See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/377751601

THE IMPACT OF NEGATIVE PUBLICITY OF BRANDS ON CONSUMER LOYALTY

Article in Journal of Consumer Behaviour · January 2022

CITATIONS READS

0 3

1 author:

Muhammad Rehan
City University of Science and Information Technology
4 PUBLICATIONS 2 CITATIONS

SEE PROFILE

All content following this page was uploaded by Muhammad Rehan on 28 January 2024.

The user has requested enhancement of the downloaded file.

THE IMPACT OF NEGATIVE PUBLICITY OF BRANDS ON

CONSUMER LOYALTY

Study of Peshawar, KP Market

BY

MUHAMMAD REHAN

DEPARTMENT OF MANAGEMENT SCIENCE CITY UNIVERSITY

OF SCIENCE & INFORMATION TECHNOLOGY PESHAWAR –
PAKISTAN

JANUARY, 2022
 Acknowledgement
I want to express my most incredible gratitude to the people who have helped me complete
this thesis.
First, I want to express my sincerest appreciation to my supervisor, Dr. Aamir Nadeem, who
has been a guiding light with his valuable insights and inspiration.
I also would like to thank my family and friends for their support and encouragement.
I want to give a special thank you to my classmates for assisting me with my work.
Finally, I thank Dr. Muhammad Zahid for inspiring me for this thesis and for aiming higher
for the future.

Peshawar, January 2022

_________________________
Muhammad Rehan

2
 UNDERTAKING
I certify that the research work titled “The Impact of Negative Publicity of Brands on Consumer
Loyalty” is my work. Also, it is certified that this work has not been presented elsewhere for
assessment. The material taken from other sources is properly acknowledged and referred to
throughout the thesis.

Name of Student: Muhammad Rehan

3
 Abstract

The objective of this study is to provide recommendations to mitigate the negative impact of a
brand on consumer loyalty in Peshawar, taking into account the consumption patterns
influenced by cultural traits in the region. The study aims to describe the factors contributing
to negative publicity and moderate the influence of celebrity endorsement and switching costs
to reduce the negative effect on consumer loyalty. Data collected through a Likert scale
questionnaire was analyzed using SPSS to test for reliability. The study also employed Partial
Least Squares (PLS) and Structural Equation Modelling (PLS-SEM) using Smart PLS 3.2.4.
PLS is preferred for its robustness, unbiased estimation, and ability to handle small sample
sizes. Future research could expand the sample and scope to include other geographic
boundaries and the remaining provinces of Pakistan. This will help to understand the impact of
cultural values on the problem and generalize the findings.

Keywords: Negative Publicity, Celebrity Endorsement, Switching Cost, Consumer Loyalty

4
 CHAPTER: 1
INTRODUCTION
This chapter presents the background of the chosen research area, followed by a problem discussion.
After that, the purpose of the thesis is stated along with the posed research question.

1.1. BACKGROUND OF THE STUDY

Customer loyalty measures are different in different fields. In 2004 (Beerlie et al.) assumed
another dimension of loyalty and recognized it as an essential dimension; it defined loyalty
as habitual and further claimed that staying with a specific brand is better for a customer
than consuming energy to change it. Cumulatively most of the researchers have
discriminated between interactive and effective loyalties.
It gives an impression that negative brand information appeals to more believed than positive
brand facts since it is progressively "demonstrative or educational" (Skowronski & Carlston,
1989; Maheswaran & Meyers-Levy, 1990). The severity of the negative information may
similarly impact customers' inspiration to process the data (Chiou et al., 2013). It is also a
belief that negative publicity influences all aspects of brands. Previous studies suggested
negative publicity harms a brand's attitude and changes consumer behavior. Negative
publicity has several results for companies since it can damage different parts of an
organization (Coombs & Holladay, 2001). Unfortunately, in the present society, corporate
scandals turn out to be increasingly clear; thus, negative publicity is generally expected in
the profitable markets. In (Xie & Peng, 2009), through broad communications, the Internet,
social networking, a corporate crisis, and negative publicity can only be held mystery for a
short time (Ward & Ostorm, 2006). The media and consumers are enthused about any news
regarding how companies work together; when a company makes a misstep, everybody will
think about it in a matter of moments. It is a broadly held belief that negative brand
information contrarily impacts all parts of a brand, yet it is indistinct if this is the situation.
Most previous research recommended negative brand information negatively affects
disposition toward a brand (Ahluwalia and Gürhan-Canli, 2000; Dawar & Pillutla, 2000;
Dentoni et al., 2011; Ullrich & Brunner, 2015). Brand value can be affected by constructive
correspondences, for example, advertising, which helps individuals to create positive brand
observations, along these lines upgrading brand esteem (Shimp & Andrews, 2013).
Advertising builds brand recall and acceptance (Keller et al., 2011) and has collective
consequences for consumer-based brand equity (Wang et al., 2009). Different types of

5
communication, including (WOM), can help individuals create a positive view of a brand
and construct purchase intention (Bambauer-Sachse & Mangold, 2011; Jalilvand & Samiei,
2012). In any case, brands likewise need to manage negative brand publicity, as consumers
get a wide assortment of brand information through conventional and new media, including
TV, radio, online social media, and cell phone applications. With the web and social media,
the speed of information spread has expanded significantly, particularly for negative brand
publicity (Ward & Ostrom, 2006). Therefore, consumers confront negative brand
information more often, more possible than they did even ten years back (Eisingerich et al.,
2011).

Furthermore, the impact of negative publicity was visible during the well-known milk brand
scandal “Milk-Pak,” a product of Nestle. In 2009, negative publicity was observed for Lays
Chips, stating that Lays chips use E631 ingredients printed on Lays potato chips wrapper
as a flavor enhancer and has pig fats, which is Haram for Muslims. Then to diminish this
negative perception, Lays hired Junaid Jamshed as a brand ambassador for an advertisement
in which he said Lays Potato Chips is Halal and manufactured in Pakistan with pure potato.
Through this message, the negative perception of Lays Chips was diminished with the help
of celebrity endorsers. As specified previously, it isn’t yet clear how the process of negative
reputation to a decrease in consumers works.

Negative publicity about brands and companies is widely prevalent in the marketplace and
is notorious for negatively impacting people’s attitudes and judgment. Due to its
determinant role in consumer attitude, researchers have put a lot of effort into finding why
negative information has such harmful effects. Some researchers assume that negative
information is given greater weight than equally positive information. Thus, negative
information is likely to exert a strong, negative impact on brand attitude formation. Other
researchers address that negative information is more diagnostic or informative than
positive information (Maheswaran & Meyers-Levy 1990; Skowronski & Carlston 1989).
Therefore, it is rational that negative publicity about a company or its products will result
in negative publicity. Such negative publicity will likely influence consumers' loyalty and
behaviour toward the company and its products, negatively affecting sales, profits, and
stock prices. However, in reality, negative publicity does not seem equally harmful to
everyone.

6
For example, a Taiwanese singer, Jeremy Liu released his MTV on YouTube in July 2011
for his new album. His quack-like voice and simple lyrics had earned him thousands of
negative buzz on the Internet. Some Internet users even said harsh words such as “The song
is terrible and harmful to ears. The Environmental Protection Administration should ban his
song” (China Times, 2011). As the criticisms climbed up, the number of views on YouTube
was soaring. His album sold out quickly, and all the media channels desperately invited him
to be on their shows. Surprisingly, he became a hot idol that year. Another example is when
the Chinese movie “Switch” was released in June 2013, it got a lot of negative buzz
regarding the actors, story, and sceneries. It even got the lowest movie rating in history by
Internet users on the famous Internet website “douban” in China. However, the movie
reached 33 million box office receipts two days after it was on stage. The above examples
show that not all people are liable for negative publicity.
Some people seem invulnerable to the influence of negative publicity. These phenomena
challenge our knowledge. Without any attempt to understand what variables make the
difference, our knowledge about the effect of adverse publicity is incomplete. However,
more research needs to focus on the characteristics of the consumer and their influence on
negative publicity (Ahluwalia, Burnkrant, & Unnava, 2000; Monga & John, 2008). The
conceptualisation of brand awareness has long been an essential variable for marketing
researchers (Andrews, Durvasula, & Akhter, 1990). Surprisingly, brand awareness's role in
negative publicity's impact has yet to be examined. The current study aims to fill the gap.
It is hypothesised that individuals’ brand awareness level might moderate the negative
publicity and consumer attitude.
A specialist's point is to make a hypothetical model together to get the net outcomes to all the
more likely to appreciate the effect of negative attention of brands on customer loyalty. The
analyst will likewise break down how to coordinate speculation, superstar underwriting,
exchanging costs, and negative impact to analyse buyers' responses about a brand they like
and to look at the net outcome. The proposition as pursued: the primary segment in a research
diagram, a writing survey, a meaning of negative attention, and earlier notoriety of a brand
(history of embarrassments). Next is celebrity endorsement and clarifying the match-up theory,
negative publicity impact, and switching costs. In last, the researcher will suggest companies
with severe results of negative publicity with the help of celebrity endorsers between negative
publicity and customer loyalty adequately, regardless of whether it can hurt individual
perception explicitly or not, explicitly to Peshawar geographic limits dependent on
consumption patterns affected by social attributes of this very district.

7
1.2.PROBLEM STATEMENT
Negative publicity is an essential component influencing consumer loyalty towards brands.
This research expects to understand the impact of negative publicity of brands on consumer
loyalty. Negative publicity can hurt individual customer faithfulness based on Peshawar
geographic boundaries dependent on consumption pattern of this location. Most studies of
negative publicity have focused on Western consumers having different cultural
backgrounds and traits, with little attention given to consumers in other countries like
Pakistan.
The researcher needs to fill this gap; the analysis is required to look at the impact of the
negative attention of brands on customers and how we can control the severe effect of
adverse publicity with the assistance of referenced factors/factors.

1.3.RESEARCH QUESTIONS

1. What are the effects of negative publicity on brands on consumer loyalty?

2. What is the moderating effect of switching costs between negative publicity and
consumer loyalty?
3. What is the moderating effect of celebrity endorsement between negative publicity and
consumer loyalty?
1.4.RESEARCH OBJECTIVES

1. To examine the effects of negative publicity of brands on consumer loyalty.

2. To examine the moderating effect of switching costs between negative publicity and
consumer loyalty.
3. To examine the moderating effect of celebrity endorsers between negative publicity
and consumer loyalty.
1.5.SIGNIFICANCE OF RESEARCH

To give commitment to analytical work about negative publicity impacts of brands on

consumer loyalty and to pick up information. This investigation plans the personality
characteristics of celebrity endorsers and the qualities of ads that are most speaking to lessen
the extreme effect of the negative impact of Brand on consumer loyalty. This study will
moderate the severe effect of adverse publicity with the help of moderators such as celebrity
endorsement and switching cost. This paper contributes to the limited and fragmented
literature on consumer response to negative publicity by examining the impact of

8
consumers’ attributions of blame to the brand under conditions of mild and severe negative
information on a range of critical brand-related outcomes. In particular, the authors find that
negative advertising impacts brand identity, brand attitudes, and purchase intentions
differently. The authors propose that brand managers use this information to inform their
marketing communications. This study will also provide recommendations and information
to moderate the negative perception of a brand that can harm consumer loyalty, specifically
to Peshawar's geographic boundaries, based on consumption patterns influenced by the
cultural traits of this region.

1.6.SCOPE OF THE STUDY

The current study's findings provide important insights for companies that celebrity
endorsement and switching costs are more effective when negative publicity occurs on their
brand, and consumers of Peshawar, KP, can buy a brand when the celebrity is associated
with a brand which is a very effective way to decrease the negativity effect on Brand. Also,
the consumer will switch a brand if there is no celebrity endorsement, and consumers of
Peshawar, KP, can bear the cost of changing a brand to another brand. In short, the findings
update the companies and other key stakeholders that adopting a flexible and positive
celebrity endorsement increases consumer loyalty in Peshawar, KP.

9
 CHAPTER: 2
LITERATURE REVIEW
The chapter will present previous research and models related to the present study. The theory is
concluded in a frame of reference at the end of this chapter. The reference structure explains and
describes the factors used for the thesis model.

2.1. Literature Review

Brands remind a customer’s overall experience with a product (Keller & Lehmann, 2006).
Past brand experience can influence consumers' willingness to pay for brands
(Bronnenbrg, Dube & Gentzkow, 2012). However, research has found that favourable
brand experiences developed into brand preference do not always increase a customer’s
purchase intention (Mishra & Datta, 2011).

Marketing literature defines the relationship between customers and brands as “brand
equity” (Wood, 2000). Silk (2006) defines brand equity as “the positive effect that the
brand has on a potential customer of a product – it reflects how much more consumers
are willing to pay for a particular brand compared with a competing brand (or with a
generic product)” (p. 100). Farquhar (1998) identified that increasing brand equity also
improved the product's value to the brand holder/maker. From the consumer perspective,
the strength of brand equity is determined by the level of consumer reaction to the brand
name (Shocker & Weitz, 1994; Keller, 1993). Brand equity is significantly influenced by
the level of consumer brand loyalty held towards a product (Khan & Mahmmod, 2012),
affects consumer purchase behaviours (Aaker, 1991), and has a positive relationship with
brand purchase intention (Aaker, 1991; Chen, Chen & Huang, 2012). However,
consumers' brand equity towards products and services can and do change. Product
problems are related to lowering brand equity (Aaker & Joachimsthaler, 2000). Such
product problems negatively impact the corporate image and consumer attitudes towards
the firm. Since brand image drives brand equity (Chen, 2010; Chen, Chen & Huang, 2012;
Heidarzadeh & Asadollahi, 2012), consumers' attitudes are changed by their perceptions
about the causes and extent of the company’s crisis and can severely damage a company
- or Brand – image (Siomkos & Malliaris, 1992). Brand image can influence a firm’s
profits and cash flow (Yoo & Donthu, 2001), as brand image influences purchase
intention (Chen, Chen & Huang, 2012). Consumers use the brand image as a cue in the
purchase decision process (Richardson, Dick & Jain, 1994), which affects and is a factor
in consumer behaviour (Burmann, Schaefer & Maloney 2008). Brand image has been

10
defined as cognitive and affective perceptions consumers have towards a brand (Dobni &
Zinkham, 1990; Roy & Banerjee, 2007) and consists of symbolic and functional beliefs
about the Brand (Low & Lamb, 2000). Research by Keller (1993) identified the brand
image as an association consumers hold in their memories regarding a product.

Branding and association have become essential to modern marketing (Bhakar, Bhakar &
Bhakar, 2013). Aaker (1991) defined brand association as “anything linked in memory to
a brand” (p. 109). As such, brand associations can be created by consumer perceptions of
a brand in terms of attitudes, attributes, and benefits (Keller, 1998). Brand associations
increase the memorability of a brand (Aaker, 1991), impact brand equity (Severi & Ling,
2013), can influence a brand’s level of differentiation over competitors (Aaker, 1996; Rio,
Vazquez & Iglesia, 2001), and can act as a method of information gathering for consumers
(van Osselaer & Janszewski, 2001). Various research studies indicate consumers develop
attachments to brands (Fournier, 1998; Keller, 2003; Schouten & McAlexander, 1995).
Brand attachment describes the strength of the connection between the customer and a
brand and affects consumer buying habits to such an extent as to impact a brand’s
profitability (Thompson, MacInnis & Park, 2005). The levels of attachment consumers
have towards a brand and their attitudes toward it are psychological constructs that impact
consumer behaviour, including purchases (Park et al., 2010).

Negative information impacts consumers more effect than positive information. This is
called negative impact. This impact makes adverse data more astounding than
constructive data (Hernard, 2002), (Herr, Kardees, & Kim, 1991) states, due to the way
that negative information is seen as more helpful than positive data in the evaluation of
individuals, items, and thoughts. Subsequently, it is obvious how proficient negative
information and apparent negative consideration can be. In earlier research, in like manner
related to negative data, (Skowronski & Donald 1987) expressed that when negative data
is associated with good perspectives, purchasers saw them as progressively significant in
contrast with negative data about viewpoints identified with organization capacities.
Moral attitudes are proposed to have more effect on the ultimate conclusion of customers
(Skowronski & Donald, 1987). In addition, negative data will impact the buy goal more
than positive data. This is specifically apparent in the administration part (Weingberger
& Dillion, 1980). A little study has been done on direct negative exposure to brands. In
any case, there is writing on the impact of negative data, which isn't the same as negative
exposure on brands. Negative publicity contrasts with negative data. When all is said and

11
 done, negative data can get through any source that isn't open media, though media is the
wellspring of negative exposure. The refinement can be found in "exposure," which
gathers data "distributed" by some media source and subsequently manoeuvres
transparently taken into notice. Therefore, negative attention is described as negative data
about an organisation or an item spread by the media or government.

To learn the effect of negative publicity on consumer loyalty, it is essential to identify

how consumer loyalty is constructed. Therefore, consumer loyalty is discussed before
negative publicity.

2.2.Consumer Loyalty

Like most business management constructs, no single theory explains consumer loyalty
(Corley and Gioia, 2011; Miles, 2012). This is hinged on the point that depending on the
research's outlook, there could be different or multiple theories explaining a phenomenon
better to understand the world (Hambrick, 2007). Hence, several approaches could explain
and support brand loyalty studies based on a specific research standpoint. From the brand
loyalty literature, Russo et al. (2016), for instance, draws on complexity theory to posit
that complexity theory tenets can help provide a more accurate understanding of what
generates customer loyalty due to the complex reality in which a phenomenon of interest
manifests itself. This is premised on the point that relationships between variables can be
non-linear in the real world, with possible abrupt switches occurring (Yang and Peterson,
2004).

Consumer loyalty has received much attention in the marketing literature during the last
five decades. Studies, for example, Cunningham (1966), Howard and Sheth (1969), and
Day (1969) brought customer loyalty to everyone’s attention and pointed out its
importance for marketing practices. (Dick and Basu 1994) state that the development,
maintenance, and enhancement of a customer’s loyalty towards the products or services
are central objectives in marketing activities. Moreover, customer loyalty is a vital
determinant in acquiring a sustainable competitive advantage (Dick & Basu, 1994;
Woodruff, 1997) and is crucial for being profitable as a company (Andreassen &
Lindestad, 1998). Holding a loyal customer can be ten times cheaper than capturing a new
one (Heskett, Sasser & Hart 1990) since loyal customers are more willing to cooperate,
complain less, and are not as price-sensitive as none loyal customers (Dick & Basu, 1994;
Zeithaml, Berry & Parasuraman, 1996; Fiol & Alcaniz, 2009; Wang, 2010). Thus,

12
 acquiring customer loyalty aims to raise sales revenues by increasing the purchase level
and stimulating consumers to buy a more extensive range of products a company offers.
The second goal can be defensive since building a relationship between the Brand and the
consumer is hoped to help maintain a sustainable consumer base (Uncles, Dowling &
Hammond, 2003). Which should potentially be able to endure some negative publicity.

The concept of loyalty continues to be a relevant subject in business research, granted that
customer loyalty enables firms to gain market advantages (Bloemer and Kasper, 1995;
Kumar et al., 2013). From a branding perspective, loyalty has been described as the
attachment a customer has toward a brand, creating the tendency to be loyal to a focal
brand, and often demonstrated by the intention to buy the Brand as a primary choice
(Oliver, 1999; Yoo and Donthu, 2001). For firms, the performance benefits of gaining
and sustaining loyal customers are wide and varied, such as increased revenues, lower
costs, and increased profitability (Lam et al., 2004; Rauyruen and Miller, 2007). The
literature suggests that patronage and declared preferences often relate to brand choice
(Narteh et al., 2012), resulting in loyalty (Punj and Hillyer, 2004). However, there are
several instances where the predictive ability of chosen brands to stimulate loyalty may
need to be revised (Horsky et al., 2004). Relying on only preference information from
consumers could thus be risky, significantly, if stated preferences diverge from actual
choices (Rahman and Azhar, 2011). It may be erroneous for marketers to deploy brand
marketing efforts based on consumer preferences without considering their eventual
nexus with actual brand loyalty.

Moreover, to generate any form of consumer brand loyalty, scholars admonish that firms
constantly undertake marketing efforts to transform brand users into loyalists, with the
five significant benefits of gaining word-of-mouth advocacy and recommendations from
them (Lin, 2010; Wallace et al., 2014).

a. Behavioral and Attitudinal loyalty

In early research on consumer loyalty, the main focus was on the behavioral aspect of
loyalty. Years later, an interest in the attitudinal side of loyalty started to grow, and
researchers realized that loyalty has two dimensions. Nowadays, there are two primary
schools of thought in defining, operationalizing, and measuring brand loyalty: behavioral
loyalty and attitudinal loyalty. This implies that loyalty is a two-dimensional concept
(Rundle-Thiele & Bennett, 2001). However, besides the two mentioned dimensions, there

13
is yet to be universally accepted definition of consumer loyalty in the literature (Pan,
Sheng & Xie, 2012). In the behavioural approach of loyalty, loyalty is measured by
behavioural characteristics (Jacoby & Chestnut, 1978). This type of consumer loyalty is
defined mainly by the pattern of past purchases. Thus, the measurement of loyalty is done
based on the repeated purchase. This data is gathered by observing patterns of consumer
buying.

Attitudinal loyalty is defined by the consumer's perception of a brand (Rundle-Thiele,

2005) and consists of a good set of stated beliefs toward the Brand (Uncles, Dowling &
Hammond, 2003). Bandyopadhyay and Martell (2007) operationalised attitudinal loyalty
by the number of positive attributes associated with the Brand. There are several ways to
measure attitudinal loyalty. Examples of measurement scales are: how much the brand is
liked, commitment to the Brand, recommendations to others, optimistic beliefs, positive
feelings, preference, and intention to purchase (Unlces et al., 2004; Rundle-Thielle, 2005;
Dick & Basu, 1994). Researchers (e.g., Dick & Basu, 1994; Anime, 1998;
Bandyopadhyay & Martell, 2007) argue that both types of loyalty complement each other
in understanding and measuring consumer loyalty and cannot be taken apart. To make
things easier, (Oliver, 1997) incorporated both the behavioral and the attitudinal school
of thought into one definition: “a deeply held commitment to re-buy or re-patronize a
preferred product or service consistently in the future, despite situational influences and
marketing efforts having the potential to cause switching behavior.” The notion that both
behavioral and additional loyalty is needed to understand loyalty can be explained by a
simple example.

Imagine someone who likes a particular motorcycle brand. This person may have a
positive attitude towards this Brand and may be loyal to it. However, this person is still
waiting to be allowed to drive a motorcycle or needs the money to buy it. The person may
have a positive attitude toward the Brand but cannot express his loyalty in purchasing the
product. This also works the other way around. Imagine someone who encounters
negative publicity, and as a result, the person loses their positive attitude towards this
particular firm or Brand. At this point, their attitude is low, but from a behavioral
perspective, they may still be loyal to the firm or Brand. One may argue that this person
is not dedicated anymore but is more or less forced to stay faithful to the company.
Bandyopadhyay and Martell (2007) call this a ‘constrained consumer.’ This is one of the
reasons why some researchers say that true loyalty cannot exist without a solid attitudinal

14
 commitment (Day, 1969; Jacoby & Chestnut, 1978; Reichheld, 1996). Repurchase
behavior is a good predictor of loyalty; however, it is insufficient to explain consumer
loyalty without positive attitudinal loyalty (Amine, 1998).

2.3.Negative Effect
Negativity impact is a suggestion grounded in impression development writing in
intelligence science and transformative hypothesis (Price, 1996). In particular,
"consumers place more weight on negative than positive information in framing by and
large assessments of a target"(Ahluwalia, Burnkrant, &Unnava, 2000). Paying attention
to harmful data more than positive data is a characteristic nature of human nature (Price,
1996). The negative effect further explains this psychological phenomenon by clarifying
duty and its relationship to breaking down negative attention. A passionate or mental
connection incites an inclination to oppose a change in light of clashing data or
experience" (Liu, Wang, and Wu, 2010). Negative publicity impact has a backward
association with responsibility level: it in all probability shows up with low-duty
purchasers since they experience a move in their mentalities and recognition substantially
more for a brand when gone up against negative, as opposed to positive, data (Liu, Wang,
and Wu, 2010). This is because customers have fewer connections to the Brand and are,
in this way, not so much joined but more prone to take the harmful data at a higher
indicative level. Conversely, profoundly dedicated customers have a nonappearance of
antagonism impact yet consider positive data increasingly reliable and suitable (Bailey,
2007).
H1: Negative publicity negatively impacts consumer loyalty

a. Prior Brand Reputation

Brand reputation can be found in two distinctive ways: a reputation for the genuine
Brand and a reputation for the company owning a brand. Regardless, in the business
and organization markets, it is highly typical that the association name is identical to
the brand name. When this is the circumstance, the company’s reputation fills in as an
umbrella brand for these characterizations (Cretu & Broodie, 2007). It isn't astonishing
that buyers will buy and trust a brand with a respectable reputation better repudiate a
brand with a poor reputation (Pan, Sheng, &Xie, 2012). Corporate reputation is hardly
seen as not precisely equivalent to mark reputation. One may state that with the

15
 company's reputation, the highlight is more on the sensitivity of the association
(Anderson &Weitz, 1992). Moreover, the trustworthiness of the association is
immovably related to this. The reliability of an association is portrayed by how many
commendable measures, for example, being moral, being direct, and keeping ensure,
are seen by a purchaser.
2.3.1.Celebrity Endorsement
Studies in marketing have been conducted to investigate how celebrity endorsement impacts
consumer purchase intentions (Wei and Lu, 2013). Also, several authors have investigated
a series of related factors, such as the effect of celebrity endorsement on product recall, the
impact of the credibility and expertise of the celebrity endorsement, and the effect of
celebrity image (Amos et al., 2008). Despite the generally positive aspects of acceptance on
the evaluation of products, which has been verified in other studies (Silva et al., 2015;
Ambroise et al., 2014), in some cases, celebrities cannot substantially help promote products
(Sliburyte, 2009). They may even damage a brand if consumers learn that an existing
endorser has committed a transgression (Um, 2013). These contrasting results demonstrate
that celebrity endorsement is complex and can influence several variables. A meta-analysis
performed by Knoll and Matthes (2017) shows that studies about the subject have found
positive, negative, and neutral effects, leaving in doubt the actual effectiveness of celebrity
endorsement. The authors demonstrated that the differences found in the studies may have
occurred due to moderating variables, such as the endorser type (Wei and Lu, 2013),
congruence (Silva et al., 2015), or endorser sex (Bergkvist and Zhou, 2016). Knoll and
Matthes (2017) suggest that future studies should focus precisely on the conditions under
which the endorsement effect may change: for example, endorsement of non-conventional
products (Myrick and Evans, 2014), endorsement of persons (Van Steenburg, 2015), non-
profit endorsement (Wheeler, 2009) and endorsement in emerging countries (Chou, 2014)
such as China, India or Brazil, all factors which will be addressed in this study.

Along with the lack of studies noted by Knoll and Matthes (2017), there has been another
global phenomenon related to endorsement and celebrity theory. With the broad
popularization of the Internet, social networks, social media, and reality television, it has
been proposed that there has been an extension of the concept of celebrity so that ordinary
people can achieve such status through these new media. As a result, the number of aspiring
celebrities has grown (Keel and Nataraajan, 2012). Therefore, knowing that the
effectiveness of celebrity endorsement effect may vary ultimately depending on the context

16
of the endorsement, and given the phenomenon of new celebrities emerging from different
media, the purpose of this paper is to analyze the effectiveness of an “acclaimed” celebrity
endorsement effect in a context where an aspiring new celebrity is being endorsed.

A celebrity endorser is an individual who acknowledges open affirmation and who uses this
affirmation in light of appropriate concern for a purchaser's right by appearing with it in an
advancement (McCracken, 1989). Further, McCracken (1989) imparted that a celebrity
endorser is an open figure who has stupefying inescapability tending to a brand in a
business. An educated endorser (for instance, an enormous name) is continuously
convincing at affecting the purchaser to buy the thing (Bardia, 2011) and to have a rousing
aura towards the Brand (Chan K. N., 2013); (Erdogan, 1999).
a.Credibility
In today’s dynamic and competitive environment, customers are continuously
exposed to a variety of brands, offering a large number of products and services that
are advertised through different marketing strategies (Kumar, 2010) to influence the
consumer behavior that leads to the final purchasing decision. This is because
consumer behavior is an important and complex area for marketers, as different
people have different needs (Kumar, 2010), and satisfying the needs of consumers is
a difficult task for most the marketers. This being the case, way back from the late-
19th century onwards, marketers around the world started using celebrities (first from
the film industry and then from sports) to endorse their products to capture the
consumer base (Roy, 2006)

Over the years, researchers worldwide have tried to assess the reason for the change
in consumer behavior due to celebrity endorsement. They have developed various
models (for example, the Source Credibility Model, the Source Attractiveness
Model, Meaning Transfer Model, and Celebrity–Product Congruence Model) to
assess which celebrity attributes are responsible for the change in consumer behavior,
leading to purchase intention. Many studies in India applied these models (either
individually or in combination) and found that, in general, seven celebrity attributes
are responsible for the change in purchase intention. These are: credibility of the
celebrity (trustworthiness and expertise), attractiveness of the celebrity (similarity,
familiarity, likeability), product match-up with the celebrity, and meaning transfer
from celebrity towards the product (Kumar, 2010; Roy, 2006, 2012a, 2012b; Roy,
Bashar, & Koh, 2012; Roy & Jain, 2016; Roy, Jain, & Rana, 2013; Roy & Moorthy,

17
2009). Studies reveal that some celebrity attributes, such as expertise, trustworthiness
(Freeman, 2015; Gupta, Kishore, & Verma, 2015), and product match-up (Roy,
2006), do not influence consumer behavior.

Around the world, celebrity is an omnipresent feature of society (Dash & Sabat,
2012), creating lasting impressions in the memories of all who cross its path
(Kurzman et al., 2007). Celebrities have been widely used for promotional activities,
as they can provoke attitudinal and emotional reactions (Kanungo & Pang, 1973), as
opposed to a non-celebrity endorser (Petty, Cacioppo, & Schumann, 1983). To be
effective, a celebrity endorser should have the credibility to attract attention (Miciak
& Shanklin, 1994), increase awareness about the endorsed product, and influence the
purchase decision of the targeted audience (Ohanian, 1990). Bowman (2002) argued
that celebrity endorsements lead to an increased level of the product recall. Chan,
Ng, and Luk (2013) found that using a celebrity in an ad increase brand awareness.
Marketers' decision about celebrity endorsements depends on the effective
segmentation, targeting, and positioning (STP) of their products in any specified
market; hence demographic profiling of the consumers becomes crucial, or else the
result of STP becomes futile. The demographic profile also provides some insights
into the attitude of the consumers; for instance, whether they will become
supporters/neutral/opposers (Subhash, Kamat, Scaglione, D’Mello, & Weiermair,
2015, 2018[forthcoming]) of any policy implemented either by the government or by
private players

Due to cut-throat competition, companies use advertising as the most effective

bombardment tool to gain a competitive advantage. As celebrities are connected
mainly to the public, celebrity endorsement is the best tool for an advertisement to
target the maximum audience. The product endorsed by a celebrity can be transferred
into a more favorable attitude and product evaluation by the consumers (Dean. &
Baswas, 2001). Its impact on financial returns is significant and helps get a
competitive advantage (Erdogan, 1999). In America, 25 % of companies use
celebrities in their ad due to their significance (Money, Shimp et al. 2006). The
audience shows interest in the endorsed product. An expert and consumer
endorsement both increase the audience's attitude towards the product (Wang, 2005
), These factors prove the importance of celebrity, but there can be other factors also
that can alter the effectiveness of celebrity like the match between the product

18
endorsed and celebrity (Till & Shimp, 1998). A celebrity endorser is any individual
who enjoys public recognition and who uses this recognition on behalf of a consumer
good by appearing with it in an advertisement (McCracken G., 1989). Further,
McCracken (1989) expressed that a celebrity endorser is an open figure with
astonishing ubiquity speaking to a brand in a commercial. People are star-struck by
film stars, cricketers, politicians, and national heroes. Why? The 170 million people
are in love with celebrities because the matching principal, admiration, comfort, and
familiarity compel them to watch the celebrities in the ads. Pepsi got the top position
in the beverages industry of Pakistan by using 90 % sports celebrities in their ad, and
the massive 500 % of sales of Telenor (talkshawk Package) proves the significance
of celebrities in Pakistan (Khan, 2009). But here, we are checking the credibility of
celebrities in Pakistan because they either respond to European or Canadian culture
(Zahaf & Anderson, 2008; Nadeem Iqbal et al., 2014) or differently.

The effectiveness of celebrity endorsement is based on how celebrity attributes

influence consumers’ purchase intention. Kumar (2010) studied all the seven
celebrity attributes provided in the earlier four models: source credibility consisting
of trustworthiness (1) and expertise (2); source attractiveness consisting of similarity
(3), familiarity (4), and likeability (5); product match-up (6) and meaning transfer
(7). The present study is based on these seven attributes. In a broad sense, source
credibility refers to a communicator’s positive characteristics that affect the
receiver’s acceptance of a message (Ohanian, 1990). It is based on the research in
social psychology by Hovland, Janis, & Kelley (1953). Trustworthiness refers to an
endorser's honesty, integrity, and believability (Erdogan et al., 2001). Consumers
generally have the impression that celebrities are a trustworthy source of information
(Goldsmith, Lafferty, & Newell, 2000). Marketers take advantage of this impression
by employing celebrities who are thus regarded as reliable, honest, believable, and
dependable among their fans and others (Lindgren & Shimp, 1997). Expertise is the
extent to which a communicator is perceived to be a source of valid assertions
(Hovland et al., 1953). To persuade the recipients of the information, the endorser’s
expertise has to have an encouraging effect on receivers; an endorser with good
knowledge and skills has a higher chance of convincing the consumers than an
endorser with less expertise (Ohanian, 1990). The source attractiveness model, as a
component of the source valance model (McGuire, 1985), is mainly based on four

19
dimensions: familiarity (knowledge about the source through exposure), likeability
(affection for the source based on the physical appearance or behavior), similarity
(resemblance between the source and the audience) and attractiveness (McCracken,
1989). Accordingly, the effectiveness of the message depends on these dimensions
(Kumar, 2010). Familiar, likable, and similar sources are more attractive and
persuasive (Ohanian, 1990). Such people evoke positive stereotypes and, as a result,
physically engaging communicators are more successful at changing beliefs and
attitudes (Baker & Churchill, 1977; Chaiken, 1979) and generating purchasing
intentions (H. H. Friedman & L. Friedman, 1979; Friedman et al., 1976; Petty, 1980;
Petty et al., 1983). Match-up suggests that an advertisement's effectiveness depends
on a perceived ‘fit’ between the endorsing celebrity and the Brand endorsed (Till &
Busler, 1998; Till & Shimp, 1998). The product match-up model suggests that the
celebrity and the product features should complement each other for effective
advertising (Kamins, 1989; Kamins, Brand, Hoeke, & Moe, 1989). The match
between the product and the celebrity depends on the attributes between product
features and the celebrity image (Misra & Beatty, 1990; Charbonneau & Garland,
2010). The model suggests that attractive celebrities, beautiful female celebrities, are
more effective at endorsing beauty products, which are believed to enhance one’s
attractiveness (Kamins, 1990). The meaning transfer is another critical attribute.
McCraken (1989) explains that a celebrity would be influential only if the clarity is
achieved by the meanings consumers associate with the endorser, which are
eventually transferred to the Brand. Since celebrity endorsement effectiveness is
determined based on various celebrity attributes seen by the consumers as prominent,
it leads to an actual purchasing decision. The present study tries to identify to what
extent consumers consider different characteristics significant when it comes to
purchasing intention.
b. Product Match-Up
The congruency between the big name and the Brand relies upon the level of match
between them (Mishra, 1990). If the superstar coordinates or is consistent with the
item publicized, it creates an abnormal state of trustworthiness (Chan K. N., 2013);
(Kamins M. a., 1994); (Kotler, 1997).
c. Physical Attractiveness
Physical engaging quality is related to an individual's first judgment passed on to
another person, which is managed by the characteristics and attributes of the endorser

20
(Bardia Y. A., 2011, for example, weight, tallness, and facial magnificence. These
wellsprings of engaging quality make the impression of validity, which is gotten from
the degree to which the VIP endorser coordinates the customer's meanings of the
modifiers "alluring," "tasteful," "attractive/excellent," "rich," and "provocative"
(Ohanian R., 1990) If customers feel that they have much similarly as VIP, they are
most likely going to believe them to be all the more engaging. Hence, the
reasonability of an advertisement that uses a VIP endorser will, without a doubt, be
improved by the physical appeal of that big name (Chan K. N., 2013); (Till B. B.,
2000); (Zahid, 2002).
d. Trust-worthiness
The reliability is identified with the legitimacy of the superstar endorser's
confirmations about the item from the client's point of view (Erdogan B., 1999);
(Hovland, 1953). Setting up a degree of dependability is the best helpful way to make
the client progressively sure about the item (Hakimi, 2011); (Ohanian R., 1990).

e. Expertise
A proficient endorser (for example, a celebrity) is progressively powerful at affecting
the consumer to purchase the item (Bardia Y. A., 2011) and to have an emotional
disposition toward the Brand (Chan K. N., 2013); (Erdogan B., 1999).
f. The Source Credibility Scale and the Source Attractiveness Scale
This study rests on two models, the Source Credibility Model (Hovland et al., 1953;
Hovland & Weiss, 1951; Johansson & Sparredal, 2002; Ohanian, 1990) and The Source
Attractiveness Model (Johansson & Sparredal, 2002; McGuire, 1968, 1985), which
further propose a new model applicable to political branding context that contends two
main attributes, namely credibility and attractiveness. Credibility is commonly
measured by expertise, while attractiveness includes physical attractiveness and
closeness.
The term “Source Credibility” refers to the endorser’s positive characteristics, which
can increase the level of acceptance and persuasion in advertising (Ohanian, 1990). The
model argues that the effectiveness of the source depends on the perceived level of
expertise and trustworthiness of an endorser (Hovland et al., 1953; Hovland & Weiss,
1951; Ohanian, 1990). According to (Hovland et al. 1953), expertise has been defined
as “the extent to which a communicator is perceived to be a source of valid assertions”
(p. 21). It is generally associated with knowledge, experience, or skills that the endorser

21
possesses in a given area (Ericsson & Håkansson, 2005). Athletes are perceived as
persuasive when endorsing sports-related products, and doctors are perceived as experts
when representing pharmaceutical products. Trustworthiness is “the degree of
confidence in the communicator’s intent to communicate the assertions he considers
most valid” (p. 21). Although early empirical work suggested that trustworthiness
served as a determinant in persuasion (e.g., Friedman & Friedman, 1978), scholars like
Ohanian (1990) argue that trustworthiness is not significantly associated with purchase
intentions. According to Ohanian (1990), the degree of trustworthiness highly depends
on: (1) the overall social trust of the communicator’s occupation and profession and (2)
the receiver’s perception of the communicator’s behavior (McGinnies & Ward, 1980;
Ohanian, 1990).
In dealing with Peng’s case, this study majorly focuses on how media content on social
media portrays the positive characteristics of Peng. Although the perception of such
images is equally important in the big picture, it is now excluded in this article due to
the specific research purpose and methodology. Following the initial model of Hovland
and his associates, The Source Attractiveness Model has its origins in the social
psychological research as a component of the “Source Valence Model” of McGuire
(1985).
By definition, attractiveness establishes an effective bond between the source and
receiver. People are more likely to be influenced by the source when they have a desire
to identify. Meanwhile, they are more likely to identify with people they find likable
(Ericsson & Håkansson, 2005; Kelman, 1961). In McGuire’s model, “likability” serves
as the first significant dimension in attractiveness, referring to “an affection for the
source due to physical appearance, behavior or other personal traits” (Erdogan, Baker,
& Tagg, 2001, p. 40). In its further development, likability has been merged as a
component of “physical attractiveness” addressed by Ohanian (1990), referring to the
desirable features and qualities in terms of facial and body appearance, dressing codes,
postures, and so on (Morrow, McElroy, Stamper, & Wilson, 1990). As the classic factor
in attractiveness, physical attractiveness mainly emphasizes the visual characteristics
of sources and celebrity figures. Still, it has been criticized for the objection of bodies,
especially to female images appealing to visual pleasure and the male gaze (Mulvey,
1989).
To a large extent, the image of how the sourced figure lives, interacts, and positions, in
reality, has yet to be neglected mainly in physical attractiveness. As both reflection and

22
supplement, this article suggests an additional dimension, namely closeness, in
attractiveness other than physical likability. Unlike physical attractiveness’s emphasis
on visual pleasure, closeness refers to establishing compelling intimacy between the
source and its receiver, usually achieved by promotion or branding strategies. It aims
to shift the focus merely on “objective” attractiveness into the interest of effective
relationships. It also pays more attention to how the sourced figure, as an exhibited
beauty and a socialized human, strategically communicates with its audience. The
concept of closeness draws first on the two dimensions other than likability in
McGuire’s Model: the audience’s supposed similarity and familiarity with the
respondent (McGuire, 1968, 1985). In the model, similarity means “a supposed
resemblance between the source and the receiver.” Familiarity refers to “knowledge of
the source through exposure” and indicates that the source is easy to recognize
(Eredogan et al., 2001).
The two dimensions, despite the absence of critical sense in McGuire’s subjectivity,
have tried to go beyond simple physical appearance and indicate a widening schema to
include virtuous characteristics such as intellectual skills, personal traits, and lifestyles
serve as the antecedents of achieving receiver’s identification with the source (Patra &
Datta, 2012). However, none is comprehensive enough to conclude the strategically
established intimacy between the source and its receiver. For example, the tendency to
secularize a politician’s public persona appears to help decrease the distance between
political figures and the public (Corner, 2003; Marshall, 1997). The emphasis on an
ordinary and private persona, as Marshall puts it, could be seen as a converged
construction of politics and entertainment, which links one’s personality to their
political ideology (Marshall, 1997).
In this sense, the portrayal of the endorser’s personal life and social network with others
should also be considered a significant component in building an affective bond with
the audience, yet cannot be adequately explained by either similarity or familiarity.
Therefore, this study decides to use “closeness” as a more complete and symbolic term
to cover the common ground of the strategical tendency to shorten the social distance
between the source and its receiver.
H2: Celebrity endorsement moderates the relationship between negative publicity
and consumer loyalty.

23
2.3.2.MATCH-UP THEORY
Coordinate speculation is the most significant hypothesis for seeing how buyers' views of
celebrity outrage work. It gives a framework that spotlights fit, or congruity, between an
endorser and Brand. Fit and consistency are tradable with articulating "organizing up" to
depict the endorser and brand relationship. The establishments of arranged speculation
were set up in made by (Mowen, Brown, and Schulman, 1979), where they constructed
ties between the endorser, the Brand, and groups using the evening out theory made by
(Heider, 1958) (Koering and Boyd, 2009)They assessed and thought how publicizing is
best when there is a strong fit between every one of the three parts. From here, the absolute
most huge investigations, for instance, achievement research coordinated by (Kahle and
Pamela, 1985)— focused on finding a match between potential endorsers and brand things
through physical connecting with quality rather than various segments, for instance, a
brand or association, on affecting great brand attitudes (Till and Busler, 2000); (Kamins
M., 1990). The congruency between the celebrity and the Brand relies upon the level of
match between them (Mishra, 1990). If the celebrity coordinates or fits with the item
publicized, it makes an abnormal state of validity (Chan K. N., 2013); (Kamins M. a.,
1994); (Kotler, 1997).
Numerous consequent studies have been growing to different domains of impact, for
example, aptitude and believability, which demonstrate that coordinate speculation is
viable in multiple parts past simply starting physical allure, discussed in the source
validity model. Concerning the root meaning of coordinate speculation, as indicated by
(Koering and Boyd, 2009), "an endorser won't beneficially affect purchaser's mentalities
and conduct except if the picture of the endorser 'coordinates' with the picture of the
brand." This implies that if buyers can characterize an association between the
representative and the Brand, the support can be viewed as inadequate and not financially
savvy because of the costly idea of celebrity support. These associations are connected to
contemplating with acquainted realizing, which is a valuable model for considering at
match hypothesis (Till &Busler, 2000). Endorsers are "increasingly successful when there
is a 'fit' between the endorser and the supported item" (Till and Busler, 2000). Related
learning exhibits how subsequently, "demeanors, picture components, or great conduct
expectations become related with the embraced brand" (Carrillat, d'Astous, & Lazure,
2013).

24
2.3.3.ATTRIBUTION THEORY
Researchers often use the Attribution theory to explain the impact of WOM (word of
mouth) and other marketing practices on persuasion. (Mizerski 1982; Chatterjee 2011;
Laczniak et al. 2001; Rifon et al. 2004; Sen & Lerman 2007). The attribution theory
describes how people make causal inferences regarding why a communicator advocates
a particular position or behaves in a certain way. (Kelley 1973; Mizerski et al. 1979;
Folkes 1988). The discounting principle in attribution theory (Kelley 1973) has been used
to examine the effects of source characteristics on persuasion. A study found that the
participants receiving an endorsement about a used car from a low trustworthy source (for
an example car mechanic who is part owner of a nearby automobile dealership) were less
likely to be persuaded by the endorsement than those receiving the endorsement from a
highly trustworthy source (example a car mechanic who has no relation to the automobile
dealership). (Mowen, 1986) According to attribution theory, consistency in provided
content signals trustworthiness (Cheung et al. 2009), and user ratings of the usefulness of
single posts and reviews can also provide valuable cues to the overall credibility of
electronic word of mouth. (Reichelt et al. 2014).

2.4.SWITCHING COST
Researchers propose three kinds of exchanging costs, to be specific: the procedural
expense (for example, time and exertion spent, monetary hazard, assessment and set up
price), money-related expense (for example, budgetary misfortune, expenses, and
advantages), and social expenses (for example mental or enthusiastic uneasiness due to
character misfortune) (Jones, Mothersbaugh, and Beatty, 2000), (Burnham, Frels, and
Mahajan, 2003). In this manner, exchanging expenses are characterized as "client
recognitions concerning time, cash and exertion related to the adjustment in specialist
organizations" (Jones, Mothersbaugh, and Beatty, 2000). When the exchange expenses
are high, shoppers may remain faithful to an organization although they are not fulfilled
with specific parts of the organization (Pan, Sheng, &Xie, 2012). Along these lines, the
apparent future exchanging expenses forestall the client from changing to a contender
(Yanamandram and White, 2006). Exchanging expenses are additionally characterized as
those costs that buyers partner with the way toward changing, starting with one provider
and then onto the next (Burnham et al., 2003; Wathne et al., 2001). At the point when
shoppers show that it isn't advantageous exchanging providers, this might be because they

25
see hindrances to it. These may incorporate looking, exchange, learning costs, limits for
dependability, propensity, enthusiastic expenses, and intellectual exertion, together with
the budgetary, social, and mental hazards. In particular, (Burnham et al. 2003) portray
three kinds of exchanging costs: procedural exchanging costs – these incorporate the
monetary risk and assessment costs and include consumption of time and exertion;
budgetary exchanging costs – include the loss of advantages and money-related assets;
and social trading costs – the loss of the individual relationship and the relationship with
the Brand, which incorporates mental and energetic trouble on account of the loss of
personality and the breaking of securities. Exchanging expenses prevents researching
new providers (Wathne et al., 2001). To the degree that people see expenses or boundaries
to leave, they will generally keep up with their provider (Burnham et al., 2003; Lee et al.,
2001). If trading costs are low, disillusionment with the organization's quality, price, or
firm will stir them to expect to switch suppliers additionally; if they are high, many
puzzled purchasers will presumably demonstrate a "false immovability." Different
assessments have shown that trading expenses are a coordinating variable that conversely
impacts the association among satisfaction and intends to keep up the relationship
(Burnham et al., 2003; Jones et al., 2000; Oliva et al., 1992; Sharma and Patterson, 2000).
As the costs rise, the effect of satisfaction on the objective to keep up the relationship
diminishes differently. As shown by Jones and Sasser (1995), if exchanging expenses are
low, there is a probability of discovering not many false loyal (prisoners) since
disappointment should bring about evolving suppliers. Then again, if exchanging
expenses are high, there is a probability of seeing numerous bogus steadfast clients
because despite being disappointed, trading costs make them more averse to change.
Regardless of the expanding enthusiasm for arbitrator factors, especially concerning the
consumer loyalty build and loyalty satisfaction connect (Olsen et al., 2005; Ranaweera et
al., 2005; Patterson, 2000), Consequently, from the hypothesis, and experimental proof
referred to over the accompanying speculation is developed.
H3: Switching cost moderates the relationship between negative publicity and
consumer loyalty.

26
2.5.CONCEPTUAL FRAMEWORK

Celebrity
Endorsement

H2
Negative Publicity Effects on
H1
Consumer
Loyalty
H3

Switching Cost

This model utilizes a comprehensive understanding of negative attention and considers

the impact of brand association and celebrity endorsement to mitigate the negative effect
on the brand. Additionally, it evaluates the cost impact of the endorsement. The level of
match between the celebrity and brand has a direct correlation with the reduction in
negativity towards the brand, as customers have higher brand commitment levels.
Furthermore, the more unique the match, the less likely the customer is to switch to a
competitor's brand or seek alternative brands. Ultimately, the results of this model
demonstrate how individual consumers perceive the brand. The model is self-developed
based on various variables, and the questionnaire used has been adapted from the study
"Customer engagement through Omni channel retailing - the effects of channel integration
quality" by Chan (2018).

27
 CHAPTER: 3

RESEARCH METHODOLOGY

In this chapter, the research method will be chosen and described. This methodology
chapter will include research purpose, research approach, research strategy, sampling,
the questionnaire, hypothesis testing, and data analysis, which will be complemented by
reliability, validity, and moderator analysis. SMART PLS will is used.

3.1. Research Purpose

The research purpose can be classified into three types, namely, exploratory, descriptive,
and explanatory. An experimental study is conducted to explore a situation or
phenomenon by analyzing literature, interviewing experts, or conducting focus groups.
In contrast, descriptive research aims to describe different conditions or events by
providing characteristics of a population or some phenomenon. This approach focuses
on telling already existing data. The third and last type is explanatory research, which is
used to analyze relationships between variables when studying a specific problem or
situation. (Saunders et al., 2009)
The main objective of this research is to describe the factors causing negative publicity
and moderate the role of celebrity endorsement and switching costs to reduce its negative
effect on consumer loyalty. To achieve this objective, three research questions have been
formulated:
RQ1: What is the effect of negative publicity of brands on consumer loyalty?
RQ2: What is the moderating effect of switching costs between negative publicity and
consumer loyalty?
RQ3: What is the moderating effect of celebrity endorsers between negative publicity
and consumer loyalty?

3.2.Research Approach
When choosing approaches, there are two alternatives: quantitative and qualitative. A
quantitative approach is when the data collection or analysis is used to generate
numerical data, while qualitative research generates or uses non-numerical data.
(Saunders et al., 2009) The data that has been collected is often dependent on the choice
between inductive and deductive methods. (Halvorsen, 1992) An inductive study will
usually not have its base in theory and models but instead in vaguer problem areas and
temporary or occasional observations. On the opposite side to inductive is the deductive

28
study that has its base in theory and hypotheses and is often tested against reality to see
if they match each other. (Halvorsen, 1992) Since this study aims to examine what factors
influence endorsers’ credibility, a deductive method suits better and therefore has been
chosen based on quantitative data through surveys (answered by consumers). Also,
previous studies within source models have used a deductive method, which has been
proven to be a good fit.

3.3.Research Strategy
Now that the research approach has been chosen, the next thing is to determine which
research strategy should be used in this thesis. There are five research strategies to choose
from: an experiment, survey, archival analysis, history, and case study. Choosing which
depends on the nature of the research questions. (Saunders et al., 2009). There are three
aspects to consider: types of research questions, control over behavioral events and focus
on contemporary events. (Saunders et al., 2009) After considering these aspects, a survey
was chosen as the research strategy for this study. A survey as a strategy gives the
researcher the chance to collect a large amount of quantitative data from a sizeable
population economically. Another pro of choosing a survey is the control it provides the
researcher over the research process. In addition, if sampling is used for the study, it can
generate findings representative of a population. There are also some implications with
surveys; they require the researcher to spend time ensuring that the sample is
representative, designing, piloting the instrument for data collection, and ensuring
enough response rates are collected. (Saunders et al. 2009) A survey as a strategy is the
best strategy alternative in research projects that involve the collection of information
from a large sample of individuals. However, it is essential to inform the respondents
that their behaviors/attitudes are collected, as this can influence their response and lead
to response bias. (Hair et al., 2007).

3.4.Research Methodology
The researcher will lead essential information started by the specialist explicitly to
address the examination issue through the consequence of surveys. A few abstract flows
were counseled to build up a hypothetical reason for the model. Web of Science and
Google Scholar were utilized to gather the actual writing. Chiefly articles distributed in
top-notch showcasing related journals were used (for example, Journal of Marketing,
Journal of Marketing Research). The criteria used to pass judgment on the quality and

29
relevance of the writing is the number of references and the kind of journal. The author
of this report has decided on the nature of the substance as a discrete component too. For
the initial segment of the model, primarily writing on the impacts of negative attention
was utilized (for example, Eisingerich et al., 2010; Ahluwalia et al., 2000), and negative
data (for instance, Mizerski, 1982; Henrad, 2002). The second piece of the model pursues
the writing on customer loyalty (for example, Dick and Basu, 1994).

3.5.Research Philosophy

The current study relies on the Positivist research philosophy that emphasizes on empirical
observation and measurement as the primary source of knowledge. This philosophical
stance entails that the only trustworthy information is objective and factual, and can be
obtained through the senses. Positivism rejects subjective interpretations and personal
experiences as valid sources of knowledge, and instead advocates for the use of scientific
methods to gather data and test hypotheses. In contrast, other research philosophies such as
Realism, Interpretivism, and Axiology have different epistemological assumptions and
ontological beliefs. Realism posits that the external world exists independently of human
perception, while Interpretivism argues that subjective meanings and interpretations are
essential for understanding social phenomena. Axiology, on the other hand, emphasizes on
values and ethics as significant components of research. However, for the present study,
Positivism has been selected as the most appropriate research philosophy due to its
emphasis on empirical evidence and objective observations. This approach aligns with the
study's objective of generating factual and reliable information through systematic data
collection and rigorous analysis.

3.6.Population and Sample

According to (Sekran, 2009), population is the gathering of individuals, occasions, or
things of premium that the analyst wishes to explore.
According to (Sekran, 2009), a sample is a part of the population. The sample of this
research is the people only in Peshawar, as many as 402 respondents (Urban/Rural) based
on their preference and usage of brands. G-Power is used to select the sample size. This
study incorporates quantitative methods and data collection is made through purposive
sampling this technique ensures inclusion of individuals who are experts and
professionals in a specific field of interest (Palinkas & Horwitz, 2015)

30
3.7.DATA COLLECTION METHOD
Having selected the survey as a research strategy, the next step is to determine the most
appropriate data collection method. The selection of the data collection method should
depend on the nature and volume of data to be collected as well as the type of research.
In the case of deductive and quantitative data collection, three categories are available:
self-completion, interviewer-completion, and observation.
A decision was made to pick questionnaires, but they can either be interviewer-
administrated or self-administrated. The interviewer administrated is as it sounds, in that
it is interview-based, and you record the interviews. Self-administrated is as heard in the
name done by the respondents themselves, and there they fill out their answers without
the presence of an instructor or interviewer, which can be facilitated through drop-
off/pick up of physical questionnaires, email, and surveys online. (Hair et al., 2007).
When deciding which questionnaire to pick out of the two, a variety of factors matters,
such as the size of the required sample, characteristics of respondents, type, and the
number of questions needed (Saunders et al., 2009). As mentioned in the research
purpose and approach of this quantitative study, data requires a large number of
respondents. This data collection method does provide sufficient data input for the study.
Therefore, this thesis has chosen self-completion as the data collection method. When
having a self-completion questionnaire, the data is collected through a structured
questionnaire with a predetermined set of questions (Saunders et al., 2009). Some
activities are considered before concluding a questionnaire: pretesting of the
questionnaire for validation, creation of a general design, and lastly, a choice of method
for questionnaire administration (Hair et al., 2007).

3.8.QUESTIONNAIRE DEVELOPMENT
Some steps are essential and need to be considered when creating a well-formulated and
constructed questionnaire. Hair et al. (2007) recommend five steps to ensure the
questionnaire is designed and evaluated correctly, questionnaire has been adapted from the
study (Chan, 2018) “Customer engagement through Omni channel retailing - the effects of
channel integration quality”.
Table 2: Steps to be followed in the Design of a Questionnaire
Step 1: Initial Considerations:
• Clarify the nature of the research problem and objectives.
• Develop research questions to meet research objectives.

31
• Define target population and sampling frame (identify potential respondents.)
• Determine sampling approach, sample size, and expected response rate.
Step 2: Clarification of Concepts
• Ensure the concept(s) can be clearly defined.
• Select the variables/indicators to represent the concepts.
• Determine the level of measurement.
Step 3: Determine Question Types, Format, and Sequence
• Determine the types of questions include and their order.
• Check the wording and coding of questions.
• Decide on the questions' grouping and the questionnaire's overall length.
• Determine the structure and layout of the questionnaire.
Step 4: Pretest and Questionnaire
• Determine the nature of the pretest for the preliminary questionnaire.
• Analyze initial data to identify limitations of the preliminary questionnaire.
• Refine the questionnaire as needed.
• Revisit some or all of the above steps, if necessary.
Step 5: Administer the Questionnaire
• Identify the “best practice” for administering the type of questionnaire utilized.
• Train and audit field workers, if required.
• Ensure a process is in place to handle completed questionnaires.
• Determine the deadline and follow-up methods.
The following steps presented in the table above are adapted from previous research and
have been formulated for this thesis to answer the formulated hypotheses. The chosen
constructs and scales used in this thesis have been used in earlier studies with satisfying
results. Also, the questionnaire has been appropriately developed, using simple wording,
avoiding ambiguity, and having an appealing format to make it as easy as possible for
the respondents to follow. (Hair et al., 2007)
a. Pretest
Before the launch, the study went through a pretest to clarify if there was anything
that could be improved and if it was difficult to interpret for the respondent. The
pretest was made by current students but also educated people that had experience in
conducting surveys in their jobs. After the pretest, some minor changes to improve
the study were made. Mainly changes in wording, questions layout, and translation.

32
3.9.DATA ANALYSIS
When analyzing data in quantitative research, examination of the data collected is
involved; this is to identify and confirm relationships. To ensure that the data can be used
in a static analysis, the quantitative data must be edited, coded, and sometimes
transformed before statistical analysis. There are two different approaches for
quantitative data analysis, and they are using descriptive statistics or using statics tests.
When using descriptive statics, it is to understand the data; the statics test is on the other
and to be able to test the hypothesis. (Hair et al., 2007)
This thesis uses the program SPSS (version 23- windows platform) when performing all
the analyses to test the formulated hypothesis and do reliability tests and inter-item
correlations. In the upcoming paragraph, the utilized method for the data analysis will be
explained in more detail.

3.10. DATA PREPARATION

After all the data was gathered, it needed to be prepared, and this was done by removing
all the surveys that needed to be completed or had unserious answers. After this was
done, the next step was to put the data in SPSS. From the 514 responses that were
completed, there were 14 responses that was removed from the sample due to obvious
response patterns (for example, answering one on all the questions). An important aspect
to consider is that problems can occur when importing data to statics program. Another
thing with this survey was that there were no forced responses, meaning that many
surveys were not finished or some answers needed to be included; this resulted that 10
respondents got removed.

3.11. STATISTICAL TECHNIQUES

When doing hypothesis testing, there are different statistical techniques to choose from.
(Hair et al.,2007). The selected methods for this thesis are Multivariate analysis,
reliability, multiple regression analysis, and t-test.

Multiple regression analysis is used to evaluate how much of the variance of the
dependent variable can be explained by the independent variables. This analysis will also
indicate how significant the relative contribution of each of the independent variables is.
(Pallant, 2016) In this thesis, multi regression analyses were used to test the proposed
hypotheses. The T-test can be used to compare the means of two variables, to be able to

33
test the formulated hypothesis. (Hair et al.,2007). Moreover, the telephonic conversation
was made with those respondents who either did not give a response.
3.12. QUALITY STANDARD
After getting the results from any concept (constructs) that are available for analysis, it
needs to be measured to see if they are accurate and consistent. If it holds over a specific
standard, it can be used for the analysis. Here is the consistency connected to the accuracy
to validity and the part of reliability. (Hair et al., 2007)

3.13. Reliability
When conducting research, the measurement that is included in the report must produce
reliable findings; this could be connected to the consistency of the research findings. To be
able to reassure there is reliability, some guidelines should be followed. To ensure scale is
reliable, the first and minimum number of items measuring a particular concept should be
at least three. Furthermore, the correlation between the items included in the scale needs to
be positive (Hair et al., 2007).
This condition is satisfied with the constructs used for this research, measuring endorser’s
credibility and consumer’s attitude towards the Brand, affected by negative publicity. All
the different constructs included in this thesis has picked from existing studies, and the only
scale used in the questionnaire is a 5 points Likert scale.
The endorser expertise, attractiveness, and trustworthiness scale by Ohanian (1990) used
in this thesis was initially similar to this thesis. Ohanian wanted to investigate factors that
affected endorsers' credibility and examined attractiveness, trustworthiness, and expertise
for what he later called as source-credibility scale.

3.14. Validity
Validity is to what degree a construct measures what it is supposed to measure. To ensure
that the report is constructed in the way it assesses measurement validity, three parts of
validity are considered. These are content validity, construct validity, and criterion validity.
(Hair et al., 2007)
For this thesis, the content validity is concerned with whether the questions in the survey
(questionnaire) will provide acceptable cover to answer the proposed hypothesis (Saunder
et al., 2009). However, more than this validation is needed to indicate, if a scale is
measuring the thing it is supposed to measure. Two additional validity will be examined to
increase the validity: constructed and criterion validation. (Hair et al., 2007)

34
Construct validity refers to scale or construct measuring the intended idea. To be able to
make sure of this criterion, the theory behind a construct is of relevance; this is due to that
it explains why the scale works and how someone interprets the results from the application.
There are two ways to ensure that the scale is a matter of fact of construct validity. The first
one is discriminant validity, and the second is convergent validity. Discriminant validity is
to what extent the construct does not correlate with other measures that is different from it.
Convergent validity, on the other hand, explains to which extent there is a positive
correlation with other measures of the same construct (Hair et al., 2007).
The third and last are criterion validity, this measure how well a construct acts as it could
be expected concerning other variables. Criterion-related validity is obtained through
different statically analyses such as correlation, and the requirement is that the received
data can be checked through statistical analyses (Saunders et al., 2009)

35
 CHAPTER: 4

PRESENTATION AND ANALYSIS OF DATA

This chapter presents the results of the survey. The data is collected through online surveys, which
were sent out via emails and social media. The output of the study was also described and analyzed
in this chapter.

Before a final analysis of the gathered data, some assumptions are made. The reason for
doing this was to provide more reliable results. These assumptions played an essential role
in the application of factor analyses and multiple regression techniques. The first
assumption must be made is to get a significant sample size. If the sample size were too
small, there would be a risk that the analytic results would not be repeatable in other surveys
and therefore lose scientific value, as it lowers the validity of the research (Pallant, 2016).
Analysis of Data gathered on Likert scale questionnaire from sample frame are run on SPSS
tested for reliability test to check whether attendants' responses are reliable. The study also
applied Partial Least Squares (PLS), Structural Equation Modelling (PLS-SEM) using
Smart PLS 3.2.4. PLS is a more robust choice with the fewest assumptions and unbiased
estimation of the small sample size. It does not require a normal distribution or a substantial
number of observations as input data (Hair, Hult, Ringle 2014). To test hypotheses, this
study used two techniques recommended by the previous literature (Hair et al., 2014; Vinzi,
Chin, Henseler, and Wang, 2010). These techniques include assessment of the
measurement model and structural model. The study applied convergent and discriminant
validity tests to test the measurement model. Convergent validity measures the correlation
between constructs. Factor loadings, composite reliability (CR), and average variance
extracted (AVE) represent measuring the convergent validity (Hair et al., 2014).

4.1. Profile of Respondents

The number of respondents used for data analysis for this research was 402. The answer
collected on the chosen constructs is some demographic data. This is on age and gender.
The table below presents the gender distribution of the respondent. The results were 141
(34.8%) females and 261 males (65.3%).

36
 GENDER
N Percent
Valid Female 141 34.8
Male 261 65.3
Total 402 100.0

4.2. Reliability Test

Reliability of scales is used to see how questions or items in a scale relate to each other. A
common way to see scales' reliability is by looking at the Cronbach alpha coefficient.
(Pallant, 2016). Because the Cronbach alpha values for this are high and above the
minimum recommendation (higher than 0.7) (Hair et al., 2007), no further internal
consistency analyses have been conducted. Answers from the respondents on all four
constructs were obtained and put in SPSS for reliability tests. The test was performed for
each construct separately to attain the estimation of “Cronbach’s Alphas.”
4.2.1. Results of Reliability Test

Construct Cronbach’s Alpha

Customer Loyalty (CL) 0.763
Negative Publicity (NP) 0.771
Celebrity Endorsement (CE) 0.780
Switching Cost (SC) 0.867

4.2.2. Interpretation of Results

The data is considered more reliable as Cronbach’s Alpha gets closer to 1.0 (Uma Sekaran,
2003). whenever the Cronbach’s Alpha is lesser than 0.60, the data is considered poor, data
with the Cronbach’s Alpha value in the range of 0.70 is considered acceptable, and data
bearing the Cronbach’s Alpha value in the range of 0.80 is thought to be good. Data with
Cronbach’s Alpha value of 0.9 and over is very good. Our data's Cronbach’s Alpha value
falls in the acceptable and good ranges.

4.3. Test for Normality of Data

Multiple normality tests are run and analyzed to test the normality of data. These SPSS tests
are helpful in giving a handful of information regarding the data, including the following.

37
Table Entitled Descriptive

Case Processing Summary

Cases
Valid Missing Total
N Percent N Percent N Percent
CL 402 100.0% 0 0.0% 402 100.0%

Descriptives
Statistic Std. Error
C Mean 5.4600 .05164
L 95% Confidence Lower 5.3585
Interval for Mean Bound
Upper 5.5615
Bound
5% Trimmed Mean 5.5375
Median 5.7500
Variance 1.067
Std. Deviation 1.03281
Minimum 1.75
Maximum 7.00
Range 5.25
Interquartile Range 1.50
Skewness -1.218 .122
Kurtosis 1.033 .243

Table (4.1)
The values of both samples, mean 5.46 and 5% trimmed mean 5.53 of the respondents’
responses on customer satisfaction fall within a 95% confidence interval of 5.35 to 5.56.
This is the interval within which a real mean is expect to fall. The reported two mean values
fall between 4 “I Agree” and 5 “I Strongly Disagree.” The values of Skewness (S) and
Kurtosis (K) are measured at 0.162 and 3.02, and a normal distribution required these
values to equal 0 and 3, respectively.
Table of Extreme Values

Extreme Values
Case Value
Number
C Highest 1 108 7.00
L 2 128 7.00
3 178 7.00

38
 4 198 7.00
5 37 6.75a
Lowest 1 328 1.75
2 299 1.75
3 48 1.75
4 257 2.25
5 6 2.25
a. Only a partial list of cases with the value 6.75 are shown in the table of upper extremes.

Table (4.2)
The highest extreme values in this case which is 4 refers to agree strongly and the extremely
low is 1, which refers to I Agree; both highest and lowest are logically acceptable.
Table of Normality Test

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
CL .189 402 .000 .871 402 .000
a. Lilliefors Significance Correction

Table (4.3)
The Shapiro-Wilk Test in these two tests is considered for smaller sample sizes. According
to these tests, if the Sig. level is greater than 0.05, the data is said to be normal, and if the
sig level is less than 0.05, data strays from a normal distribution. Here in our case, the data
according to Shapiro-Wilk Test is normally distributed at a p-value >0.05.

39
Normal Q-Q Plot

This test suggests that the points will be closer to the diagonal line for normal data
distribution. If the data is not distributed normally, the points will stay away from the
diagonal line.
The graph concludes that data majorly appears normally distributed as it is projected closer
to the diagonal line.

Detrended Normal Q-Q Plot

The detrended Normal Q-Q Plot further clarifies the position regarding data.

40
4.4. Histogram

The histogram here represents that majority of respondents’ responses fall between 4 to 5

4.5. Stem and Leaf Test

CL Stem-and-Leaf Plot

Frequency Stem & Leaf

5.00 Extremes (=<2.3)

.00 2 .
6.00 2 . 577
7.00 3 . 222
21.00 3 . 5555555577
22.00 4 . 0000222222
46.00 4 . 55555555555557777777777
24.00 5 . 002222222222
73.00 5 . 555555555555555555555777777777777777
161.00 6 .
000000000000000000000000000000022222222222222222222222222222222222222222
22222222
31.00 6 . 555555555777777
4.00 7 . 00

Stem width: 1.00

Each leaf: 2 case(s)

41
4.6. Outlier Test

4.7.Detection of Multicollinearity

Multicollinearity is fine relating to the model specification or its estimation. Instead, it is a

problem originating from the nature of the data as it exists/happens when one independent
variable affects another independent variable. In general research analysis routines, one
cannot eliminate multicollinearity. Therefore, its detection should mean locating which one
or two explanatory variables are causing the problem and to what degree multicollinearity
exists among them. Such detection of the problem-causing variables may help reduce the
severity of the problem.
Many measures could be adopted to measure the degree of multicollinearity. In this study,
we will use the following methods to detect multicollinearity.
a. Correlation between independent variables
b. Klien’s Rule of Thumb
c. Tolerance and V IF.

According to one of the standard assumptions of the “Ordinary Least Squares (OLS)”
estimation of the explanatory variables, Xi should not linearly correlate or affect each other,
which means one explanatory variable should not correlate with other explanatory
variables. If they do, the problem in econometrics is referred to as multicollinearity

42
 Correlations
Correlations
NP CE SC
NP Pearson 1 .092* .102*
Correlation
Sig. (2-tailed) .037 .020
N 514 514 514
CE Pearson .092* 1 .298**
Correlation
Sig. (2-tailed) .037 .000
N 514 514 514
SC Pearson .102* .298** 1
Correlation
Sig. (2-tailed) .020 .000
N 514 514 514
*. Correlation is significant at the 0.05 level (2-tailed).
**. Correlation is significant at the 0.01 level (2-tailed).
Table (4.4)
The output reflects a mild correlation between the explanatory X variables.

4.7. Measurement Model Construct Reliability and Validity

Table 4.5: Measurement Model Construct Reliability and Validity

Cronbach's rho_A Composite AVE
Alpha Reliability
Celebrity Endorsement 0.780 0.797 0.848 0.529
Consumer Loyalty 0.763 0.762 0.849 0.585
Negative Publicity 0.771 0.772 0.845 0.523
Switching Cost 0.867 0.873 0.900 0.602
According to the literature, the benchmark for factor loadings is (>0.7) concerning an
individual item representing a construct. Similarly, composite reliability should be (>0.7)
and AVE (>0.5) as standards. If the values for the measures are above the described cutoff
values, then there is sufficient convergent validity (Hair et al., 2014; Vinzi et al., 2010).
The results of the measures above reported in Table 4.5 show that all the values are above
the cutoff values, and thus the measurement model is appropriate for all the constructs.

43
4.8.Fornell-Larcker Criterion
Table 4.6: Fornell-Larcker Criterion
CE CL NP SC
Celebrity 0.728
Endorsement
Consumer Loyalty 0.536 0.765
Negative Publicity 0.391 0.421 0.773
Switching Cost 0.521 0.719 0.646 0.776
Note: The square roots of AVE are shown diagonally in bold

4.9.Discriminant Validity (HTMT0.90 Criterion)

Table 4.7: Discriminant Validity (HTMT0.90 Criterion)

CE CL NP SC

Celebrity
Endorsement
Consumer Loyalty 0.678
Negative Publicity 0.498 0.543
Switching Cost 0.616 0.875 0.791
Note: (HTMT0.90 Criterion)

For the assessment of the measurement model, the study used discriminant validity.
Discriminant validity shows the degree to which a construct differs from its counterparts
(Hair et al., 2014). The criterion for assessing discriminant validity is that the values in the
diagonal should be greater than all values in the respective row and column (Hair et al.,
2014; Vinzi et al., 2010). Table 4.6 reports that all the diagonal values are greater than
others; thus, the measurements have discriminant validity. Moreover, the HTMT 0.90
criterions have also proved discriminant validity, as all the values reported in Table 4.7 are
below 0.90.
4.10. Factor Loadings and Cross Loadings

Table 4.8: Factor Loading and Cross Loadings

CE CL NP SC
CE1 0.616 0.307 0.342 0.335
CE2 0.739 0.351 0.247 0.368
CE3 0.734 0.321 0.138 0.261
CE4 0.8 0.49 0.372 0.469
CE5 0.737 0.433 0.293 0.416
CL1 0.342 l0.713 0.389 0.609
CL2 0.436 0.778 0.265 0.485
CL3 0.432 0.798 0.364 0.554

44
 CL4 0.433 0.768 0.262 0.539
NP1 0.327 0.267 0.681 0.47
NP3 0.339 0.294 0.754 0.4
NP4 0.281 0.306 0.78 0.494
NP5 0.289 0.318 0.69 0.395
NP6 0.19 0.329 0.706 0.567
SC1 0.323 0.482 0.458 0.66
SC2 0.405 0.563 0.493 0.771
SC3 0.418 0.474 0.537 0.757
SC4 0.431 0.581 0.528 0.854
SC5 0.407 0.599 0.457 0.794
SC6 0.433 0.62 0.538 0.805

Items Celebrity Consumer Negative Switching

Endorsement Loyalty Publicity Cost
CE1 0.616
CE2 0.739
CE3 0.734
CE4 0.800
CE5 0.737
CL1 0.713
CL2 0.778
CL3 0.798
CL4 0.768
NP1 0.681
NP3 0.754
NP4 0.780
NP5 0.690
NP6 0.706
SC1 0.660
SC2 0.771
SC3 0.757
SC4 0.854
SC5 0.794
SC6 0.805

Factor loadings, composite reliability (CR), and average variance extracted (AVE)
represent measuring the convergent validity (Hair et al., 2014). According to the literature,
the benchmark for factor loadings is (>0.7) regarding an item representing a construct.
Similarly, composite reliability should be (>0.7) and AVE (>0.5) as standards. If the values
for the measures are above the described cutoff values, then there is sufficient convergent
validity (Hair et al., 2014; Vinzi et al., 2010). The results of the measures above reported
in Table 4.8 show that all the values are above the cutoff values, and thus the measurement
model is appropriate for all the constructs.

45
4.11. Assessment of Structural Model (Hypothesis Testing)

Table 5: Assessment of the Structural Model (Hypothesis Testing)

Relationships Std.beta Std.Error t-values p-values LL CI UL CI

Celebrity 0.214 0.056 3.848 0.000 0.105 0.318 Supported

Endorsement
-> Consumer
Loyalty
Celebrity - 0.041 1.323 0.186 -0.125 0.048 Not
Endorsement 0.055 Supported
X Negative
Publicity ->
Consumer
Loyalty
Negative - 0.044 1.495 0.136 -0.148 0.024 Not
Publicity -> 0.066 Supported
Consumer
Loyalty
Switching 0.694 0.054 12.898 0.000 0.593 0.792 Supported
Cost ->
Consumer
Loyalty
Switching 0.095 0.032 2.979 0.003 0.025 0.154 Supported
Cost X
Negative
Publicity ->
Consumer
Loyalty
Note: 99% (0.01) and 95% (0.05) Significance Levels

4.12. HYPOTHESIS
The first hypothesis, H1 of the study, proposed a significant positive relationship between
celebrity endorsement and consumer loyalty. The findings (β = 0.214, t-value = 3.848, p <
0.05) for the relationship reported in Table 5 support H1 of the study. The findings indicate
that celebrity endorsement help decrease the severe negative publicity effect of a brand,
and consumers of Peshawar, KP, are willing to buy a product that is endorsed by a celebrity.
The findings are consistent with Chaudhary, U., & Asthana, A. (2015) but inconsistent with
many previous studies (Mookda, Rarina & Khan, Sanan & Intasuwan, Sirichanok &
Chotchoung, Sirikanya. (2020). The inconsistency of findings may be an outcome of the
difference in cultures, values, and perceptions of these studies. The findings indicate that a
celebrity endorser with attributes such as attractiveness, trustworthiness, and familiarity
positively influences consumers’ perception of quality, purchase intentions, and consumer
loyalty (Kofi Osei-Frimpong, Georgina Donkor & Nana Owusu-Frimpong (2019).

46
The second hypothesis, H2 of the study, assumed an insignificant negative relationship
between celebrity endorsement and negative publicity impact on consumer loyalty. The
statistics reported in Table 5 show that celebrity endorsement with negative publicity has
an insignificant but positive (β = -0.055, t-value = 1.323, p > 0.05) association with
consumer loyalty which does not support H2. The findings consistent with (Kofi Osei-
Frimpong, Georgina Donkor & Nana Owusu-Frimpong (2019) explain that celebrity
endorsement positively impacts consumer loyalty when there is negative publicity.
However, celebrity endorsers’ negative publicity had no moderation effect on consumer
loyalty. This is logical, particularly in the cultural context of KP, where people do not
accept when negative publicity occurs.

The third hypothesis, H3 of the study, is also an insignificant negative relationship

between negative publicity and consumer loyalty. The statistics reported in Table 5 show
that celebrity endorsement with negative publicity has an insignificant and negative (β = -
0.055, t-value = 1.323, p > 0.05) association which rejects H3 of the study. The findings
consistent with Yu, M., Liu, F., Lee, J., and Soutar, G. (2018) explain that negative
publicity negatively influences all aspects of a brand and consumer loyalty. Further, the
relationship between brand image and brand attitudes and intentions depends on the level
of information severity affecting consumer loyalty.

The fourth hypothesis, H4 of the study, also proposed a significant positive coefficient of
switching costs in relation to consumer loyalty. The statistics (β = 0.694, t-value = 12.898,
p < 0.05) in Table 5 show a significant positive association that support H4 of the study.
The findings consistent with El-Manstrly, D. (2016) have a probable explanation:
switching costs moderate, in different ways, the relationships between customer loyalty,
trust, and perceived value.

The fifth hypothesis, H5, projected a positive and significant relationship between
switching costs and negative publicity, which impact consumer loyalty. The findings (β =
0.095, t-value = 2.979, p < 0.05) for the relationship as reported in Table 5 show a
significant positive association which support H5 of the study. The findings of this study
show that the switching cost factor directly affects loyalty and has a moderator effect on
both negative publicity and consumer loyalty. Therefore, it plays a crucial role in winning

47
consumer loyalty. In short, it is a quasi-moderator. However, switching costs are measured
as a unidimensional factor, but switching costs contain psychological, financial, and
procedural sub‐dimensions. The study is similar to Aydin, S., Özer, G., and Arasil,
Ö. (2005).

To conclude, this study established five hypotheses (H1 to H5) for investigating the impact
of negative publicity of brands on consumer loyalty. The study empirically examined these
hypotheses in the context that negative publicity can affect consumer loyalty based on
Peshawar geographic boundaries dependent on utilization examples impacted by social
traits of this very location. The study found that three out of five hypotheses are accepted
while the rest of the two are rejected.

48
 CHAPTER: 5

CONCLUSION AND RECOMENDATION

In this chapter, the conclusions for the research will be presented, and the formulated
research will be answered to fulfill the purpose of this study; the chapter will also include
implications and recommendations for the theory, practitioners, and further research, as
well as the identified limitations for this research.

5.1. Conclusion
Based on the results and studies discussed in the previous chapters, negative publicity has a potential
effect on consumers overall loyalty towards brands or firms. However, this effect is not straight
forward and may only sometimes be harmful. As visualized in the model, the effect of negative
publicity on consumer loyalty is moderated by several factors, namely; celebrity endorsement and
switching cost. The effects of the presented moderators can in their presence decrease or in their
absence increase the impact of negative publicity on consumer overall loyalty. Knowing that
negative publicity has an impact on consumer loyalty and knowing that loyalty is a necessary
construct, I propose that there is a potential link between negative publicity and loyalty. In other
words, negative publicity influences consumer loyalty. Switching costs can be the last resort for a
company to keep their customers loyal and, therefore, is the previous moderating factor in the
model. The model proposes that negative publicity affects consumer attitudes toward a firm or
brand, which may influence consumer loyalty. However, this effect is moderated by many factors
which can, in their presence, form a barricade against negative publicity in influencing consumer
overall attitude. In conclusion, throughout the negative change in attitude toward a brand or
company caused by negative publicity, negative publicity has a potential adverse effect on
consumer loyalty.
The current study's findings provide important insights for practitioners that celebrity
endorsement and switching costs are more effective when negative publicity occurs, and
consumers of Peshawar, KP, can buy a brand when the celebrity is associated with a brand,
which is a very effective way to decrease the negativity effect on Brand. Also, the consumer
will switch a brand if there is no celebrity endorsement, and consumers of Peshawar, KP,
can bear the cost of changing a brand to another brand. In short, the findings update the
companies and key stakeholders that adopting a flexible and positive celebrity endorsement
increases consumer loyalty in Peshawar, KP. The insignificant results of negative publicity
effects on brands can hurt consumer loyalty towards the brand. Given this, companies of
various brands should take some steps against negative publicity when it occurs. They can

49
decrease the effect by hiring a celebrity, keeping in mind the match-up product and
trustworthiness to increase the loyalty of consumers in this region. Besides this, if
companies of such brands fail to maintain their consumer loyalty when negative publicity
occurs, the consumers of this region will switch to another related brand. During data
collection, it has also been revealed that consumers in KP are loyal to the Brand until and
unless any scandal or negative publicity does not affects them.

5.2. Recommendations
The findings of this study provided information on what companies should focus on to
focus on their brand credibility and maintain consumer loyalty towards their Brand when
negative publicity occurs. This study can contribute by locating the factors of endorser
credibility that the YouTuber (the endorser) can focus on to build higher brand credibility.
As this study also indicates a direct and positive relationship between some of the factors
of celebrity endorser credibility and consumers' attitude towards the negative publicity of
a brand, this study also gives companies information on what attributes (characteristics) in
an endorser they should look at when picking a celebrity endorser for their advertisement
to moderate the severe negative impact of a brand on consumer loyalty.

5.3. Limitation and Future Direction

Like all other human activities, this study also has some obvious limitations that unveil
opportunities for further research. First, this study used a small sample size due to a weak
response and time and money constraints. Future studies may increase the sample size to
get a clearer picture of the subject matter. Second, the current study only focused on
Peshawar in the KP province of Pakistan. Studies in the future may extend their sample and
scope to other geographic boundaries and the rest of the country's three provinces. Besides
others, this will assist the impact of cultural values on the subject and the generalization of
the findings. Third, this study has a purely quantitative approach. Therefore, further studies
should be carried out to focus on a qualitative aspect of the subject for more robust findings.

50
 REFERENCES

Ahluwalia, Burnkrant, & Unnava. (2000, May). Consumer Response to Negative

Publicity: The Moderating Role of Commitment. Journal of Marketing
Research, 37, 203-214.

Ahluwalia, R. and Gürhan-Canli, Z. (2000), “The effects of extensions on the family

brand name: an accessibility diagnosticity perspective,” Journal of Consumer
Research, Vol. 27 No. 3, pp. 371-381.

Amos, C. H. (2008). Exploring the Relationship between Celebrity Endorser Effects

and Advertising Effectiveness. International Journal of Advertising, 27(2),
209- 234.

Anderson, E., & Weitz, B. (1992). The Use of Pledges to Build and Sustain
Commitment in Distribution Channels. Journal of Marketing Research, 29, 18-
34.
Aydin, S., Özer, G. and Arasil, Ö. (2005), "Customer loyalty and the effect of switching
costs as a moderator variable: A case in the Turkish mobile phone
market," Marketing Intelligence & Planning, Vol. 23 No. 1, pp. 89-103.
Bailey, A. (2007). Public Information and Consumer Skepticism Effects on
Celebrity Endorsements: Studies among Young Consumers. Journal of
Marketing Communications, 13(2), 85-107.
Bambauer-Sachse, S. and Mangold, S. (2011), “Brand equity dilution through
negative online word-of-mouth communication,” Journal of Retailing
and Consumer Services, Vol. 18No. 1, pp. 38-45.

Bardia, Y. A. (2011). Investigate the impact of celebrity endorsement on brand image.

European Journal of Scientific c Research, 58(1), 116-132.

Burnham, T., Frels, J., & Mahajan, V. (2003). Consumer switching costs: a
typology, antecedents, and consequences. Journal Academy of Marketing
Science, 31(2), 109-126.
Carrillat, F., d'Astous, A., &Lazure, J. (2013). For Better, for Worse? What to Do
When Celebrity Endorsements Go Bad. Journal of Advertising Research,
53(1), 15-30.

Chan. (2018). Customer engagement through Omni channel retailing - the effects of
channel integration quality. Industrial Marketing.

Chan, K. N. (2013). Impact of celebrity endorsement in advertising on brand image

among Chinese adolescents. Young Consumers, 14(2), 167-179.

Chan, K. N. (2013). Impact of celebrity endorsement in advertising on brand image

among Chinese adolescents. Young Consumers, 14(2), 167-179.

51
Campbell, C., L. F. Pitt, M. Parent, and P. R. Berthon, “Understanding Consumer
Conversations Around Ads in a Web 2.0 World,” Journal of Advertising, Vol. 40,
No. 1:87-102, 2011.
Chaudhary, U., & Asthana, A. (2015). “ Impact of celebrity endorsements on consumer
brand loyalty: Does it matter ? ”.
Chiou, J.S., Chi-Fen Hsu, A., and Hsieh, C.-H. (2013), “How negative online
information affects consumers’ brand evaluation: the moderating effects of
Brand attachment and source credibility,” Online Information Review, Vol. 37
No. 6, pp. 910-926.
Coombs, W., & H. S. (2001). An extended examination of the crises: A fusion of the
relational management and symbolic approaches.” Journal of Public Relations
Research. Journal of Public Relations Research, 13, 321-340.

Cretu, A., &Broodie, R. (2007). The influence of brand image and company reputation
where manufacturers market to small firms: 36, 230-240. A customer value
perspective

Cunningham, N., & Bright, L. F. (2012). The Tweet Is in Your Court: Measuring
Attitude Towards Athlete Endorsements in Social Media. International Journal
of Integrated Marketing Communications, 4(2), 73-87.
Dawar, N. and Pillutla, M. (2000), “Impact of product-harm crises on brand
equity: the moderating role of consumer expectations,” Journal of
Marketing Research, Vol. 37 No. 2, pp. 215-226.
Dentoni, D., Tonsor, G., Calantone, R. and Peterson, H. (2011), “Animal
welfare’ practices along the food chain: how do negative and positive
information affect consumers,” Journal of Food Products Marketing,
Vol. 17 Nos 2/3, pp. 279-302.
Eisingerich, A.B., Rubera, G., Seifert, M. and Bhardwaj, G. (2011), “Doing good
and doing better despite negative information? The role of corporate
social responsibility in consumer resistance to negative information”,
Journal of Service Research, Vol. 14 No. 1, pp. 60-75.
El-Manstrly, D. (2016), "Enhancing customer loyalty: critical switching cost
factors," Journal of Service Management, Vol. 27 No. 2.
Erdogan, B. (1999). Celebrity endorsement: a literature review. Journal of
Marketing Management, 15(3), 291-314.
Hair, J.F., Hult, G.T., Ringle, C. M. &. 2014. Partial Least Squares Structural Equation
Modeling (PLS-SEM).
Heider, F. (1958). The psychology of interpersonal relations.

Hernard. (2002). Negative Publicity: What Companies need to know about public
reactions.
47(4), 8-12.

52
Herr, P., Kardees, F., & Kim, J. (1991). Effects of Word-of-Mouth and Product-
Attribute Information on Persuasion: An Accessibility-Diagnosticity
Perspective. Journal of Consumer Research, 17, 454-462.
Hovland, C. I., Janis, I. K., & Harold H., K. (1953). Communication and Persuasion.
Yale University. New Haven: CT: Yale University Press.
Jalilvand, M.R. and Samiei, N. (2012), “The effect of electronic word of mouth on
brand image and purchase intention: an empirical study in the automobile
industry in Iran,” Marketing Intelligence & Planning, Vol. 30 No. 4, pp. 460-
476.
Jones, M., Mothersbaugh, D., & Beatty, S. (2000). Switching barriers and
repurchase intentions in services. Journal of Retailing, 76(2), 259-274.
Kahle, L., & Pamela, M. (1985, March 11). Physical Attractiveness of the Celebrity
Endorser: A Social Adaptation Perspective. Journal of Consumer Research,
954- 961.

Kamins, M. (1990). An Investigation into the "Match-Up" Hypothesis in Celebrity

Advertising: When Beauty May be Only Skin Deep. Journal of Advertising, 19(1), 4-
13.

Kamins, M. a. (1994). Congruence between spokespersons and product type: a match-up

hypothesis perspective. Journal of Psychology and Marketing, 11(6), 569-586.
Keel, A., & Nataraajan, R. (2012). Celebrity endorsements and beyond New avenues for
celebrity branding. Psychology & Marketing, 29(9), 690-703.
Keller, K.L., Parameswaran, M. and Jacob, I. (2011), Strategic Brand Management:
Building, Measuring, and Managing Brand Equity, Pearson, Chennai.

Koering, S., & Boyd, T. (2009). To Catch a Tiger or Let Him Go: The Match-up
Effect and Athlete Endorsers for Sport and Non-Sport Brands. 18(1), 15-37.

Kofi Osei-Frimpong, Georgina Donkor & Nana Owusu-Frimpong (2019) The Impact of
Celebrity Endorsement on Consumer Purchase Intention: An Emerging Market
Perspective, Journal of Marketing Theory and Practice, 27:1, 103-121

Kotler, P. (1997). Marketing Management: Analysis, Planning, Implementation, and

Control.
Nj: Englewood Cliffs.

Lee, J., Lee, J. and Feick, L. (2001), “The impact of switching costs on the customer
satisfaction-loyalty link: mobile phone service in France,” Journal of Services
Marketing, Vol. 15 No. 1, pp. 35-48.

Liu, T., Wang, C., & Wu, L. (2010). Moderators of the negativity effect: Commitment,
identification, and consumer sensitivity to corporate social performance. 27(1),

53
 54-70.

McCracken, G. (1989). Who is the celebrity endorser? Cultural foundation of the

endorsement process. Journal of Consumer Research, 16(3), 310-321.
McGuire, W. J. (1985). Attitudes and Attitude Change. In G. L. Aronson, Handbook of
SocialPsychology (Vol. 2, pp. 233-346). New York: Random House.
Mishra, S. A. (1990). Celebrity spokesperson and brand congruence. Journal of Business
Research, 21(2), 159-173.

Mizerski, R. (1982). An attributional explanation of the disproportionate

influence of unfavorable information. Journal of Consumer
Research, 9, 301-310.

Mookda, Rarina & Khan, Sanan & Intasuwan, Sirichanok & Chotchoung, Sirikanya. (2020).
The Effect of Celebrity Endorsement on Consumer Purchase Intention: The
Mediating Role of Brand Loyalty. International Journal of Scientific and Research
Publications (IJSRP). 10. 1016-1021. 10.29322/IJSRP.10.06.2020.p102123.

Mowen, J., Brown, S., & Schulman, M. (1979). Theoretical and empirical extensions of
endorser effectiveness. Marketing educators conference proceedings (pp. 258-
263). Chicago: American Marketing Association.
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.
Oliva, T.A., Olivier, R.L. and MacMillan, I.C. (1992), “A catastrophe model for
developing service satisfaction strategies,” Journal of Marketing, Vol. 56 No. 3,
pp. 83-95.
Pallant, J. (2016). SPSS survival manual: A step-by-step guide to data analysis using SPSS.
McGraw-Hill International.
Pan, Y., Sheng, S., & Xie, F. (2012). Antecedents of customer loyalty: An empirical
synthesis and re-examination. Journal of Retailing and Consumer Services, 19,
150- 158.

Pikas, B. S. (2012). Assessing the Qualities of Athlete Endorsers: A Study of Consumer

Preferences for the 3 Qualities of Sports Endorsers Attractiveness,
Trustworthiness, Expertise. Journal of Marketing, 6(3), 42-55.

Price, L. J. (1996). Understanding the Negativity Effect: The Role of Processing Focus.
7(1), 53-62. Richey, M., Koeings, R., Richey, H., & Fortin, R. (1975). Negative
Salience in-Impressions of Character: Effects of Unequal Proportions of Positive
and Negative Information. Journal of Social Psychology, 97, 233-241.
Reichelt, J., Sievert, J., & Jacob, F. (2014). How credibility affects eWOM reading: The

54
 influences of expertise, trustworthiness, and similarity on practical and social
functions. Journal of Marketing Communications, 20(1-2), 65-81
Sharma, N. and Patterson, P.G. (2000), “Switching costs, alternative attractiveness and
experience as moderators of relationship commitment in professional consumer
services,” International Journal of Service Industry Management, Vol. 11 No. 5,
pp. 470-90.

Shimp, T. and Andrews, J.C. (2013), Advertising Promotion and Other Aspects of
Integrated Marketing Communications, Cengage Learning, Boston, MA.

Sekran, U. a. (2009). A skill-building approach. In Sekran, Research Methods for

Business.
Willey: Fifth Edition.

Skowronski, J., & Donald, E. (1987). Social judgment and social memory: the role of
cue diagnosticity in negativity, positivity, and extremity biases. Journal of
Personality and Social Psychology, 52(4), 689-699.
Skowronski, J.J. and Carlston, D.E. (1989), “Negativity and extremity biases in
impression formation: a review of explanations,” Psychological Bulletin, Vol.
105 No. 1, pp. 131-142.

Till, B., & Busler, M. (2000). The Match-Up Hypothesis: Physical Attractiveness,
Expertise, and the Role of Fit on Brand Attitude, Purchase Intent and Brand
Beliefs. Journal of Advertising, 29(3), 1-13.
Ullrich, S. and Brunner, C.B. (2015), “Negative online consumer reviews: effects of
different responses”, Journal of Product & Brand Management, Vol. 24No. 1,
pp. 66-77.
V. Esposito Vinzi, W.W.Chin, J. Henseler, and H. Wang. 2010. Handbook of Partial Least
Squares: Concepts, Methods and Applications. Springer.

Ward, J., &Ostorm, A. (2006). Complaining to the Masses: The Role of Protest
Framing in Customer-Created Complaint Web Sites. Journal of Consumer
Research, 33, 220- 230.

Wang, F., Zhang, X.P.S. and Ouyang, M. (2009), “Does advertising create sustained
firm value? The capitalization of brand intangible”, Journal of the Academy of
Marketing Science, Vol. 37 No. 2, pp. 130-143.
Wathne, K., Biong, H. and Heide, J.B. (2001), “Choice of the supplier in embedded
markets: relationship and marketing program effects,” Journal of Marketing,
Vol. 65, April, pp. 54-66.
Weingberger, M., & Dillion, W. (1980). The Effects of Unfavorable Product
Rating Information. 7, 528-532.

55
Xie, Y., &Peng, S. (2009). How to repair customer trust after negative publicity: The
roles of competence, integrity, benevolence, and forgiveness. Psychology and
Marketing,26(7), 572-589.

Yanamandram, V., & White, L. (2006). Switching barriers in business-to-business

services: a qualitative study. International Journal of Service Industry
Management, 17(2), 158- 192.
Yu, M., Liu, F., Lee, J. and Soutar, G. (2018), "The influence of negative publicity on brand
equity: attribution, image, attitude, and purchase intention," Journal of Product &
Brand Management, Vol. 27 No. 4, pp. 440-451.

56
 QUESTIONNAIRE

Section - I
Age No of Years
Region Peshawar
Gender 0 for F and 1 for M
Section - II
Strongly Disagree Neutral Agree Strongly
Disagree Agree
Negative Publicity 1 2 3 4 5
I believe that negative publicity of the brand will effect
buying decision
I believe that negative publicity of the brand will effect
spreading good word of mouth about the brand
I believe that negative publicity of the brand will effect
reputation of the brand
I believe that negative publicity of the brand will effect
consumer referrals
I feel more attached to a brand which I am using it
I prefer to buy a product because it has good reputation of a
brand

Switching Cost
It is risky to change my brand to a new brand
I would feel frustrated if I terminate my relationship with my
brand
Do you switch to other brand due to increase in price
I prefer to switch a brand because it has bad reputation
Do you switch to other brand due to negative publicity
Would you switch from your regular product to a new which
is affected by negative event or incident?

Celebrity Endorsement
Do you think that ads having celebrities are more effective
than those which don’t?
Do you think ads having real life couples are more effective
than those which don’t?
I buy a product because celebrity associated with the
product is trustworthy person to represent the product
I buy a product because celebrity associated with the
product is reliable person
I buy a product because celebrity is expert to represent the
product in advertisement
I buy a product because celebrity associated with the
product is attractive

Customer Loyalty
Do you recommend your brand to your friends/colleagues
Product support installation and user guidence is demanding
factor of After Sale Services which leads in Brand satisfaction
Product technical support in warranty claims is a sensitive
element and it is highly valued
Do you prefer a Brand over others due to their great quality
of after sales services

57

View publication stats