THE EFFECT OF STAR ENDORSER ATTRACTION, CREDIBILITY, AND EXPERTISE

Background: Advertising media are frequently employet as promotial media to disseminate information about produducts. The aspect of celebrity endorser's profile is their level of attractiveness, another aspect their credibility and domain knowledge. Purpose: This study aims to examine the effect of attractiveness, credibility, and expertise of celebrity endorsers on purchasing decisions. Method : Experimental research techniques are 59 participants were utilize as the experimental and 60 participants as the control. The data for this study were derived from responses of participants who had received various treatments prior to the distribution of the questionnair. Result : The variable attraction (X1) of the celebrity endorser to the purchase decision (Y) has a value of 0.011 and a significant t-count value of 4.891 at = 0.05 (one-way) t-table value of 1.986. The t-count value > the t-table. The credibility variable (X2) of the celebrity endorser to the purchase decision (Y) has a value of 0.085 and a significant t-count value of 3.426 at = 0.05 (one-way) table t value of 1.986. The t-count value > the t-table. The skill variable (X3) of the celebrity endorser to the purchase decision (Y) has a value of 0.051 and a t-count value of 2.416 is significant at = 0.05 (one-way) t-table value 1986. T-count value > t -table. Conclusion: The results in this study showed that the celebrity endorser variable (X1), credibility variable (X2) and skill variable (X3) had a positive and significant correlation effect on purchasing decisions (Y). on the effect of attractiveness, credibility and expertise of


INTRODUCTION
The development of free trade has led to increasingly fierce business competition. This requires companies to be more creative in carrying out their business activities. This era is marked by the development of various products and services accompanied by new innovations. Various efforts have been made by companies to increase company sales and satisfy the needs of their customers (Heruwati & Khasanah, 2010). Marketing strategies are needed and one of the marketing strategies currently widely used by companies or producers to inform their products to consumers is by promotion. Promotion is an element in the marketing mix that the company uses to market its needs.
Promotion is seen as a one-way flow of information or persuasion made to direct a person or organization to action to create exchanges in marketing (Swastha, 2008). Promotional media that is often used to convey information about products is advertising media. Advertising is one of the media used by companies, it can be classified according to its purpose, namely to provide information, persuade and remember (Kotler, Keller, Brady, Goodman, & Hansen, 2019). Advertising is an efficient way to communicate products because it is able to reach many buyers who are geographically dispersed.
Creative ways to advertise using celebrity endorsers. According to (Kotler et al., 2019) celebrity endorser is the use of sources (sources) as attractive or popular figures in advertisements, this is one creative way to convey messages to get higher attention and can be remembered. Using celebrity supporters must go through several considerations, including the popularity of celebrities with a product whether the chosen celebrity can represent the character of the product being advertised (Frans, 2005). Celebrity is as a source of advertising or information about brands and product attributes that are fun, convincing and attractive general public attention. Using celebrities as endorsers is indeed easier to influence the psychological consumption of consumers. The use of celebrities in an advertisement involves attractiveness and credibility which is unique in itself (Sebayang & Siahaan, 2008).
Celebrity attractiveness is a dimension in celebrity endorsers, the dimensions included in celebrity endorsers are credibility and expertise (Krisjanti, 2006) mean that attractiveness is a number of elements contained in celebrities which include a number of characteristics that can be seen by the audience in supporters such as physical attractiveness, intelligence, personality traits and lifestyle. Celebrities have an attraction in sharing ways that can attract various target audiences to make purchasing decisions, the higher the attractiveness of the Celebrity Endorser, the higher the purchase decision to buy or use the advertised product.
While what is meant by credibility is the skill and value of trust in convincing others about the product being promoted. Credibility celebrities can be trusted and are considered to have insight on certain issues, such as brand reliability, will be the most able to convince others to take an action purchasing decision (Tasruddin, 2015). The higher the credibility of the Celebrity Endorser, the higher the purchase decision to buy or use the advertised product. And what is meant by expertise is the ability of the advertising source, where in this case the celebrity as an advertisement star to make a valid statement about the characteristics of the product being advertised. This supports the relationship with the product or brand they represent (Krisjanti, 2006). So as to attract interest or purchase decisions. The higher the expertise possessed by the Celebrity Endorser, the higher the purchase decision to buy or use the advertised product. In choosing an endorser, it is not arbitrary. There are several considerations to focus on. For example, an endorser is thought to have a strong character, not exposed to negative issues and has a good image in the eyes of the public. So, consumers can be confident and believe in the products brought by the endorser. Indirectly, loyalty arises in the endorser to maintain and promote the brand that has been built by the company. Various kinds of products are promoted by Celebrity Endorser including food, clothing, electronics, and cosmetics. And cosmetics are one of the products that use Celebrity Endorser a lot because cosmetics need media/people to show the results of these cosmetics. Cosmetics are care substances used to enhance the appearance or smell of the human body. Cosmetics are generally a mixture of various chemical compounds, some are made from natural ingredients and can not only be used by women but can also be used by men, cosmetic products also vary from perfume, powder, lipstick, eyeliner etc. The results of research on the effect of attractiveness, credibility and expertise of Celebrity Endorser on purchasing decisions also show different results. The attractiveness, credibility and expertise of Celebrity Endorser have a positive effect on purchasing decisions (Indrayani, Suwendra, Yulianthini, & SE, 2015). Credibility, attractiveness and expertise have a positive effect on purchasing decisions (Nopiyanti, Pratiwi, & Ismail, 2015). The attractiveness, credibility and expertise of Celebrity Endorser have a positive and significant effect on purchasing decisions (Utami, Eka, & Nofiawaty, 2020).
Based on the description above, the author takes the research title "The Effect of Attractiveness, Credibility, and Expertise of Celebrity Endorsers on Purchase Decisions". Problem Statements: The formulation of the problem of this research is the increasingly fierce business competition that affects the increasing demand for a product, this encourages companies to take promotional actions, one of which is through Celebrity Endorser. And prove that the promo using Celebrity Endorser can increase sales.
Research objectives: The objectives to be achieved in this research are as follows: Testing the effect of the attractiveness of Celebrity Endorser on buying decision, testing the influence of Celebrity Endorser credibility on purchasing decisions, testing the influence of Celebrity Endorser expertise on purchasing decisions. Benefits of research: The benefits expected from this research are as follows: Theoretical benefits. The results of this study are expected to provide useful material for the development of science in the field of marketing management related to attractiveness and credibility which also influences consumer decisions in purchasing.
For researchers, this research is one proof that researchers have been able to apply the knowledge in the form of theories obtained during the author's studies into the world of work as well as a place to explore additional knowledge in the field regarding matters related to marketing management. especially consumer decisions in purchasing. For the University, this research is expected to contribute to the literature in the field of marketing. In addition, it is also expected to enrich the development of science in the field of marketing as well as input for further research.

RESEARCH METHODS
Experimental research is research in which at least one variable is manipulated to study cause-and-effect relationships, therefore, experimental research is closely related to testing a hypothesis in order to find the effect, relationship, or difference in changes in the group that is subjected to treatment (Apuke, 2017). Experimental research is research that seeks to find the effect of certain variables on other variables with tight control. Experimental research is a systematic, logical, and thorough research in controlling conditions. Experiments can be interpreted as research methods used to find the effect of certain treatments on others under controlled conditions (Sugiyono, 2015).
The research variable is an attribute or nature or value of people, objects or activities that have certain variations that are determined by the study and then draw conclusions (Sugiyono, 2016). variables are divided into two types, namely independent variables (independent) and determined variables (dependent).
Data analysis method is the treatment of the experiments carried out, to see the correlation of each variable in several experimental groups, and the analytical method used to test the treatment is analysis of variance. ANOVA is used to analyze the difference in variance between two or more groups with one dependent variable. In addition, a covariate test was conducted for the respondent's identity variables, namely gender and age. This test is to identify whether there is a possibility that the variable acts as a new variable that strengthens or weakens the relationship between the independent variable and the dependent variable.
Validity test is used to measure the validity or validity of a questionnaire. A questionnaire is said to be valid if the questions on the questionnaire are able to reveal something that will be measured by the questionnaire (Brown & Moore, 2012). In this study, the validity test used CFA (Corfirmatory Factor Analysis).
The Valid Loading Factor is at least 0.4 and does not correlate with other question items. While the reliability test is used to find out how far the measurement results remain consistent when repeated measurements are made of the same symptoms using the same measuring instrument. Reliable means with the same measuring instrument and the same symptoms, but in different conditions and situations, the measurements show consistent results (Field & Golubitsky, 2009).
Reliability is measured using items corrected for total correlation and Cronbach's alpha which reflects the internal consistency of the measuring instrument. The rule of thumb used is Cronbach's Alpha, each item must be greater than 0.6. In this reliability test it is possible to eliminate items.
This study uses a partial hypothesis test ("t" test), the t-test is used to determine the contribution of each independent variable partially to the dependent variable, using the test of each independent variable regression coefficient (attractiveness, credibility and expertise) whether or not it has a significant effect on the dependent variable (Sugiyono, 2016). The size of the hypothesis support can use the comparison of the T-table and Tcount values. If the T-count is higher than the T-table value with = 5%, then the hypothesis is supported or accepted (Ghozali, 2018).

RESULTS AND DISCUSSION
Participants in this study were Makassar, South Sulawesi-Indonesia, all 119 of the distributed questionnaires can be used for additional analysis after screening based on responders' responses, and each participant fills out a full profile. As a result, the study's overall response rate was 100%. Respondents for the distribution and participants' responses for the attractiveness variable included 4 respondents for all participants who were 17 years old or 3.4%, 4 respondents who were 22 years old or 3.4%, 34 respondents who were 20 years old or 28.6%, 27 respondents who were 18 years old or 22.7%, 30 respondents who were 19 years old or 25.2%, and 20 respondents who were 21 years old or 16.8%. The answer to the question "The appearance of the ad star in the commercial is really gorgeous" got the highest average score. The average response from participants was 5.56. The average of the lowest responses is "the product's ad star has the same personality (personality) as I want," while the typical response to this query is 3.74. As Shown Table 1.

Volume 2, Nomor 10, Oktober 2022 p-ISSN 2774-7018 ; e-ISSN 2774-700X
http://sosains.greenvest.co.id Participants' responses to the credibility variable's distribution. "The advertising star has a compatibility with the product; hence it has a positive value for the advertised product," is the question item with the highest average value. The average participant gave a score of 5.00 in response to the query. While "advertising stars are honest in transmitting advertising messages" is the average of the least responses. The typical response to this query is 3.76. As show on Table 2 Table 2  The distribution and participant responses for the skill variable are shown in Table 4.9. "The celebrity has experience as an advertisement star" is the question item with the highest average score. The average response from participants was 5.23. The least popular response, although being the average, is "The ad star has frequently appeared in the media to pitch other items." There are 4.46 average responses to questions. Distribution and participants' responses for the variables influencing purchasing decisions. The query, however, "After seeing the star of the product advertisement, I have contemplation of the benefits of buying the product," has the greatest average score. 4.13 is the typical participation response to the query. While the most common response is "I am certain to buy the goods after witnessing the advertising celebrity promote it." The typical response to this query is 3.55. displayed in Table 4.   The probability result of 0.000 is less than 0.05, all correlation values for the total variable X and the total variable Y are positive, as shown in Tables 6 and  Table 7. Based on the validity test results shown in the table above. This is in keeping with the decision-making premise, which stipulates that if the findings of the probability 0.05 correlation are positive. The results of the reliability test and the reliability statistics in Table 8, the reliability coefficient [Cronbach's Alpha] value per variable was 0.770 for [X] and 0.825 for [Y]. The test results are consistent with the reliability test's basic decision-making, which states that if Cronbach's Alpha value is > 0.70, the questionnaire will be declared reliable or consistent. The results of the reliability test can be continued to carry out in the next test, i.e., the classical assumption test.

Volume 2, Nomor 10, Oktober 2022 p-ISSN 2774-7018 ; e-ISSN 2774-700X
The basic decision-making of the reliability test is if the Alpha value of Cronbach is > 0.70, the questionnaire will be revealed well-founded or consistent, conversely, if the Alpha value of Cronbach is < 0.70, the questionnaire is considered inconsistent.

CONCLUSION
Based on the results of data analysis and the results of hypothesis testing on research questions, the conclusions of this study The attractiveness variable (X1) celebrity endorser has a positive and significant effect on purchasing decisions (Y). This is evidenced by the results of hypothesis testing which shows that the relationship has a value of 0.011 and a t-count value of 4.891 which is Volume 2, Nomor 10, Oktober 2022 p-ISSN 2774-7018 ; e-ISSN 2774-700X significant at = 0.05 (one way) t-table value of 1.986. Value of t-count > t-table. The credibility variable (X2) celebrity endorser has a positive and significant effect on purchasing decisions (Y). This is evidenced by the results of hypothesis testing which shows that the relationship has a value of 0.085 and a t-count value of 3.426 which is significant at = 0.05 (one way) t table value of 1.986. Value of tcount > t-table. The skill variable (X3) celebrity endorser has a positive and significant effect on purchasing decisions (Y). This is evidenced by the results of hypothesis testing which shows that the relationship has a value of 0.051 and the t-count value of 2.416 is significant at = 0.05 (one tailed) t-table value 1986. The value of t-count > t-table.
Further research is recommended to use a larger number of samples considering that this study uses a minimal number of samples. in future research, it is possible to add latent variables and dimensions to existing models and to models that are undergoing development. Periodically evaluate the image of celebrity endorser advertisements so as to get results whether the celebrity endorser needs to be maintained or must choose a new celebrity endorser in accordance with the developing trend by monitoring his personal life whether it is still in demand by consumers or not, so that consumers are not bored with the character. BIBLIOGRAPHY