View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by International Institute for Science, Technology and Education (IISTE): E-Journals

European Journal of Business and Management www.iiste.org

ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

The Influence of Internet Celebrity on Purchase Decision and

Materialism: The Mediating Role of Para-social Relationships and
Identification
Hamza Kaka Abdul Wahab 1 Meng Tao2
1. School of Business Administration, Dongbei University of Finance and Economics, 217 Jianshan
Street, Shahekou District, Dalian 116025, Liaoning, China
2. International Business College, Dongbei University of Finance and Economics, 217 Jianshan Street,
Shahekou District, Dalian 116025, Liaoning, China
* E-mail of the corresponding author: [email protected]
Abstract
The objective of this paper is to examine the influence of internet celebrities on customer purchase decision and
materialism, with special attention on parasocial relationships and identification as mediators and how they play a
role in celebrity influence on purchase decision and materialism.
413 female students from three universities in Dalian, Liaoning Province, China were sampled randomly using
structured questionnaires. Structural equation modelling was used with the help of SmartPls3.
The highlight of the study is the mediating roles of para-social relationships and identification between internet
celebrities and purchase decision and materialism. Again, the study portrayed that internet celebrity significantly
influence parasocial relationships, whereas parasocial relationships in turn influence identification with the
celebrity. More so, identification significantly influences purchase decision and materialism. Furthermore, internet
celebrity ultimately influences purchase decision and materialism through the mediating roles of para-social
relationships and identification.
The study made available some significant findings for marketers and academics. Marketers can use the internet
celebrities as third parties to communicate to their target audience through their blogs and pages on the internet by
endorsing their brands. Academics can explore how para-social relationships and identification mediates between
internet celebrities and purchase decision and materialism.
To the knowledge of the authors, this study is the first of its kind in an attempt to examine the influence of internet
celebrity on purchase decision and materialism using parasocial relationships and identification as mediators.
Keywords- Internet celebrity, Para-social relationships, Identification, Purchase decision and materialism,
Structural equation modelling
Paper type: Research paper
DOI: 10.7176/EJBM/11-15-20
Publication date:May 31st 2019

1. Introduction

Boorstin (1961) indicated that, a celebrity is a person who is famous, irrespective of whether the fame was achieved
via sports, politics, entertainment, medicine or having a close relationship with other celebrities. Moreover, some
findings have established how celebrities influence attitudes of the public, for instance, how consumers behave in
their quest to make a purchase (Tripp et al.,, 1994; Till and Shimp, 1998). Alperstein (1991) indicated that,
messages from celebrities are now part of the individual’s social construction of reality.

Yue & Cheung, (2000) posited that celebrity worship has become the order of the day amongst young people
around the world, especially teenage and adolescent girls. Josselson, (1991) posited that this could arise from the
fact that they have the burning desire for intimacy and identification. Moreover, worshipping of celebrities could
have a huge influence in the shaping of followers’ attitudes, behaviours and values (Schultze et al, 1991).

Celebrities are being revered and held in high regard by female teenagers and adolescents alike (McCutcheon et
al, 2002; McCutcheon et al,2003 and Maltby et al, 2004) and this idolisation of celebrities stem from the fact that

183
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

they (celebrities) are seen to be elegant and classy than ordinary individuals. These celebrities often appear to be
well groomed by wearing makeup, undergoing cosmetic surgery and wearing elegant clothes. More so, they flaunt
their wealth via the cars they own, expensive jewellery and houses.

Teenagers and adolescents females more often than not get tempted to imitate the celebrities they idolised by
dressing like them and also purchasing the brands or products endorsed by these celebrities (Austin et al.,2008;
Chan et al, 2013). By virtue of celebrities’ influence on their followers (Chan & Zhang, 2007; Fraser and Brown,
2002; Wen, 2017), marketers and advertisers alike usually capitalise on this to entreat teenagers and adolescents
to purchase and consume products endorsed by their favourite celebrities (Austin et al., 2008; Chan et al, 2013).

Goldberg et al., (2003) posited that teenagers and adolescents adoption of materialistic values goes a long way to
have an impact on the balance between private and public choices they often make in their life time. It is very
crucial to note that materialism is seen as a negative value by virtue of the fact that it negates interpersonal
relationships and it is also related to happiness and subjective well-being in a negative sense (Kasser, 2002).
Furthermore, it is also related to material values and has a collective orientation toward family and religious values
(Burrough and Rindfleisch, 2002).

According to the generalised social comparison theory, individuals tend to compare the personal materials they
possess with the ones owned by prominent individuals to establish their status in society (Saunders, 2001). A point
worthy of note is that apart from comparing themselves to their peers, teenagers and adolescents may also compare
themselves to celebrities.
In the context of China, studying materialism and celebrity worship is different from the Western way of life. Zhao
(1997) argue that as a result of the long standing values regarding families and human relations, mass consumption
in Chinese society is quite contrary to that of the Western societies.

Despite the numerous studies in the area of celebrity influence (Chan & Zhang, 2007; Fraser and Brown, 2002;
Wen, 2017), celebrity endorsement (Austin et al, 2008; Chan et al, 2013), and celebrity worship (McCutcheon et
al, 2002; McCutcheon et al, 2003 and Maltby et al, 2004), there is a gap left to be filled; thus, it is the role internet
celebrities play on their followers’ purchase decision and materialism through identification and parasocial
relationship.

The issue, then, is whether parasocial relationships and identification with the internet celebrity idol contributes to
purchase decision and materialism amongst female teenagers and adolescents in China.

2. Literature Review

2.1 Social Influence

Influence as a theory had long been studied by scholars and researchers in the fields of marketing, political science,
sociology and communication (Rogers, 1962; Katz and Lazars Feld, 1955). Influence plays a crucial role in our
day to day activities and how society operates. For instance, one could see observations on how fashion spreads
(Gladwell 2002) and how people vote (Berry and Keller 2003). More so, China is seen as having a high-power
distance culture because of its Confucianism background (Looner et al., 1980, Hofstede and Bond, 1988).
Influence leads to social impact, and Latane (1981) defines social impact as an influence exerted on a person’s
way of thinking, how the person feels or how the person behaves which is triggered by the implied, real, or
imaginary existence or activities of others. From this definition, it is apparent that individuals can get influenced
by the actions of others in the society to whom they aspire.
Taking the social influence theory into consideration, social influence processes cannot create uniformity of
opinion on their own. According to Festinger’s (1950) influential theory of group’s uniformity pressures, pressures
ought to encompass pressures rejection of strays as a contributory process. More so, in the empirical sense, usually,
numerous groups have divergence of opinions, which brings about division on basic issues that can be supported
factually. Most often, even those on the minority side hold on to their views and try to convince the majority. By
this, people can easily be influenced irrespective of where they belong. Latane and Wolf (1981), also stated that
the differences between those on the majority side and the minority side basically manifest their different size and
status instead of qualitative differences in modes of influence.

184
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

2.2 Social Media Unique Attributes

Modern media technologies consist of the Websites, Internet and numerous social media platforms such as Weibo,
WeChat, QQ, and Youku in China; whereas in the West, there are Twitter, Pinterest, Facebook, WhatsApp,
Instagram and Tumblr. These social media platforms make it easy and quick for users to establish and share what
is called user-generated messages and, instant communication with different users on numerous devices held in
the hand (Sundar and Limperos 2013). According to Eveland (2003), interactivity is the crucial feature that
separates conventional mass media from contemporary social media technologies; it is important to note that, users
are both sources and receivers as well, and virtually simultaneous also.
Again, social media users easily get accustomed to technology, and content can revolve around the self. Joinson
and Paine (2007) found out that, self-disclosure has long played an important role in computer-mediated
communication. This is so because, social media are hugely more personal channels compared to the conservative
impersonal mass media. This manifests itself in the crowd of personalized customization of sites, blogs and
digitized pictures that portray the self, friends, or strangers a person admires personally on Pinterest, Snap chat,
and Instagram with visual story applications (Wortham and Goel, 2013).
Thirdly, Amichai-Hamburger (2007) suggested that, social media platforms provide for people who are like-
minded, and therefore homogeneous in nature, whereas mass media presents a wide array of audiences who are
heterogeneous; therefore it is no wonder that social media is seen as the media of peers.
Furthermore, according to Sundar et al. (2012), by just providing users of digital communicative technologies the
capability to customize, shape and direct interactions online, modern media changes mass media receivers who
were previously passive into active communicators, which boost personal agency, autonomy, and self-efficacy.
Last, but not least, Green et al (2004) and Barak (2007) suggested that, social media platforms are viewed as
interpersonal procedures that present users with the leverage to use or view videos, animation, and transformative
multi-media cues. These cues present emotional presence, providing the user with an opportunity to
psychologically get use to some internet domains to spur the discontinuation of attitude change and belief.
2.3 Internet celebrity and Para-social Relationships
It is believed that fame is a psychological concept which is tantamount to object relations theory (Sheldon and
Elliot, 1998) and also diverse regarding its scope. Other studies have established how celebrities influence attitudes
of the public, such as consumer behaviour (Tripp Jensen, and Carlson, 1994, Till and Shimp, 1998).
Para-social relationships are relationships created by an admirer of a celebrity which are always one-sided (Horton
and Wohl, 1956), thus, it flows from the follower to the celebrity with the celebrity not aware of such relationships.
These one-sided relationships are established as a result of the usage of the media often by interacting with the
celebrity (Peres and Rubin, 1989).
The celebrity influence model which was put forward by Brown et al., (2003); Brown and de Matviuk, (2010) can
be used to expound the available relationship between the influence of internet celebrities and purchase decision
and materialism. The celebrity influence model basically relied on the following psychological orientations:
identification and parasocial relationship. According to Horton and Wohl, (1956), Hartmann and Gold Hoorn,
(2011)), and Horton and Strauss, (1957), para-social relationship brings about intimate relationship which is
perceived by an individual to have with media celebrities.
The celebrity influence model proposed that exposure to the media is the basis of establishing parasocial
relationships. The chances of meeting celebrities in person by young people are very slim, so they often get to link
up with the celebrities via the media and the internet is not an exception. It was suggested that in order to keep in
touch with the celebrities, they follow them on their social media blogs to read their latest posts, look at their
photos, and gossips about them (celebrities) and access other updates (Redmond and Holmes, 2007).
By virtue of the social relationship with the internet celebrities, followers of these celebrities view themselves as
having some level of closeness with the celebrities (Brown et al., 2003; Brown and de Matviuk, 2010; Kosenko et
al., 2016).
Other research findings prove that the relationships created between the celebrities and the followers due to the
exposure and influence through social media are well established (Brown et al., 2003; Brown and de Matviuk,
2010; Kosenko et al., 2016). Nevertheless, Giles, (2002) maintains, that the followers’ perception of having
intimate relationships with social media celebrities is nothing but a mere illusion, since it is established and
maintained only by the follower without the celebrities being aware of it. We there by formulated this hypothesis:

185
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

H1: Influence of internet celebrity leads to parasocial relationship with the celebrity.
2.4 Para-social Relationships and Identification
Numerous researchers have argued that the longer an individual uses or spends time on the internet, the less
effective the individual’s interpersonal relationship would be (Kraut et al., 1998 and Sanders et al., 2000). Scholars
have posited that users of the internet attain social values with the help of online communities and they establish
relationships with their peers as a result (Eastin and LaRose, 2005; Shaw and Gant, 2002; Xie, 2008).
Relationships with social media celebrities can make the follower develop some empathy towards the celebrities
and therefore breeds the ground for celebrity identification (Wen, 2017; Chia and Poo, 2009; Fraser and Brown,
2002; Cheney, 1983). Further, celebrity identification makes the followers believe that they share something in
common with the celebrity (Cohen, 2001). We therefore propose this hypothesis:
H2: Parasocial relationships with the Internet celebrity has a positive effect on identification with the celebrity.
2.5 Identification, Purchase Decision and Materialism
Consumers of the media react to members in the media in numerous ways, and identification happens to be one of
those ways (Hoffner & Cantor, 1991). It is without doubt that, followers of the media personalities respond by
either admiring or hating the media personalities, feeling attached to them (Newton & Buck, 1985; Newton, Buck
& Woelfel,1986), looking for things they have in common or differences between the personalities and themselves
(Reeves & Miller, 1978), and also, yearn to be like the media personalities (Hoffner, 1996).
Numerous scholars have researched around the area of the impact of celebrity endorsers on consumers’ behaviour
and such research dwelled on attractiveness or credibility as a yardstick to measure how effective a message is
(Kamins, Brand, Hoeke & Moe, 1989). However, the theories of Bandura, Kelman and Burke postulated that
identification is another factor playing a role in celebrity effects (Basil, 1996).
According to Burke (1950), identification has to do with the link between a character and audience member. This
theory is from a drama point of view, but Kelman (1961) also propounded a theory of identification regarding
opinion change. Kelman (1961) suggested that an individual’s adoption of another person’s behaviour or attitude
is to satisfy his/her self-defining relationship with the other individual; this then leads to identifying him/herself
with the person. On the other hand, Bandura’s (1977) social cognitive theory regarding celebrity identification,
pointed out that this could bring about behavioural adaptation. In a nutshell, this theory has to do with the
suggestion that an individual’s behaviour is mediated as a result of identifying with a celebrity model.
Audience most often than not get carried away by the characters portrayed in movies, and television programs and
therefore, tend to identify themselves with the characters, and this is no different from internet celebrities. Basil
(1996) established that, audience who identify with celebrities who promote health messages heighten the adoption
of the messages. More so, a study was carried out about ethnography audience and it was established that, TV
audience often emphasise on their feelings and reactions to characters, which includes the proclamation of their
strong identification with characters (Liebes & Katz, 1990). It is crucial to note that, audience who identify with
celebrities in the media get the chance to experience social reality from different views and it also goes a long way
in shaping the development of self-identity and social attitudes (Erikson, 1968). Findings by scholars, suggests,
that children who communicate most often with friends tend to be materialistic (Moschis and Churchil 1978;
Churchil and Moschis 1979) and are easily opened to being influenced (Achenreiner, 1979). So since the followers
of the celebrities see them as their peers as a result of para-social relationships created, this then leads to
identification with the celebrity. We therefore formulated this hypothesis:
H3: Identification with the internet celebrity as a result of parasocial relationships has a positive effect on
purchase decision and materialism.
3. Methodology
This part of the research entails the discussion of techniques for data collection and methods for sampling. The
section again describes the techniques for data analysis and constructs measurement.
3.1 The design of questionnaire and appraisal of scale
A structured questionnaire comprising of 24 items was designed to measure identification, para-social relationships
and to evaluate influence of internet celebrities on teenage and adolescent women attitude towards purchase
decision and materialism. The questionnaire had four parts and comprise of demography of the respondents,

186
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

regarding their ages and sexual orientations. The second part of the questionnaire had to do with identification
with the internet celebrity. While the third part contain questions about parasocial relationships with the celebrity.
Some of the items used to measure identification and para-social relationships were adapted from Celebrity-
Persona Identification Scale (Bocarnea and Brown, 2007) and (CPPI), Celebrity-Persona Parasocial Interaction
Scale (Bocarnea and Brown, 2007) respectively with some changes made to them. Fourth part of the questionnaire
also dealt with the respondents purchase decision and materialism. 5 point Likert scale was used and answers
ranging from 1=strongly disagree to 5= strongly agree.
3.2 Sampling procedures
The sample was made up of female students from three universities from Dalian in Liaoning province of China.
What informed our decision for sampling female students from universities was as a result of the belief that they
identify themselves with celebrities, like to imitate celebrities which then make them to be materialistic (Chan et
al., 2008). Demographically, the respondents fell under different types of demographic dispensations, such as age
and sexual orientation. We sampled 550 students through simple random sampling. We used paper and pencil
survey to identify potential respondents. Though it has its own disadvantages such as cost, interviewers bias and
longer period of collection of data (Aaker et al., 2000), we employed this method because it paves the way to
clarify some of the questions to the respondents, allows for speed, sample control, anonymity, and quality control
(Kassim and Abdullah, 2010). In the end, 413 useable questionnaires were obtained in the course of a month and
four days within the month of June and July 2018, representing 75% of the sampled population. The respondents
were between the ages of 18 years to 29 years.
3.3 Data analysis
Structural equation modeling (SEM) technique was employed supported by SmartPls 3 (Wende & Becker, 2015).
This study mirrored the two step approach propounded by Anderson and Gerbing, (1998). To establish the
construct validity of the model, we carried out a confirmatory factor analysis (CFA). The structural model was
examined to test the hypotheses and to test how fit the model is.
4. Analysis and results
This section of the study has to do with descriptive analysis of data. More so, it encompasses results’ interpretation
and findings.

IDEN
H3

IIC H2 PDM

H1

PSR

Figure 1.

Proposed conceptual model

187
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Table I. Measurement of constructs

Constructs Items Cronbach α

Influence of internet 0.746

Celebrities (IIC1) Influence of internet celebrities lead to para-social

relationships with the celebrities

(IIC2) Exposure to your internet celebrity idol will predict

your parasocial relationships with her

(IIC3) I understand the emotions the celebrity experience

(IIC4) Exposure to your internet celebrity idol through

para-social relationships will predict identification with her

Para-social (PSRI) The para-social relationships you have with your celebrity 0.723

Relationships idol on the internet leads to identification with her

(PSR2) Parasocial relationships with your celebrity idol on the

Internet makes you aspire to look like her

(PSR3) Parasocial relationships with your celebrity idol on the

Internet makes you aspire to think like her

(PSR4) Parasocial relationships with your celebrity idol on the

Internet makes you aspire to behave like her

Identification (ID1) Your heavy dependence on the internet makes you have 0.778

some feelings for your internet celebrity idol

(ID2) Your heavy dependence on the internet makes you feel

related to her

(ID3) Your heavy dependence on the internet makes you see her

as a personal role model

(ID4) Identification with your internet celebrity increase when

your para-social relationship with her develops

(ID5) Identification with your internet celebrity idol will make you

imitate her

(ID6) Identification with your internet celebrity idol will make you

188
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

adopt her values, beliefs and behaviours

Purchase (PDM1) I will purchase a product endorsed by my internet celebrity idol 0.737

Decision and (PDM2) I approve of persons being influenced by their celebrity idols to

Materialism to make a purchase of products being used by their celebrity idols

(PDM3) I believe in the quality of products my internet celebrity idol endorses

(PDM4) Identification with your internet celebrity idol has a positive

relationship with purchase decision and materialism

(PDM5) Interacting with internet celebrities has a positive effect on

purchase decision and materialism

(ID6) If I become materialistic as a result of the influence by my celebrity idol,

I would be embarrassed to tell people other than family and close friends

4.1 Demographic Characteristics of respondents and controlled variables

Table II portrays the respondents’ characteristics and controlled variables. Age and sexual orientation of the
respondents are the highest points. The respondents’ age ranged from 18 to 29. Majority of respondents were
between 18 to 20 years old, representing (45.0 percent), whereas the lowest were between 24 to 26 years old,
representing (12.8 percent). With regard to sexual orientations, 85.7 percent of the respondents were heterosexuals,
whereas 3.9 percent were homosexuals (Lesbians). When asked about the celebrity they identified with the most
on the internet, majority of the respondents chose Papi Tyan (41.9 percent), whereas the lowest was Guo Meimei
with 6.3 percent. With hours spent on social media in a day, majority of the respondents claimed they spent
between 1 to 2 hours in a day (43.1 percent), whereas the respondents with the lowest percentage of 14.0 spent
between 7 to 8 hours a day surfing the internet.

Table 2

Demographic characteristics Frequency % Cumulative (%)

Age

18-20 years 186 45 45

21-23 years 112 27.1 72.2

24-26 years 53 12.8 85

27-29 years 62 15.0 100

Sexual Orientation

Bisexual 43 10.4 10.4

Homosexual 16 3.9 14.3

Heterosexual 354 85.7 100

Beautiful Celebrity

Naicha Meimei 158 38.3 38.3

189
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Papi Tyan 133 32.2 70.5

Zhang Dayie 87 21.1 91.5

Gao Meimei 35 8.5 100

Identified Celebrity

Papi Tyan 173 41.9 41.9

Naicha Meimei 124 30 70

Zhang Dayie 88 21.3 21.8

Gao Meimei 26 6.3 100

Hours spent on the internet

1-2 hours 178 43.1 43.1

3-4 hours 112 27.1 70.2

5-6 hours 65 15.7 86.0

7-8 hours 58 14 100

Note: N = 413

4.2 The Results of Measurement Model

To establish the goodness of fit of the measurement model, the unidimentionality technique was employed,
reliability of the items, content validity, convergent validity and discriminant validity were confirmed to have met
the thresholds set for them.

4.2.1 Unidimensionality technique: According to Awang (2010), in order to achieve the unidimentionality, all the
measuring items ought to have the acceptable factor loadings for the individual latent variables. To adhere to the
unidimentionality technique postulated by Awang (2010), we deleted items whose factor loadings were below 0.60
until we arrived at the acceptable factor loadings for the many constructs.

4.2.2 Items reliability: Considering the confirmatory factor analysis (CFA) results of the study in Table III, the
individual items loaded above the level posited by Chin et al (1997), which is 0.60. Moreover, the internal
reliability results when using values of Cronbach α ranged from 0.723 to 0.778, exceeding the 0.70 threshold
suggested by Nunnally and Bernstein (1994). Again, all the latent constructs values of the composite reliability
ranged from 0.817 to 0.870 which were by far above the threshold level of 0.70 postulated by (Hair et al., 2010).

4.2.3 Content validity: Factor loadings are used to make sure that all the individual items created to help in the
measurement of a construct are loaded very high and also have significant impact on the constructs whiles
measuring them as suggested by (Chin, 1998 and Hair et al., 2010). Figure 2 depicts that all the individual items
on the different constructs were significantly and highly loading on each construct which ultimately confirms the
content validity of the measurement model.

4.2.4 Convergent validity: Taking the literature of Structural Equation Modelling (SEM) into consideration,
convergent validity can be referred to as the level to which a set of indicators converges in measuring the concept
which is being considered (Bagozzi and Yi, 1988; Hair et al., 2010). From the definition, we then used the
Cronbach alpha’s coefficient, items reliability, average variance extracted (AVE) and internal consistency to aid
in confirming the convergent validity.

190
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

More so, the average variances extracted manifest the whole amount of the shared variance among the indicators,
measuring a particular latent construct ranging from 0.527 to 0.691, overshadowing the acceptable level of 0.50
suggested by Bagozzi and Yi (988) and Hair et al., (2010). All the latent constructs had composite reliability of at
least 0.817 and lowest AVE of 0.527.

Table III CFA Results of Measures

Constructs Items Internal reliability Factor loading Convergent validity

Cronbach α Composite reliability (CR)a AVEb

Internet Celebrity IIC1 0.746 0.854 0.853 0.660

IIC2 0.832

IIC4 0.748

Parasocial Relationships PSR1 0.723 0.767 0.817 0.527

PSR2 0.756

PSR3 0.677

PSR4 0.701

Identification IDEN3 0.778 0.862 0.870 0.691

IDEN4 0.878

IDEN5 0.747

Purchase Decision

& materialism PDM1 0.737 0.774 0.851 0.655

PDM4 0.835

PDM5 0.818

Note: aCR= (Σ factor loading)2/{( Σ factor loading)2)+ Σ (variance of error)}; bAVE = Σ (factor loading)2/ { Σ
((factor loading)2 + Σ (variance of error)}. It is very important to note that all the loadings are significant at 0.001
significant level.

4.2.5 (GoF)

The overall measure for model fit for PLS-SEM has its own flaws because it does not generate overall Goodness
of Fit (GoF); however, to check the exploratory strength of the model, rule of R2 can be used (Henseler et al.,
2016). We therefore used the diagnostic tool suggested by Tenenhaus et al., (2005) as the Goodness of Fit (GoF)
index to ascertain the model fit. We did this by computing the average mean value of the AVE values and the
average R2 values with this equation (GoF=√ (AVE × R2)). There is no threshold values set by Tenenhaus et al.,
(2005) in their Goodness of Fit ( GoF) index, but these thresholds were postulated by Wetzels et al., (2009) to
ascertain the analysis of GoF; Small= 0.1, medium= 0.25 and large= 0.36. Having gotten 0.566 as our GoF index
value we therefore conclude that the model has a good fit.

191
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Table IV. Goodness of Fit (GoF)

Constructs AVE R2

Internet Celebrity 0.660

Para-social Relationships 0.527 0.542

Identification 0.691 0.606

Purchase decision and Materialism 0.655 0.378

Average scores 0.633 0.508

AVE ×R2 0.321

GoF = √ (AVE × R2 ) 0.566

4.2.6 Discriminant validity: This refers to the extent at which a set of variables of a certain construct vary from
other constructs in a model. According to Compeau et al., (1999), the variance shared among a set of items
measuring a construct is higher compared to the variance shared with other constructs in a model. Again, according
to the method postulated by Fornell and Larcker (1981), the discriminant validity can be established as a result of
comparison between the square root of the average variance extracted (AVE) values and the correlations among
the constructs.

In a nutshell, there are two criteria involved when evaluating the discriminant validity of a model. First, the square
root of each latent variable must be higher than the correlation between that and the latent variables remaining
(Chin, 1998), whereas the other way is the proposed latent variables which are to be measured must load highly
against other latent variables (Chin et al., 2003). Table 4 depicts that the square roots of the AVE which are
represented diagonally are higher than other values in their columns and rows. These results therefore affirm that
the model’s discriminant validity is enough and also confirm the adequacy of the reliability, convergent validity
and discriminant validity of the measurement model.

Table 5. Discriminant validity and descriptive analysis

Constructs 1 2 3 4 Min Max Mean SD

1. IDEN 0.831 3.00 15.00 10.289 2.468

2. ICC 0.761 0.813 4.00 29.00 14.606 3,444

3. PDM 0.615 0.690 0.809 3.00 15.00 11.702 2.413

4. PSR 0.779 0.736 0.555 0.726 4.00 20.00 12.676 3.074

Note: The bolded values on the diagonal represent the square root of the average variances extracted, whereas the
other values in the rows and columns represent the correlations among the latent constructs.

IDEN: Identification; ICC: Influence of internet celebrity; PDM: Purchase decision and materialism; PSR: Para-
social relationships.

4.3 Results of structural model

After the model’s reliability and validity had been ascertained, we then used SEM to test the hypotheses by virtue
of the fact that there is statistical efficiency from the technique. And also, as a result of its ability to evaluate the
relationships meticulously, it has therefore presented a transition from exploratory to confirmatory analysis (Hair
et al., 2006). In line with Cohen’s (1988) postulations, standardized path coefficients with (β) with absolute values

192
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

of less than 0.10 may be regarded as “small” effect, values in the region of 0.30 may indicate “medium” effect,
whereas coefficients with absolute value of 0.50 or more is regarded as “large” effect. The analyses results are
discussed below.

4.3.1 Analyses results

After testing the hypotheses, we came up with the following results depicted in table V. From the table it can be
deduced that internet celebrity has a strong relationship with para-social relationships. Therefore, the coefficient
of influence of internet celebrity (β=0.736, t-value=40.097, p<0.001) is significant with para-social relationships.
Hence H1 was accepted as suggested in the study. This therefore buttressed our hypothesis that influence of social
media celebrity leads to para-social relationships with the celebrity.

Again, the para-social relationships established with the internet celebrity that leads to identification was
statistically supported (β=0.779, t-value=45.237, p < 0.001). This therefore supported H2 as postulated in the
study.

H3 also portrays another significant relationship between identification and purchase decision & materialism, (β=
0.615, t-value = 19.919, P < 0.001). From this evidence, H3 was statistically supported, so we therefore accepted
H3.

Figure2. Path analysis model

193
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Table 6. Hypothesized direct effect

Hypotheses Hypothesized direct effect β- value t-value p-value Decision

H1 Influence of internet celebrity 0.736 40.097 0.000 Accepted

leads to para-social relationships

with the celebrity.

H2 Para-social relationships with the 0.779 45.237 0.000 Supported

Internet celebrity has a positive

effect on identification with the celebrity.

H3 Identification with the internet celebrity 0.615 19.919 0.000 Accepted

as a result of para-social relationships has

a positive effect on purchase decision and

materialism.

Note *** p< .001, ** p< .01, * p< .05

5. Limitations

Care ought to be taken not to generalise these findings since it is only related to female teenagers and adolescents
in China and for that matter the respondents were students in three universities in Dalian, Liaoning Province.
Moreover, another limitation has to do with the fact that though the items reliability, convergent validity, content
validity, Cronbach alpha and discriminant validity were all satisfactory, however, the SRMR and NFI were not
when we checked for the model fit. So we used the Goodness of Fit (GoF) index propounded by Tenenhaus et al.,
(2005) to ascertain the model fit.

6. Discussion and managerial implications

6.1 Discussion

The highlight of the study is the mediating roles para-social relationships and identification play between internet
celebrities and purchase decision and the materialistic behaviours of their followers.

The study found that internet celebrities have an influence on the purchase decision and materialism of teenage
and adolescent females through para-social relationships and identification. Internet celebrities have a significant
direct impact on para-social relationships with the followers. This buttresses the findings of previous studies on
celebrity influence and para-social relationships (See for e.g. Horton and Wohl, 1956; Horton and Strauss, 1957;
Gold and Hoorn, 2011). Moreover, the study found that para-social relationships with the internet celebrity idol
leads to identification with the celebrity. This is also in line with the findings of previous studies (See Cohen,
2001; Fraser and Brown, 2002; Chia and Poo, 2009). Furthermore, it was found that there is a significant
relationship between identification with the celebrity and purchase decision and materialism. This is because,
teenage and adolescent females relate and see themselves as having similar characteristics as their idolised
celebrities. They then tend to imitate whatever the celebrities do since they see them as their peers (see Moschis
and Churchil 1978; Churchil and Moschis 1979; Achenreiner, 1979; Basil 1996)

6.2 Managerial Implications

The importance of our findings pave the way for marketing managers, NGO’s and Governments. For instance, as
a result of the influence of the internet celebrities on purchase decision and materialism, marketing managers ought
to take advantage to use them in their commercials in a bid to promote their products and services to the target
audience.

194
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Again, because of the identification the followers have with the internet celebrities, marketers and firms who are
into female items should use the internet celebrities to endorse their products and services online instead of using
the well-established traditional celebrities such as actresses and sports women who will demand huge sum of
money to endorse a product or service.

Moreover, firms must take advantage of the power of social media to channel their commercials there since it is
the cheapest compared to television and magazine. They can also embark on ambush marketing by providing the
internet celebrities with some of their products to use and share a video of themselves using the products on their
blogs or pages.

More so, since the followers of the internet celebrities see themselves as having a relationship with the celebrities
as a result of the para-social relationships they established, NGO’s and governments ought to tap into this one
sided relationship by using the internet celebrities to reach out to the youth in order to sensitise them about issues
regarding prostitution, cyber bullying, cyber scamming, depression, exposure to pornographic sites, drug and
alcohol abuse among other factors affecting the youth.

7. Conclusions and direction for future research

The main contribution to this research is mediating roles para-social relationships and celebrity identification play
in investigating the relationship between internet celebrities’ influence on purchase decision and materialism as
against previous studies that look at the relationship between para-social relationships and identification of the
traditional celebrities.

We address this gap in the model and suggested that internet celebrities ought to influence purchase decision and
materialism via para-social relationships and identification. Taking the objective of the study into consideration,
influence of internet celebrities has a significant relationship with purchase decision and materialism through the
mediating roles of para-social relationships and identification. This then suggests that marketers and governments
alike should tap into this opportunity to use the internet celebrities as third parties to send across messages to their
target audience via social media. Again, the findings of this study contribute to the literature of para-social
relationships and identification and show how they influence purchase decision and materialism.

Future research should focus on investigating males by comparing them with the females to know which gender
is greatly influenced by the internet celebrities. Moreover, the current model can be enhanced by adding social
media platform as a moderator between internet celebrities and para-social relationships.

References

Aaker, D. A., Kumar, V., & Day, G. S. (2008). Marketing research. John Wiley & Sons, New York, NY

Alperstein, N. M. (1991). Imaginary social relationships with celebrities appearing in television

commercials. Journal of Broadcasting & Electronic Media, 35(1), 43-58.

Amichai-Hamburger, Y. (2007). Personality, individual differences and Internet use. Oxford handbook of Internet
psychology, 187-204.

Austin, E. W., Vord, R. V. D., Pinkleton, B. E., & Epstein, E. (2008). Celebrity endorsements and their potential
to motivate young voters. Mass communication and society, 11(4), 420-436.

Awang Z., Ahmad J. H., & Zin, N. M. (2010). Modelling job satisfaction and work commitment among lecturers:
A case of UiTM Kelantan. Journal of Statistical Modeling and Analytics, 1(2), 45-59.

Bagozzi R.P. and Yi, Y. (1988), “On the evaluation of structural equation model”, Journal of Academy of
Marketing Science, Vol. 16 No. 1, pp. 74-94.

Barak, A. (2007). Phantom emotions: Psychological determinants of emotional experiences on the

Internet. Oxford handbook of Internet psychology, 303-329.

195
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Basil, M. D. (1996). Identification as a mediator of celebrity effects. Journal of Broadcasting & Electronic
Media, 40(4), 478-495.

Berry J., and Keller, E. 2003. The Influentials: One American in Ten Tells the Other Nine How to Vote, Where to
Eat, and What to Buy. Free Press.

Bocarnea M. C., & Brown, W. J. (2007). Celebrity-Persona Parasocial Interaction Scale. In R. A. Reynolds, R.
Woods, & J. D. Baker (Eds.), Handbook of research on electronic surveys and measurements (pp. 309–
312). Hershey, PA: Idea Group Reference.

Boorstin, D. J. (1961). The image. New York: Atheneum.

Brown, W. J., Basil, M. D., & Bocarnea, M. C. (2003). The influence of famous athletes on health beliefs and
practices: Mark McGwire, child abuse prevention, and androstenedione. Journal of Health
Communication, 8(1), 41-57.

Brown, W. J., & De Matviuk, M. A. C. (2010). Sports celebrities and public health: Diego Maradona's influence
on drug use prevention. Journal of health communication, 15(4), 358-373.

Burroughs, J. E., & Rindfleisch, A. (2002). Materialism and well-being: A conflicting values perspective. Journal
of Consumer research, 29(3), 348-370.

Chan, K., & Zhang, C. (2007). Living in a celebrity-mediated social world: The Chinese experience. Young
Consumers, 8(2), 139-152.

Chan, K., & Prendergast, G. P. (2008). Social comparison, imitation of celebrity models and materialism among
Chinese youth. International Journal of Advertising, 27(5), 799-826.

Chan, K., Leung Ng, Y., & Luk, E. K. (2013). Impact of celebrity endorsement in advertising on brand image
among Chinese adolescents. Young Consumers, 14(2), 167-179.

Chang, T. K., Southwell, B. G., Lee, H. M., & Hong, Y. (2012). Jurisdictional protectionism in online news:
American journalists and their perceptions of hyperlinks. New Media & Society, 14(4), 684-700.

Chia, S. C., & Poo, Y. L. (2009). Media, celebrities, and fans: An examination of adolescents' media usage and
involvement with entertainment celebrities. Journalism & Mass Communication Quarterly, 86(1), 23-44.

Cheney, G. (1983). The rhetoric of identification and the study of organizational communication. Quarterly
journal of speech, 69(2), 143-158.

Chin W.W., Gopal, A. and Salisbury, W.D. (1997), “Advancing the theory of adaptive structuration: the
development of a scale to measure faithfulness of appropriation”, Information Systems Research, Vol. 8
No. 4, pp. 342-367.

Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media
characters. Mass communication & society, 4(3), 245-264.

Eastin, M. S., & LaRose, R. (2005). Alt. support: Modeling social support online. Computers in Human
Behaviour, 21(6), 977-992.

Erikson, E. H. (1968). Life cycle. International encyclopedia of the social sciences, 9, 286-292.

Eveland Jr, W. P. (2003). A “mix of attributes” approach to the study of media effects and new communication
technologies. Journal of Communication, 53(3), 395-410.

Festinger, L. (1950). Informal social communication. Psychological review, 57(5), 271.

196
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Fornell C. and Larcker, D.F. (1981), “Evaluating structural equation models with unobservable variables and
measurement error”, Journal of Marketing Research, Vol. 18 No. 1, pp. 39-50.

Fraser, B. P., & Brown, W. J. (2002). Media, celebrities, and social influence: Identification with Elvis
Presley. Mass Communication & Society, 5(2), 183-206.

Giles, D. C. (2002). Parasocial interaction: A review of the literature and a model for future research. Media
psychology, 4(3), 279-305.

Gladwell M. 2002. The Tipping Point: How Little Things Can Make a Big Difference. Back Bay Books.

Goldberg, M. E., Gorn, G. J., Peracchio, L. A., & Bomossy, G. (2003). Materialism among youth. Journal of
Consumer Psychology, 13(3), 278-288.

Hair J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2010), Multivariate Data Analysis, Prentice
Hall, Upper Saddle River, NJ.

Henseler, J., Hubona, G., Ray, P.A., 2016. Using PLS path modeling in new technology research: updated
guidelines. Ind. Manag. Data Syst. 116, 2–20. https://doi.org/10. 1108/IMDS-09-2015-0382.

Hoffner, C., & Cantor, J. (1991). Perceiving and responding to mass media characters. Responding to the screen:
Reception and reaction processes, 63-101.

Hoffner, C. (1996). Children's wishful identification and parasocial interaction with favorite television
characters. Journal of Broadcasting & Electronic Media, 40(3), 389-402.

Hofstede, G., & Bond, M. H. (1988). The Confucius connection: From cultural roots to economic
growth. Organizational dynamics, 16(4), 5-21.

Horton, D., & Wohl, R. R. (1956). Mass communication and para-social interaction: Observations on intimacy
at a distance. Psychiatry, 19, 215–229.

Horton, D., & Strauss, A. (1957). Interaction in audience-participation shows. American Journal of
Sociology, 62(6), 579-587.

Joinson, A. N., & Paine, C. B. (2007). Self-disclosure, privacy and the Internet. The Oxford handbook of Internet
psychology, 2374252.

Josselson, A. (1991). Finding Herself: Pathways to Identify Development in Women. San Francisco, CA: Jossey-
Bass.

Kamins, M. A., Brand, M. J., Hoeke, S. A., & Moe, J. C. (1989). Two-sided versus one-sided celebrity
endorsements: The impact on advertising effectiveness and credibility. Journal of advertising, 18(2), 4-10.

Kassim, N., & Asiah Abdullah, N. (2010). The effect of perceived service quality dimensions on customer
satisfaction, trust, and loyalty in e-commerce settings: A cross cultural analysis. Asia pacific journal of
marketing and logistics, 22(3), 351-371.

Kasser, T., & Ahuvia, A. (2002). Materialistic values and well‐being in business students. European Journal of
Social Psychology, 32(1), 137-146.

Katz, E., & Paul, F. (1955). Lazarsfeld. 1955. Personal Influence: The part played by people in the Flow of Mass
communications. Glencoe, Illinois: The Free Press. Katz Personal Influence: The Part Played by People
in the Flow of Mass Communication.

197
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Kelman, H. C. (1961). Three processes of social influence. Public Opinion Quarterly, 25, 57-78.

Kosenko, K. A., Binder, A. R., & Hurley, R. (2016). Celebrity influence and identification: a test of the Angelina
effect. Journal of health communication, 21(3), 318-326.

Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukophadhyay, T., & Scherlis, W. (1998). Internet paradox:
A social technology that reduces social involvement and psychological well-being?. American
psychologist, 53(9), 1017.

Latané, B. (1981). The psychology of social impact. American psychologist, 36(4), 343.

Latané, B., & Wolf, S. (1981). The social impact of majorities and minorities. Psychological Review, 88(5), 438.

Liebes, T., & Katz, E. (1990). The export of meaning: Cross-cultural readigns of Dallas.

Lonner, W. J., Berry, J. W., & Hofstede, G. H. (1980). Culture's consequences: International differences in w
Newton, B. J., & Buck, E. B. (1985). Television as significant other: Its relationship to self-descriptors in
five countries. Journal of Cross-Cultural Psychology, 16(3), 289-312.ork-related values.

Maltby, J., Day, L., McCutcheon, L. E., Gillett, R., Houran, J., & Ashe, D. D. (2004). Personality and coping: A
context for examining celebrity worship and mental health. British Journal of Psychology, 95(4), 411-428.

McCutcheon, L. E., Lange, R., & Houran, J. (2002). Conceptualization and measurement of celebrity
worship. British journal of psychology, 93(1), 67-87.

McCutcheon, L. E., Ashe, D. D., Houran, J., & Maltby, J. (2003). A cognitive profile of individuals who tend to
worship celebrities. The Journal of Psychology, 137(4), 309-322.

Newton, B. J., & Buck, E. B. (1985). Television as significant other: Its relationship to self-descriptors in five
countries. Journal of Cross-Cultural Psychology, 16(3), 289-312.

Newton, B. J., Buck, E. B., & Woelfel, J. A. (1986). Metric multidimensional scaling of viewers' perceptions of
TV in five countries. Human Organization, 45(2), 162-170.

Perse, E.M. and Rubin, R.B. (1989), “Attribution in social and parasocial relationships”, Communication
Research, Vol. 16 No. 1, pp. 59-77.

Reeves, B., & Miller, M. M. (1978). A multidimensional measure of children's identification with television
characters. Journal of Broadcasting & Electronic Media, 22(1), 71-86.

Rogers, Everett M., Diffusion of Innovations, The Free Press of Glencoe, New York. 1962.

Sanders, C. E., Field, T. M., Miguel, D., & Kaplan, M. (2000). The relationship of Internet use to depression and
social isolation among adolescents. Adolescence, 35(138), 237.

Saunders S. (2001). Fromm’s marketing character and Rokeach values. Social Behavior and Personality, 29(2),
191-196.

Schultze, Q. J., Anker, R. M., Schultze, Q., Bratt, J. D., Zuidervaart, L., & Worst, J. (1991). Dancing in the dark:
Youth, popular culture, and the electronic media. Wm. B. Eerdmans Publishing.

Sheldon, K. M., & Elliot, A. J. (1998). Not all personal goals are personal: Comparing autonomous and controlled
reasons for goals as predictors of effort and attainment. Personality and Social Psychology Bulletin, 24(5),
546-557.

Sundar, S. S., & Limperos, A. M. (2013). Uses and grats 2.0: New gratifications for new media. Journal of
Broadcasting & Electronic Media, 57(4), 504-525.

198
European Journal of Business and Management www.iiste.org
ISSN 2222-1905 (Paper) ISSN 2222-2839 (Online) DOI: 10.7176/EJBM
Vol.11, No.15, 2019

Sundar, S. S., Oh, J., Kang, H., & Sreenivasan, A. (2012). How does technology persuade?: theoretical
mechanisms for persuasive technologies. In The Sage handbook of persuasion: Developments in theory and
practice (pp. 388-404). SAGE Publications Inc..

Till, B. D., & Shimp, T. A. (1998). Endorsers in advertising: The case of negative celebrity information. Journal
of advertising, 27(1), 67-82.

Tenenhaus, M., Vinzi, V. E., Chatelin, Y. M., & Lauro, C. (2005). PLS path modeling. Computational statistics
& data analysis, 48(1), 159-205.

Tripp, C., Jensen, T. D., & Carlson, L. (1994). The effects of multiple product endorsements by celebrities on
consumers' attitudes and intentions. Journal of consumer research, 20(4), 535-547.

Wahab, A. K. H, & M, T (2018), The impact of social media celebrity on cosmetic surgery: Perspective of
parasocial relationships and identification. Asia Pacific Journal of Research, 1(XCII), 19-29rterly, 177-
195.

Wen N. (2017). Celebrity Influence and Young People’s Attitudes toward Cosmetic Surgery in Singapore: The
Role of Parasocial Relationships and Identification. International Journal of Communication, 11, 19.

Wortham, J., Goel, V., & Perlroth, N. (2013). Facebook, still dominant, strives to keep cachet. New York Times, 18.

Xie, B. (2008). The mutual shaping of online and offline social relationships. Information Research: An
International Electronic Journal, 13(3).

Yue X.D. and Cheung, C.K. (2000), ‘‘Selection of favourite idols and models among Chinese young people: a
comparative study in Hong Kong and Nanjing’’, International Journal of Behavioural Development, Vol.
24, pp. 91-8.

Zhao, B. (1997) Consumerism, Confucianism, communism: Making sense of China today, New Left Review,
222(Mar-Apr), pp.43-59

199