Introduction

Instagram and other social media platforms have turned into marketing and promotion channels for businesses. This is because images and videos are more visually appealing than text-based content when it comes to attracting viewers' attention (Hanson, 2018; Hashim et al., 2020). With Instagram becoming a popular medium, particularly among young customers, many businesses are paying well-known celebrities to post sponsored content on their Instagram accounts. It is expected that the fame and likability of these well-known celebrities, who have a significant fan base, will pass onto the endorsed brand.

In reality, in many situations, fans follow celebrities not just because they like them, but also because of the unique and entertaining content they share on their social media channels. As a result, these well-known celebrities have aroused consumers' interest in following their lifestyles, attitudes, and personal characteristics as displayed on social media platforms such as Instagram and Facebook.

As a result, businesses have spent a lot of money to employ celebrities as endorsers, assuming that hiring celebrities will increase the credibility of the campaign and hence provide better results. According to Munnukka et al. (2016), celebrities have a significant influence on product and brand associations. Celebrities may affect the purchasing of health and beauty products by acting as role models for young consumers (Ahmad et al, 2015). Advertisements with well-known celebrities are thought to be more effective (McCormic, 2016), and their role in influencing customers is said to be more crucial.

The purpose of this article is to illustrate how celebrity Instagram endorsement affects customer purchase intention. Specifically, the study aims to investigate how celebrity credibility (as defined by the attributes of source credibility given by Ohanian, 1990) impacts consumers’ purchase intention. Our study intends to answer the following research questions (RQ):

RQ1: Do celebrity endorsed Instagram advertisements impact purchase intention of followers?

RQ2: Does this impact of celebrity endorsed Instagram sponsored content vary for different brands?

To determine the influence on purchase intention, this study uses survey data from young fans of celebrity Virat Kohli (an Indian cricketer) for two distinct brands – Wrogn (apparel brand) and Unacademy (e-learning platform). It is reported that Cricketer Virat Kohli earns Rs 1.3 crore per Instagram post (Source: www.economictimes.com , July, 2009). With such a huge amount paid to celebrities to post a sponsored content on their Instagram account, it becomes worthwhile to examine whether such an endorsement will bring the desired results for the brand.

Literature Review

Source Credibility and Source Attractiveness Models

The research on celebrity endorsement is based on a ground-breaking study on source trustworthiness conducted by Hovland and his colleagues in 1953 (Arshad et al., 2017). The source attractiveness model and the source credibility model are the foundations of their research. The source credibility model considered the source's expertise and trustworthiness, whereas the source attractiveness model considered the source's attractiveness.

The degree to which a source is recognised as a dependable source of information is defined as source expertise. The degree to which the source is perceived as familiar, likeable, similar, and appealing is characterised as source attractiveness, whereas source trustworthiness is defined as the degree of confidence in the communicator's willingness to communicate the statements he or she believes are the most valid (Ohanian, 1990; Erdogan, 2001; Dwivedi et al.,2016).

Both the source attractiveness model as well as the source credibility model explains many aspects of source credibility. Pornpitakpan (2004) found that an endorser's trustworthiness and attractiveness, when measured on these characteristics, had a positive relationship with purchase intent. The Ohanian Model of source credibility (Ohanian, 1990) was used to express the persuasiveness of a message during the communication process with a target segment that is influenced by source characteristics (Corina, 2006).

Our study uses Ohanian's Source Credibility Model as its foundation for conducting its investigation to measure the impact of celebrity credibility on the purchase intention of brands promoted on Instagram. As per the model, three aspects of source credibility are expertise, attractiveness, and trustworthiness of the source. Some cues from the five-dimensional PATER Model of celebrity endorsement (Gupta et al., 2017) have also been used.

Attractiveness, Trustworthiness and Expertise of a Celebrity

Consumers' brand attitudes and purchase intent are strongly influenced by the attractiveness of a message source, according to previous studies (Pornpitakpan, 2003; Chan et al., 2013). An attractive celebrity is more likely to receive good comments and attention from the public since customers are primarily attracted by their looks. An attractive celebrity endorser would have a stronger effect on customer attitudes about the company and its products (Ohanian, 1991). Furthermore, if a celebrity is more knowledgeable and experienced in the endorsed sector, he or she can quickly earn trust, favourable recognition, and even inspire the desire to purchase (Braunsberger, 1996).

If endorsers have knowledge, experience, or abilities, Erdogan (1999) claims that they are viewed as having expertise. It also implies that the message source is regarded as a reliable source of facts (Erdogan, 1999). Previous study has shown that an expert message source, such as a celebrity endorser, can influence attitudes and purchase intention (Ohanian, 1991).

Celebrity Endorsement and Purchase Intention

The conscious choice of someone to buy a certain brand is described as purchase intention (Spears & Singh, 2004). Schiffman & Kanuk (2000) contend that when consumers pay attention to a product that has been recommended by endorsers, their interactions with it may allow them to assess customers' desire to buy. Celebrity endorsement has a substantial influence on buyer's intention in favour of a certain product (Kapitan & Silvera, 2016; Gupta et al., 2015; Chan et al., 2013, Pornpitakpan, 2003; Ohanian, 1991).

On the basis of the above discussion we lay down the following hypotheses:

H1: The impact of celebrity credibility on purchase intention for the brand Wrogn is positive and significant.

H2: The impact of celebrity credibility on purchase intention for the brand Unacademy is positive and significant.

The goal is to see if a celebrity Instagram endorsement by the same celebrity (Virat Kohli in our example) produces comparable or different effects for different brands in different sectors, especially when their target market is the same.

Research Methodology

Sample Size, Sampling Design, and Measurement Instrument

This study employs survey data from young followers of celebrity Virat Kohli for two different brands – Wrogn (apparel brand) and Unacademy (e-learning platform) to find out the impact on purchase intention. In both these studies, Study 1 (Wrogn) and Study 2 (Unacademy), a questionnaire was utilised to collect primary data from college students and young professionals in the age group of 18-30 years, which is also the target market for both these brands in India. The questionnaire was personally administered to acquire responses after showing the short video ads of both the brands (which were run on Instagram as the sponsored content posted by Virat Kohli) to respondents; and the qualifying question for the respondents was whether they are followers of celebrity Virat Kohli on Instagram. Thus, we used convenience and judgemental sampling methods to collect data for our study. We collected 219 responses for study 1 (Wrogn) and 121 responses for study 2 (Unacademy).

The demographics of the respondents were covered in the first section of the questionnaire. The demographic characteristics of the respondents are shown in Table 1.

Table 1 Demographic Profile of Respondents
Demographic Factors		Study1	Study2
Age	18-30	219	121
Gender	Male	123	79
	Female	96	42
Education	Undergraduates	61	43
	Graduates	55	32
	Postgraduates	78	31
	Higher degree	25	15
Total		219	121

To investigate the relationship between a celebrity endorser's credibility and the influence on purchase intention, we used the Ohanian model of celebrity credibility (Ohanian, 1990). As a result, celebrity credibility was assessed on three dimensions: the celebrity’s attractiveness, trustworthiness, and expertise.

Questions on the items that made up the study's constructs - attractiveness, trustworthiness, expertise, and purchase intention for the brand – were included in the second half of the questionnaire. A total of 18 items (see Tables 2 and 4) were classified as questions based on these criteria. These questions were answered on a seven-point Likert scale ranging from 1 to 7, with 1 denoting "strongly disagree" and 7 denoting "strongly agree."

Research Techniques Used for the Study

EFA was used to check the construct validity of the constructs. To determine sampling adequacy, the Kaiser Meyer Olkin (KMO) test was employed. The model's validity (convergent and discriminant) and reliability were assessed using CFA. The influence of celebrity credibility on brand purchase intent was assessed using SEM.

Data Analysis and Results

Study 1: Impact of Virat Kohli Wrogn ad on Purchase Intention

Figure 1 shows the first ad chosen for our study. It is for the apparel brand ‘Wrogn’. It can be seen from the figure, that this ad received more than 26 lakhs views. It is also pertinent to note here that there is a sponsorship disclosure in this particular Instagram post as we can make out from the terms ‘Paid partnership with staywrogn’.

Exploratory Factor Analysis

The result of the KMO test was 0.917, which is deemed excellent (Kaiser, 1974). The construct validity of the constructs was confirmed by the results of the principal component rotated matrix. Table 2 shows the factor loadings. The total variance explained was 81.17 percent, exceeding the necessary 50 percent threshold.

Validity and Reliability Analysis

The reliability of the constructs was determined using Cronbach's alpha (Nunnally, 1978). Table 2 shows that all alpha values are more than 0.70, indicating that the constructs are reliable. The convergent and discriminant validity of our constructs were determined using confirmatory factor analysis (CFA).

Table 2 shows that α>0.7, AVE>0.5, and α>AVE are all confirmed. As a result, the constructs' convergent validity is demonstrated (Hair et al., 2010). Discriminant validity of the constructs is also proven as AVE>MSV and AVE>ASV (Table 3). (Hair et al., 2010).

Table 2  Reliability and Convergent Validity – Unacademy AD
Constructs	Items	Factor Loadings	Cronbach Alpha (a)	AVE	Composite Reliability (CR)
Attractiveness (Ohanian, 1990; Gupta et al, 2017)	A1: Virat Kohli is Attractive A2: Virat Kohli is Classy A3: Virat Kohli is Handsome A4: Virat Kohli is a Style Icon A5: Virat Kohli is Sexy	0.745 0.814 0.836 0.847 0.829	0.934	0.776	0.946
Trustworthiness (Ohanian, 1990)	T1: Virat Kohli is Dependable T2: Virat Kohli is Honest T3: Virat Kohli is Reliable T4: Virat Kohli is Sincere T5: Virat Kohli is Trustworthy	0.724 0.847 0.878 0.845 0.641	0.949	0.794	0.953
Expertise (Adapted from Ohanian, 1990)	EX1: Virat Kohli is an expert in this field EX2: Virat Kohli has an experience with this brand EX3: Virat Kohli is knowledgeable about this brand EX4: Virat Kohli is qualified to endorse this brand EX5: Virat Kohli has a personal experience with this brand	0.821 0.852 0.882 0.877 0.829	0.936	0.721	0.930
Purchase Intention (Adapted from Spears &Singh, 2004)	PI1: I have a high purchase intention of buying this brand after seeing this ad PI2: I will probably buy this brand after seeing this ad PI3: I have bought this brand after seeing this ad	0.809 0.864 0.844	0.796	0.601	0.819

Table  3 Discriminant Validity – Wrogn AD
	CR	AVE	MSV	ASV	Expertise	Attractiveness	Trustworthiness	Purchase Intention
Expertise	0.930	0.728	0.331	0.174	0.853
Attractiveness	0.946	0.779	0.462	0.270	0.436	0.882
Trustworthiness	0.953	0.802	0.462	0.273	0.575	0.680	0.896
Purchase Intention	0.819	0.601	0.158	0.062	-0.018	0.398	0.164	0.775

Structural Equation Modelling- Path Analysis

Structural equation modelling was used to investigate the impact of celebrity credibility on purchase intention. The standardised beta coefficient value (β)=0.23, p<0.05 (at 95 percent confidence level) indicating that the credibility of celebrity Virat Kohli has a positive and significant impact on the purchase intention for brand Wrogn (H1 is accepted). On checking the model fit, we found the following values. Chi-square=292.888, df=128, CMIN/df=2.288, GFI=0.874, CFI=0.957, RMSEA=0.077 (all values permissible as per Hair et al., 2010 and Baumgarther & Homburg, 1996). This validates our model as shown in Figure 2.

Study 2: Impact of Virat Kohli Unacademy ad on Purchase Intention

Figure 3 depicts the second advertisement used in our research. It is for the e-learning portal ‘Unacademy.' This ad garnered more than 17 lakh likes, as can be seen in the figure. It is also worth noting that the words 'Paid partnership with Unacademy' indicate that this Instagram post has a sponsorship disclosure.

Exploratory Factor Analysis

The KMO test yielded a score of 0.842, which is considered exceptional (Kaiser, 1974). The results of the principal component rotated matrix corroborated the construct validity of the constructs. The factor loadings can be seen in Table 4. The total variation explained was 79.85 percent, well above the required 50%.

Validity and Reliability Analysis

The reliability of the constructs was determined using Cronbach's alpha (Nunnally, 1978). Table 4 shows that all alpha values are more than 0.70, indicating that the constructs are reliable. The convergent and discriminant validity of our constructs were determined using Confirmatory Factor Analysis (CFA).

Table 4 shows that α>0.7, AVE>0.5, and α>AVE are all confirmed. As a result, the constructs' convergent validity is demonstrated (Hair et al., 2010). Discriminant validity of the constructs is also proven as AVE>MSV and AVE>ASV (Table 5) (Hair et al., 2010).

Table 4  Reliability and Convergent Validity – Unacademy AD
Constructs	Items	Factor Loadings	Cronbach Alpha (a)	AVE	Composite Reliability (CR)
Attractiveness (Ohanian, 1990; Gupta et al., 2017)	A1: Virat Kohli is Attractive A2: Virat Kohli is Classy A3: Virat Kohli is Handsome A4: Virat Kohli is a Style Icon A5: Virat Kohli is Sexy	0.745 0.910 0.868 0.889 0.735	0.903	0.699	0.924
Trustworthiness (Ohanian, 1990)	T1: Virat Kohli is Dependable T2: Virat Kohli is Honest T3: Virat Kohli is Reliable T4: Virat Kohli is Sincere T5: Virat Kohli is Trustworthy	0.689 0.793 0.865 0.875 0.660	0.905	0.626	0.902
Expertise (Adapted from Ohanian, 1990)	EX1: Virat Kohli is an expert in this field EX2: Virat Kohli is experienced  in this field EX3: Virat Kohli is knowledgeable in this field EX4: Virat Kohli is qualified to endorse this product EX5: Virat Kohli has a personal experience with this brand	0.895 0.864 0.855 0.908 0.875	0.947	0.759	0.941
Purchase Intention (Adapted from Spears &Singh, 2004)	PI1: I have a high purchase intention of enrolling into a course after seeing this ad PI2: I will probably enrol into a course after seeing this ad PI3: I have enrolled into a course after seeing this ad	0.875 0.907 0.917	0.816	0.729	0.890

Table 5 also needs to be corrected (please keep the same size of the table as you have kept in Table 3 above). The formatting is required as the numerical values are not reflected properly. I am including Table 5 here for your ready reference

Table  5 Discriminant Validity – Unacademy AD
	CR	AVE	MSV	ASV	Expertise	Attractiveness	Trustworthiness	Purchase Intention
Expertise	0.941	0.761	0.264	0.141	0.873
Attractiveness	0.924	0.711	0.239	0.122	0.357	0.843
Trustworthiness	0.902	0.652	0.264	0.169	0.514	0.489	0.808
Purchase Intention	0.890	0.730	0.030	0.011	0.173	-0.018	0.052	0.854

Structural Equation Modelling – Path Analysis

Structural equation modelling was used to investigate the impact of celebrity credibility on purchase intent. The standardised beta coefficient value (β)=0.09, p=0.428 (at 95 percent confidence level) indicates that celebrity Virat Kohli's credibility has no significant influence on purchase intention for brand Unacademy (Figure 4). Thus, H2 is rejected. On checking the model fit, we found the following values. Chi-square=245.391, df=126, CMIN/df=1.948, GFI=0.822, CFI=0.936, RMSEA=0.089 (all values permissible as per Hair et al., 2010 and Baumgarther & Homburg, 1996). This validates our model.

Discussion

In today's digital world, businesses see social media marketing as a must-have approach for promoting their brands through digital communication. Despite the current increase in interest in celebrity endorsements on Instagram as a corporate communication strategy, more study is needed to understand how they might affect customers in different product categories.

This research contributes to the body of knowledge by evaluating the use of celebrity endorsement on Instagram as a social media marketing tactic and its influence on brand purchase intention. To find this underlying relationship, the study looked at the impact of Virat

Kohli's brand endorsement on Instagram for two different brands: Wrogn (an apparel brand) and Unacademy (an e-learning platform).

Exploratory factor analysis was used to extract the various factors or attributes of celebrity credibility as given by Ohanian (1990). The survey questionnaire included 18 Likert scale items that were used to assess the various aspects of celebrity credibility as well as their influence on the variable ‘purchase intention’. Confirmatory factor analysis aided in the confirmation of the constructs' validity. The study's model was put to test to see how well it fits the data. Structural equation modelling was used to analyse the impact of celebrity credibility on brand purchase intention.

The results of the study reveal that celebrity credibility (as evaluated by his attractiveness, trustworthiness, and expertise) has a favourable and significant impact on purchase intention for the brand Wrogn, but not for the brand Unacademy. Thus, celebrity Instagram endorsements may have variable effects on purchase intent for various brands.

If we try to figure out the reasons for this, it's probable that Virat Kohli is a celebrity who is well-known in the cricketing world as well as a fashion icon. It is possible to attract audiences by having a nice appearance. When it comes to an apparel brand, such as a t-shirt (brand Wrogn in our study), his followers may be affected by him and his choice as shown in the Instagram endorsement, and may wish to buy the brand. When it comes to an e-learning portal like Unacademy, however, his fans may appreciate the advertisement but are not affected by his choice of learning platform for studying various courses. This implies that there are some deeper insights into such endorsements in terms of the consumer cognitive process and the different celebrity credibility characteristics at play. Future scholars may be interested in such studies. This will contribute to the growing body of knowledge about celebrity social media endorsements. This also demands that marketers pick the proper celebrity endorser for their social media endorsements after conducting research to find the optimal fit between celebrity, product, and message delivered (Oh & Park, 2020; Kim & Lee, 2010; Lee, 2017).

Limitations and Directions for Future Research

Our study is limited to studying the impact of celebrity Virat Kohli’s endorsement on Instagram for two different brands-Wrogn (an apparel brand) and Unacademy (an e-learning platform). Also, we considered studying the impact of these Instagram endorsements on the young followers of Virat Kohli in the age group of 18-30 years, as this also is the target market for two brands in question. Future studies can consider different respondent groups and might investigate at the moderating effects of age and gender on celebrity endorsements' impact on customer purchase intent. Future research might also examine the effects of celebrity endorsements on products with varying degrees of involvement.

Furthermore, each celebrity has his/her own distinct characteristics, which vary in relevance to the product depending on the sector in which he or she is promoting. So, while Virat Kohli may have his own unique qualities as a celebrity endorser, another endorser may have distinct qualities that make him or her a good fit for a certain brand. Future research can look at the characteristics that people associate with different celebrities.

Conducting a Sentiment Analysis of the comments left by followers on such sponsored posts could be an intriguing study. Consumer perceptions about the celebrity, the sponsored brand, the relationship between the two, whether the followers like it or not, and so on will be revealed through this form of study.

The impact of sponsorship disclosure and non-disclosure may also be investigated by comparing two product endorsements, one with and one without sponsorship disclosure.

Conclusion

Celebrity sponsored ads on Instagram are on the increase. The prospect of reaching out to a huge number of customers is enormous due to the large number of followers that renowned celebrities have on social media platforms. Furthermore, in the case of celebrity’s’ followers who like and comment on the sponsored posts, these advertisements are seen more frequently than a one-time ad on a media like television, which would be wasted if the target audience did not see it.

Because of the high expenses of celebrity endorsements on Instagram, marketers must be aware of which product endorsements work best for certain celebrities. The influence of celebrity Virat Kohli's Instagram endorsements for two separate products – Wrogn (apparel) and Unacademy (e-learning platform) – on the purchase intention of the followers of the famous cricketer on Instagram was investigated.

Results revealed that celebrity Virat Kohli’s credibility has a favourable and significant impact on purchase intention for the brand Wrogn, but not for the brand Unacademy. The possible causes are discussed. Thus, celebrity Instagram endorsements may have variable effects on purchase intent for various brands.

References

1. Ahmad, A.N., Rahman, A.A., & Rahman, S. (2015). Assessing knowledge and religiosity on consumer behavior towards halal food and cosmetic products. International Journal of Social Science and Humanity, 5, 10- 14.
2. Arshad, S., Ikram, M., Yahya, M., & Nisar, Q.A. (2017). Does celebrity endorsement influence the corporate loyalty: mediating role of corporate credibility? International Journal of Social Science, Humanities and Education, 1, 308–319.
3. Baumgarther, H., & Homburg, C. (1996). Applications of structural equation modelling in marketing and consumer research: A review. International Journal of Research in Marketing, 13(2), 139-161.
4. Braunsberger, K. (1996). The effects of source and product characteristics on persuasion. Arlington, TX: Doctoral dissertation, University of Texas at Arlington.
5. Chan, K.N.G.Y.L., & Luk, E. (2013). Impact of celebrity endorsement in advertising on brand image among chinese adolescents. Young Consumers, 14(2), 167-179.
6. Corina, S. (2006). Marketing communication in online social programs: Ohanian model of source credibility. Journal of Empirical Generalisations in Marketing, 1(1), 778-784.
7. Dwivedi, A., Johnson, L., & McDonald, R. (2016). Celebrity endorsements, self-brand connection and relationship quality. International Journal of Advertising, 35, 486–503.
8. Erdogan, B. (2001). Selecting celebrity endorsers: The practitioner’s perspective. Journal of Advertising Research, 41, 39-49.
9. Erdogan, B.Z. (1999). Celebrity endorsement: A literature review. Journal of Marketing Management, 15(4), 291-314.
10. Gupta, R., Kishor, N., & Verma, D.P.S. (2017). Construction and validation of a five-dimensional celebrity endorsement scale: Introducing the PATER model. British Journal of Marketing Studies, 5(4), 15-35.
11. Hair, J., Black, W., Babin, B., & Anderson, R. (2010). Multivariate Data Analysis, (7th edition): Prentice-Hall, Inc. Upper Saddle River, NJ, USA
12. Hanson, L. (2018). The fashion brand’s guide to creating excellent visual content.
13. Hashim, N.A.A.N., Awang, Z., Yusoff, A.M., Safri, F.H.M., Fatt, B.S., Velayuthan, S.K., Hashim, H., & Novianti, S. (2020). Validating the measuring instrument for determinants of tourist's preferences toward revisit intention: A study of genting highland. Journal of Advanced Research in Dynamical and Control Systems, 12(7), 2236-2240.
14. Hovland, C.I., Janis, I.L., & Kelley, H.H. (1953). Communication and persuasion. New Haven, CT: Yale University Press.
15. Kaiser, H. F., & Rice, J. (1974). Little Jiffy Mark IV. Educational and Psychological Measurement, 34(1), 111-17.
16. Kapitan, S., & Silvera, D.H. (2016). From digital media influencers to celebrity endorsers: Attributions drive endorser effectiveness. Marketing Letters, 27(3), 553-567.
17. Kim, Y.E., & Lee, J.W. (2010). Relationship between corporate image and customer loyalty in mobile communications service markets. African Journal of Business Management, (18), 4035–4041.
18. Lee, J.W. (2017). Critical factors affecting consumer acceptance of online health communication: An application of service quality models. Journal of Asian Finance, Economics and Business, 4(3), 85–94.
19. McCormic, K. (2016). Celebrity endorsements: Influence of a product-endorser match on Millennials attitudes and purchase intentions. Journal of Retailing and Consumer Services, 32(2), 39–45.
20. Munnukka, J., Uusitalo, O., & Toivonen, H. (2016). Credibility of a peer endorser and advertising effectiveness. Journal of Consumer Marketing, 33, 182–192.
21. Nunnally, J.C. (1978). Assessment of Reliability. In: Psychometric Theory, (2nd edition). New York: McGraw-Hill.
22. Oh, A.H., & Park, H.Y. (2020). The effect of airline’s professional models on brand loyalty: Focusing on mediating effect of brand attitude. Journal of Asian Finance, Economics and Business, 7(5), 155–166.
23. 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.
24. Ohanian, R. (1991). The impact of celebrity spokespersons’ perceived image on consumers’ intention to purchase. Journal of Advertising Research, 31(1), 46–54.
25. Pornpitakpan, C. (2004). The persuasiveness of source credibility: A critical review of five decades' evidence. Journal of applied social psychology, 34(2), 243-281.
26. Schiffman, L.G., & Kanuk, L.L. (2000). Consumer behavior 7. London, UK: Prentice Hall.
27. Spears, N. & Singh, S.N. (2004). Measuring attitude toward the brand and purchase intentions. Journal of Current Issues and Research in Advertising, 26(2), 53-66.