International Journal of Economics and Finance; Vol. 9, No.

1; 2017
ISSN 1916-971X E-ISSN 1916-9728
Published by Canadian Center of Science and Education

The Effects of Celebrity Endorsement on Customer’s Attitude toward

Brand and Purchase Intention
Nguyen Minh Ha1 & Nguyen Hung Lam1
1
Graduate School, Ho Chi Minh City Open University, Vietnam
Correspondence: Nguyen Minh Ha, Graduate School, Ho Chi Minh City Open University, 97 Vo Van Tan Street,
District 3, Ho Chi Minh City, Vietnam. Tel: 84-8-3930-0947. E-mail: [email protected] or
[email protected]

Received: July 9, 2016 Accepted: July 29, 2016 Online Published: December 14, 2016
doi:10.5539/ijef.v9n1p64 URL: http://dx.doi.org/10.5539/ijef.v9n1p64

Abstract
Celebrity endorsement has become a popular choice in advertising due to the its originality and the celebrity’s
attractiveness that make a big impact on brand awareness and customer behavior. This study analyzes the effect
of celebrity endorsement on customer’s attitude toward brand as well as the effect of customer’s attitude toward
brand on customer’s purchase intention in Vietnam. A survey was conducted with 306 individuals in Vietnam.
After conducting the explanatory factor analysis (EFA), and multi-variable regression analysis, results indicate
that customer’s attitude toward brand is positively affected by 03 factors: celebrity match-up congruence with
the brand/product, celebrity trustworthiness, and celebrity expertise. Attitude toward brand also has a positive
impact on customer’s purchase intention.
Keywords: celebrity endorsement, customer’s attitude toward brand, purchase intention
1. Introduction
In recent years, the development of commercial communication and of rise-up of live TV shows has attracted
lots of attention from the public, especially young generation. Marketing strategies of companies focus mostly on
promoting products to the market with core objectives as to persuade their customers; since the customers have
got a lot of knowledge, references, and choices before making a purchase decision, competition has also became
more severely. There are a number of ways to promote brands, but employing celebrity as an aid to the brand has
become popular in all over the world (Friedman et al., 1979; Kamins, 1989). This is because advertisements in
which celebrity appears are generating effective outcomes in making the brand identity and retaining customers’
attention, that is the mandatory objective of any commerce (Erdogan, 1999; Kamins & Gupta, 1994; Kaikati,
1987; Patti & Frazer, 1988). Belch and Belch (2004, p. 174) showed that “in today’s television viewing
environment and the “stopping power” of celebrity endorsed commercials are more remarkable”. However, if
they later make up a negative image of themselves, brand image will also be affected. Therefore, it is very
important to select appropriate celebrity to represent a brand.
There have been a lot of studies in the world on the effect of celebrity on brand promotion activities, but this
field hasn’t been addressed adequately in Vietnam. The questions are that which factors of celebrity endorsement
will have effect on customer’s attitude toward brand and how is the relationship between customer’s attitude
toward brand and purchase intention in Vietnam?
2. Literature Review
There are a lot of definitions of celebrity. According to Young and Pinsky (2006, p. 464) “individuals who have
achieved a significant level of fame that makes them well known in society”. The celebrity has rose to become a
powerful force in the 21st century and hold an important role in the contemporary culture (Koernig & Boyd,
2009; Lord & Putrevu, 2009). A celebrity is a person whose name can attract public attention, ignite public
interest, and create individual values from the public (Kotler, Keller, & Jha, 2007). However, perhaps the most
impressive and widely referenced definition is Daniel Boorstin’s (1982, p. 49), in which he defined: as “The
celebrity is a person who is known for his well-knownness”.
Early definition of celebrity endorsement is mentioned by Freiden (1984). According to him, celebrity
endorsement means celebrity in direct connection to an advertised product. According to McCracken (1989, p.

64
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

310): “An individual who enjoys public recognition and who uses this recognition on behalf of a customer good
by appearing with it in an advertisement”. According to Kotler et al. (2007), celebrity endorsement is one of
communication channels that are used by celebrity as a means of expressing their words to promote the brand on
basis of their fame and personalities.
2.1 Meaning Transfer Model
According to McCracken (1989), this model is developed to illustrate celebrity endorsement process. Advertising
is one of means to transfer individual meaning to the brands. This model is divided into three stages. First stage
is the development of celebrity image and description of cultural meaning of the society. In this stage, it is
assessed whether subject, people and context are suitable to the celebrity. The second stage is relevant to the
celebrity’s transfer of meanings from brand endorsement to the product. In the final stage, brand image is
transferred to the customers.

Figure 1. Meaning transfer model

Note. Path of meaning movement. Stage of meaning movement.
Source: McCracken, 1989, p. 45.

2.2 Customer’s Attitude Toward Brand

Customer’s attitude toward brand is predisposition that focuses on favorable or unfavorable impact on a specific
brand after watching an advertisement on that brand (Phelps & Hoy, 1996). According to Lutz et al. (1983),
customer’s attitude toward brand is the customer’s emotional reaction toward a brand advertisement. It is
associated with the customer’s feeling if his/her purchase intention toward the brand is positive or negative,
favorable or unfavorable.
2.3 The Elaboration Likelihood Model (ELM)
Elaboration Likelihood Model (ELM) is a model that consists of two phases of response toward advertising
incentive. It explains how attitudes are formed on basis of the degree of participation. Current attitude may be
changed and it is assumed that when a customer receives a message, he starts processing it. There are two
possible directions: Central route used for persuasion if customer participation is high, or peripheral route used
for persuasion if customer participation is low. The model has two fundamental factors, motivation and ability to
process communication. Motivation means the customer’s readiness, participation, and needs. Ability means the
knowledge, qualification, and capacity to process information (Petty et al., 1983).

Figure 2. Elaboration Likelihood Model (ELM)

Source: Petty et al., 1983.

65
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

2.4 The Relation between Source Credibility Model and Customer’s Attitude toward Brand
Hovland & Weiss (1953) introduced a source credibility model, which is further studied by Ohanian (1990)
(cited by Armando, 2014). According to source credibility model, “the effectiveness of a message depends on the
apparent level of expertise and trustworthiness of the endorser” (Hovland & Weiss, 1953, p. 20)
2.4.1 Celebrity Trustworthiness
Trustworthiness refers to “the honesty, integrity and believability of an endorser” (Erdogan et al., 2001, p. 40). A
celebrity is considered as trustworthy (Goldsmith et al., 2000) and his/her trustworthiness is described as a
summary of values that create positive features and increase the acceptance of the message (Erdogan, 1999).
Trustworthiness is the most useful and effective tool to make the customer be more confident and reliable on the
brand (Ohanian, 1990). A hypothesis is provided:
Hypothesis 1 (H1): The more celebrity trustworthiness is perceived by the customer, the more positive customer’s
attitude toward brand will be.
2.4.2 Celebrity Expertise
Hovland et al. (1953) defines expertise as the level of knowledge and experience that a person may obtain in a
specific field that is acknowledged as valid. The more persuasive a celebrity’s expertise is (Aaker, 1997), the
more purchase decisions will be generated (Ohanian, 1991). Speck, Schumann, and Thompson (1988) affirms
that celebrity is considered as an expert in a specific field, resulting in a higher brand endorsement than a
celebrity without expertise (Hoekman & Bosmans, 2010). Following hypothesis is provided:
Hypothesis 2 (H2): The more celebrity expertise is perceived by the customer, the more positive customer’s
attitude toward brand will be.
2.5 The Relation between Source Attractiveness Model and Customer’s Attitude toward Brand
Source attractiveness model is developed by McGuire (1985), he holds that an individual message is accepted
and affected by the similarity between the receiver and the sender together with the familiarity and likeliness.
The meaning of source attractiveness model is referenced to be a famous philosopher, Aristotle: “beauty is a
greater recommendation than any letter of introduction”. Aristotle wants to emphasize that the most importance
is beauty and attractiveness (Hoekman & Bosmans, 2010).
2.5.1 Celebrity Attractiveness
Physical attractiveness transited via a person’s weight, height, and facial beauty is the very first expressions
perceived by another (Bardia et al., 2011). This concept does not only means physical attractiveness. It also
requires mental skills, personality, lifestyle, and art talents (Erdogan, 1999). A celebrity is attractive because
he/she has built up a popular image among the public. His/her attractiveness increases the persuasiveness toward
the customers as they want to be like the celebrity that they love (Cohen & Golden, 1972). A hypothesis is given
Hypothesis 3 (H3): The more celebrity attractiveness is perceived by the customer, the more positive customer’s
attitude toward brand will be.
2.5.2 Celebrity Similarity
Similarity is described as “a supposed resemblance between the source and the receiver of the message”
(McGuire, 1985). In other words, a customer may similarize himself with the endorser. People will be more
easily influenced when they find the similarity between them and the endorser. If the celebrity and the customers
share popular factors, such as similar interest or lifestyle, a better association will be formed (Erdogan, 1999).
Following hypothesis is provided:
Hypothesis 4 (H4): The more celebrity similarity is perceived by the customer, the more positive customer’s
attitude toward brand will be.
2.5.3 Celebrity Liking
Likeability is the “affection for the source as a result of the source’s physical appearance and behaviour”
(McGuire, 1985, p. 239). In addition, McGuire also holds that when customers like a celebrity, they will like
brands associated with the celebrity. A hypothesis is provided:
Hypothesis 5 (H5): The more celebrity liking is perceived by the customer, the more positive customer’s attitude
toward brand will be.
2.5.4 Celebrity Familiarity
Familiarity means the feeling of similarity by means of emotions and contact with a celebrity (Erdogan, 1999;

66
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

Belch & Belch, 2004). Celebrity familiarity will have a more positive impact when the customer himself finds
that he/she is similar to the celebrity. This is called the mere exposure effect (Zajonc, 1968). When the customers
have short contacts with the celebrity and contact interval becomes longer, the effects of familiarity will improve
customer’s attitude toward brand. On the contrary, the effect is negative when they have long contacts and
contact interval becomes shorter (Bornstein, 1989). A hypothesis is given by:
Hypothesis 6 (H6): The more celebrity familiarity is perceived by the customer, the more positive customer’s
attitude toward brand will be.
2.6 The Relationship between Brand/Product Match – up Hypothesis Model and Customer’s Attitude toward
Brand
According to Forkan (1980); Kamins (1989), brand/product match – up hypothesis model means that celebrity
image and product message must be similar and matched up in order for the advertisement to be effective.
2.6.1 Celebrity Match-up Congruence With The Brand/Product
A number of studies conducted by Cooper (1984) and Forkan (1980) indicate that celebrity match-up congruence
with the brand/product has a significant play. When a product is advertised by a celebrity with suitable image
that is highly relevant to the product, the confidence will be higher on the advertisement and the celebrity
compared to a product image promoted by a less famous, less relevant person (Kotler, 1997). A hypothesis is
given:
Hypothesis 7 (H7): The more celebrity match-up congruence with the brand/product is perceived by the customer,
the more positive customer’s attitude toward brand will be.
2.6.2 The Relationship between Customer’s Attitude toward Brand and Purchase Intention
Customer’s purchase intention addresses the predisposition to purchase a certain brand or product (Belch &
Belch, 2004). Purchase intention also tells about the possibility that a person will purchase a product (Phelps &
Hoy, 1996). Amos, et al. (2008) hold that the customer’s positive attitude toward celebrity endorsement will
improve his/her purchase intention. Many studies also indicate that customer’s attitude toward brand has a
positive and significant impact on purchase intention (Mitchell & Olson, 1981; Gresham & Shimp, 1985; Batra
& Ray, 1986; Phelps & Hoy, 1996). A hypothesis is provided:
Hypothesis 8 (H8): The more positive customer’s attitude toward brand is, the more positive purchase intention
will be.
3. Methodology and Research Model
3.1 Methodology
This study is conducted in two major stages. A qualitative study is conducted by face-to-face direct interview on
04 marketing experts and a group discussion is carried out among 12 customers of 18 years old or more, which is
intended to modify, supplement, and complete the measurement scale. A quantitative study is carried out to
collect data by using a questionnaire survey based on Likert rating scale with 5 options, including “1-Totally
Disagree”, “2-Disagree”, “3-Neutral”, “4-Agree”, “5-Totally Agree”, used to measure observation variables for
each factor.
3.2 Proposed Research Model
From theories and previous studies, an analysis is conducted on the effect of the celebrity endorsement
consisting of 07 factors, which are: celebrity trustworthiness, celebrity attractiveness, celebrity expertise,
celebrity similarity, celebrity liking, celebrity familiarity, celebrity match-up congruence with the brand/product
are independent varieties. Customer’s attitude toward brand means temporary variable. Purchase intention means
dependent variable.

67
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

Figure 3. Proposed research model

3.3 Research Data

A total of 534 questionnaire sheets were directly distributed to the customers of 18 years old or more in Vietnam
or via online method. A total of 392 sheets were returned and, after being screened, 306 sheets were valid for
data analysis, achieving a ratio of 78.06% compared to actual data.
4. Research Results
4.1 Descritptive Statistics by Characteristics
For gender, there are 184 female and 122 male respondents, accounting for 60.1% and 39.9%, respectively, of
which 61.4% are single, 38.6% are married. For age, respondents of 18–25 years old account for 28.1%, of 26–
35 years old account for 58.5%, 36–45 years old account for 7.5%, and of more than 45 years old account for
5.9%. For education, university respondents account for 45.1%, post-graduate ones account for 28.4%, primary
school ones account for 2.6%, secondary school ones account for 7.2%, high school ones account for 8.5%, and
college-intermediary school ones account for 8.2%.
4.2 Descriptive Statistics of Model Variables

Table 1. Descriptive statistic of quantitative variables

Content Min Max Mean Standard Deviation
Celebrity Trustworthiness
CT1 You believe in celebrity’s brand choice. 1 5 3.15 0.904
CT2 You think that the celebrity is an honest person. 1 5 2.92 0.866
CT3 You think that the celebrity provides reliable source of information. 1 5 2.94 0.878
CT4 You think that the celebrity is a sincere person. 1 5 2.93 0.875
CT5 You think that the celebrity is a trustworthy person. 1 5 2.96 0.892
Celebrity Expertise
CE1 You think that the celebrity is an expert in the field that he/she represents. 1 5 2.60 0.964
CE2 You think that the celebrity has experience in using the brand. 1 5 3.02 0.887

68
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

CE3 You think that the celebrity has a lot of knowledge about this brand. 1 5 2.92 0.884
CE4 You think that the celebrity has got high professional qualification. 1 5 2.77 0.908
CE5 You think that the celebrity has skilled this brand. 1 5 3.01 0.905
Celebrity Attractiveness
CA1 You think that the celebrity has got a strong attractiveness. 1 5 4.01 0.696
CA2 You think that the celebrity is a very classy. 1 5 3.36 0.881
CA3 You think that the celebrity has a very pretty face. 1 5 3.81 0.831
CA4 You think that the celebrity has a very elegant fashion style. 2 5 3.87 0.801
CA5 You think that the celebrity has a very attractive appearance. 1 5 3.82 0.864
CA6 You think that the celebrity has a very persuasive voice. 1 5 3.46 0.846
CA7 You think that the celebrity has a very professional manner. 2 5 3.81 0.778
Celebrity Similarity
CS1 You think that the celebrity and you share the same culture. 1 5 3.01 0.861
CS2 You think that the celebrity and you share similar lifestyle. 1 5 2.56 0.817
CS3 You think that the celebrity and you share similar interests. 1 5 2.89 0.873
CS4 You think that the celebrity and you share similar perspectives. 1 5 2.60 0.771
CS5 You think that the celebrity and you share similar likings. 1 5 2.67 0.886
Celebrity Liking
CL1 You like the celebrity’s behaviors. 1 5 3.36 0.757
CL2 You like the celebrity’s appearance. 1 5 3.75 0.762
CL3 You think that the celebrity is very popular. 1 5 4.07 0.721
CL4 You like the celebrity’s voice. 1 5 3.34 0.800
CL5 You like the celebrity’s fashion style. 1 5 3.64 0.818
CL6 You like the celebrity’s professional manner. 1 5 3.71 0.799
CL7 Overall, you like the celebrity. 1 5 3.66 0.806
Celebrity Familiarity
CF1 You often see the celebrity on the TV. 1 5 4.15 0.625
CF2 You often see the celebrity on the stage or in the cinema. 1 5 3.31 0.950
CF3 You often see the celebrity at events or festivals. 1 5 3.52 0.877
CF4 You often see the celebrity on the advertising boards. 2 5 4.06 0.627
CF5 You often see the celebrity on the newspapers. 1 5 4.10 0.609
CF6 You often see the celebrity in person. 1 5 2.57 0.994
CF7 You often see the celebrity on the Internet. 1 5 4.15 0.672
CF8 You often listen to the celebrity over the radio. 1 5 3.07 0.989
Celebrity Match-up Congruence with The Brand / Product
CM1 You often see the celebrity in the advertisements of this brand. 1 5 3.93 0.694
CM2 You think that celebrity image suits this brand. 1 5 3.74 0.736
CM3 You think that this brand is totally suitable for the celebrity to represent. 1 5 3.67 0.779
CM4 You think that the celebrity that represents this brand is trustworthy. 1 5 3.43 0.787
CM5 You believe that the celebrity is using this brand. 1 5 3.05 0.943
Customer’s Attitude toward Brand
AB1 You believe that this brand is a good one. 1 5 3.54 0.724
AB2 You think that this brand is very interesting. 1 5 3.50 0.712
AB3 You like this brand. 1 5 3.44 0.763
AB4 You think that this brand has a good quality. 2 5 3.46 0.751
AB5 You are satisfied with this brand. 1 5 3.46 0.734
AB6 You are confident in this brand. 1 5 3.42 0.766
Purchase Intention
PI1 You will seek more information on this product. 1 5 3.59 0.806
PI2 You will actively seek for this product. 1 5 3.41 0.857
PI3 You will try this product when you see it. 1 5 3.68 0.762
PI4 You will purchase this product. 1 5 3.34 0.823
Celebrity appearance in the advertisement has motivated you to purchase
PI5 1 5 3.21 1.009
this product.

Celebrity trustworthiness: CT1 has the highest mean value (3.15), while CT2 has the lowest mean value (2.92).
So, the customers are confident on celebrity’s brand choice, but they don’t think that the celebrity is honest.
Celebrity expertise: CE2 has the highest mean value (3.02), while CE1 has the lowest mean value (2.60). Hence,
the customers think that the celebrity has a lot of experience in using this brand, but they don’t think that the
celebrity is an expert in the field that he/she represents.

69
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

Celebrity attractiveness: CA1 has the highest mean value (4.01), while CA2 has the lowest mean value (3.36).
Hence, the customers think that the celebrity has got a strong attractiveness, but they don’t think that the
celebrity is a very skilled elite.
Celebrity similarity: CS1 has the highest mean value (3.01), while CS2 has the lowest mean value (2.56). Hence,
the customers think that the celebrity and they share the same culture, but they don’t think that the celebrity and
them share similar lifestyle.
Celebrity liking: CL3 has the highest mean value (4.07), while CL4 has the lowest mean value (3.34). Hence, the
customers think that the celebrity is very popular, but they don’t like his/her voice.
Celebrity familiarity: CF1 and CF7 has the highest mean value (4.15), while CF6 has the lowest mean value
(2.57). Hence, the customers agree that they often see the celebrity on the TV and Internet, but they don’t agree
that they often see the celebrity in person.
Celebrity match-up congruence with the brand / product: CM1 has the highest mean value (3.93), while has the
lowest mean value (3.05). Hence, the customers often see the celebrity in brand advertisements, but they don’t
believe that the celebrity is using this product.
Customer’s attitude toward brand: AB1 has the highest mean value (3.54), while AB6 has the lowest mean value
(3.42). Hence, the customers believe that the brand represented by the celebrity is a good one, but they don’t
believe in this brand.
Purchase intention: PI3 has the highest mean value (3.68), while PI5 has the lowest mean value (3.21). Hence,
the customers agree that they will try this product when they see it, but they don’t think that celebrity appearance
in the advertisement has motivated them to purchase the product.
4.3 Testing the Cronbach’s Alpha
Celebrity trustworthiness: Removing CT1 has the Cronbach’s Alpha if item deleted is 0.908 larger than the
Cronbach’s Alpha coefficient of the 1st scale test value of 0.907. In the 2nd test, the Cronbach’s Alpha
coefficient is 0.908, and item-total correlation for all variables are satisfied (higher than 0.3). Therefore,
remaining 4 observation variables are included in factor analysis (EFA).
Celebrity expertise: The Cronbach’s Alpha coefficient is 0.817, and item-total correlation for all variables are
satisfied (higher than 0.3). Therefore, the 5 observation variables are included in EFA.
Celebrity attractiveness: The Cronbach’s Alpha coefficient is 0.810, and item-total correlation for all variables
are satisfied (higher than 0.3). Therefore, the 7 observation variables are included in EFA.
Celebrity similarity: removing CS1 has the Cronbach’s Alpha if item deleted is 0.867 larger than the Cronbach’s
Alpha coefficient of the 1st scale test value of 0.851. In the 2nd test, the Cronbach’s Alpha coefficient is 0.867,
and item-total correlation for all variables are satisfied (higher than 0.3). Therefore, remaining 4 observation
variables are included in factor analysis (EFA).
Celebrity liking: The Cronbach’s Alpha coefficient is 0.833, and item-total correlation for all variables are
satisfied (higher than 0.3). Therefore, the 7 observation variables are included in EFA.
Celebrity familiarity: 3 varieties which are eliminated because the correlation coefficients of item-total is not
satisfied (lower than 0.3) include CF1(0.273), CF6 (0.243), and CF8 (0.297). After eliminating such varieties, the
Cronbach’s Alpha coefficient has been increased. Hence, the 2nd test shall be conducted, the Cronbach’s Alpha
coefficient is 0.684, and the item-total correlation for all variables are satisfied (higher than 0.3). Therefore,
remaining 5 observation variables are included in EFA.
Celebrity match-up congruence with the brand / product: CM1 and CM5 are eliminated because they have the
Cronbach’s Alpha if item deleted are 0.787 and 0.801 respectively, and higher than the Cronbach’s Alpha
coefficient of the 1st test scale of 0.784. In the 2nd test, CM4 is eliminated because it has the Cronbach’s Alpha
if item deleted is 0.863 higher than the Cronbach’s Alpha coefficient of the 2nd scale test value of 0.822. In the
3rd test, the Cronbach’s Alpha coefficient is 0.863, and item-total correlation for all variables are satisfied
(higher than 0.3). Therefore, remaining 2 observation variables are included in EFA
Customer’s attitude toward brand: AB2 is eliminated because it has the Cronbach’s Alpha if item deleted is
0.930, which is higher than the 1st scale test value of 0.923. In the 2nd test, AB1 is eliminated because it has the
Cronbach’s Alpha if item deleted is 0.931, which is higher than the Cronbach’s Alpha coefficient of the 2nd
scale test value of 0.930. In the 3rd test, the Cronbach’s Alpha coefficient is 0.931, and item-total correlation for
all variables are satisfied (higher than 0.3). Therefore, remaining 4 observation variables are included in EFA.

70
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

Purchase intention: PI5 is eliminated because it has the Cronbach’s Alpha if item deleted is 0.865, which is
higher than the Cronbach’s Alpha coefficient of the 1st scale test value of 0.860. In the 2nd test, the Cronbach’s
Alpha coefficient is 0.865, and item-total correlation for all variables are satisfied (higher than 0.3). Therefore,
remaining 4 observation variables are included in EFA.

Table 2. The Cronbach’s Alpha coefficient of the variables in the model

Variable Corrected Item-Total Correlation Cronbach’s Alpha if Item Deleted Cronbach’s Alpha
Celebrity Trustworthiness
CT2 0.803 0.877
CT3 0.734 0.901
0.908
CT4 0.828 0.868
CT5 0.803 0.877
Celebrity Expertise
CE1 0.546 0.800
CE2 0.612 0.779
CE3 0.706 0.752 0.817
CE4 0.563 0.794
CE5 0.616 0.778
Celebrity Attractiveness
CA1 0.462 0.798
CA2 0.469 0.800
CA3 0.661 0.764
CA4 0.656 0.765 0.810
CA5 0.582 0.778
CA6 0.488 0.795
CA7 0.513 0.790
Celebrity Similarity
CS2 0.705 0.836
CS3 0.696 0.840
0.867
CS4 0.784 0.807
CS5 0.697 0.840
Celebrity Liking
CL1 0.523 0.819
CL2 0.578 0.811
CL3 0.533 0.818
CL4 0.562 0.814 0.833
CL5 0.602 0.807
CL6 0.604 0.807
CL7 0.666 0.796
Celebrity Familiarity
CF2 0.389 0.671
CF3 0.430 0.641
CF4 0.513 0.610 0.684
CF5 0.503 0.616
CF7 0.436 0.636
Celebrity Match-up Congruence with The Brand / Product
CM2 0.760
0.863
CM3 0.760
Customer’s Attitude toward Brand
AB3 0.817 0.917
AB4 0.814 0.918
0.931
AB5 0.867 0.901
AB6 0.857 0.904
Purchase Intention
PI1 0.733 0.820
PI2 0.778 0.800
0.865
PI3 0.631 0.860
PI4 0.719 0.826

71
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

4.4 Explanatory Factor Analysis (EFA)

Explanatory factor analysis is conducted on whether celebrity endorsement has an impact on customer ’s attitude
toward brand. After the 1st analysis, CA1, CA2, CL1, CL3 CL7, CL2, CL5, CE1, CF4, CL6, CL4 are eliminated.
After the 2nd analysis, CA6, CA7, CE4 are eliminated. After 3 nd analysis, CF5 and CF7 are eliminated. After 4th
analysis, factor loading values are satisfactory with KMO coefficient of 0.844 with significance level of the
Bartlett test of 0.000.

Table 3. Results of the Explanatory Factor Analysis (EFA)

Conducting times Eliminating the variable KMO coefficient Significance of Bartlett test
CA1, CA2, CL1, CL3, CL7, CL2,
1st time 0.883 0.000
CL5, CE1, CF4, CL6, CL4
2nd time CA6, CA7, CE4 0.848 0.000
3rd time CF5, CF7 0.834 0.000
4th time 0.844 0.000

Component
Composition
1 2 3 4 5 6
CT4 0.874
CT5 0.856
Celebrity Trustworthiness
CT2 0.852
CT3 0.789
CS4 0.860
CS2 0.802
Celebrity Similarity
CS5 0.758
CS3 0.757
CA5 0.863
Celebrity Attractiveness CA4 0.860
CA3 0.847
CE2 0.815
Celebrity Expertise CE3 0.799
CE5 0.769
CM2 0.905
Celebrity Match-up Congruence with The Brand / Product
CM3 0.879
CF3 0.850
Celebrity Familiarity
CF2 0.823
Eigenvalues 5.791 2.416 1.705 1.512 1.234 1.108
% of Variance 32.172 13.422 9.472 8.403 6.857 6.156
% Cumulative 32.172 45.594 55.066 63.469 70.326 76.482

Based on results presented in Table 3, 18 observation variables are satisfactory and divided into 6 factor groups.
Factors are unchanged from original ones. However, “celebrity liking” factor is eliminated and hypotheses and
research model are corrected as follows:

72
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

Figure 4. Research model correction

4.5 Regression Analysis

According to results in Table 4, Analysis of the fitness of the regression model on customer’s attitude toward
brand, Adjusted R2 get value is 0.273, which is varied by 27.3% in term of “customer’s attitude toward brand”, it
can be explained by independent factors in the model. F = 20.058 with significant level of 0.000. Therefore, the
regression model is considered as overall fit.

Table 4. Results of regression analysis the model of customers’ attitude towards brand
Unstandardized Standardized
Model t Sig. VIF
Coefficients (B) Coefficients ()
(Constant) 1.126 4.210 0.000
Celebrity Trustworthiness 0.182*** 0.206 3.491 0.001 1.455
Celebrity Expertise 0.165*** 0.181 3.148 0.002 1.385
Celebrity Similarity 0.050 0.052 0.856 0.393 1.548
Celebrity Attractiveness 0.036 0.038 0.717 0.474 1.197
Celebrity Match-up Congruence with The Brand/Product 0.278*** 0.288 5.223 0.000 1.275
Celebrity Familiarity -0.002 -0.003 -0.055 0.956 1.034
R 0.536
R2 0.287
Adjusted R2 0.273
F (Anova) 20.058
Sig. (Anova) 0.000
Durbin - Watson 2.069
Note. Dependent variable: Customer’s attitude toward brand. ***: Results at the significance level of 1%.

According to the Table 5, the match-up congruence of the regression model of purchase intention, Adjusted R2
get value is 0.483, means 48.3% upon the variability of the purchase intention can be explained by the customers’

73
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

attitude towards brand. F = 285.814 with significant level of 0.000; therefore, the regression model is considered
as overall fit.

Table 5. Results of regression analysis the model of purchase intention

Unstandardized Standardized
Model t Sig. VIF
Coefficients (B) Coefficients ()
(Constant) 1.106 7.646 0.000
Customers’ attitude towards brand 0.696*** 0.696 16.906 0.000 1.000
R 0.696
R2 0.485
Adjusted R2 0.483
F (Anova) 285.814
Sig. (Anova) 0.000
Durbin - Watson 1.973
Note. Dependent variable: Purchase intention. ***: Results at the significance level of 1%.

4.6 Discussion of Results

4.6.1 Celebrity Trustworthiness
This variable is statistically significant at 1%, with value  = 0.206> 0, this means that H1 hypothesis is
supported. Trustworthiness refers to “the honesty, integrity and believability of an endorser” (Erdogan et al.,
2001). The celebrity is considered as a trustworthy person (Goldsmith et al., 2000). A study by Pham & Nguyen
(2015) indicates that “celebrity trustworthiness” has a positive impact on “customer’s attitude toward the
advertisement”. Results of this study also find positive impact of “celebrity trustworthiness” on “customer’s
attitude toward brand”. This can be explained as follows: celebrity appearance in commerces in Vietnam has
become popular because the celebrity can influence the public. However, “scandals” in their personal lives have
caused the public wonder and lose trusts in the celebrity’s ethics. Only a few celebrities are acknowledged for
their talents and ethics, and they are respected and relied upon by most of the public and colleagues. Therefore,
celebrity endorsement will have more impact on the customers than non-celebrity trustworthy.
4.6.2 Celebrity Expertise
This factor is statistically significant at 1%, with value  = 0.181> 0, this means that H2 hypothesis is supported.
The expertise mentions the level of knowledge and experience that a person may obtain in a specific field that is
acknowledged as valid (Hovland et al., 1953). The more persuasive a celebrity’s expertise is (Aaker, 1997), the
more purchase decisions will be generated (Ohanian, 1991). A study by Pham and Nguyen (2015) indicates that
“celebrity expertise” has a positive impact on “customer’s attitude toward the advertisement”. Results of this
study also find positive impact of “celebrity expertise” on “customer’s attitude toward brand”. This can be
explained as follows: With their attractiveness and influencing ability, celebrities are highly paid for their
appearance in advertisements although the brand is not related to their expertises. Therefore, customers feel that
they are not assured, and their confidence in the brand is wondered. So, enterprises need to select the celebrities
that have expertises related to the brand to be advertised so that customers will be more confident and accept it
more positively.
4.6.3 Celebrity Attractiveness
This variable is statistically significant at 5%, with value  = 0.038, this means that H3 hypothesis is not
supported. A study by Pham & Nguyen (2015) indicates that “celebrity attractiveness” has a positive impact on
“customer’s attitude toward the advertisement”. However, this study only considers the “customers’ attitude
towards brand” and the its results indicate no positive effect of the “celebrity attractiveness” factor on the
“customer’s attitude toward brand”. This can be explained as follows: Most of today’s celebrities have pretty,
attractive and ideal appearance to make their advertisements more impressive. Therefore, celebrity attractiveness
will generate attention toward the customers and make advertisements more attractive, however, attractiveness
has no impact on customer’s attitude toward brand.
4.6.4 Celebrity Similarity
This factor is statistically significant at 5%, with value  = 0.052, this means that H4 hypothesis is not supported.
A study by Pham and Nguyen (2015) indicates that “celebrity similarity” has a positive impact on “customer’s
attitude toward the advertisement”. However, this study only considers the “customers’ attitude towards brand”

74
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

and the its results indicate no positive effect of the “celebrity similarity” factor on the “customer’s attitude
toward brand”. This can be explained as follows: Celebrities have ideal appearance and outstanding talents. All
fans want to become like the ones that they like, from their lifestyle, likings, fashion style, etc. They learn after
celebrities and think that they share similar values, to a certain extent, with celebrities. However, similarity only
makes advertisements with celebrity endorsement receive more attention and be remembered, but it does not
affect customer’s attitude toward brand.
4.6.5 Celebrity Familiarity
This variable is statistically significant at 5%, with value  = 0.003, meaning that H’5 hypothesis is not
supported. The study by Pham and Nguyen (2015) does not consider “celebrity familiarity” factor. A study by
Shahrokh and Arefi (2013) indicates that there is a positive impact of “celebrity familiarity” on “source
attractiveness model”, and thereby resulting in a positive impact on “the effectiveness of celebrity endorsement”.
However, this study only considers the “customers’ attitude towards brand” and the its results indicate no
positive effect of the “celebrity familiarity” factor on the “customer’s attitude toward brand”. This can be
explained as follows: This celebrity will be covered everywhere so that customers can easily see the ones they
love just by clicking or navigating a remote control. It is celebrity familiarity that will help the customers
easily memorize and be impressed at the advertisements with celebrity endorsement, but familiarity has no
impact on customer’s attitude toward brand.
4.6.6 Celebrity Match-up Congruence with The Brand/Product
This factor is statistically significant at 1%, with value  = 0.288 > 0, meaning that H’6 hypothesis is not
supported. Celebrity match-up congruence with the brand / product may create the absolute confidence through
the homogeneous process (Langmeyer & Walker, 1991), and cause positive effect on the customers’ attitude
towards advertisment, brand / product and purchase intention (Kirmani & Shiv, 1998). A study by Pham and
Nguyen (2015) indicates that “celebrity match-up congruence with the brand / product” has a positive impact on
“customer’s attitude toward the advertisement”. Results of this study also find positive impact of “celebrity
match-up congruence with the brand / product” on “customer’s attitude toward brand”. This can be explained as
follows: everyday, customers can watch a lots of advertisements with celebrity, even the same one in different
advertisements. When a brand / product is advertised by a celebrity with suitable image that is highly relevant to
the brand / product, the confidence will be higher on the advertisement and the celebrity compared to a brand /
product image promoted by a less famous, less relevant person.
4.6.7 Customer’s Attitude toward Brand
This variable is statistically significant at 1%, with value  = 0.696 > 0, meaning that H’7 hypothesis is not
supported. Customer’s attitude toward brand is predisposition that focuses on favorable or unfavorable impact on
a specific brand after watching an advertisement on that brand (Phelps & Hoy, 1996). The study by Pham and
Nguyen (2015) does not consider the effect of “customer’s attitude toward brand” factor on the “purchase
intention”. Studies of Qurat and Mahira (2012), Aycha and Kaouther (2010) and Mazzini et al. (2014) indicate
the positive effect of “customer’s attitude toward brand” factor on the “purchase intention”. Results of this study
also affirm the correctness of previous studies. This impact is positive and considerable in Vietnamese market.
Attitude is used as factor to forecast customer’s intention and behavior (Fishbien & Ajzen, 1975; Ajzen, 1991).
When customers have more positive attitude toward brand, they will more likely intend to purchase the products.
5. Conclusions and Recommendations
5.1 Conclusions
The final study model includes 7 factors of celebrity endorsement with impact on customer’s attitude toward
brand. Also, factors of attitude toward brand have impacts on customer’s purchase intention. After testing the
reliability of the measurement scale and conducting explanatory factor analysis, results are that celebrity liking
factor is eliminated while other factors are retained, resulting in a correction of study model hypotheses. Results
of the multi-variable regression analysis indicate that customer’s attitude toward brand in Vietnam is positively
affected by 03 factors: celebrity match-up congruence with the brand / product, celebrity trustworthiness, and
celebrity expertises. Results also indicate that purchase intention is strongly and positively impacted by
customer’s attitude toward brand.
5.2 Suggestions of Policy Implications
Based on findings, the study suggests some policy implications as follows:
Firstly, enterprises should pay careful attention to selecting celebrities for promoting the brand. If the celebrity

75
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

has made any dispute statement, improper behavior, and a negative scandal, it will negatively affect the brand
and reputation of the enterprises. Therefore, it is necessary to select a trustworthy celebrity that is trusted by the
public.
Secondly, not every celebrity endorses a brand that is within his/her expertise. Therefore, customers usually
wonder that celebrity endorsement is just intended to make the ads attractive, but product quality is not
persuasive because it is beyond his/her expertise. Thus, if an enterprise luckily chooses a celebrity that is an
expert in the brand that it wants to promote, the persuasiveness and customer’s attitude toward brand will be
more positively impacted.
Thirdly, when enterprises can formulate a meaningful message that is suitable to the celebrity and brand, this will
have a positive impact on customer’s attitude toward brand. So, enterprises need to select a suitable celebrity for
their brand images and advertising message that they want to transfer, so that customers have better perception of
the advertisements and positive attitude toward the brand.
References
Amos, C., Holmes, G., & Strutton, D. (2008). Exploring The Relationship between Celebrity Endorser Effects
and Advertising Effectiveness. A Quantitative Synthesis of Effect Size. International Journal of Advertising,
27(2), 209-234. http://dx.doi.org/10.1080/02650487.2008.11073052
Aycha, M. G., & Kaouther, S. B. R. (2010). The Persuasive Effectiveness of Famous and Non Famous Endorsers
in Advertising. IBIMA Business Review, 14. http://dx.doi.org/10.5171/2010.474771
Broostin, D. (1982). The Image: A Guide to Pseudo-Events in America. New York: Athenneum, 49.
Erdogan, B. Z. (1999). Celebrity endorsement: A literature review. Journal of Marketing Management, 14,
291-314. http://dx.doi.org/10.1362/026725799784870379
Freiden, J. B. (1984). Advertising Spokesperson Effects: An Examination of Endorser Type and Gender on Two
Audiences. Journal of Advertising Research, 24(5), 33-41.
Friedman, H., Termini, S., & Washington, R. (1976). The Effectiveness of Advertisements Utilizing Four Types
of Endorsers. Journal of Advertising, 5(3), 22-24. http://dx.doi.org/10.1080/00913367.1976.10672647
Goldsmith, R., Lafferty, B., & Newell, S. (2000). The Impact of Corporate Credibility and Celebrity Credibility
on Consumer Reaction to Advertisements and Brands. Journal of Advertising, 29(3), 43-54.
http://dx.doi.org/10.1080/00913367.2000.10673616
Gresham, L. G., & Shimp, T. A. (1985). Attitude toward The Advertisement and Brand Attitudes: A Classical
Conditioning Perspective. Journal of Advertising, 14(1), 10-18.
http://dx.doi.org/10.1080/00913367.1985.10672924
Hoekman, M. L. L., & Bosmans, A. M. M. (2010). Celebrity Endorsement. How Does Celebrity Endorsement
Influence The Attitude towards The Brand and How Does Negative Publicity Affect This Relationship?
Bachelor Thesis, ANR. 343396, Tilburg University, Netherlands.
Hovland, C. I., & Weiss, W. (1953). The Influence of Source Credibility on Communication Effectiveness.
Public Opinion Quarterly, 15, 635-650. http://dx.doi.org/10.1086/266350
Kaikati, J. G. (1987). Celebrity Advertising: A Review and Synthesis. International Journal of Advertising, (6),
99-105. http://dx.doi.org/10.1080/02650487.1987.11107007?src=recsys
Kamins, M. A., & Gupta, K. (1994). Congruence between Spokesperson and Product Type: A Matchup
Hypothesis Perspective. Psychology and Marketing, 11(6), 569-586.
http://dx.doi.org/10.1002/mar.4220110605
Kotler, P., Keller, K., & Jha, M. (2007). Marketing Management. A South Asian perspective (12th ed.). Pearson
Education, 340.
Mazzini, M., Musa, R., Mohamed, R. N., & Borhan, H. (2014). Celebrity Entrepreneur Endorsement and
Advertising Effectiveness. Procedia – Social and Behavioral Sciences, 130, 11-20.
http://dx.doi.org/10.1016/j.sbspro.2014.04.002
McCracken, G. (1989). Who Is The Celebrity Endorser? Cultural Foundation of The Endorsement Process.
Journal of Consumer Research, 16, 310-321. http://dx.doi.org/10.1086/209217
McGuire, W. J. (1985). Attitudes and Attitude Change. In G. Lindzey, & E. Aronson (Eds), Handbook of Social
Psychology (Vol. 2, pp. 233-346). New York: Random House.

76
ijef.ccsenet.org International Journal of Economics and Finance Vol. 9, No. 1; 2017

Ohanian, R. (1990). Construction and Validation of A Scale to Measure Celebrity Endorsers’ Perceived
Expertise, Trustworthiness, and Attractiveness. Journal of Advertising, 19(3), 39-52.
http://dx.doi.org/10.1080/00913367.1990.10673191
Ohanian, R. (1991). The Impact of Celebrity Spokesperson’s Perceived Image on Consumers’ Intention to
Purchase. Journal of Advertising Research, 31(1), 46-52.
Pham, T. M. L., & Nguyen, T. V. (2015). The Effect of Celebrity Endorsement in Advertising on Consumer
Attitude. Journal of Economics and Development, (215), 76-86.
Phelps, J. E., & Hoy, M. G. (1996). The Aad-Ab-PI Relationship in Children: The Impact of Brand Familiarity
and Measurement Timing. Psychology and Marketing, 13(1), 77-101.
http://dx.doi.org/10.1002/(SICI)1520-6793(199601)13:1<77::AID-MAR5>3.0.CO;2-M
Pornpitakpan, C. (2003). The Effect of Celebrity Endorsers’ Perceived Credibility on Product Purchase Intention:
The Case of Singaporeans. Journal of International Consumer Marketing, 16(2), 55-74.
http://dx.doi.org/10.1300/J046v16n02_04
Qurat, U. A., & Mahira, R. (2012). Impact of Celebrity Advertisement on Customers’ Brand Perception and
Purchase Intention. Asian Journal of Business and Management Sciences, 1(11), 53-67.
Shahrokh, Z. D., & Arefi, A. (2013). Identify Factors Influencing on The Effectiveness of Celebrities
Advertising. Journal of Applied Environmental and Biological Sciences, 3(12), 112-121.

Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution
license (http://creativecommons.org/licenses/by/4.0/).

77