
11.9%
Postgraduate
Occupation
Employee
200
65
59
42
5
53.9%
17.5%
15.9%
11.3%
1.3%
Student
Manager/ director
Self-employed
Other
Income
Below 12 million VND
12 - just less than 20 million VND
20 - just less than 30 million VND
Above 30 million VND
219
99
59.0%
26.7%
8.6%
32
21
5.7%
Source: The researcher’s data analysis
4.2. Scale assessment
Since the scales were adopted from previous studies, it is not necessary to evaluate the
scales using the exploratory factor analysis method because the theoretical structures of the latent
variables have been determined and it may be inappropriate to make subjective decisions such as
the number of factors to be extracted or discovered (Hair, Black, Babin, & Anderson, 2010). As
the current study use SmartPLS 3.2.7, the SmartPLS model includes a measurement model and
structure model (Nguyen & Vu, 2020). After estimating the SmartPLS model, the study first
examined the item reliability (outer loadings), the convergent validity and the discriminant validity
of the scales. Both Cronbach's alpha and combined reliability (CR) coefficients were used to
evaluate internal consistency with threshold > 0.7, and a few observed variables may have loadings
54
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
just > 0.5 to be considered acceptable; and convergent validity is achieved when the average
variance extracted (AVE) is greater than 0.5 (Hair, Hult, Ringle, & Sarstedt, 2016). The results in
Table 2 show that most the scales have Cronbach's Alpha, CR and AVE values above the
respective required thresholds, demonstrating the reliability and convergent validity of the scales
measuring the concepts.
Table 2
The reliability and convergent validity
Concepts
Factor loadings Cronbach's
Alpha
CR
AVE
[0.812 - 0.877]
[0.844 - 0.884]
[0.749 - 0.865]
[0.676 - 0.754]
[0.690 - 0.827]
0.864
0.835
0.874
0.814
0.851
0.908
0.901
0.909
0.865
0.889
0.711
0.752
0.667
0.517
0.573
Intention to follow
IF
IR
Intention to recommend
Message quality
MQ
OL
PS
Online opinion leadership
Parasocial interaction
Source: The researcher’s data analysis
Hair et al. (2016) proposed that the discriminant validity is assessed using the criteria of
Heterotrait-Monotrait (HTMT) values of less than 0.9. Results in Table 3 shows that HTMT values
of concept pairs are less than 0.9, the scales of the concepts achieve discriminant validity.
Table 3
Heterotrait-Monotrait Ratio (HTMT)
Online
Intention to Intention to Message
opinion
follow
recommend quality
leadership
IF
IR
MQ
OL
0.883
0.773
0.690
0.830
Intention to recommend
Message quality
IR
0.791
0.685
0.846
MQ
0.686
0.740
Online opinion leadership OL
0.722
Parasocial interaction
PS
Source: The researcher’s data analysis
4.3. Structural model and hypothesis testing
The result of the multi-collinearity test from the SmartPLS output shows that the maximum
VIF value of latent variables is 1.566; thereby multi-collinearity being not an issue in this research.
The coefficients of determination (R²) of the intention to follow advice (IF) and the intention to
recommend (IR) are 0.455 and 0.459 respectively, showing that with the research data, the model
explains 45.5%. and 45.9% respectively of the variation of intention to follow (IF) and intention
to recommend (IR). In addition, the R² value of message quality (MQ) is 0.477, indicating that
47.7% variation of message quality (MQ) is explained by two concepts of online opinion
leadership (OL) and parasocial interaction (PS). Besides R², another means of assessing predictive
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
55
accuracy that is Q² was also calculated using SmartPLS blindfolding procedure with omission
distance of six. Results also show that Q² of intention to follow (IF), intention to recommend (IR)
and message quality (MQ) are 0.297, 0.319 and 0.287, respectively; all are larger than zero,
indicating that these endogenous variables have predictive relevance with their respective
predecessors (Hair, Jeffrey, Sarstedt, & Christian, 2019).
The results of testing the structural model by the bootstrapping method (Figure 1) show
that the online opinion leadership (OL) has a significant, positive impact on the message quality
(MQ) (β = 0.315; p <0.001) and parasocial interaction (PS) also has a significant, positive impact
on the message quality (MQ) (β = 0.455; p <0.001). Therefore, hypotheses H1 and H2 are
supported. The results also show that the message quality (MQ) has a significant, positive impact
on the intention to follow (IF) (β = 0.667; p <0.001) as well as on the intention to recommend (IR)
(β = 0.682; p < 0.001). Therefore, hypothesis H3 and H4 are supported.
Figure 2. Testing of structural equation model (using bootstrapping)
Note: *** Statistically significant at p<0.001
With all hypotheses having been supported based on estimation using the bootstrapping
method, according to Preacher and Hayes (2008), it is possible to conclude the mediating role of
message quality in the relationship between the source characteristics and consumers’ behavioural
intentions. Table 4 presents the results of the indirect effects of the SmartPLS-based structural
equation modelling by bootstrapping. The results provided further evidence of the mediating role
of message quality (all p-values below 0.05), with total indirect effects of parasocial interaction
on both behavioural intentions via message quality being larger than those of online opinion
leadership.
Table 4
Indirect effects
Sample Mean Standard Deviation T Statistics
P Values
0.000
OL -> MQ -> IF
PS -> MQ -> IF
OL -> MQ -> IR
PS -> MQ -> IR
0.215
0.309
0.216
0.311
0.040
0.047
0.041
0.047
5.260
6.516
5.175
6.641
0.000
0.000
0.000
Source: The researcher’s data analysis
56
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
4.4. Discussions
Research results show that online opinion leadership and parasocial interaction have a
positive effect on message quality; and in turn, message quality positively affects both the intention
follow advice (i.e., to respond to the product information embedded in the message) spread by the
online opinion leader and the intention relating the online opinion leader (to recommend to others).
More generally, in this research, the source and message characteristics both play a role in driving
consumers’ behavioural intentions. This overall finding is consistent with the studies such as
Cheung and Thadani (2012) examining both source and message influence on customer
behavioural intentions; as well as is consistent with studies into the influence of source
characteristics such as online opinion leadership (Casaló et al., 2020) or parasocial interaction
(Sokolova & Kefi, 2020) on behavioural intentions.
Unlike many previous studies, this study examines the causal relationships between source
and message characteristics in line with the causality between two routes of persuasion stipulated
in the ELM theory (Petty & Cacioppo, 1986). The findings in this study are broadly consistent
with Le et al. (2020) in that perceptions of source characteristics lead to the perceptions of message
quality and subsequently to the consumer’s behavioural intention. However, Le et al. (2020)
considered interpersonal relationship characteristic of information source being represented by
source intimacy - a close relationship or bond with the source as if in a reciprocal relationship. In
the context of the Internet and social media, it is not always possible that the online opinion leaders
interact or have real relationships with their followers, so this study considers the interpersonal
relationship characteristic of the source to be represented by perceptions of parasocial relationship
in which the source may not know the followers but can still have an influence on their behaviour
(Labrecque, 2014).
In this study, message quality acts as a mediating variable in the relationship between
source characteristics and customer behavioural intention. According to Kenny (2008), it is not
necessary to assess whether there is partial or full intermediation. Although some studies found
that source characteristics can have a direct influence on behavioural intentions (Casaló et al.,
2020; Cheung & Thadani, 2012; Sokolova & Kefi, 2020); this study focuses on the causal
mechanism in line with ELM theory and the parsimony requirement of theory.
5. Conclusion
This research contributes to the advances of information processing theories related to
consumers who process persuasive cues and information about the products or services exposed
to them through the online environment. The result on the mediating role of the message quality
also provides support for the causality of the information processing routes in accordance with the
ELM model when consumers are exposed to a huge amount of information on the online platforms
and are unable to process the information thoroughly. Moreover, the research also contributes
evidence on information procession theories and the application of ELM theory in the emerging
market context, which is an important marketing context where current research is scant.
The results of this research also provide implications for managers to build digital
marketing strategies incorporating influencer marketing and EWOM marketing. Although UGCs
are non-commercial and influencers may not be for monetary benefits, marketers can still influence
influencers on social platforms and their generated and disseminated content (arguments).
Specifically, marketers need to detect influencers on online social platforms and target them with
appropriate communication strategies, so that this audience will be the creator and distributor of
UGCs affecting final consumers. In addition, the finding of the mediating role of message quality
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
57
shows that once consumers move from the peripheral cue processing route to the central
information merit processing route, the marketers must also monitor and assess the consumer’s
perceptions of the message quality through the online opinion leaders. The result also shows that
a micro-influencer when conveying a quality message about the business's products or services
not only benefits the business but also themselves.
Like any other study, this study has some limitations that may be considered for venues for
future research. This study was a cross-sectional survey, so in the future, it is possible to combine
other methods such as longitudinal studies to increase the generalisability of the results.
Questionnaires were sent to potential respondents in the Ho Chi Minh City area, so to generalise
and confirm the results found in this study, future studies may need to be expanded to a sample in
the larger geographic area such as other regions in Vietnam and other countries. In addition,
research may also need to be considered with other product and service categories. Finally, in this
study, the message factor was operationalised as message quality, and future research may consider
other message characteristics that may also influence consumer behavioural intentions.
References
Abidin, C. (2016). Visibility labour: Engaging with Influencers’ fashion brands and #OOTD
advertorial campaigns on Instagram. Media International Australia, 161(1), 86-100.
Al-Ansi, A., Olya, H. G. T., & Han, H. (2019). Effect of general risk on trust, satisfaction, and
recommendation intention for halal food. International Journal of Hospitality Management,
83, 210-219.
Anderson, E. W. (1998). Customer satisfaction and word of mouth. Journal of Service Research,
1(1), 5-17.
Bhattacherjee, A., & Sanford, C. (2006). Influence processes for information technology
acceptance: An Elaboration Likelihood Model. MIS Quarterly, 30(4), 805-825.
Bijen, Y. J. (2017). #AD: The effects of an influencer, comments and product combination on
brand image. (Master's thesis). University of Twente, Enschede, Netherlands.
Casaló, L. V., Flavián, C., & Ibáñez-Sánchez, S. (2020). Influencers on Instagram: Antecedents
and consequences of opinion leadership. Journal of Business Research, 117, 510-519.
Chatterjee, P. (2011). Drivers of new product recommending and referral behaviour on social
network sites. International Journal of Advertising, 30(1), 77-101.
Cheung, C. M. K., & Thadani, D. R. (2012). The impact of electronic word-of-mouth
communication: A literature analysis and integrative model. Decision Support Systems,
54(1), 461-470.
De Veirman, M., Cauberghe, V., & Hudders, L. (2017). Marketing through Instagram influencers:
The impact of number of followers and product divergence on brand attitude. International
Journal of Advertising, 36(5), 798-828.
Depouilly, X. (2019). Vietnam in the digital era: Evolution, impacts and social considerations.
Retrieved
November
1,
2020,
from
content/uploads/1912-Digital-Vietnam-2019-by-Indochina-Research.pdf
Djafarova, E., & Trofimenko, O. (2019). ‘Instafamous’ - Credibility and self-presentation of micro-
celebrities on social media. Information, Communication & Society, 22(10), 1432-1446.
58
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
Gentina, E., Butori, R., & Heath, T. B. (2014). Unique but integrated: The role of individuation
and assimilation processes in teen opinion leadership. Journal of Business Research, 67(2),
83-91.
Goldenberg, J., Lehmann, D. R., Shidlovski, D., & Barak, M. M. (2006). The role of expert versus
social opinion leaders in new product adoption. Marketing Science Institute Reports
(Working Paper Series), 6(4), 67-84.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th
ed.). Upper Saddle River, NJ: Prentice Hall.
Hair, J. F., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2016). A primer on Partial Least Squares
Structural Equation Modeling (PLS-SEM). Thousand Oaks, CA: Sage Publications.
Hair, J. F., Jeffrey, J. R., Sarstedt, M., & Christian, M. R. (2019). When to use and how to report
the results of PLS-SEM. European Business Review, 31(1), 2-24.
Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronic word-of-
mouth via consumer-opinion platforms: What motivates consumers to articulate themselves
on the Internet? Journal of Interactive Marketing, 18(1), 38-52.
Horton, D., & Wohl, R. R. (1956). Mass communication and para-social interaction. Psychiatry,
19(3), 215-229.
Hovland, C. I. (1948). Social communication. Proceedings of the American Philosophical Society,
92(5), 371-375.
Ilicic, J., & Webster, C. M. (2014). Investigating consumer-brand relational authenticity. Journal
of Brand Management, 21(4), 342-363.
Iyengar, R., Van den Bulte, C., & Valente, T. W. (2010). Opinion leadership and social contagion
in new product diffusion. Marketing Science, 30(2), 195-212.
Kenny, D. A. (2008). Reflections on mediation. Organizational Research Methods, 11(2),
353-358.
Labrecque, L. I. (2014). Fostering consumer-brand relationships in social media environments:
The role of parasocial interaction. Journal of Interactive Marketing, 28(2), 134-148.
Le, T. D., Robinson, L. J., & Dobele, A. R. (2020). Word-of-mouth information processing routes:
The mediating role of message and source characteristics. Journal of Consumer Behaviour,
19(2), 171-181.
Li, Y.-M., Lee, Y.-L., & Lien, N.-J. (2012). Online social advertising via influential endorsers.
International Journal of Electronic Commerce, 16(3), 119-154.
Mahapatra, S., & Mishra, A. (2017). Acceptance and forwarding of electronic word of mouth.
Marketing Intelligence & Planning, 35(5), 594-610.
Min, J. H. J., Chang, H. J. J., Jai, T.-M. C., & Ziegler, M. (2019). The effects of celebrity-brand
congruence and publicity on consumer attitudes and buying behavior. Fashion and Textiles,
6(1), 1-19.
Nguyen, H. M., & Vu, T. H. (2020). Phân tích dữ liệu áp dụng mô hình PLS - SEM [Data analysis
using PLS - SEM model]. Ho Chi Minh, Vietnam: NXB Kinh Tế Thành Phố Hồ Chí Minh.
Park, C. S., & Kaye, B. K. (2017). The tweet goes on: Interconnection of Twitter opinion leadership,
network size, and civic engagement. Computers in Human Behavior, 69, 174-180.
Kieu Anh Tai et al. Ho Chi Minh City Open University Journal of Science, 11(1), 47-59
59
Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and peripheral
routes to attitude change. New York, NY: Springer-Verlag.
Petty, R. E., Cacioppo, J. T., & Schumann, D. (1983). Central and peripheral routes to advertising
effectiveness: The moderating role of involvement. Journal of Consumer Research, 10(2),
135-146.
Preacher, K., & Hayes, A. (2008). Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3),
879-891.
Sheth, J. N. (2011). Impact of emerging markets on marketing: Rethinking existing perspectives
and practices. Journal of Marketing, 75(4), 166-182.
Sokolova, K., & Kefi, H. (2020). Instagram and YouTube bloggers promote it, why should I buy?
How credibility and parasocial interaction influence purchase intentions. Journal of
Retailing and Consumer Services, 53, Article 101962.
Stokburger-Sauer, N. E., & Hoyer, W. D. (2009). Consumer advisors revisited: What drives those
with market mavenism and opinion leadership tendencies and why? Journal of Consumer
Behaviour, 8(2/3), 100-115.
Susarla, A., Oh, J.-H., & Tan, Y. (2016). Influentials, imitables, or susceptibles? Virality and
Word-of-Mouth conversations in online social networks. Journal of Management
Information Systems, 33(1), 139-170.
Sweeney, J. C., Geoffrey, N. S., & Mazzarol, T. (2012). Word of mouth: Measuring the power of
individual messages. European Journal of Marketing, 46(1/2), 237-257.
Trusov, M., Bucklin, R. E., & Pauwels, K. (2009). Effects of Word-of-Mouth versus traditional
marketing: Findings from an Internet social networking site. Journal of Marketing, 73(5),
90-102.
We Are Social, & Hootsuite. (2019). Digital 2019 - Vietnam. Retrieved November 1, 2020, from
13 trang
16600
Free
Bạn đang xem tài liệu "The impact of online micro-influencers on behavioural intentions: The case of mobile phone customers in Ho Chi Minh city", để tải tài liệu gốc về máy hãy click vào nút Download ở trên
File đính kèm:
- the_impact_of_online_micro_influencers_on_behavioural_intent.pdf
