Celebrity Endorsement’s Effects on Youths’ Cosmetic Buying Behavior: A Study in a Malaysian University

The Malaysian cosmetic industry is currently operating in a competitive market. For brands to stand out and gain a competitive advantage over others, marketers are turning towards the use of celebrity endorsement in advertisements. Therefore, it is important for advertisers of cosmetic brands to be aware of the properties in celebrity endorsement that can influence their customer’s buying behavior. The objective of this study is to investigate the significant effects of celebrity credibility, attractiveness and popularity on female youth’s cosmetic buying behavior. This quantitative cross sectional study utilizes data obtained from a personally administered questionnaire. A sample of 253 female youths from a university was chosen using the convenience sampling method. Various statistical analyses were conducted via the SPSS statistical package. The findings indicate that the properties of celebrity attractiveness and popularity had positive and significant effects on consumer buying behavior, however, celebrity credibility did not. This suggests that female youths are more concerned about the physical attractiveness and likeability of the endorsers when making decisions, whilst not concerned about the credibility of the endorser such as expertise and trustworthiness. The study concludes with the discussion and implications as well as limitations and suggestions for future re-search.


Introduction
The cosmetic industry is a multi-million dollar industry worldwide. It is no different in Malaysia. The robust beauty industry in Malaysia saw consumer expenditure on cosmetics estimated at US$58 million for the year 2018. Revenue is expected to show an annual growth rate (CAGR 2018-2022) of 9.4% resulting in a market volume of US$83m in 2022 (Statista, n.d).
The Malaysian cosmetic market is highly competitive, with many leading brands competing in the market, notably those from France and Switzerland. Increasingly, South Korea, Japan, Spain, Italy, Australia and New Zealand have increased their foothold in Malaysia. Swiss and French brands are generally perceived as the top end of the scale in terms of pricing, prestige and efficacy, whilst Spain an Italian cosmetics are regarded as a cheaper alternative. Australian and New Zealand brands are considered favorably by Malaysian due to their geographical proximity and eco-friendly image. The Japanese has been first choice for all things related to skin whitening and the popularity of K-pop and K-drama, has made South Korean cosmetics attractive, especially to the younger female consumers. On top of imported cosmetic brands, there are also more than 210 local cosmetic manufacturers in Malaysia who conform to the Good Manufacturing Practices (GMP) requirement in accordance to the ASEAN Guidelines for Cosmetics (U.S Department of Commerce, 2016).
Clearly, the Malaysian cosmetic market is saturated with plenty of choices for consumers to choose from. Thus, brands competing in this market need to stand out from their competition and attract the attention of consumers to purchase their products. Advertisements placed in the media and magazines are the most effective ways to increase the purchasing rate of cosmetics products among consumers (Hassali et al., 2015). Of late, celebrity endorsements in advertisements have been used as an effective marketing strategy to increase sales.
The majority of cosmetic product users are female with youths tending to be more interested in maintaining their self-image (Mansor et al., 2010). Poghosyan (2015), state that the younger consumers especially, will be easily influenced by the celebrity endorsement in advertisements. Celebrity endorsement works particularly well with younger consumers because they relate closely to endorser qualities (Liu et al., 2007), and use these celebrity endorsers as role model models (Martin and Bush, 2000).
As cosmetic brands are competing for sales using different advertising strategies, this study proposes that female youth's cosmetic buying behavior could be influenced by the properties of celebrity endorsement which are celebrity credibility, celebrity attractiveness and celebrity popularity. Kotler and Armstrong (2012), defined consumer buying behavior as how individuals select, purchase, use and dispose of products or services to meet their demands and needs. Proctor and Stone (2002), postulate that the consumer buying process is complex and marketers that understand this process will be able to determine better marketing decisions. In a similar vein, Schiffman et al. (2001), state that marketers will be able to predict how consumers are likely to react to various external informational and environmental cues and to use them to craft their marketing strategies to their advantage. Marketers that have a better understanding of factors that influence their customers buying behavior will have a greater competitive advantage in the marketplace.

Consumer Buying Behavior and Celebrity Endorsement
To influence consumer's buying behavior, marketers generally use advertisements, a form of paid messages by the company or brand. Celebrity endorsement has emerged as one of the most successful and effective advertising tool. It has become a trend for companies and consumers to perceive celebrity endorsement as a winning prescription for increasing sales (Nasir et al., 2016). Keller (2012), defines celebrity endorsement as a type of a brand promotion where a famous person is used in the marketing campaign to advertise the product or service by using his or her fame and place in society.
Celebrity endorsement works through the process of meaning transfer (McCracken, 1989), The symbolic properties belonging to the celebrity endorser will be transferred to the specific product, and then from the product to the consumers.
There are many benefits for using celebrity endorsement in marketing a product. Previous research indicate that celebrity endorsement has an impact on the effectiveness of an advertisement, assists in brand awareness and recall, generate a higher likelihood of the consumer choosing and buying the endorsed brand (Spry et al., 2011). Khong and Wu (2013), suggest that celebrity endorsement makes a product more noticeable, appealing and more compelling to the consumer. Ahmed et al. (2014), state that there is a significant positive relationship between celebrity endorsement and buying behavior. Their study found that consumers were more prone to buy products endorsed by celebrities as they were influenced by the celebrities in the advertisements.
Based on this premise, this study proposes that the properties of celebrity credibility, attractiveness and popularity have significant effects on consumer buying behavior.

Celebrity Credibility
According to Ohanian (1990), the celebrity's credibility refers to the extent to which the celebrity is perceived as possessing expertise and trustworthiness relevant to the communication topic at hand and can be trusted to given an objective opinion on the subject.
If the information is from a credible source (the celebrity) can influence beliefs, attitudes, opinions and/or behavior through the internalization process (Erdogan, 1999), Similarly, according to Elmira and Chloe (2017), celebrities are able to transform an unknown product into a well-known product through persuasion techniques and generating positive associations via advertisements using credible celebrities.
A number of studies have found that celebrity credibility has a significant positive effect on consumer buying behavior. The study conducted by Ahmed et al. (2014), amongst students in a university in Pakistan confirmed celebrity credibility as a predictor of consumer buying behavior. The study by Adam and Hussain (2017), on female consumer's cosmetic buying behavior, found that the credibility of the celebrity was important as the expertise and trustworthiness of the endorser made the consumer trust the brand more, thus influencing their purchase intention. Similarly, the study by Chang et al. (2012), found that credible endorsers were perceived with higher levels of trust and familiarity by the consumers, thus, influencing their buying behavior positively towards the brand. Zahaf and Anderson (2008), and Goldsmith et al. (2000), concur, that the higher the celebrity's credibility, the higher the influence on consumer's buying behavior.
Based on these findings, this study constructed the following hypothesis: H1 Celebrity credibility has a significant and positive effect on consumer buying behavior.

Celebrity Attractiveness
Attractiveness refers to the extent that engage and captivate people's attention. It isn't limited to just the measure of the body, but encompasses all psychical attributes such as hair shading, facial highlights and outward physical appearance (Au-Yeung, 2012).
Besides this, attractiveness constitute any quantity of upright characteristics that customers may recognize in the endorser such as intellectual capacities, personality attributes, lifestyle properties, athletic prowess and competence of endorsers (Shimp, 2003).
Attractiveness affects consumer's mindsets because they want to look alluring, chic, beautiful and glamorous like the celebrity and they believe that by purchasing and using the product, it will help them look attractive too. This is especially applicable in cosmetic advertisements (Malik and Qureshi, 2016).
According to Wan (2018), celebrities who are popular, famous, trendy, attractive and/or fashionable are powerful and influential. A study conducted by Priscilla and Michael (2017), show that female respondents increased their interest in an advertising when the celebrity was attractive.
Existing research indicate physically attractive celebrities are more influential in influencing consumer's buying decision. The study by Ahmed et al. (2015), amongst university students in Pakistan found that celebrity attractiveness influenced consumer buying behavior, whilst Liu et al. (2007), study amongst students in Peking University, China concluded that the attractiveness of the celebrity had significant influence on purchase intentions on sport products. The study by Malik and Qureshi (2016), concur, and concluded that celebrity attractiveness had significant effects on consumer buying behavior.
Based on the body of past research, this study proposes the following hypothesis: H2 Celebrity attractiveness has a significant and positive effect on consumer buying behavior.

Celebrity Popularity
Popularity is the fact that something or someone is liked, enjoyed, or supported by many people (Cambridge Dictionary, 2016). Scott and Judge (2009), states that popularity is an obvious phenomenon that significantly influences the way individuals are viewed and treated by others. This indicates that the higher the degree of someone being popular, the higher the positive feelings of audiences towards the person. If the information is from a popular celebrity, it can influence the consumer's buying behavior of the endorsed brand.
According to Liu et al. (2007), "celebrity chasing" is popular in China, especially amongst the young Chinese consumers. As long as the person is popular and famous, consumers will like him or her very much. This might be the reason why consumers in China care more about "who they are" rather than "what they do." A celebrity's popularity is a major consideration when brands choose which celebrity to endorse their products. This is because past studies have concluded that the achievement of higher sales is related with the higher degree of attention and recall, which is linked to the celebrity's popularity. The study by Rai and Sharma (2013), in India concluded that the popularity of the celebrity had a significant effect on consumer buying behavior. Deepa and Giridhar (2018), in their study amongst college students in India found that celebrity popularity has a significant and positive influence on student's buying behavior. Lastly, the study by Dhaka and Bdgujar (2017), also conducted in India, found similar results, by concluding that celebrity popularity had a significant positive effect on consumer buying behavior.
Based on the body of research and findings, this study constructed the following hypothesis: H3 Celebrity popularity has a significant and positive effect on consumer buying behavior.

Figure-1. Research framework
The research framework of this study is as illustrated in Figure 1. The framework illustrates the relationships between the variables and hypotheses of this study.

Research Methodology
The research design for this study was a quantitative cross sectional design. As postulated by Mansor et al. (2010), and Poghosyan (2015), young female consumers are more prone to purchase and use cosmetic for beautification purposes. The Malaysian (National Youth Development Policy, 1997), categorized youths between the ages of 15-40 years old. Hence, the population of study consisted of female students enrolled in a university in Perak. Purposive convenience sampling design was used to collect the 253 samples. The inclusion criteria for the sample were that they be female youths who have at least one experience in purchasing cosmetics before.
A structured personally administered questionnaire was used to collect the data. Reliable and valid scales were adapted from previous research and used to capture the responses of the respondents. The scales were anchored on a five point Likert scale with (strongly disagree = 1 to strongly agree = 5).
Out of 300 questionnaires distributed, all were returned as they were collected on the spot by the researcher upon completion. Out of the 300, only 253 were answered completely and usable for data analysis. The data was analysed using various statistical analyses to determine its validity, reliability and to answer the hypotheses of this study.

Profile of Respondents
The profile of the respondents is presented in Table 1, based on frequency distributions and percentages. Majority of the respondents, were aged between 20 to 25 years of age. 98% of the respondents were Chinese and lastly, 68% of the respondents were from the Public Relations program.

Factor Analysis
Factor analysis via principal component analysis with varimax rotation was used to validate the construct validity of the questionnaire. The objective of conducting factor analysis was to ensure that the measurement dimensions were operationally defined, to ensure that the most appropriate items in the measurements were used as well as to reduce unsuitable items that might affect the validity of the study. As suggested by Hair et al. (2010), the factor loadings for all independent variables were above the accepted level of 0.5, while cross loadings for all independent variables were below the accepted level of 0.35. The analysis extracted four factors with eigenvalues greater than 1. There were celebrity credibility (Eigenvalue = 2.21, variance explained = 55.22%). Celebrity attractiveness (Eigenvalue = 2.89, variance explained = 57.86%), celebrity popularity (Eigenvalue = 2.53, variance explained = 50.51%) and consumer buying behavior (Eigenvalue = 2.19, variance explained = 54.79%).

Reliability Analysis
The reliability of the measures was tested after factor analysis was conducted. The results in Table 2 shows that the items in the measurements are reliable as the Cronbach's alpha coefficient score is above 0.70, as recommended by DeVellis (2012).  Table 3 shows the results of the regression analysis which was used to test the hypotheses of this study. The results indicated that the F value of 12.78 is significant at a confidence level of 0.05, and R² 0.13 indicated that the three independent variables could explain 13% of the variance in consumer buying behavior. As recommended by Pallant (2013), the result show that there is no serious multicollinearity problem in the model which is indicated by the small values of the VIF (Variance Inflation factor) ranging from 1.17 to 1.28, which is well below the recommended 10.0. The tolerance levels are also higher than 0.10, ranging from 0.78 to 0.85. The Durbin Watson value of 1. 658 falls within the acceptable range of 1.5 to 2.5, implying that there are no serious auto-correlation problems in the error terms.

Multiple Regression Analysis
In terms of effect size, based on Cohen (1988), conventions, combined effect of this magnitude by the three independent variables in this study can be considered medium.
The results of the regression analysis have found that celebrity popularity and attractiveness have significant positive effects on consumer buying behavior, whilst celebrity credibility did not.

Discussion and Implications
This study has identified that celebrity attractiveness and popularity had a significant effect on consumer buying behavior. This finding is broadly consistent with past research such as Ahmed et al. (2015); Deepa and Giridhar (2018); Dhaka and Bdgujar (2017), and Malik and Qureshi (2016). This findings suggest that female youths are more concerned with the physical attractiveness and likeability of the celebrity endorsers, and would be influenced to purchase the brands endorsed. As postulated by McCracken (1989), celebrity endorsement works through the process of meaning transfer. The properties of attractiveness and popularity had been transmitted from the celebrity to the brand being endorsed. Hence, female youths who purchased the cosmetic brand would feel attractive and popular after using the product.
In spite of strong previous research that support celebrity credibility as a predictor of consumer buying behavior, the results of this study did not support this hypothesis. This finding suggests that female youths are less concerned with the credibility of the endorser when making their purchase intentions towards a cosmetic brand; hence the trustworthiness and expertise of the endorser are not important to them. The study by Mansor et al. (2010), found that Malaysian female consumers were less concerned about the effects of the cosmetic on their skin and the ingredients of the cosmetic product, than the brand of the cosmetic. This suggests that if the consumer is not concerned about these factors, it would be highly unlikely that the trustworthiness of the endorser would be important to them.
The findings from this study have several implications. As huge amounts of money are spent by marketers for advertising campaigns using celebrity endorsements, it is imperative that the correct celebrity be chosen to ensure return on investment. Cosmetic brands must choose the correct celebrities for their target audience, especially since every celebrity endorser transfers their meanings onto the product endorsed. Thus, if marketers of cosmetic brands are targeting to sell to female youths, the properties of attractiveness and popularity of the celebrity is paramount, whilst credibility is of less importance. This study finds that Malaysian female youths tend to identify with current hip and popular celebrities which they perceive as being beautiful, have charming personalities and lead exciting and glamorous lifestyles, all of which they wish they could emulate.

Limitations and Suggestions for Future Research
The limitations of this research were largely associated to the generalizability of the findings and the cross sectional nature of the research design. There are several directions for future research. Different age groups and inclusion of male gender can be incorporated for even generalized response. Future research should also look at other variables not included in this study such as the celebrity's gender, profession, performance and image. As this study was conducted using the cross sectional design, future research could be conducted as a longitudinal study so that relationships between consumer buying behavior and celebrity endorsement be more accurately revealed. The framework for this study could be utilized with other product categories to increase the generalizability of the results.

Conclusion
The level of competition amongst cosmetic brands in Malaysia is on the rise, and it is imperative for management of these brands to determine suitable advertising strategies to overcome this problem. Even though celebrity endorsement has proven to be effective, it is not the magic bullet that it has been made out to be.
This study concluded that the properties of celebrity popularity were the most influential on female youth's cosmetic buying behavior, followed by celebrity attractiveness. Celebrity credibility was found to have no effect on female youth's cosmetic buying behavior.
This study has significance for advertisers of cosmetic products. These findings explain the impact of certain properties of celebrity endorsement on consumer buying behavior, hence providing marketing managers of cosmetic brands with a deeper understanding of their customer's decision making and buying process. Thus, they will be better equipped in devising more profitable advertising campaigns in the future which utilizes celebrity endorsements.