Social Media Influencer in Advertising: The Role of Attractiveness, Expertise and Trustworthiness

The cosmetic industry has shown rapid growth worldwide and is seen as a profitable business yet highly competitive. A popular grab-attention strategy being adopted worldwide by the brand manufacturers is celebrity endorsement to improve the effectiveness of the advertisement. Despite its popularity, companies increasingly abandoning it in favor of social media influencers due to the popularity of social media and online stores. However, the effectiveness of this new way of communication using an influencer is not well understood. To address this gap, this research investigated the influence of social media influencer credibility (attractiveness, trustworthiness, and expertise) on advertising effectiveness (attitude toward the product, attitude toward the advertisement, and purchase intention to ensure if it is suitable to hire a social media influencer to advertise cosmetic products in Malaysia over a celebrity. The finding can serve as a reference to help cosmetic products marketers to develop effective ads using source credibility to communicate with their customers and stand out from the surrounding media clutter.


Introduction
Globalizations of the market, online stores, and e-commerce platforms made foreign and local cosmetic products available widely in any market worldwide and hence increase the intensity of competition and new product introduction. With the increasing rivalry, most marketers opt for celebrity advertising which is popular yet a challenging form of communication to promote their brand. Celebrity endorsement has been found to exert a quadruple effect such as enhance communication effectiveness by eliminating noise (Muda, Musa & Putit, 2012), brand switching (Deighton. J, 1994), better recall of the brand name (Petty, Cacioppo & Schumann, 1983) as well as favorably enhance advertising effectiveness in a way towards the ad itself, product and ultimately purchase intention (Chekima & Chekima, 2019, Schouten et al., 2020Bergkvist and Zhou 2016;Atkin and Block 1983).
However, in the age of social media and online purchasing, utilizing "traditional or conventional" celebrities comprises of actors, athletes, singers, and supermodels seem to be fading because companies increasingly abandoning it in favor of social media influencers (Schouten et al., 2020;Marwick 2015). Through different social media platforms such as Youtube, Instagram, Tik Tok, Facebook, social media influencers are aptly employed to engage, publicize product information as well as communicate the latest promotions and latest information regularly with their online followers through live video, vblog, text, or poster posts, product review, etc.. (Markethub 2016;Liu et al. 2012). Social media influencer gains much of the visibility, popularity, and large follower base by enthusiastically sharing self-generated content of a different product or topics including beauty, health, fashion, investment and food (Khamis, Ang, and Welling, 2017), thus turning them to be perceived or branded by followers as an expert on social media platforms (Lin, Bruning, and Swarna 2018).
Despite the increasing deployment of social media influencers for product and service endorsement, there is a lack or limited knowledge on their marketing value in terms of understanding its credibility effectiveness towards the ad, product, and purchase intention of cosmetic products in Malaysia. Most preceding studies have focused on celebrity endorsement with inconsistent results depending on the context and product and thus make it pertinent to revisit the effectiveness of credibility sources from the perspective of social media influencer where online and social media purchasing is dominating the market and have evolved the behavior of the consumer. Another lack is in the context where previous studies have focused more on the western setting. Therefore this study aims to investigate the impact of social media influencer credibility on advertising effectiveness (attitudes toward the advertisement and product, and purchase intention) in Malaysia concerning cosmetic products. Influencer credibility was operationalized as consisting of three constructs which are attractiveness, expertise, and trustworthiness. The cosmetic industry is considered a highly profitable business in Malaysia which was valued at $764.9 million in 2020 and is expected to increase by 8.1% annually to $1,288.7 million by 2027 (Market Research, 2020). Presently, this industry is heavily dependent on social media communication and thus makes it feasible to validate this model.

Underpinning Theory
The Source Credibility Theory posits that the value of a message could be influenced by the endorser's acknowledgment level of appeal, trustworthiness, and expertise (Hovland and Weiss, 1951;McGuire, 1985;Ohanian, 1991). The degree that the source is known as having noteworthy experience or adequate information to not present a one-sided choice is alluded to as validity (Carroll, 2008;O'Mahony and Meenaghan, 1997). Prior discoveries demonstrated that source validity is essential for the accomplishment of endorser support because the degree of influence of a message generally lays on the believability of the source (Wei and Li, 2013). As indicated by Ohanian (1990), the credibility of the source is normally used to stand out to the positive attributes of communicators which would impact the beneficiary's affirmation of the correspondence.
Additionally, because of globalization and a bounty of item decisions, customers presently can undoubtedly switch among brands and items when contrasted with the past and along these lines expanded degrees of rivalry and variety in the makeup business. To demote this from occurring, scholars of product differentiation (Schouten et al, 2020;Hannan et al., 1990) prompted that organizations may confront less rivalry by separating themselves from others through big name or influencer underwriting. Besides, it is powerful to utilize endorsers from the dependable source towards buyers which have a bothersome understanding of the item (Schouten et al, 2019;Singh and Banarjee, 2018;Thomas and Johnson, 2017;Daneshvary and Schwer, 2000). Consequently, the tenable source can neutralize these sentiments and have a positive result which is a greater message acceptance (Byrne et al., 2003).

Attractiveness
McCroskey and McCain (1974) put out that attractiveness is linked with how sources are perceived and related social values. In the context of effective advertising, source attractiveness largely depends on the source's similarity, familiarity, and likeability to the receivers (Ohanian, 1991;McGuire, 1985). The concept of similarity is the supposed resemblance between the receiver (social media followers) and source, while familiarity is measured knowledge of the source based on their exposure; and likeability is liking for the source as a result of the source's physical facial beauty and behavior (McGuire, 1985).
The physical engaging quality of online media influencers is seen to have a high inclination in driving the acknowledgment pace of publicizing and saw more alluring when they have a lot of similarities (Lim et al. 2017). In view of Bardia et al. (2011), source physical appeal is interrelated to an individual's soonest judgment given to the following person, which is driven by the attributes and characteristics of the endorser for instance facial beauty, height, and weight. Ranjabarian et al., (2010) expressed that to expand the affirmation or awareness towards an ad or product, source endorsement could be utilized as a significant variable. Furthermore, it is established that endorsers "which are considered attractive will be more likely to lead the consumer purchase intention" (Van Der Waldt et al., 2009). Therefore, the physical appeal of the source could be exploited to heighten the advertisements' effectiveness (Schouten et al, 2020;Singh and Banarjee, 2018;Thomas and Johnson, 2017;Chan et al., 2013). Given the pervasiveness of source credibility on consumer evaluation towards the ad and behavior, this study posits that: H1: social media influencer attractiveness has a positive influence on attitude toward the ad H2: social media influencer attractiveness has a positive influence on attitude toward the ad brand. H3: social media influencer attractiveness has a positive influence on purchase intention of cosmetic product.

Expertise
Expertise has been portrayed by McCracken (1989) as "the apparent capacity of the source to make legitimate affirmations". As such, the endorser is viewed as capable enough to convey legitimate and exact information or conscious of a specific subject (Hovland et al., 1953). An endorser that is more powerful is found to have a more elevated level of skill (Bardia et al., 2011;Aaker and Myers, 1987) and would influence a greater purchase intention (Chan et al., 2013;Erdogan, 1999;Ohanian 1991).
Earlier studies have highlighted that the beneficiary's buying choice and perceived trustworthiness could be affected by source skill in the product (Smith et al., 2006). Similarly, Dean and Biswas (2001) express that source influence in promoting practical and expensive brands, endorser's skill rises as a pertinent factor. Along these lines, it is less powerful to utilize ordinary messages and advertorials moved by the association than to utilize a source with applicable aptitude as an endorser (Charbonneau and Garland, 2005). Also, as indicated by the aftereffects of the examination by Chaovalit (2014), between credibility, there is a significant connection regarding expertise and buying behavior.
Likewise, influencers' expertise or relevance to the product is acknowledged by online buyers as important to endorsement success as well (Djafarova and Rushworth 2017). Schouten et al. (2019) further affirmed that the effect of the product-endorser match on credibility is more noticeable for influencers compared to regular celebrity endorsers because they have identified themselves successfully as representative of a known domain of interest, such as 'gaming vlogger', 'fitness vlogger', 'makeup vlogger' or 'fashion blogger', and share product information on a routine basis with their online followers (Balog et al. 2008). The study in hand evaluates the impact of influencers' ability in bringing and endorsing cosmetics in which they usually are users themselves of such product/brand and understand it well before sharing information with their followers. In comparison to celebrities, in most cases, they are just endorsing the product without prior adoption of the product and more depending on their popularity in a certain field which doesn't reflect product-fit with the product. Thus they convey the information based on input from the marketers only and might be perceived as not knowledgeable of the product. Therefore, a good fit between the influencer and the product could be of higher importance than for celebrities, which results in the following hypothesis: H4: Social media influencer expertise has a positive influence on attitude toward the ad H5: Social media influencer expertise has a positive influence on attitude toward the ad brand. H6: Social media influencer expertise has a positive influence on purchase intention of cosmetic product

Trustworthiness
Trustworthiness is characterized as the apparent eagerness of the source of influencer to make lawful statements in the perspective on the consumer or followers and gives truthful and honest information about a product (Ohanian, 1990;McCracken, 1989). As O'Mahony and Meenaghan (1997) point out, trust is the credibility of the source to make the essential highlights of the impacts of changes in customer disposition. Without trust, different properties possessed by the endorser can't be compelling in changing customer perspectives (Miller and Basehart, 1969). Parts of trust are significant in the idea of credibility likewise has been ended up being powerful in encouraging consumers (Moore, Hausknecht, and Thamodaran 1988) and the motivation of disposition (McGinnies and Ward, 1980). This is in accordance with Atkin and Block (1983), which propose that referred sources or endorsers are acknowledged as more reliable than regular ones.
According to Chao et al. (2015) and Wei and Li (2013), source trustworthiness influences consumers' purchase intention and backing the constructive outcome on endorsement viability. Additionally, a decent conviction towards the source can goodly affect the impression of clients' worth and further influence their buying decision (Erdem and Swait, 2004). Evidently, a positive relationship happens among the trustworthiness, fascination, and product-related with the quality of the contention; for example, when solid item related contentions mark the promoting duplicate, the impact of trustworthiness is less observable than it is for feeble item related contentions (Chan et al., 2013;Priester and Petty, 2003). Thusly, influencers who will be chosen must be a person whom purchasers can identify with and are perceived as genuine, fair-minded, and legitimate (Temperley and Tangen, 2006). This is commonly reflected with regards to this examination, whereby, web-based media influencers' adherents see them as an individual like them with whom they share comparable qualities and interests just as increment the transparency of the product depending on the shared content on social media. Based on these, the following hypothesis is deduced: H7: Social media influencer trustworthiness has a positive influence on attitude toward the ad H8: Social media influencer trustworthiness has a positive influence on attitude toward the ad brand. H9: Social media influencer trustworthiness has a positive influence on purchase intention of cosmetic product

Measurement items
The first part of the questionnaire was constituted by respondents demographics followed by items of advertising effectiveness including an adopted from Bruner & Hensel (1992), attitude toward the brand adopted from MacKenzie & Lutz (1989), and purchase intention adopted from Yi (1990) while attractiveness, expertise, and trustworthiness were adopted from McCracken (1989). All items were anchored on a Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Data Collection
A convenience sampling was used to elicit responses from consumers who have been exposed to social media influencers' ads. The sample size was decided based on G*Power software to calculate the minimum sample size required based on statistical power (Faul et al. 2009). With 6 latent variables, an effect size of 0.15, and power at 0.80, a minimum of 98 is required. A total of 150 questionnaires were distributed and 126 questionnaires (84%) were returned in which exceeded the minimum required number of questionnaires. This also fulfills the recommendation in structural equation modeling (SEM) to use the "10 times" rule of thumb (Hair et al. 2017). This rule suggests that PLS-SEM only requires a sample size of 10 times the largest number of paths or relationships in the structural model, which would call for a minimum sample size of 90 for this model (that is, 10 x 9 structural paths = 90 individuals). Respondents were sampled from several states in Malaysia and the participation of respondents in the study was voluntarily.

Common Method Variance
Common method bias was analyzed since the information about the indicator and measured factors were gathered from a sole source (Podsakoff et al. 2003). A few remedies have been recommended to recognize this issue including Harman's one-factor test. A factorial examination without rotation utilizing SPSS completed and the outcome shown that the principal factor represented just 29.74% of the variance, much lower than the majority signifying common bias isn't an issue in this research.

Profile of Respondents
Based on the data collected, the majority of the respondents are female which represents 64% while male respondents were 36%. The age of respondents ranges from 18 to 57 years.

Measurement Model
Structural Equaling Modelling (SEM) software PLS 3.2.8 was employed for the data analysis purpose. PLS-SEM technique was chosen instead of CB-SEM, is because of the predictive oriented objective of this research. (Hair et al. 2017). The measurement model assessment involves the estimate of reliability and validity. The convergent validity is assessed by observing indicators' loading, Average Variance Extracted (AVE), and Composite Reliability (CR) (Hair et al. 2017). The summarized result in Table 1 confirms that all the loadings values were higher than 0.4, the CR's above 0.7, and AVE scores were larger than 0.5 as recommended by Hair et al. (2017). Hence, the convergent validity of the model's deemed acceptable. Discriminant validity was assessed using the Fornell and Larcker (1981) approach and results indicated that the constructs established discriminant validity.
Moreover, a blindfolding technique was executed to examine the model's predictive relevance (Q2). According to Hair et al. (2014), values of 0.02, 0.15, and 0.35 signify small, medium, and large predictive relevance of an exogenous toward endogenous construct. The Q2 for ad attitude, brand attitude, and purchase intention was .342, .357, and .412 respectively indicating a large prediction relevance of the model.

Conclusion
This study revealed the effects of source credibility such as source attractiveness, expertise, and trustworthiness influence positively consumer attitudes towards ad and brand subsequently purchase intention of cosmetic products. This finding corroborates with previous studies by Chekima & Chekima (2019), Schouten et al., (2020), Singh and Banarjee (2018), Thomas and Johnson (2017), and Muda et al., (2014). These results validated the source credibility from the perspective of social media influencer and its ability to exert a favorable outcome of consumers toward the ad. This research demonstrated that respondents were more likely to trust the social media influencer as well as feel delighted with their product-fit and attractiveness with whom they perceived as a resemblance to themselves.
The finding of this study could help cosmetic marketers to consider selecting social media influencer who possesses tailored credibility sources to gain a competitive advantage in the midst of high rivalry market. These social media influencers must be able to grab the attention of the target audience with effective communication.
Doing so can assist with maximizing the influence of advertising endeavors to create positive social behavior and subsequently increment their sales