Impact of Celebrity Endorsement on Consumer Buying Behaviour in the State of Goa

Abstract

Celebrities have been involved in endorsing various activities since the eighteenth century. This article focuses on five aspects: to examine the present status of literature available on the effect of celebrity endorsement on consumer buying behaviour; to identify whether there is any association between the three groups of consumers (supporters/neutral/opposers of celebrity endorsement); applying multi-logit analysis to identify the probability of a consumer in becoming a supporter/neutral/opposer of the celebrity endorsement based on demographic profiles of consumers; to examine the celebrity attributes likely to influence consumer purchase intentions; and, finally, to examine the perception of the Goan consumers about celebrity endorsements. Literature review revealed that no major studies have been carried out on the impact of celebrity endorsement on Goan consumers’ buying behaviour. Hence, a survey using structured questionnaire was administered among 200 consumers during June 2015 to April 2016, with a response rate of 86 per cent (received responses were 171). It was found that a significant association exists between celebrity supporters/neutral/opposers with respect to age, gender, occupation and income level, except in case of education and location. Consumers also find celebrity endorsements more attractive and influential as compared to non-celebrity endorsements. Moreover, the tested attributes show positive relationship with purchase intention, except in case of expertise and trustworthiness of celebrity. In general, it can be concluded that celebrity attributes do impact the purchase intention of consumers.

Keywords

Brand ambassador celebrity endorsement consumer buying behaviour multi-logit analysis Goa

Introduction

Celebrities are involved in endorsing various activities since the eighteenth century (Roy, 2006).¹ With each day passing, the number of celebrities involved in endorsing products and services also increases. Perhaps it has become one of the most important tools of advertising. Celebrities are considered to bring attention towards a particular advertisement due to their likeability and attractiveness (Erdogan, Baker, & Tagg, 2001). Marketers use celebrities to influence the purchase decision of consumers in order to increase sales and expand market share (Kumar, 2010). In the words of Silvera and Austad (2004), celebrities are those people who enjoy public recognition among a large group of people and possess distinctive qualities like attractiveness and trustworthiness. Celebrity endorsement business is a billion-dollar industry today (Kambitsis, Harahousou, Theodorakis, & Chatzibeis, 2002). According to Dash and Sabat (2012), India is the only place where celebrities are not only loved but also worshipped. The top 10 celebrity endorsers in India, who comprise of film and sports personalities (see Table 1), indicates that they endorse almost all types of products (durable as well as non-durable). The service industry is also not an exception to celebrity endorsement. Celebrity is a person involved with film industry, television, sports, politics, business, or who is an artist or a person from the military (McCracken, 1989). In this modern age of marketing, celebrity may also be an animated character like Fred Flintstone or an animal (Miciak & Shanklin, 1994). Celebrity endorser is an individual who is known by the public for his or her achievements in areas other than that of the product class endorsed (H. H. Friedman & L. Friedman, 1979; Friedman, Termini, & Washington, 1976) and who uses this public recognition on the behalf of a consumer good by appearing with it in an advertisement (McCracken, 1989).

Literature Review

Background

The present study analysed around 262 research papers published during the period 1959–2015, collected through a detailed search carried out using various sources such as Emerald Full Text, Elsevier, JSTOR, Springer, Taylor & Francis, Oxford University Press, SSRN and Google Scholar. The literature review shows that around 66 per cent of the research work was carried out during the years 2011–2015. Majority of the papers were published in open source journals (39%) followed by Taylor & Francis (17%), JSTOR (17%) and Elsevier (12%). It was found that most of the studies were carried out in North America (45%), followed by Europe (23%) and Asia (21%), but very few on India and especially no detailed study on Goa. Thus, there exists a gap and the present study tries to fill it by adding valuable knowledge and new perspectives, thereby presents possibilities for further research. This article offers inputs for different stakeholders, especially academic institutions, marketers, marketing agencies, academicians, government and NGOs working in the region under study.

Table 1.

Top 10 Celebrity Endorsers

Big Endorser	Brand
Shah Rukh Khan	Airtel, Videocon, Emami, Hyundai, Pepsodent, Dish TV, Tag Heuer, Pepsi, Santro, LML, Tata Tea, Nokia, Mayur Suitings, ICICI Bank, Belmonte, Navratna Oil, Sona Chandi Chyawanprash, Smoodles, Cinthol soap, Bagpiper club soda, Clinic All Clear, Sunfeast, Nerolac Paints, Genus Powder, Colgate, Linc Pens, Frooti, Sprite, Gitanjali Jewellers
Salman Khan	ThumsUp, Revital, Wheel, Suzuki Motorcycles, Yatra.com, Dixcy Scott, Splash, Relaxo, Limca, Chlormint, Graviera Suiting, Red Tape, Britannia, Astral Pipes, Maya Jewels, Rotomac, Mountain Dew, Force Motors, Mauka – For the Pro Kabbadi League
Amitabh Bachchan	Parker Pens, Dabur, Dairy Milk, Boroplus, ICICI, Kalyan Jewellers, Gujarat Tourism, Navratna Oil, Maggie, Pepsi, Nerolac, Reid & Taylor, Polio awareness, Maruti Suzuki, Zen Mobile, Eco Friendly Cleaning Product, Real Estate, Max New York Life Insurance, Tanishq’s Latest TVC, Luxor Nano Clean, Sunglasses, Cookies, JustDial
Mahendra Singh Dhoni	Aircel, Pepsi, Boost, Reebok Bat, Sonata Watches, TVS, Amrapali Group, Ashok Layland, Dabur Chyawanprash, Exide Battery, Gulf Oil, Lafarge Customer Service, Lays Wafers, Maxx Mobile, McDowell’s No 1 Soda, Siyarams Suits, Sony Bravia
Aamir Khan	Samsung Mobiles, Tata Sky, Coca-Cola, Titan Watches, Snapdeal, Titan, Etisalat Mobile, Godrej, Incredible India, Mahindra Bikes, Ministry of Women’s Development, UNICEF
Akshay Kumar	Relaxo, Dollar Club, Micromax Mobile, Eveready, Manappuram Gold Loan, Honda India, Sugar Free, Rasna, Canadian Tourism Commission, Bagpiper Whisky, Honda Dream Yuga, Sure DEO, Levis Jeans, LG Electronics, Sparx Footwear, ThumsUp
Virat Kholi	Adidas, Boost, Clear Shampoo, Fair & Lovely, Nike, Fastrack, TVS, Toyota, Border Security Force, Celkon Mobiles, Cinthol, Herbalife Products, Munch, Red Chief Shoes, Sangam Clothing, Pepsi, Flying Machine Ad
Sachin Tendulkar	Britannia Biscuits, Pepsi, Adidas, TVS, Castrol, MRF Tires, ESPN, Aviva Insurance, Sunfeast, Canon Camera, Jaypee Group Cement, Coca-Cola, Luminous Power Back Up Solutions, UAE based travel portal Musafir.com, Audemars Piguet Watch, Amit Enterprises in Grand Haytt, Kaspersky Internet Security, Boost, VISA, Philips, Sanyo BPL
Deepika Padukone	Coca-Cola, Axis Bank, Kellogg’s, Garnier, Vogue, Tissot, Van Heusen, Lifestyle Melange, Parachute, HP, Nescafe, Lux, Fiama Di Wills, YAMAHA, Kingfisher (Kingfisher Calendar Girl), ORBIT, Sony Cybershot, Neutorgena Fine Fairness, Garnier
Hrithik Roshan	Acer Laptops, Coca-Cola, Hero Honda Bikes, Iball Silde, Intel, Mountain Dew, Myntra, Liberty Shoes, RADO Watches, Reliance Mobile, J Hampstead Photoshoot, Joyalukkas Jeweller’s, Sony Ericsson Mobiles

Sources: Top 10 celebrity endorsement (2015), Forbesindia.com and authors own compilation.

Most of the research, as a general practice, used primary data collected through structured questionnaires with 5-point Likert scale, which were administered on a sample size of 100–300 consumers selected through convenience sampling method. The most commonly used data analysis techniques were descriptive statistics, correlation and regression analysis, ANOVA and factor analysis for data reduction. This study also uses similar sample size and techniques for data analysis, but applies multi-logit analysis, which may help to predict the likelihood of the consumers becoming a supporter/neutral/opposer towards celebrity endorsement, based on the demographic characteristics of consumers.

Celebrity Endorsement and Consumer Buying Behaviour

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 behaviour that leads to the final purchasing decision. This is because consumer behaviour 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 of 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 around the world have tried to assess the reason for the change in consumer behaviour due to celebrity endorsement and 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 of the celebrity attributes are responsible for the change in consumer behaviour, leading to purchase intention. Many studies in India applied these models (either individually or in combination) and found that, in general, there are seven celebrity attributes responsible for the change in purchase intention. These are: credibility of the celebrity (trust-worthiness 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, 2009). Studies reveal that some of the celebrity attributes, such as expertise, trustworthiness (Freeman, 2015; Gupta, Kishore, & Verma, 2015) and product match-up (Roy, 2006) do not influence consumer behaviour.

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). In order 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 product recall. Chan, Ng and Luk (2013) found that using a celebrity in an ad increase the 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. Demographic profile also provides some insights to 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. Serious efforts are needed for profiling consumers before going for celebrity endorsement. This leads to the development of the first research question (RQ 1) and the first hypothesis (H₁):

RQ 1a: Is it possible to assess the effect of demographic variables displayed by consumer categories (supporters/neutral/opposers) on the issue of celebrity endorsement?

RQ 1b: Is it possible to predict the effect of demographic variables displayed by consumer categories (supporters/neutral/opposers) on the issue of celebrity endorsement?

H₁: No significant difference exists between attitudes towards celebrity endorsement (supporters/neutral/ opposers) with respect to age, gender, education, occupation, location and income level.

Influence of Celebrity Endorsement on Purchase Intention

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.

Source credibility, in a broad sense, refers to a communicator’s positive characteristics that affect the receiver’s acceptance of a message (Ohanian, 1990), and is based on the research in social psychology by Hovland, Janis, & Kelley (1953). Trustworthiness refers to the honesty, integrity and believability of an endorser (Erdogan et al., 2001). Consumers generally have the impression that celebrities are a trustworthy source of information (Goldsmith, Lafferty, & Newell, 2000). Marketers take the advantage of this impression by employing celebrities who are thus regarded as trustworthy, honest, believable and dependable among their fans and even among 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). In order 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 have a higher chance of convincing the consumers than an endorser with less expertise (Ohanian, 1990).

Source attractiveness model, as a component of the source valance model (McGuire, 1985), is mainly based on four dimensions: familiarity (knowledge about the source through exposure), likeability (affection for the source based on the physical appearance or behaviour), 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, likeable and/or similar sources are seen as more attractive and persuasive (Ohanian, 1990). Such people evoke positive stereotypes and, as a result, physically attractive 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 the effectiveness of an advertisement depends on the existence of 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 common attributes between product features and the celebrity image (Misra & Beatty, 1990; Charbonneau & Garland, 2010). The model suggests that attractive celebrities, especially attractive female celebrities, are more effective at endorsing beauty products, which are believed to enhance one’s attractiveness (Kamins, 1990).

Meaning transfer is another critical attribute and McCracken (1989) explains that a celebrity would be effective only if 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 which are seen by the consumers as prominent, it leads to actual purchasing decision. The present study tries to identify to what extent consumers consider various attributes as significant when it comes to purchase intention. Thus, the following research questions and hypothesis are formulated:

RQ 2: Is it possible to identify the effect of celebrity attributes towards consumer purchase intention, and if so, which are the most influencing attribute(s)?

H₂: There is no significant impact of celebrity attributes (expertise, trustworthiness, similarity, familiarity, likeability, match-up and meaning transfer) on purchase intention.

RQ 3: Is it possible to examine and identify the perception of consumers towards a celebrity endorsing a product or a service?

Research Methodology

Data were collected from June 2015 to April 2016 by using a structured questionnaire. The survey was based on convenience sample of about 200 respondents in and around the state of Goa, of which 171 responded (response rate of 86%). The survey questionnaire consisted of three parts. Based on earlier studies (Renton, 2006, 2009), the questionnaire was developed after making some modifications to suit the local market conditions as well as the consumer profile based on discussions with academicians, industry experts.

A pilot survey was also undertaken before the actual survey. The pilot survey consisted of three individuals from academics (faculty members, research scholars and masters-level students) and industry (senior managers as well as junior-level managers), which was mainly carried out for ensuring whether the revised questionnaire was clear and conveyed the meaning to the consumers, so that they may be able to respond correctly. All the 15 members responded, stating that the questions asked were relevant and clear with respect to the objective of the survey.

Part A of the questionnaire was aimed at assessing the basic demographic information of the respondents: age (up to 18/18–30/30–50/above 50); gender (male/female); education (up to 10/10–12/graduation/postgraduation); occupation (student/business/housewife/employed); location (north Goa/south Goa) and income level (less than 20,000/20–40,000/40–60,000/above 60,000). Three more aspects were also included in this part: first, respondents’ attitude towards celebrity endorsement, categorizing them into supporters/neutral/opposers; second, understanding their level of knowledge about the type of celebrity (film or sports); and third, the type of product that is preferred because of celebrity endorsements. Respondents were informed to identify themselves as supporters/neutral/opposers, based on their past experience during their purchase decisions. If they felt that celebrity endorsement was acceptable, they might identify themselves as a supporter. If celebrity endorsement was unacceptable to them, they could identify themselves as opposers. If they did not have any opinion, they could identify themselves as neutral. They were asked to name the celebrities, recollecting from the advertisements they have seen earlier, and this information was used to create the list of celebrities (Table 1). They were also asked to identify the product category (durable and non-durable) that gets preference because of celebrity endorsement when it comes to purchasing.

Part B was aimed at understanding the seven celebrity attributes that influence the buying behaviour of consumers, along with their purchase intentions. The scale used for celebrity attributes in earlier studies was either a 7-point semantic differential scale (Renton, 2006) or a 5-point Likert scale (Renton, 2009). For the present study, a 5-point Likert’s scale was used to assess all the seven celebrity attributes. This also helped in bringing uniformity and ease of understanding among consumers in Goa. Similarly, with respect to purchase intentions, earlier studies used a 7-point Likert’s scale (Renton, 2006), but again a 5-point scale was used for the present study. Participants were asked to rate the extent to which they agreed or disagreed with the seven statements related to the seven celebrity attributes and also their purchase intention on a 5-point Likert’s scale (1 for strongly disagree and 5 for strongly agree). The reliability of the statements connecting the seven celebrity attributes showed a Cronbach’s alpha (α) of 0.770, while the three statements of purchase intentions showed only 0.538. However, when the seven attribute statements and three purchase intention statements were tested together, the result showed 0.797, which is well within the acceptable limit and is considered good (Diekhoff, 1992; Nunally, 1978).

Part C relates to the consumers’ perception towards celebrity endorsement, where the respondents are asked to express their perception with respect to the medium of advertisements used (multiple response were allowed), frequency of watching the advertisements, most attractive advertisements, recall capability based on advertisements, usage of the products by the endorsing celebrities themselves and the level of influence of celebrity endorsement. This section also includes a part where consumers are asked to identify whether they will be influenced by any negative publicity of the celebrity while making purchase decisions.

Based on the existing literature review on the impact of celebrity endorsements on consumer buying behaviour, the present study identified a few research questions. Appropriate hypotheses were also formulated. Statistical techniques used in similar studies elsewhere (Renton, 2006, 2009; Subhash et al., 2015, 2018[forthcoming]) were used. Chi-square test was applied to find an answer to RQ 1, where association between the three categories of consumers, namely supporters, neutral and opposers, and their demographic characteristics were analysed.

Demographic characteristics related with consumers attitude could be a useful tool for marketers. To achieve this, a multi-logit model was run. If any demographic characteristic showed a significant difference between the clusters, logit analysis permits to derive predictive parameters on the significant variables. In this multi-logit analysis, the variable of three clusters (CL_3i; supporters/neutral/opposers) was treated as the dependent variable and demographic characteristics as independent variables; multi-logit regression was applied to determine the factor that explained the pertinence of a concrete type of cluster. In multinomial logit notation, the model was written as

Multinomial Logit CL_iα + βj xk

where CL_3i is the odds of occurrence on cluster i over the other clusters 2; α = the intercept parameter; βj = the vector of slope parameter; and xk = the explanatory demographic variables (Subhash et al., 2015, 2018[forthcoming]).

Multiple correlation and regression analysis was applied to find the answer to RQ 2, which tried to identify the most and the least important among the seven celebrity attributes and purchase intention. The model used was:

PIβ₀+ β₁(EX) + β₂(TW) + β₃(SM) + β₄(FM) + β₅(LK) + β₆(MU) + β₇(MT) + ε

where PI is purchase intention, EX is expertise, TW is trustworthiness, SM is similarity, FM is familiarity, LK is likeability, MU is match-up and MT is meaning transfer. The result would provide required information on the most and the least influencing celebrity attributes with respect to purchase intention. And finally, using cross tabulation, taking age group as a common denominator, consumer’s perception towards celerity endorsement was analysed to find the answers to RQ 3 with respect to the perception of consumers towards a celebrity endorsing a product or service.

Analysis and Discussion

Impact of Celebrity Endorsement on Demographic Profile

This section is in three parts. First, we applied chi-square to see whether there was any association between the attitude of consumers (supporters, neutral, opposers) and their demographic characteristics. Second, we wanted to see the possibility of predictability of the attitude of consumer becoming a supporter, neutral or opposer towards celebrity endorsement. Third, using basic tabulation, we tried to ascertain the type of celebrity, the most prominent celebrity and the category of product influenced by celebrity endorsements.

Demographic Profiling

The demographic profiling of the respondents (see Table 2) provided the break-up of consumers based on their attitude towards celebrity endorsement, that is, being a supporter, neutral or opposer. With respect to age, gender, occupation and income level, there exists a significant association between the consumer being a supporter, a neutral, or an opposer. Education and location are the two variables that showed no significant association with the attitude towards celebrity endorsement. Majority of the consumers under study (53%) were in the age group of 18–30 (62% among supporters, 43% among neutral, 61% among opposers). With respect to gender, women (68%) were more (73% among supporters, 61% among neutral, 74% among opposers). With respect to occupation, majority of the consumers (53%) were students (56% among supporters, 57% among neutral, 30% among opposers). Finally, income level showed that the majority (57%) were in the low-income group (58% among supporters, 62% among neutral, 34% among opposers). The education status shows that graduates are more, that is, 37 per cent among supporters, 35 per cent among neutral, and 44 per cent among opposers.. Geographical concentration showed that 54 per cent are from North Goa (49% among supporters, 55% among neutral and 65% among opposers).

With respect to age (χ² = 13.198, p = 0.040), customers falling under the age group of less than 30 years are found to be supporters to celebrity endorsement than those under the age group of above 30. Genderwise (χ² = 7.498, p = 0.024), female consumers tend to be more supportive than male consumers. Students (χ² = 16.387, p = 0.012) show more supportive attitude towards celebrity endorsement than other categories of occupation. The low income group consumers (χ² = 17.068, p = 0.009) are inclined towards becoming supporters of celebrity endorsing than medium or high-income consumers. The two demographic characteristics which do not show any influence on consumers attitude towards celebrity endorsement are education (χ² = 3.559, p = 0.736) and location (χ² = 1.803, p > 0.05).

It can be seen from the result that except for education and location, all other demographic variables are significant, that is, irrespective of their education level and the location, youngsters till 30 years of age, females, students and employed and middle-class consumers are more inclined towards celebrity endorsement. Based on the observations made here, the formulated hypothesis (H₁) of RQ1a is rejected with respect to age, gender, occupation and income level, with an exception of education and location.

Marginal Coefficient

Marginal effects where performed to further investigate the effects of the demographic variables over the dependent variable for each group significantly different between the clusters, that is, the attitude towards celebrity endorsement (supporters/neutral/opposers). A multi-logt model was applied to understand the change in cluster membership with a variation in demographic factor. In this multi-logit analysis, the three clusters, that is, supporter, neutral and oppose, were taken as dependent variable, demographic characteristics being the independent variable.

As shown by the results in Table 3, the marginal effect for variables age, gender and income level is significant at 95 per cent level of confidence for the group of opposers. This means that if a consumer belongs to the age group 18–30, the probability of his/her being an opposer of celebrity endorsement is increased by 26.6 per cent as compared to those belonging to the age group of up to 18 years. Similarly, the probability of a male consumer being an opposer of celebrity endorsement is decreased by 11.2 per cent as compared to a female consumer. With respect to a consumer belonging in the income group of above ₹60,000, the probability of his/her being an opposer of celebrity endorsement decreases by 15.66 per cent as compared to a consumer belonging to the income group of less than ₹20,000.

Table 2.

Demographic Profiling of the Respondents (N = 171)

Demographic Characteristics/Variables		Attitude towards Celebrity Endorsement						Total (%)	χ2 (p)
		Supporters		Neutral		Opposers
		#	%	#	%	#	%
Age	Up to 18	13	18	22	29	0	0	20	13.198 (0.040)*
	18–30	44	62	33	43	14	61	53
	30–50	12	17	18	23	7	30	22
	Above 50	2	3	4	5	2	9	5
Gender	Female	52	73	47	61	17	74	68	7.498(0.024)*
	Male	19	27	30	39	6	26	32
	Up to 10	6	8	8	10	1	4	9	3.559(0.736)
	10–12	16	23	24	30	7	30	27
Education	Graduation	26	37	27	35	10	44	37
	Post Graduation	23	32	18	23	5	22	27
Occupation	Student	40	56	44	57	7	30	53	16.387(0.012)*
	Business	2	3	6	8	5	22	8
	Housewife	5	7	5	6	5	22	9
	Employed	24	34	22	29	6	26	30
Location	North Goa	35	49	42	55	15	65	54	1.803(0.406)
	South Goa	36	51	35	45	8	35	46
Income Level	20,000 and less	41	58	48	62	8	34	57	17.068(0.009)*
	20,000–40,000	23	32	18	23	5	22	27
	40,000–60,000	5	7	6	8	5	22	9
	60,000 and more	2	3	5	7	5	22	7
Total		71		77		23

Source: Primary data.

Notes: # Number of Consumers; * Significant at 0.05.

With respect to the occupation of the consumers, probability of a businessperson being a supporter of celebrity endorsement decreases by 33 per cent as compared to students; whereas, the probability of a businessperson being an opposer of celebrity endorsement increases by 55.8 per cent as compared to students.

Though this is a conflicting result, it still can be taken as a decisive criterion while targeting those consumers who fall under business category, as the probability of their becoming opposer seems to be more than becoming a supporter. This is just an indication and the results needs to be tested among other groups of respondents to assess the validity. However, there is a greater probability of being an opposer than being a supporter or neutral, indicating a general perception that whatever celebrities are endorsing is not entirely true and fair of what they say about the product they vouch for.

Most Influential/Favourite Celebrities and Types of Products Endorsed

Respondents were asked about the type of celebrity (film or sports) who motivates them to purchase the product, who are the most favourites celebrities and which category of product is most influenced by celebrity endorsement. First, the type of celebrity was analysed and the result (see Figure 1) shows that 54 per cent of female consumers consider film stars to be more motivating than sports personalities (11%) in terms of purchase of products. But for male consumers, a sports personality has greater influence (20%) than film stars (15%) when it comes to purchase. The result indicates a significant difference (α² = 31.177, p = 0.000*) between male and female consumers when it comes to the type of celebrity. The overall result reveals that 69 per cent of both female and male respondents consider that film stars influence them more to purchase the product than sports personalities (31%).

Table 3.

Marginal Coefficients

Demographic Characteristics		Attitude towards Celebrity Endorsement
Demographic Characteristics		Supporters	Neutral	Opposers
Age	Up to 18	BC	BC	BC
	18–30	–0.052	–0.213	0.266*
	30–50	–0.167	0.093	0.74
	Above 50	–0.191	–0.0119	0.203
Gender	Female	BC	BC	BC
	Male	0.080	0.031	–0.112*
	Up to 10	BC	BC	BC
	10–12	–0.100	–0.0566	0.157
Education	Graduation	–0.020	0.752	–0.054
	Post Graduation	0.049	0.054	–0.103
	Student	BC	BC	BC
Occupation	Business	–0.33*	–0.222	0.558*
	Housewife	0.035	–0.223	0.188
	Employed	0.086	–0.132	0.046
Location	North Goa	BC	BC	BC
	South Goa	0.193	–0.118	–0.075
	Less than 20,000	BC	BC	BC
Income Level	20,000–40,000	–0.035	0.029	0.006
	40,000–60,000	0.22	–0.188	–0.035
	Above 60,000	0.21	–0.057	–0.1566*

Source: Primary data.

Notes: Marginal Coefficients, *p <0.05; BC = Base Category.

Male Celebrities	#	Female Celebrities	#
Salman Khan	40	Deepika Padukone	30
M S Dhoni	31	Kareena Kapoor	21
Sachin Tendulkar	25	Saina Nehwal	18
Virat Kholi	24	Shraddha Kapoor	17
Shahrukh Khan	14	Kajol	15
Akshay Kumar	11	Aishwarya Rai	13
Ranbir Kapoor	10	Sania Mirza	11
Amitabh Bachchan	9	Katrina Kaif	8
Aamir Khan	8	Madhuri Dixit	5
Varun Dhawan	3
John Abharam	3
Hrithik Roshan	2
Shahid Kapoor	2

Variables	Mean	Rank	S.D.	Rank
( PI ) Purchase Intention	3.19		1.29
( EX ) Expertise	2.31	7	1.002	2
( TW ) Trustworthiness	2.72	6	1.007	3
( SM ) Similarity	2.96	5	1.134	5
( FM ) Familiarity	3.05	4	1.159	6
( LK ) Likeability	3.33	1	1.063	4
( MU ) Match Up	3.17	3	0.958	1
( MT ) Meaning Transfer	3.27	2	1.192	7
Grand Mean	2.97

Variables	Correlation	Adjusted R²	β	t	sig
( EX ) Expertise	.133	0.012	0.013	.155	0.877
( TW ) Trustworthiness	.110	0.006	–0.116	–1.387	0.167
( SM ) Similarity	.417**	0.169	0.179	2.214	0.028**
( FM ) Familiarity	.568**	0.319	0.210	2.182	0.031**
( LK ) Likeability	.483**	0.229	0.242	2.607	0.010**
( MU ) Matchup	.437**	0.186	0.236	2.632	0.009**
( MT ) Meaning Transfer	.592**	0.347	0.240	2.525	0.013**

#	Particulars [N = 171] *
1	Medium of advertisement more likely to convey celebrity endorsed advertisement
	TV advertisements	26	91	33	6	156 (61)
	Magazines	10	14	11	0	35 (14)
	Internet	6	21	6	1	34 (13)
	Radio	0	3	5	0	8 (3)
	Newspapers	3	11	5	2	21 (8)
2	Frequency of watching celebrity endorsed advertisements
	Once a day	12	27	15	2	56 (33)
	2 to 7 times a day	15	43	15	4	77 (45)
	More than 7 times/day	6	23	7	2	38 (22)
3	Type of advertisement which attracts more
	Celebrity endorsed	32	77	33	7	149 (87)
	Non celebrity endorsed	1	16	4	1	22 (13)
4	Presence of celebrity endorsed advertisement helps recognizing/recalling products
	Strongly disagree	3	8	4	0	15 (9)
	Disagree	8	8	6	1	23 (13)
	Neutral	8	21	9	3	41 (24)
	Agree	10	32	11	3	56 (33)
	Strongly agree	4	24	7	1	36 (21)
5	The products endorsed by the celebrities are actually being used by the celebrities
	Strongly disagree	16	27	9	1	53 (31)
	Disagree	1	16	4	2	23 (13)
	Neutral	9	19	9	3	40 (23)
	Agree	7	20	9	2	38 (22)
	Strongly agree	2	9	6	0	17 (11)
6	Effectiveness/influence of celebrity endorsement when it comes to purchasing
	Very much influential	2	13	10	2	27 (16)
	Influential	11	35	12	5	63 (37)
	Somewhat influential	12	20	7	0	39 (23)
	Not at all influential	8	25	8	1	42 (24)
7	Negative publicity of the celebrity endorser influences purchasing decision
	Yes	9	21	5	3	38 (22)
	No	18	40	16	3	77 (45)
	Sometimes	6	32	16	2	56 (33)

Figure 1.
Source: Primary data.

Figure 2.
Source: Primary data.

In the second aspect, consumers were asked about the category of product (durable or non-durable) they think is most influenced by celebrity endorsement. The cross tabulated analysis on the basis of age group and type of product (see Figure 2) shows that people of all age groups consider durable products (70%) endorsed by celebrity to be highly influential, which is statistically significant (0.0384 < 0.05). Non-durable products, on the other hand, are not considered to be highly influential, probably because such products are used by consumers for a shorter duration. Most of the celebrities identified by the consumers were endorsing durable products.

Third, when the consumers were asked to recollect the names of the celebrities from film and sports fields who influenced them in purchasing a product or service, they identified (see Table 4) more male celebrities (13 names) than female celebrities (9 names). Of the total names recollected (22), only four names were from the sports field and 19 were from the films, indicating the bigger influence of film personalities in the lives of common people. It was found that most of the respondents considered Salman Khan as their favourite celebrity from Bollywood, while from the sports field, they liked M. S. Dhoni most. It could be because of their lifestyle, coolness, and so on. It is evident that respondents do not like young celebrities as much as they like Salman Khan and MS Dhoni. In case of female celebrities, it was found that most of the respondents considered Deepika Padukone as their favourite celebrity from Bollywood and from sports they like Saina Nehwal as well as Sania Mirza. It seems that Deepika Padukone is a more popular actress than others.

Table 4.
Male and Female Celebrities

Male Celebrities # Female Celebrities #

Salman Khan 40 Deepika Padukone 30

M S Dhoni 31 Kareena Kapoor 21

Sachin Tendulkar 25 Saina Nehwal 18

Virat Kholi 24 Shraddha Kapoor 17

Shahrukh Khan 14 Kajol 15

Akshay Kumar 11 Aishwarya Rai 13

Ranbir Kapoor 10 Sania Mirza 11

Amitabh Bachchan 9 Katrina Kaif 8

Aamir Khan 8 Madhuri Dixit 5

Varun Dhawan 3

John Abharam 3

Hrithik Roshan 2

Shahid Kapoor 2

Source: Primary data.

Note: # = number of consumers recollected the names of the particular celebrity.

Factors Affecting Purchase Intention

To understand and assess the impact of the seven celebrity endorsement attributes, that is, trustworthiness, expertise, similarity, familiarity, likeability, product match-up and meaning transfer on consumer purchase intentions, statistical measures like descriptive statistics and correlation and regression analysis are used. The result was divided into three parts: first, using descriptive statistics, ranking was carried out to see which of the attributes was more significant for consumers’ purchase intentions. In the second part, correlation matrix was used to understand the strength of relationship between the dependent variable purchase intention and the seven celebrity endorsement attributes. In the third part, multiple regression analysis was carried out to assess the most and the least influential celebrity attributes.

Descriptive Statistics

Descriptive statistics result (see Table 5) was used to find out which of the seven celebrity attributes highly influenced the purchase intention of the consumers. The mean value was high (3.33) in case of likeability among consumers, indicating that the level of the celebrity’s popularity among consumers is very high, which in turn can influence the purchase decision of the consumers. The second highest attribute was meaning transfer, having a mean value of 3.27, indicating that celebrities transfer meaning to a certain product with the help of their lifestyle. This was followed by match between the celebrity and the product and familiarity, with mean values of 3.17 and 3.05, respectively. The remaining three attributes, that is, similarity, trustworthiness and expertise, had values below the grand mean (2.97) value. Similarity and trustworthiness were at 2.96 and 2.72, respectively. Expertise is considered to be the least important attribute as its mean value was lower than other attributes, at only 2.31, which indicates that the level of knowledge, skill or expertise of any celebrity identified by the consumers based on their experience with respect to any product or service endorsed is normally not so high and consumers are least bothered about these attributes when it comes to celebrity endorsement and purchase intentions.

Table 5.
Descriptive Statistics of Celebrity Attributes

Variables Mean Rank S.D. Rank

( PI ) Purchase Intention 3.19 1.29

( EX ) Expertise 2.31 7 1.002 2

( TW ) Trustworthiness 2.72 6 1.007 3

( SM ) Similarity 2.96 5 1.134 5

( FM ) Familiarity 3.05 4 1.159 6

( LK ) Likeability 3.33 1 1.063 4

( MU ) Match Up 3.17 3 0.958 1

( MT ) Meaning Transfer 3.27 2 1.192 7

Grand Mean 2.97

Source: Primary data.

Impact of Celebrity Attributes on Purchase Intention

In this section, correlation and regression analysis was applied to find out the degree of relation as well as to assess whether there is any significant impact of the seven celebrity attributes on purchase intention. The summary result of regression analysis (see Table 6) showed r² 0.684 with an adjusted r² 0.44, which suggest that only 44 per cent of variations with respect to purchase intention are explained by the seven celebrity attributes (independent variables) and 56 per cent of variations are unexplained, which means that there must be other variables that may influence the purchase intention of consumers. It also indicates the moderate reliability of the data and the scope for future research. There can be various other dimensions of the attributes of celebrity endorsement. A study about the impact of celebrity endorsement on a different set of sample consumers could as well be taken up.

Results also show that among the seven celebrity attributes, only five (similarity, 0.417**; familiarity, 0.568**; likeability, 0.483**; match-up, 0.437**; and meaning transfer, 0.592**) have a significant correlation (moderate to average degree of positive correlation) when it comes to purchase intention, which means that expertise (0.133) and trustworthiness (0.110) do not have any significant relation (very low degree of positive correlation). Hence, these are considered as the least important attributes. The result is similar to that of the regression coefficients.

The results of multiple regression indicate the impact of the attributes on purchase intension, which allows comparison between the seven attributes and identify the most and the least influencing attributes. The regression results (adjusted r²) reveal that two attributes have low variability explanation, that is, only 1.2 per cent of purchase intention variability is explained by the attribute of expertise, whereas only 0.6 per cent of purchase intention variability is explained in case of trustworthiness. These two statistically insignificant celebrity attributes (expertise 0.877 > 0.05 and trustworthiness 0.167 > 0.05) lead to the conclusion that consumers are fully aware that the celebrities do not have expertise on the product they endorse and they do not believe in what the celebrities say about the product when it comes to purchasing a product.

However, the purchase intention variability is explained more in the case of the remaining five attributes than expertise and trustworthiness. The result is as follows: meaning transfer (34.7%), familiarity (31.9%), likeability (22.9%), match-up (18.6%) and similarity (16.9%). The highest value of beta coefficient is for likeability (0.010 < 0.05), followed by meaning transfer (0.013 < 0.05), matchup (0.009 < 0.05), familiarity (0.031 < 0.05) and similarity (0.028 < 0.05). All these five statistically significant celebrity attributes also lead to the conclusion that the Goan consumers consider them to be influencing their purchase decisions. The significant level for all these five independent variables is less than 0.05, which indicates that meaning transfer, match between the celebrity and the product, similarity, likeability and familiarity are statistically significant and can be used for the prediction of purchase intention, whereas expertise and trustworthiness of celebrity cannot.

Table 6.
Effects of Attributes on Purchase Intention

Variables Correlation Adjusted R² β t sig

( EX ) Expertise .133 0.012 0.013 .155 0.877

( TW ) Trustworthiness .110 0.006 –0.116 –1.387 0.167

( SM ) Similarity .417** 0.169 0.179 2.214 0.028**

( FM ) Familiarity .568** 0.319 0.210 2.182 0.031**

( LK ) Likeability .483** 0.229 0.242 2.607 0.010**

( MU ) Matchup .437** 0.186 0.236 2.632 0.009**

( MT ) Meaning Transfer .592** 0.347 0.240 2.525 0.013**

Source: Primary data.

Notes: **p-jvalue < 0.05, R² = 0.684, Adjusted R² = 0.44.

The hypothesis framed for testing the impact of attributes of celebrity on purchase decision is partially rejected, with the exception of expertise and trustworthiness. This means, there is a significant impact of the five celebrity attributes, namely similarity, familiarity, likeability, match-up and meaning transfer, on purchase intention, whereas celebrities’ expertise and trustworthiness do not have any influence on purchase intention. We can conclude that most of the attributes of a celebrity do have an impact on the purchase intentions of the consumers.

Perception Level of Consumers towards Celebrity Endorsement

In this section, the survey results are analysed using cross tabulation with age groups, which indicates some interesting facets of the perception levels of consumers towards the celebrity, based on the medium of advertisement that influences them more. This is further broken down to how often consumers watch the advertisements, attractiveness of advertisement, how often consumers can recall them, how often celebrities use the products they endorsed, the level of influence of celebrities and the impact of negative publicity of the celebrity endorser. The cross tabulated information is provided in Table 7. The respondents were asked which medium they came across most for celebrity-endorsed advertisements: TV, magazines, the Internet, radio or newspapers. This was then cross tabulated with age group. The most preferred medium of advertisement was TV across all age groups (61%), followed with magazines, the Internet, radio and newspapers. With respect to the frequency of watching advertisement on TV, consumers responded that they watch various advertisements on TV for at least 2–7 times a day, indicating the bombardment of advertisement by the marketers on TV to capture the attention of consumers.

With respect to the attractiveness of an advertisement, majority of the consumers opined that celebrity endorsed advertisement (87%) were more attractive than those not endorsed by a celebrity (13%). Thus, it is evident that the respondents of all age groups are more attracted towards celebrity endorsed advertisements than non-celebrity endorsed. In all the three cases, youngsters in the age group of 18–30 formed the majority group. With respect to recalling the product at a later stage during purchase, consumers across all age groups agree (33%) or strongly agree (21%) that the presence of celebrities in the advertisement helps them to recognize the product/services and recall them promptly when it comes to purchasing.

About those products/services endorsed by the celebrities and the usage of the same by the celebrities, consumers across almost all age groups strongly disagree (31%) or disagree (13%) with the statement that celebrities do not themselves use the products they endorse in their real life. Only 33 per cent of the respondents either agree (11%) or strongly agree (22%) with the statement and around 23 per cent are undecided or neutral. Hence, it can be concluded that the celebrities do not themselves use the products in their real life. This is a clear indication that though consumers know that the celebrities themselves do not use the products endorsed by them; they endorse them nonetheless because of monetary reasons. It is a complete violation of ethics by the celebrities, thereby falsifying the claims made by them about the products/services in making people purchase the product without really bothering about whether the product/service is good or bad or even harmful to the consumers. This also raises a critical question on business ethics and social responsibility issues and indicates a flaw in the code of conduct practised by advertising agencies, and it amounts to violation of the fundamental right of every consumer about getting fair, true and correct information about any product being advertised.

Table 7.
Age Profile and Perception Level of Consumers towards Celebrity Endorsement

# Particulars [N = 171] * Age
Total (%) **

Up to 18 18–30 30–50 Above 50

1 Medium of advertisement more likely to convey celebrity endorsed advertisement

TV advertisements 26 91 33 6 156 (61)

Magazines 10 14 11 0 35 (14)

Internet 6 21 6 1 34 (13)

Radio 0 3 5 0 8 (3)

Newspapers 3 11 5 2 21 (8)

2 Frequency of watching celebrity endorsed advertisements

Once a day 12 27 15 2 56 (33)

2 to 7 times a day 15 43 15 4 77 (45)

More than 7 times/day 6 23 7 2 38 (22)

3 Type of advertisement which attracts more

Celebrity endorsed 32 77 33 7 149 (87)

Non celebrity endorsed 1 16 4 1 22 (13)

4 Presence of celebrity endorsed advertisement helps recognizing/recalling products

Strongly disagree 3 8 4 0 15 (9)

Disagree 8 8 6 1 23 (13)

Neutral 8 21 9 3 41 (24)

Agree 10 32 11 3 56 (33)

Strongly agree 4 24 7 1 36 (21)

5 The products endorsed by the celebrities are actually being used by the celebrities

Strongly disagree 16 27 9 1 53 (31)

Disagree 1 16 4 2 23 (13)

Neutral 9 19 9 3 40 (23)

Agree 7 20 9 2 38 (22)

Strongly agree 2 9 6 0 17 (11)

6 Effectiveness/influence of celebrity endorsement when it comes to purchasing

Very much influential 2 13 10 2 27 (16)

Influential 11 35 12 5 63 (37)

Somewhat influential 12 20 7 0 39 (23)

Not at all influential 8 25 8 1 42 (24)

7 Negative publicity of the celebrity endorser influences purchasing decision

Yes 9 21 5 3 38 (22)

No 18 40 16 3 77 (45)

Sometimes 6 32 16 2 56 (33)

Source: Primary data.

Notes: # Number

* Except for the first item where N = 254 due to multi-response option given, rest all six items are having N = 171.

** Values given inside the parenthesis are percentage.

It is clear that consumers are well aware of the fact that celebrities do not really buy and use the products/services endorsed by them. Hence, it is rather strange that the influence of the celebrities is so strong that consumers blindly follow them and buy the products/services endorsed by them. Majority of the consumers opined that celebrity endorsement is very much influential (16%) or influential (37%) or somewhat influential (23%). Only a few consumers opined that celebrity endorsement is not at all influential (24%). The result also reveals that majority (78%) of the consumers across all age groups are not at all bothered about the negative publicity (45% no influence and 33% sometimes), and no matter what the buzz around about the particular celebrity, consumers are ready to accept the endorsement. It may lead to making the consumer a potential buyer and sometimes they actually buy the product/service. Overall, it suggests that people of all age groups are not affected by negative information about the celebrity and such negative publicity does not have any effect on their purchase decision.

Conclusion

Celebrity endorsement has totally changed the nature of advertising in recent years, and has become one of the most important tools of advertising. Such endorsements can help the marketer to better understand consumer reactions towards a particular celebrity. Since majority of the respondents of the present study consists of students, the result is slightly skewed towards youngsters who support celebrity endorsements more than the elders do. Both male and female respondents supported celebrity endorsements. The result of multi-logit also indicates that, in Goa, the probability of being an opposer decreases, which means that there is more chance of becoming a supporter of celebrity endorsements. Goan consumers also have the know-ledge that whatever the celebrities endorse and claim are not entirely true. This means that celebrity attributes such as expertise and trustworthiness are statistically insignificant to influence the purchase decision of consumers. This gives an insight for expanding the study in the state of Goa as well as other states of India to arrive at a comprehensive conclusion. Of the total 22 celebrities identified by Goan consumers (see Table 4), majority are from the film industry (11 male and 7 female celebrities) than from sports (only 2 male and 2 female celebrities), which is similar to the all-India trend (see Table 1) of complete dominance of film personalities (6 male and 1 female celebrities) in the advertising industry, except for the sport cricket (3 male celebrities). Consumers also opined that celebrity endorsement influenced them mainly for purchasing durable products. The most significant celebrity attributes when it comes to purchase intentions among Goan consumers were likeability, meaning transfer, match between the celebrity and the product, familiarity and similarity. Majority also agreed that TV advertising is the most influential medium and they are attracted only if the advertise involved film celebrities. They also opined that recognizing and recalling celebrity endorsement is easy, which allows them to take purchase decisions without much difficulty.

However, the most significant aspect of the study is that Goan consumers know very well that (see Table 7, Item 5) the celebrities do not really use the products they endorse (44% of consumers said so). It was found that 76 per cent of consumers were willing to buy a product just because it was endorsed by their favourite celebrity (see Table 7, Item 6). Hence, the results of this study indicate that companies must select only those celebrities who are likely to create high creditability in the minds of consumers. Companies believe that celebrities do influence purchase decision of consumers in one or the other way. If their favourite celebrity is involved in any controversy (negative publicity), 78 per cent of consumers tend to ignore it and go ahead with the purchase of a product endorsed by them (see Table 7, item 7). Companies must realize that having a celebrity in an advertisement can create a favourable impact on consumers and this, in turn, can influence them to buy the product.

Increasing the scope of similar studies focusing on the entire state of Goa and covering all consumers from all categories of occupations may provide more insights and interesting facets with respect to the impact of celebrity endorsements and purchasing decisions, which may be useful for various stakeholders.

Footnotes

Note

References

1.
Baker M. & Churchill G. (1977). The impact of physically attractive models on advertising evaluations. Journal of Marketing Research, 14(4), 538–556.

2.
Bowman J. (2002). Facing advertising reality. Media Asia, 7(6), 14–15.

3.
Chaiken S. (1979). Communicator physical attractiveness and persuasion. Journal of Personality and Social Psychology, 37(8), 1387–1397.

4.
Chan K. Ng Y. L. & Luk E. K. (2013). Impact of celebrity endorsement in advertising on brand image among Chinese adolescents. Young Consumers, 14(2), 167–179.

5.
Charbonneau J. & Garland R. (2010). Product effect on endorser image: The potential for reverse image transfer. Asia Pacific Journal of Marketing and Logistics, 22(1), 101–110.

6.
Dash S. K. & Sabat D. P. (2012). The impact of celebrity endorsed TV commercials on demographic dynamics of attitude: An Indian context. International Journal of Research in Management & Technology, 2(2), 192–204.

7.
Diekhoff G. (1992). Statistics for the social and behavioral sciences: Univariate, bivariate and multivariate. Dubuque, IA: William Brown C. Publishers.

8.
Erdogan B. Baker M. & Tagg S. (2001). Selecting celebrity endorsers: The practitioner’s perspective. Journal of Advertising Research, 41(3), 39–48.

9.
Freeman K. S. (2015). Wither the impact of celebrity endorsement. Paper presented at the International Conference on Communications, Media, Technology and Design, Dubai, UAE, 16–18 May 2015.

10.
Friedman H. H. & Friedman L. (1979). Endorser effectiveness by product type. Journal of Advertising Research, 19(5), 67–71.

11.
Friedman H. H. Termini S. & Washington R. (1976). The effectiveness of advertisements utilizing four types of endorsers. Journal of Advertising, 5(3), 22–24.

12.
Goldsmith R. E. Lafferty B. A. & Newell S. J. (2000). The impact of corporate credibility and celebrity credibility on consumer reaction to advertisements and brands. Journal of Advertising, 29(3), 43–54.

13.
Gupta R. Kishore N. & Verma D. P. S. (2015). Impact of celebrity endorsements on consumers purchase intentions: A study of Indian consumers. Australian Journal of Business and Management Research, 5(3), 1–15.

14.
Hovland C. I. Janis I. K. & Kelley H. H. (1953). Communication and persuasion. New Haven, CT: Yale University Press.

15.
Kambitsis C. Harahousou Y. Theodorakis N. & Chatzibeis G. (2002). Sports advertising in print media: The case of 2000 Olympic games. Corporate Communication: An International Journal, 7(3), 155–161.

16.
Kamins M. (1989). Celebrity and non-celebrity advertising in a two sided context. Journal of Advertising Research, 29(3), 34–42.

17.
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.

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

19.
Kanungo R. N. & Pang S. (1973). Effects of human models on perceived product quality. Journal of Applied Psychology, 57(2), 172–178.

20.
Kumar A. (2010). Celebrity endorsements and its impact on consumer buying behaviour. SSRN Electronic Journal. doi: 10.2139/ssrn.1802531

21.
Kurzman C. Anderson C. Clinton Key C. Lee Y. O. Moloney M. Silver A. & Van Ryn M. W. (2007). Celebrity status. Sociological Theory, 25(4), 347–67.

22.
Lindgren J. H. & Shimp T. A. (1997). Advertising, promotion, and supplemental aspects of integrated marketing communications. Fort Worth, TX: Dryden Press.

23.
McCracken G. (1989). Who is the celebrity endorser? Cultural foundations of the endorsement process. Journal of Consumer Research, 16(3), 310–321.

24.
McGuire W. J. (1985). The nature of attitudes and attitude change. In Lindzey G. & Aronson E. (Eds.), Handbook of social psychology (2nd ed., Vol. 31, pp. 136–314). New York, NY: Random House.

25.
Miciak A. R. & Shanklin W. L. (1994). Choosing celebrity endorsers. Marketing Management, 3(3), 50–60.

26.
Misra S. & Beatty S. (1990). Celebrity spokesperson and brand congruence. Journal of Business Research, 21(2), 159–173.

27.
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.

28.
Petty R. A. (1980). Effects of issue involvement on attitudes in an advertising context. In Gom G. G. & Goldberg M. E. (Eds.). Proceedings of the Division 23 Program (pp. 75–79). Montreal, Canada: American Psychological Association.

29.
Petty R. E. Cacioppo J. T. & Schumann D. (1983). Central and peripheral routes to advertising: The moderating role of involvement. Journal of Consumer Research, 10(2), 135–146.

30.
Renton K. (2006). The relationship of celebrity advertisement to consumer attitudes and purchase intentions (Master’s Thesis submitted to the Department of Textiles and Consumer Sciences). College of Human Sciences, The Florida State University.

31.
Renton K. (2009). Impact of athletic endorsements on consumers purchase intentions (Doctoral Thesis submitted to the Department of Textiles and Consumer Sciences). College of Human Sciences, The Florida State University.

32.
Roy S. (2006). An exploratory study in celebrity endorsements. Journal of Creative Communications, 1(2), 139–153.

33.
Roy S. (2012a). To use the obvious choice: Investigating the relative effectiveness of an overexposed celebrity. Journal of Research for Consumers, 22, 41–69.

34.
Roy S. (2012b). Investigating endorser personality effects on brand personality: Causation and reverse causation in India. Journal of Brand Strategy, 1(2), 164–179.

35.
Roy S. & Jain V. (2016). The meaning transfer process in celebrity endorsements: A quantitative exploration. In Obal M. W. Krey N. & Bushardt C. (Eds.), Let’s get engaged! Crossing the threshold of marketing’s engagement era. Developments in Marketing Science: Proceedings of the Academy of Marketing Science (pp. 539–544). Cham (Zug), Switzerland: Springer.

36.
Roy S. & Moorthi Y. L. R. (2009). Celebrity endorsements and brand, personality (IIMB Working Paper No. 2009–09–289). Bangalore: Indian Institute of Management. http://iimb.ac.in/sites/default/files/WP.IIMB_.289.pdf

37.
Roy S. Bashar G. & Koh A. C. (2012). Predicting the effectiveness of celebrity endorsements using the balance theory. Journal of Customer Behaviour, 11(1), 33–52.

38.
Roy S. Jain V. & Rana P. (2013). The moderating role of consumer personality and source credibility in celebrity endorsements. Asia-Pacific Journal of Business Administration, 5(1), 72–88.

39.
Silvera D. & Austad B. (2004). Factors predicting the effectiveness of celebrity endorsement advertisements. European Journal of Marketing, 38(11/12), 1509–1526.

40.
Subhash K. B. Kamat K. Scaglione M. D’Mello C. & Weiermair K. (2015). Residents’ perception towards film tourism in Goa: A multi-logit analysis. Paper presented at the 3rd World Research Summit for Tourism and Hospitality & 1st USA–China Tourism Research Summit, jointly organised by UCF Rosen college of Hospitality Management, Orlando, FL, USA and Elsevier. Florida, USA, 15–19 December.

41.
Subhash K. B. Kamat K. Scaglione M. D’Mello C. & Weiermair K. (Forthcoming, 2018). Film tourism and its impact on residents quality of life: A multi logit analysis. In Campón-Cerro A. M. Hernández-Mogollón J. M. & Folgado-Fernández J. A. (Eds.), Best practices in hospitality and tourism marketing and management: A Quality-of-life perspective. Springer (In Press).

42.
Till B. D. & Busler M. (1998). Matching products with endorsers: Attractiveness versus expertise. The Journal of Consumer Marketing, 15(6), 576–586.

43.
Till B. D. & Shimp T. A. (1998). Endorsers in advertising: The case of negative celebrity information. Journal of Advertising, 27(1), 67–82.