The Effect of Celebrity Endorsement Towards Consumer Purchase Intention

Chapter I

Introduction

• Background of Study

 Tight business competition is now demanding and forcing a company to have advantages such as superior product quality, excellent in the services offered, and so forth. The company's​​ benefits must be well managed so that consumers can know and understand the advantages of a company's products or services. Advertising is one of the right ways for companies to introduce their products or services. Advertising is considered very effective​​ in increasing consumer awareness of the existence of a product, service or brand.

 Kasali (2010) defines advertising as a message offering a public product through media. Meanwhile, according to Morisan M.A (2015: 18), advertising is one of the best known​​ and most talked-about forms of promotion; this relates to broad reach. From the understanding, according to Kasali, it can be interpreted that advertising is the primary source of information that can influence the public or consumers in making purchasing​​ decisions. This advertising activity is one way most companies certainly do it because advertising has a​​ broad reach, persuading, and decisive factor in influencing consumer tastes and intentions. Advertising is an effective and efficient element in building the image of a product or service used to increase consumer purchasing decision-making (Kurniawan and Kunto, 2014).

 In advertising a product, many media types can be done, for example, print and electronic media. One of the advertising media that is often used by companies today is through the internet. The development of technology is very rapid; mainly, the internet provides many companies' contributions to advertising their products. With the internet and the many social media that have sprung up like Facebook and Instagram, it has become easier for companies to promote and introduce their products to the general public. The use of the internet as an advertising medium is carried out considering that company advertisements must be visible to the public with a wide area and reach the public quickly, inexpensive, and attractive. Several other companies also use artists to support and introduce their products through social networks like Instagram. Celebrity endorsement is a common advertising strategy.

 Marketing and advertising are two inseparable things to convey a persuasive message (Ohana, 1990). In introducing a product or​​ service often, companies use artists to help advertise their products, which we often hear with the term celebrity endorsement.​​ Celebrity endorsement is considered an effective way for companies to introduce their products or services to the public, hopefully increasing consumer purchase intentions and leading to actual purchases (Kurniawan and Kunto, 2014).​​ 

 Royan (2014) states that celebrities' use in advertising can help shape two things, namely forming a brand personality or increasing sales volume in the short term. Celebrities are concrete manifestations of various images or associations that consumers think of a brand. Celebrity endorsements are someone famous or someone who has the visibility, credibility, Attractiveness, and strength in communicating messages in a product to attract consumers' attention, increasing the purpose of buying consumers for the advertised products​​ (Wijaya, 2015). If a brand's word is associated with a beautiful and young brand, then celebrities must represent all of these associations. For a marketer, brand personality is critical because, like humans, a character makes a product's brand look different from other brands. While competitors will easily imitate features and prices, personality is more difficult to replicate.

 The role of celebrities in advertising a product is very influential if a star has credibility supported by Expertise,​​ Trustworthiness, and liking (Royan, 2014). In marketing, celebrities' testimonials become a benchmark for consumers in determining what products they will buy (Ohanian, 1990). The selection of stars in conveying messages about products is expected to impact brand awareness and recognition. According to Hardiman (2016: 38), an endorser is interpreted as the person chosen to represent the image of a product, usually from among potential consumer figures who have prominent characters and strong aa product's image endorsements that have promising careers, high popularity, and have a lot of fans will be straightforward in increasing consumer awareness of endorsed products. The suitability of the image and popularity of celebrities must be following the brand that​​ is being carried. For example, a girl group artist from South Korea, BlackPink, which has recently been in the spotlight, has a strong and energetic character that has made several companies interested in using them as iconic products of companies such as​​ Pepsi, Samsung, Adidas. Because BlackPink has many fans, consumers will quickly experience the products of the companies they are working with, especially fans. Due to the high popularity of​​ celebrities, their collaborative products are highly anticipated, especially fans. A celebrity usually does a lot of advertisements that can change consumers' perceptions of stars as a product or brand endorser.

 Companies should be more focused and selective in deciding celebrities who will become endorsers of their products or brands. Companies must be able to review who they are, the size of the product range, the image they want to build related to the product, the compatibility between the product image and the endorser must be done carefully by the company to determine the maximum and appropriate results. Gupta et al. (2015) state that one of the models used to evaluate celebrities as endorsers is to measure endorsers' credibility by using three dimensions, namely Attractiveness, Expertise, and Trustworthiness. Attractiveness is related to the physical attractiveness or celebrity characters that stand out, such as personality, lifestyle, attitudes, etc. Expertise is associated with the experience, knowledge, or skills possessed by a celebrity or endorser for the products it offers. At the same time, Trustworthiness is related to honesty, integrity, responsibility. It can be trusted by celebrities (the endorser) as a source.

 Understanding the importance of the relationship between celebrities and products to increase​​ consumer purchase intentions of their brands or products triggers company management to use appropriate stars and under the company's products or brands as endorsers. One of the celebrity endorsements of the widely used homeland is Karin Novilda, with the​​ social media account Instagram @awkarin. The beauty and Attractiveness of Karin Novilda would not doubt that this is what makes many companies, especially on social media Instagram, use and offer to collaborate with Karin Novilda. Of course, Karin's popularity is expected to attract the public's interest to buy company products advertised through Instagram social media.

Figure 1.1

Karin Novilda products and advertising models

(Source: @awkarin, accessed 31 March 2020)

  Millennials (also known as​​ Generation Y, Gen Y, or Generous) are demographic groups after Generation X (Gen-X). There is no definite time limit for the beginning and end of this group. Experts and researchers usually use the early 1980s as the beginning of this group's birth and the​​ mid-1990s to early 2000s as the end of birth​​ (https://id.wikipedia.org/wiki/Generasi_Z).

  Generation Y or millennial is the generation after generation X whose life begins to be surrounded by increasingly evolving technology. Whatever the​​ activity, most of them are related to cyberspace, technology, the internet, and social media, especially Instagram. Since childhood, they are familiar with technology and familiar with sophisticated gadgets that indirectly affect their personality. The use​​ of technology and the internet is developing quite rapidly, making the internet a medium for getting information more quickly, especially for the current generation. 88% of internet users use the internet to get information related to shopping, 61% choose​​ to shop online, and 38% choose to use conventional methods (http://www.republika.co. Indo). Technology or social media can change the lifestyle, especially for the millennial generation in shopping. Nowadays, many consumers shop online through social media like Instagram because it is considered easy, fast, and saves time.

  Salzman (2010) in Ogle et al. (2013) found evidence that adolescents in America are a sizeable potential market because they use many internet media to find information and buy products. Seeing the lifestyle changes that occur in the current generation in​​ using technology and social media makes companies or other marketers adopt social media, especially Instagram, to sell products and use celebrity endorsement to attract consumer attention. Of course, this is an exciting thing to discuss in studying the behaviour of Generation millenial in Indonesia, especially those who live in Surakarta.

• ​​ Formulation of​​ The Problem

Based on the background above, the problems formulated in this study are as follows:

• Does the Attractiveness of celebrity endorsers have an impact on purchase intention?

• Does the Expertise of celebrity endorsers have an impact on purchase intention?

• Does the Trustworthiness of celebrity endorsers have an impact on purchase intention?

• ​​ Research​​ Purposes

The objectives of this research are:

• To analyze the influence of Karin Novilda's Attractiveness as a celebrity endorsement on consumer purchase intentions in products advertised by her on Instagram social media.

• To analyze the influence of Karin Novilda's Expertise as a celebrity endorsement on consumer purchase intentions in products advertised by her on Instagram social media.

• To analyze the influence of Karin Novilda's Trustworthiness as a celebrity endorsement on consumer purchase intentions in products advertised by her on Instagram social media.

• ​​ Benefits of Research

The results of this study are expected to provide help to several parties, including :

• For Companies

 This study's results are expected to help several​​ companies or marketers determine celebrity endorsement used by knowing the variables associated with buying intentions in consumers, which is the basis for choosing the company's right marketing strategy.

• For Writers

 This research is one of the media in applying marketing management theories that have been obtained on the lecture bench in the real business world.

• For Other Parties

 This study's results can undoubtedly be read and used as a reference for other parties interested in adding an insight or those in need, especially in marketing management.

• ​​ Systematic​​ Research

Chapter I: Introduction

  This chapter explains the problem's outline to be investigated, such as background, problem formulation, problem boundary, research objectives, research benefits,​​ and systematic research.

Chapter II: Literature Review

  This chapter contains a description of the theories concerned and supports celebrity endorsement and purchase intentions in consumers and the research framework, previous studies, and hypotheses.

Chapter III: Research Methodology

  This chapter contains the methods used in this study, consisting of population, form and time of research, samples and sampling methods, data collection methods, data measurement methods, data testing methods, and data analysis methods.

Chapter IV: Data Analysis and Discussion

  This chapter discusses the analysis and testing of data on the respondents' answers from filling out the questionnaire distributed and the research results.

Chapter V: Closing

  This chapter​​ contains conclusions, suggestions proposed by the authors, and research implications that can be used as material for other interested parties.

Chapter​​ II

Literature Review

• • ​​ Consumer Behavior

According to Sunyoto (2012), consumer behaviour can be​​ defined as the activities of individuals involved in obtaining and using goods or services, including the decision-making process in preparation for the determination of these activities.​​ 

Gunawan in Awaliyah (2017: 14) defines consumer behaviour is a study of how individuals, groups, and the organization or institution chooses, buys, and uses goods, services, ideas, or experiences to satisfy consumers' needs and desires. Whereas Kotler and Armstrong (2018: 158) state that consumer buyers' behaviour refers​​ to individual end​​ consumers and households who buy goods and services for personal consumption.​​ 

Many definitions of behaviour, but the same only different formulation, consumer behaviour has a particular interest for people with various reasons to influence or change the action, including people whose primary interest is marketing. AMA (American Marketing Association) defines consumer behaviour as a dynamic interaction between cognition, affection, behaviour, and the environment in which humans carry out exchange activities in their lives. This definition contains three essential things, namely:

• Consumer behaviour is dynamic, so it isn't easy to guess.

• Involves interactions: cognition, affection, behaviour, and events around the consumer's environment and involves exchanges, such as exchanging a seller's goods for buyer's money.

Based on the description above, consumer behaviour can be concluded as an activity carried out by individuals to obtain goods and services and the decision to purchase and use goods​​ and services.

• Factors That Affect Consumer Behavior

The factors that influence consumer behaviour have a significant influence on consumers in choosing products. According to Supranto and Limakrisna (2007), external factors include culture, sub-culture, demographics, social status, reference groups, family, and marketing activities. Internal factors include perception, learning, memory, motives, personality, emotions, and attitudes.

According to Kotler and Armstrong (2018: 159-173), factors that​​ influence consumer behaviour consist of cultural, social, personal, and psychological.

• Cultural Factor

• ​​ Culture

 Cultural, sub-cultural, and social classes greatly influence behaviour consumer purchases. Culture is a fundamental cause of desire and one's behaviour. Each culture consists of several smaller sub-cultures that provide more specific identification and socialization for their members.

• ​​ Subculture

 A group of people with shared value-based system experiences and general life situations, every culture contains smaller subcultures or groups of people with value systems based on experience and public life situations. Subcultures include nationalities, racial groups, religions, and geographical territories.​​ 

 Many subcultures form essential​​ market segments, and marketers often design products and marketing programs tailored to their needs.

• ​​ Social Classes

 A relatively permanent and regular division in a society, its members have the same values, interests, and behaviours. Class social is not​​ determined by a single factor, such as income, but measured as a combination of work, income, education, wealth, and other variables. In some social systems, members of different classes were raised for specific roles and could not change their social position.

• ​​ Social Factor

Social factors that influence purchasing behaviour, such as:

• ​​ Reference Group

A person's reference group is that all money groups have a direct (face to face) or indirect effect on their attitude or behaviour.

• ​​ Family

The family is the most critical consumer purchasing organization in the community, and family members are the leading influential reference groups. There are two families inside the buyer's life, namely: family orientation consisting of parents and siblings, procreation family, namely spouse, and children.

• ​​ Social Roles and Status

People participate in many groups, families, clubs, and organizations. Groups are often important sources of information in helping to define behavioural norms. We can determine a person's position in each group where he is a member-based on role and status.

• ​​ Personal Factor

 When personal characteristics also influence purchasing decisions. Individual factors include age and stages in the buyer's life cycle, work and economic​​ circumstances, personality and self-concept, lifestyle, and values.

d.​​ Physicological Factor

 Purchasing decisions are also influenced by psychological factors such as motivation, perception, learning, trust, and behaviour.

• Consumer Behavior Model

  Consumer behaviour is fundamental to be understood by a company to market its products correctly. Every consumer's behaviour has many differences, but on the other hand, it has many similarities, so that it needs to be a concern for marketers.​​ 

  A marketer who understands consumer behaviour will estimate how likely a consumer's attitude to the information it receives. Kotler and Keller (2016: 187) state that the model of consumer behaviour is presented in the figure as follows:

Figure 2.1

Consumer Behavior Model

Source: Kotler and Keller (2016: 187)

2.2 Marketing Mix

A marketing mix is a tool that can be used by companies as a strategy to get a positive response from the target market. A marketing mix is several tools that can be used by​​ companies as a strategy to get a positive response from the target market. According to Malau (2017: 10), the marketing mix integrates planning marketing activities through four components, namely products, prices, promotions, and distributors that are combined.

The opinions of leading experts also support this:

 ​​ ​​ ​​​​ "The primary marketing mix tools are classified into four broad groups, called the four P's of marketing: product, price, place, and promotion. To deliver on value propositions, the firm must first create a need-satisfying market offering (product). It must then decide how much it will charge for the offering (price) and how it will make the offer available to target consumers (place). Finally, it must engage target consumers, communicate about​​ the offering, and consumers of the offer's merits (promotion) (Kotler and Armstrong, 2018: 38).

According to Jerome Mc. Carthy is quoted by Kotler and Keller (2016: 47), "Various marketing activities become a marketing mix tool of four broad types, which he calls marketing 4Ps: product, price, place, and promotion."

The following are the elements of the marketing mix referred to from Kotler and Armstrong (2018, 77-78) :

• Product is a combination of goods and services offered to the target market to meet consumers' needs and desires.

• Price is the amount of money that consumers must pay to obtain a product.

• Promotion refers to the company's activities in communicating its advantages and inviting the target market to buy the product.

• The place is part of the company's activities that make products available to consumers.

• • Advertisement

Companies not only make products that are good and attractive to me, but they must also be able to inform consumers about the functions and advantages of the products they​​ offer. Therefore, companies must be experts in making promotions. Promotion is intended to attract attention and get new customers, and can retain existing customers. One of the promotional tools is advertising.

Advertising is one of the marketing stages,​​ which are interrelated and crucial as the other marketing stages. Suppose the use of the advertising stage is forgotten. In that case, it will make the company's market unstable and weak, causing the company to collapse because​​ there is no introduction to​​ the company's needs to the public. That's what makes advertising is one of the essential stages of marketing.

• Definition of Advertising

Advertising, according to the AMA (American Marketing Association) in Malau (2017: 85), "Advertising is a form of communication that is set up to disseminate information about the uses, advantages, or advantages of a product to generate a desire to make a purchase."

According to Kotler and Keller (2016: 631), advertising is any paid form non-personal presentations and promoting ideas, goods, or services by an identified sponsor. Meanwhile, according to Morisan M.A (2015:18), advertising is one of the best known and most widely discussed forms of promotion; this is likely due to its broad reach.​​ 

Based on several definitions​​ of the above advertisement, it can be concluded that the ad is all kinds of paid communication in which the sponsor will distribute and present the ad, which contains the company's products.

• ​​ Advertising Purpose

The purpose of the relationship is a​​ specific communication task by defining specific audiences or potential customers that we want to​​ reach within a particular time. The overall goal of advertising is to help build customer relationships by communicating value to customers.

There are five advertising objectives: informative advertising, persuasive advertising, advertising reminders, and advertising reinforcement. For more details, here is an explanation of the purpose of advertising, according to Kotler and Keller (2016, 609):

• Informative Advertising

 To create brand awareness and knowledge about new products or new features of existing products.

• Persuasive Advertising

 To create preferences and confidence in purchasing a product from a product or service. Some persuasive ads are comparative ads, which compare the attributes of two or more brands.

• Advertising​​ Reminders

 To stimulate repeated purchases of products and services.

• Advertising Reinforcement

 To convince buyers that they made the right choice.​​ 

• Advertising Indicator

Required by​​ marketing managers in advertising according to Kotler and Keller (2016: 609) or known as 5M, namely:

• Mission (sales goals, advertising objectives)

• Give information

• Inviting

• Remind

• Encourage the decision to buy a product.

• Message

 The message carried by the​​ advertisement must communicate the added value possessed by a product by using words and images. Advertising is also a storytelling medium that tells buyers what the product is like and how it is.

• Media

 The media are chosen based on their ability to reach a significant target market without ignoring its cost. The media is the best means of distributing advertisements.

• Money (Funding)

 Usually, the advertising budget is determined by the decision on each function's price, marketing reach, frequency, and desired effect. This budget must also take into account the bills that must be paid by the company for the costs of making these advertisements and other costs.​​ 

 ​​ ​​ ​​​​ But the best ad is not a high-cost ad, but an ad that can represent a company to introduce a​​ product, and of course, the ad must be as attractive as possible to attract prospective buyers' attention.

• Measurement

 The measure referred to here is the benchmark obtained after the ad campaigns. Is the ad successful, or does it even worsen the product's brand image? Successful product sales and a positive public response to the work are the keys to successful advertising.

• • ​​ Celebrity Endorser

In advertising a product, celebrities' role is vital in smooth marketing activities; celebrities'​​ existence can give and build appeal to a product with the target market. Celebrities can indirectly create a self-image process on consumers, and it influences or is bound to the advertised products. Usually, consumers will associate a celebrity's image with the product that promotes when going to buy a product. A star for consumers can also be an imaginary user when they want to determine the product they want to buy or need.

• Definition of Celebrity Endorser

According to Hardiman (2016: 38), a​​ celebrity endorser is interpreted as the chosen person representing the product picture's image (product picture) of the potential customers who have prominent characters and strong appeal.​​ 

Meanwhile, according to Suryadi (2016), a celebrity endorser is an icon often referred to as a direct source for delivering messages or entrusting products or services in promotional activities that support product message delivery. Advertising is a form of providing a​​ product or service message to consumers. It's just​​ that the problem is who will advertise the product, whether sportsman, presenter, celebrity, or layman. Of course, it must be chosen carefully and correctly. Following the company's standards and image, companies can conduct individual evaluations in determining the celebrities who will represent their products. The choice of celebrities in delivering messages is undoubtedly expected to have an immediate impact on brand awareness and brand recognition.

In marketing activities, the use of endorsers increases​​ product sales and increases brand awareness, and represents products with a brand personality that has been carried; this is expected to be well mobilized so that consumers are interested in buying approved products, Khoiruman (2015).

In general, celebrities are concrete manifestations of various images or associations that consumers think of a product brand. Suppose the appearance of a product you want to bring is energetic, full of stamina, young, and passionate. Of course, a celebrity must be able to represent all of these elements. For marketers, brand personality is essential because it is a point or characteristic inherent in the product brand, and when compared with other brands, it must​​ be very different. Features and prices are indeed effortless for competitors to copy. Still, personality is more challenging to emulate because brand personality is the same as a human personality, which is certainly different from one another.

Celebrities can indeed increase the brand value of the product being advertised, but the celebrity itself can also decrease a brand's cost. It should be more attention by companies in determining the celebrity who will promote their products. Celebrities for consumers are imaginary users, where consumers certainly want to look like celebrities who advertise the products they buy. Usually, just before consumers buy a product, they will associate the product with the image of celebrities who promote the product.

• The Purpose of Using Celebrity Endorser

Celebrity endorser is not only​​ used to communicate a product, but the purpose of using celebrity endorser according to Ankasanicara in the et al. (2017:213) is as follows:

• ​​ Explicit (support this product)

 ​​ ​​ ​​ ​​ ​​ ​​​​ Endorser not only plays the advertisement for this product but, at the same time, supports it to be closer to​​ consumers. The use of endorsers for a product expects to attach the product's identity to consumers' minds when purchasing.

• ​​ Implicit (using this product)

 The supporters of using the product to be introduced to consumers are expected to instil a good and strong image in consumers' minds when buying a product.

• ​​ Imperative (recommend using this product)

 Endorsers perform optimally on advertisements that offer this product. It is expected that consumers can receive the message to use this product, giving rise to a positive image of the product.

• ​​ Co-presentational (using the product daily)

 Endorser not only advertises this product but also uses it daily. It is expected that consumers will be influenced to use this​​ product in their daily lives.

• The Role of Celebrity Endorse

Public celebrity endorsers are known for their achievements and the products they support, celebrities as advertising models that companies can use in promoting their products.​​ 

According to Rama​​ Kertamukti, M.Sn. (2015), celebrity endorsers function as follows:

• Testimonials, if the celebrity personally uses the product, can give testimony about the advertised product or brand's quality and benefits.

• An endorsement is when a celebrity is asked to​​ star in a product advertisement where he is personally not an expert in that field.

• Actors, celebrities are asked to promote a specific product or brand related to the role that he is starring in a particular program.

• A spokesperson, a celebrity who promotes a product, brand, or company within a certain period, is included in the spokesperson role group. Their appearance will be associated with the brand or product they represent.

Based on the description above about celebrity endorsers' role, it is known that celebrity endorsers have several positions consisting of​​ testimonials, endorsements, actors, and speakers where celebrities are used as advertising models in promoting their products to consumers.

• Endorser Credibility

Endorser credibility is the ability of a figure or public figure known by the public used as a medium to convey a message to consumers, which later is expected to deliver the message that can lead to purchasing intentions (Hansudoh in Mahadewi, 2014).​​ 

Three dimensions that shape the credibility of celebrities, according to Dwivedi (2015), namely:

• Attractiveness

 Refers to consumers' perceptions of an endorser's physical Attraction, indicator forming attraction that is good appearance, attractive, and attractive (Palentina, 2017; Dwivedi, 2015; Stephanie, 2013; Spry, 2011).

• Expertise

 Expertise is the consumer's perception of a supporter's knowledge, experience, or ability with a supported product.​​ Expertise indicators are when the endorser is qualified, knowledgeable, expert, has a good and​​ skilled experience (Palentina, 2017; Dwivedi, 2015; Stephanie, 2013; Spry, 2011).

• Trustworthiness

 Refers to the honesty, integrity, and level of trust associated with an endorser, indicators of faith are that an endorser must be trusted, honest, and​​ reliable (Palentina, 2017; Dwivedi, 2015; Stephanie, 2013; Spry, 2011).

• • ​​ Purchase Intention

• Definition of Purchase Intentions

Kotler and Keller (2012) define intentions as consumers' decisions regarding brands' preferences in a collection of choices. Zarrad and Debabi (2015) express intention as a motivating factor influencing a person's behaviour to do something. Meanwhile, according to Sigit (2006), the intention is to realize one's desire or intention to act. According to some of these definitions, it can be concluded that the purchase intention is someone's impulse or someone's decision in determining the goods or services they will buy and use.

 ​​​​ A marketer will try hard to measure the intention to purchase by consumers and know what factors influence these intentions. According to Kotler and Keller (2016:151), purchase intention is a customer purchase decision-all of their experiences in learning, choosing, using, even disposing of a product. Meanwhile, according to Assael (1998), as Setyawan and Susila​​ (2014) quoted, marketers will always test elements of the marketing mix that might affect purchase intention, for example, by testing product concepts, advertising strategies, packing, or brands.​​ 

 ​​​​ According to Schiffman et al. (2015), purchase​​ intention is often referred to as a conative component in attitudes related to consumers' likelihood to buy certain products. Theory of Planned Behavior put forward by Ajzen (1988), as quoted by Asakdiyah (2005), states that when consumers have a right attitude towards the product or service they receive, consumers tend to be interested in behaviour making purchases. It is certainly profitable for the company because it strengthens the relationship between customers and the company (Asakdiyah, 2005). Thus,​​ purchase intention can be seen as an interest in behaving well in making a purchase.

• Measurement of Purchase Intention

There are many ways you can do to find out the buying interest in a product or service brand. Ferdinand (2016) states that a person's purchase intention can be measured by the tendency of consumers to buy and always look for product-related information.

• Factors That Influence Purchase Intention

Assael (2012) states that a marketer will always test elements of the marketing mix that might influence consumer purchase intentions on a company's product or service, for example, by testing product concepts, advertising strategies, packing or brands, and existing risks.​​ 

A marketer will try hard to measure purchase intentions to consumers and determine what factors influence the purchase intention. Gupta et al. (2015) also state that advertising using celebrity endorsers with a high credibility level can also stimulate consumer purchase intentions.

According to Kotler and Keller (2016), several factors that shape consumer purchase intentions:

• Other people's attitudes, the extent to which other people's attitudes reduce alternatives that someone likes will depend on two things: the intensity of others' negative​​ nature against option by consumers and consumers' motivation to obey others' wishes.

• Situation factors that are not anticipated, these factors will change consumers' attitudes in making purchases. It depends on the customer's thinking, whether he is confident in deciding whether to buy an item or not.

• • Research Framework

To make it easier to understand the relationship between Attractiveness, Expertise, and Trustworthiness and the dependent variable (Purchase intention) is described in a research framework as follows:

Figure 2.2

Research Framework

• • Hypothesis

The use of celebrities in advertisements is an effective way because celebrities' existence considers attracting the public's attention, affecting the product brand. Because the use of stars can make a product look more attractive and prominent. A celebrity has achievements in their respective fields, whether actors, presenters, athletes, or others. A celebrity's existence is considered very important in marketing because it can increase sales volume; this can see how celebrities' popularity and appeal. Especially stars on the rise will undoubtedly get the public's attention very much, which will later influence consumers' purchasing decisions.

According to research Chi et al. (2009), Gupta et al. (2015), the​​ Attractiveness of celebrity endorsers in advertising fosters consumer purchase intentions of advertised products so that the appearance, character, and popularity of the artist in delivering advertisements can make the appearance of the ad more attractive. Which later can affect public perceptions of consumers to make decisions in making purchases.

Based on the theoretical concept and research results of Gupta et al. (2015); Chi et al. (2009), the authors propose the following research hypothesis:

H1: Attractiveness Has A Positive and Significant Effect on Consumer Purchase Intentions of Products Advertised by Karin Novilda on Instagram.

The use of endorsers in an advertisement must provide adequate support to improve consumer purchasing decision making. Endorsers who have​​ Expertise have better abilities to persuade than endorsers who lack​​ Expertise. The ability to influence is an excellent way to convince consumers of the products offered. It expects that the product image is right and can be accepted by consumers, which is later wish to be purchased by consumers. 

An excellent endorser's ability to influence consumer behaviour will undoubtedly make consumers have products, services, or brands that are advertising (Sertoglu et al., 2014). The use of celebrities as endorsers expect to positively impact consumers, the interstellar relationship, and its products.

Based on the theoretical concept and research results of Gupta et al. (2015); (Sertoglu et al., 2014), the authors propose the following research hypothesis:

H2:​​ Expertise Has A Positive and Significant Effect on Consumer Purchase Intentions of Products Advertised by Karin Novilda on Instagram.

Royan (2014) states that a celebrity will be very influential if trustworthy factors support credibility.​​ One of the factors that influence consumers' purchase intentions of celebrities' advertisements is celebrities' ability to be trusted in providing information in ads. According to Temporal and Lee (2001), as Elyasa (2011) quoted, endorsers' main task is to create good associations between endorsers and advertised products. Therefore, the celebrity used must have good knowledge of the product or service. Celebrities with good experience will certainly deliver advertising messages to make consumers more comfortable understanding the product, how the product works, and others related to the product.

Marketers need to understand how a celebrity's ability to be trusted can make consumers believe in the advertised product. It is one of the keys to an advertisement's success. One​​ form of celebrity's ability is to​​ know about the advertised product. Knowledge is useful in communicating a product to consumers and will certainly make consumers believe. The results of research by Gupta et al. (2015); Hassan and Jamil (2014) provide tangible evidence that consumer confidence in a celebrity affects increasing consumer purchase intentions.

Based on the theoretical concept and research results of Gupta et al. (2015); Hassan and Jamil (2014), the authors propose the following research hypothesis: 

H3: Trustworthiness Has A Positive and Significant Effect on Consumer Purchase Intentions of Products Advertised by Karin Novilda on Instagram

Chapter​​ iii

Research Methodology

• Types of Research

The method used in this research is​​ quantitative. Quantitative research is a method that uses data in the form of numbers as a tool to analyze and conduct research studies, especially regarding what has been examined (Kasiram, 2008).

• Place and​​ Time of​​ Research

This research was conduct in Surakarta, especially in people who saw product advertisements on Instagram with an ad star Karin Novilda, the research conduct in April 2020.

• Population, Samples, and Sampling Methods

The population is a generalization area consisting of objects or​​ subjects with specific characteristics and qualities determined by researchers to be studied and then drawn conclusions (Sugiyono, 2012). This study's population is the entire people of Surakarta, especially those in the millennial generation (17-30's years old) and have seen advertisements on Instagram advertised by Karin Novilda.

The sample is a portion of the population's number and characteristics (Sugiyanto, 2012). The sampling technique used in this study is purposive sampling. The selection is made by selecting subjects based on specific criteria that have been determined by researchers and consider to have a close relationship with population characteristics. The sample used in this study is people belonging to generation millennials or those aged around 17-30's and have seen​​ advertising products advertised by Karin Novilda at least 3-5 times in the past month.

There are many forms or types of sampling methods used to determine representative sample sizes in a study. Still, according to Hair et al. (2010), the number of representative research samples is at least 30 times the number of research variables (4×30=120). Based on this, the minimum number of representative samples is 120 respondents.​​ 

• ​​ Method of​​ Collecting​​ Data

Data in this study were collected using a questionnaire. A questionnaire is a data collection technique by making a list of written questions to distributed to respondents with several alternative answers (Sugiyono, 2012).

The research questionnaire distribution is done online or through Google form, where respondents can access it at the address https://forms.gle/YJNxhgM6r241Ht72A. The researcher also sent selected respondents who had previously known that the respondent had seen an advertisement on Instagram starring Karin Novilda. The​​ questionnaire in this study develops from several previous studies conducted by Gupta et al. (2015).

• Data Measurement Method

The questionnaire's data processing is by giving a weighting assessment of each question will use a Likert Scale. With a Likert scale, the variables to be measured translate into indicator variables. The assessment categories and the weights of the possible respondents' answers are as follows (Sekaran, 2012):

• Strongly Agree :​​  5

• Agree   : 4

• Neutral   : 3

• Disagree  : 2

• Strongly Disagree : 1

• Data​​ Analysis​​ Method​​ 

• ​​ Percentage Analysis​​ 

 Percentage analysis uses to determine the respondent's profile. The method used is to present the respondent's answer to the​​ respondent's data response question. The percentage analysis formula is as​​ follows (Dajan, 1993):

Formula : 

n1N x 100%

n1: Number of respondents who chose answers.

N: The total number of respondents.

Percentage analysis was performed using the help of the SPSS computer program.

• ​​ Validity Test

 Valid has the meaning as the tools can be​​ used to measure what should be measured (Sugiyono, 1999). Testing is finished by correlating each item's scores (X) using Pearson correlation analysis (Correlate Bivariate).

 According to Sujarweni Wiratna (2015), criteria in testing validity refer to the​​ formula df = n - 2 with Sig 5%. The outcome condition is that if r count > r table, then the question item is valid and vice versa.

• ​​ Reliability Test

 A reliability test is used to measure the consistency of measuring instruments in measuring a​​ concept or can also be used to measure respondents' character in answering statement items in questionnaires or research instruments (Abdillah and Jogiyanto, 2015).​​ 

 The reliability index indicates how trusted, and dependable a measuring device is and how​​ consistent the measurement results are when measured twice or more in any conditions, using the same tool (Sekaran, 2003). The results show how far a measuring instrument can give to measure the reliability of the measuring device. The technique used is Alpha Cronbach. If the value of Cronbach's Alpha is greater (>) than 0.60, then it is considered reliable (Nunnaly, 1978).

• ​​ Assumption Classic Test​​ 

• Normality Test

 The normality test aims to assess data distribution in a group of data or variables, whether​​ the distribution of data is normally distributed or not. The normality test is useful for determining which data are normally distributed or taken from​​ the normal population. Based on Singgih Santoso (2001), guidelines for deciding the normality test are​​ when the value of Sig. or significant is greater than 0.05, then the distribution is normal (symmetrical).

• Multicolinierity Test

 A multicollinearity test is used to determine whether the independent variable experiences multicollinearity or not. These methodologies involve calculating the tolerance test and variance inflation factor (VIF) (Kleinbaum et al., 1988). To assess whether the independent variable experiences multicollinearity, the following measurements are taken:​​ 

• None of the tolerance levels is​​ ≤ 0.01

• All VIF values are well below 10.​​ 

 Tolerance measures the variability of the selected independent variable that is not explained by other independent variables. Low tolerance value equals the high VIF value for VIF = 1/tolerance.

• Heteroscedasticity Test

 Heteroscedasticity is used to test whether there is a difference in residual variance from observation to another period. Many statistical methods can be used to determine whether a model is free from heteroscedasticity problems or not, such as the​​ Graph test, Park test, Glejser test, Spearman test, rank Correlation, and Lagrange Multiplier test (LM test). In this study, the Lagrange Multiplier Test (LM test) was used.

 Handling models that experience symptoms of heteroscedasticity can be done​​ with standard logarithmic transformation (LN) data and WLS (Weight Least Square) models: dividing all data by estimation of​​ 'y', dividing all variables by one of the independent variables, or dividing all variables by one of the root variables independent​​ (Setiaji, 2004: 31). Lagrange Multiplier test (LM test) is done to deal with symptoms that experience heteroscedasticity.​​ 

The regression equation is:​​ e2=a+by2+u

• The value of Chi-Square is 9.2

• If​​ R2×​​ N <​​ 9.2,​​ then the standard error (e) does not experience heteroscedasticity.

• ​​ Regression Analysis

 Multiple linear regression analysis is used to test the effect of Attractiveness, Expertise, and Trustworthiness on Purchase intentions. The formula for multiple linear regression analysis is as follows (Kuncoro, 2012):

Y =​​ a+b1X1+b2X2+b3X3+e

Y : Purchase Intention   X2 : Trustworthiness

a  : Constant    X3 : Expertise

b1-3 : Regression Coefficient​​   ∝ : 0,05

X1 : Attractiveness   e​​  : Standard Error

• ​​ F test

 The F test will be used to test whether Attractiveness, Expertise, and Trustworthiness can predict changes in purchase intention. In the F test, information about the Attractiveness, Expertise, and Trustworthiness predictions of purchase intention is indicated by the determination coefficient (adj. R Square). The steps of the F test are as follows (Kuncoro, 2012):

• ​​ Determine Ho and Ha

• Ho÷β1=βn=0

Attractiveness, Expertise, and Trustworthiness are significantly unable to predict changes in purchase intentions.

• Ha÷β1≠βn≠0

Attractiveness, Expertise, and Trustworthiness are significantly able to predict changes in​​ purchase intentions.

• ​​ Reception Ho and Ha

Ho is accepted if probability (p)​​ >0,05

Ha is accepted if probability (p)​​ ≤0,05

 Multiple linear regression analysis was performed using SPSS program computer assistance.

• ​​ T-test

 To examine the influence of Attractiveness,​​ Expertise, and Trustworthiness on purchase intentions is used t-test. The t-test will be obtained information about the direction of the impact of Attractiveness, Expertise, and Trustworthiness on purchase intentions indicated by the regression coefficient​​ (beta). The steps of the t-test are as follows (Kuncoro, 2012):

• Determine Ho and Ha

• Ho÷β1 atau βn=0

Attractiveness, Expertise, and Trustworthiness do not have a significant influence on purchase intentions.

• Ha÷β1 atau βn≠0

Attractiveness, Expertise, and Trustworthiness have a​​ significant influence on purchase intentions.

• Reception Ho and Ha

Ho is accepted if probability (p)​​ >0,05

Ha is accepted if probability (p)​​ ≤0,05.​​ 

 Multiple linear regression analysis was performed using SPSS program computer assistance.

Chapter IV

Data​​ Analysis​​ and Discussion

 Chapter four will present the results of the research conducted. The objectives of this research are :

• To analyze the influence of Karin Novilda's Attractiveness as a celebrity endorsement on consumer purchase intentions in​​ products advertised by her on Instagram social media.

• To analyze the influence of Karin Novilda's Expertise as a celebrity endorsement on consumer purchase intentions in products advertised by her on Instagram social media.

• To analyze the influence of Karin Novilda's Trustworthiness as a celebrity endorsement on consumer purchase intentions in products advertised by her on Instagram social media.

 The research data was collected through distributing questionnaires to 120 people who had seen one of the adverts on Instagram with Karin Novilda star adverts at least three times in the past month. Data analysis methods used in this study consisted of descriptive analysis using Percentage analysis, Validity test, Reliability test,​​ Assumption Classic​​ test,​​ and Multiple Linear Regression analysis. Data analysis was performed using SPSS program computer assistance to simplify the analysis process and maintain the research results' accuracy.

• ​​ Demographic Characteristics of Respondents

 The percentage analysis in this study was used to determine the demographic characteristics of the respondents. The results of the​​ percentage analysis that have been carried out are as follows: (Appendix III)

• Characteristics of Respondents by Gender

Table 4.1

Distribution of Respondents by Gender

Source: Primary data processing 2020

 The percentage analysis results on the characteristics of Gender know that the majority of respondents (77.5%) are female, while the​​ rest (22.5%) are male.

• Characteristics of Respondents by Age

Table 4.2

Distribution of Respondents by Age

Source: Primary data processing 2020

 The results of percentage analysis on age characteristics note that the majority of respondents (98.3%) were aged between 18 to 20 years, (0.8%) of respondents under 17 years of age, and (0.8%) of other respondents aged 30 years and over.

• Characteristics of Respondents by Profession

Table 4.3

Distribution of Respondents by Profession

Source: Primary data processing 2020

 The​​ results of the percentage analysis on job characteristics are known that the majority of respondents (93.3%) work as students or colleger, (3.3%) respondents​​ work as private/public employees, and (3.3%) other respondents work as entrepreneurs.

• ​​ Research Instrument Test

• Validity Test

 This test is performed using Pearson Correlation Analysis (Correlate Bivariate), which links the scores of each item to the questionnaire questions from four variables, four variables consisting of​​ Attractiveness, Expertise, Trustworthiness, and Purchase Intention.

Table 4.4​​ Validity Test

Source:​​ processed data by SPSS 13, 2020

 In accordance with the data on the table, it shows the correlation of each question item in the Attractiveness. The value of r table for the sig. 5% with df​​ n-2=120-2=118​​ is 0.179. The results of Pearson Correlation for each item question of ATT1, ATT2, ATT3, ATT4 and​​ ATT5 are 0.761, 0.769, 0.787, 0.780, and 0.779. Because the value of r count​​ > r table; 0.761 > 0.179, 0.769 > 0.179, 0.787 > 0.179, 0.780 > 0.179 and 0.779 > 0.179, so each question items in the Attractiveness can be considered as valid.

 In accordance with the data on the table, it shows the correlation of each question item in the Expertise. The value of r table for the sig. 5% with df​​ n-2=120-2=118​​ is 0.179. The results of Pearson Correlation for each item question of EXP1, EXP2, EXP3, EXP4 and EXP5 are 0.867, 0.814, 0.884, 0.839, and 0.824. Because the value of r count > r table; 0.867 > 0.179, 0.814 > 0.179, 0.884 > 0.179, 0.839 > 0.179 and 0.824 > 0.179, so each question items in the Expertise can be considered as valid.

 In accordance with the data on the table,​​ it shows the correlation of each question item in the Trustworthiness. The value of r table for the sig. 5% with df​​ n-2=120-2=118​​ is 0.179. The results of Pearson Correlation for each item question of TRU1, TRU2, TRU3, TRU4 and TRU5 are 0.773, 0.878, 0.890, 0.741, and​​ 0.772. Because the value of r count > r table; 0.773 > 0.179, 0.878 > 0.179, 0.890 > 0.179, 0.741​​ > 0.179 and 0.772 > 0.179, so each question items in the Trustworthiness can be considered as valid.

 The data on the table shows the correlation of each question item in the Purchase Intention. The value of r table for the Sig. 5% with df​​ n-2=120-2=118​​ is 0.179. The results of the Pearson Correlation for each item question of PIN1, PIN2, and PIN3 are 0.918, 0.948, and 0.906. Because the value of r count > r table; 0.918 >​​ 0.179, 0.948 > 0.179, and 0.906 > 0.179, so each question items in the Purchase Intention can be considered as valid.

• ​​ Reliability Test

 Alpha Cronbach is the technique used in this test. If the value of Alpha Cronbach is greater​​ >​​ than 0.60, then it is considered reliable (Nunnaly, 1978). The results of the test show how far the measuring instrument can be rendered to gauge the reliability of the gauge. In the process, this research uses valid cases, about 120 subjects (100%). There is no case exclude in the process.

Table 4.5​​ Reliability Test

Source: Processed data with SPSS 13, 2020

According to the value of Alpha Cronbach in table 4.5, the amount​​ of AC for Attractiveness is 0.832, which the amount is more than 0.60. The amount of AC for Expertise is 0.899, which value is more than 0.60. The amount of AC for Trustworthiness is 0.868, which value is more than 0.60. The AC for Purchase Intention is 0.914, which amount is more than 0.60. These all outcomes show that the entire things from all factors mentioned in each of Attractiveness, Expertise, Trustworthiness, and Purchase Intention are reliable.

• ​​ Respondent's Response

 The respondent's responses were submitted for getting the information that can be used as the result of this research. The questionnaire​​ distributed to the respondents used to collect and analyze their response to the targeted sample.​​ 

 The research uses the Likert Scale, collected with the highest score weighted, five, and the lowest weight is 1. The total respondents are 120​​ people. To make the results more​​ straightforward and more transparent to be understood, the researcher has developed a description standard as​​ follows:

First, the lowest score in a question will​​ be:​​ 

• The lowest weighted score x number of​​ samples​​ =1×120=120 

And for the highest score in an item will be:​​ 

• The highest weighted score x number of samples​​ =5×120=600​​ 

Thus,​​ 

 Range of the scale ​​ =nm-1m

    ​​ =1205-15=96

Hence, the distribution table for the range arranged as follows:​​ 

120-216​​ = Very Low​​ 

216-312​​ = Low​​ 

312-408​​ = Sufficient​​ 

408-504​​ = High​​ 

504-600​​ = Very High

The data of all responses for the independent variable (X) is presented below:

Table 4.6​​ Respondents' Response based on​​ Attractiveness

Source: processed data with SPSS 13, 2020

 Table 4.6​​ shows that​​ the highest result from Karin Novilda's Attractiveness as a celebrity endorsement is at​​ indicator number "4" of 4.69%, and the lowest is on indicator number "1" of 3.78%.

Table 4.7​​ Respondents' Response based on​​ Expertise

Source: processed data with SPSS 13, 2020

 Based on table 4.7, it can be seen that the highest result from Karin Novilda's Expertise as a celebrity endorsement is at indicator number "2" of 4.53%, and the lowest is on indicator number "5" of 4.24%.

Table 4.8​​ Respondents' Response based on​​ Trustworthiness

Source: processed data with SPSS 13, 2020

 Based on table 4.8,​​ it​​ can be seen that the highest result from Karin Novilda's Trustworthiness as celebrity endorsement is at indicator number "5" of 4.58%, and the lowest is on indicator number "4" of 3.77%.

Table 4.9​​ Respondents' Response based on​​ Purchase Intention

Source: processed data with SPSS 13, 2020

 Based on table 4.9, the highest result from Karin Novilda as​​ celebrity endorsement is at indicator number "3" of 3.69%, and the lowest is on indicator number "1" of 3.28%.

• ​​ Assumption Classic Test

• Normality

Table 4.10​​ Normality​​ Test

Source: Processed data from SPSS 13, 2020

 According to table 4.10, the p-value results from one K-S sample are around 1.049 and Asymp. Sig. the value is 0.221​​ > 0.050, this means that the data meet the requirements for normal​​ distribution.

• ​​ Multicollinearity test

Table 4.11​​ Multicollinearity Test

Sources: Processed data with SPSS 13, 2020

 Based on the results of the​​ multicollinearity test in table 4.11, it was found that the tolerance level of the independent variables, namely Attractiveness, Expertise and Trustworthiness (0.295, 0.224, 0.291) was more than 0.01. Meanwhile, the VIF value of Attractiveness, Expertise,​​ Trustworthiness (3,389, 4,464, 3,434) is smaller than 10.​​ The results of the value on VIF indicate that each variable does not occur multicollinearity to the purchase intention variable in the regression model.

• Heteroscedasticity Test

 Heteroscedasticity is used to test whether there is a difference in residual variance from observation to another​​ period. Many statistical methods can be used to determine whether a model is free from heteroscedasticity problems or not, such as the​​ Graph test,​​ Park test,​​ Glejser test, Spearman test, rank​​ Correlation, and Lagrange Multiplier test (LM test). In this study, the Lagrange Multiplier Test (LM test) was used.​​ Lagrange Multiplier test (LM test) is done to deal with symptoms that experience​​ heteroscedasticity. The regression equation is:​​ e2=a+by2+u.

Table 4.12​​ Heteroscedasticity​​ Test

Sources: Processed data with SPSS 13, 2020

 From the results of the​​ Heteroscedasticity Test that has been carried out data processing with the help of SPSS 13.0. The results can be seen that the​​ R2​​ value is 0.010, while the N​​ in this study is 120. Then LM =​​ R2×​​ N​​ (0.010 x 120 = 1.2). Because the LM value is smaller than 9.2 (1.2 <​​ 9.2), it can be concluded that in this regression model, the standard error (e) does not experience symptoms of heteroscedasticity.

• ​​ Regression Analysis

 Multiple linear​​ regression analysis​​ was used to test of Karin Novilda's credibility (Attractiveness, Expertise, and Trustworthiness) on consumer Purchase Intentions on products advertised by Karin Novilda on Instagram. In the regression analysis model, the F test will be performed.​​ 

 The summary of the results of the multiple linear regression analysis​​ that has been carried out is as follows: (Appendix V)

Table 4.13 Regression Analysis

Source: Primary data processing 2020

• F test

 The F test is used to test whether​​ Attractiveness, Expertise, and Trustworthiness can predict changes in Purchase intentions. The F test will be obtained information about the predictive ability of Attractiveness, Expertise, and Trustworthiness of Purchase intentions indicated by the value​​ of the coefficient of determination (adj. R Square).

 Provisions for the significance of the F test in the multiple linear regression analysis models in this study are as follows :

• Accept Ho: If probability (p) > 0.05, which means​​ Attractiveness, Expertise, and Trustworthiness are significantly unable to predict changes in Purchase intention.

• Accept Ha: If probability (p) ≤ 0.05, which means Attractiveness, Expertise, and Trustworthiness are significantly able to predict changes in​​ Purchase intention.

 F test results obtained an F-count value of 22.026 probability (p) 0.000. Based on the F test conditions where the probability value (p) ≤ 0.05 concluded that Ha was accepted, which means, Attractiveness, Expertise, and Trustworthiness​​ were significantly able to predict changes in consumer Purchase intentions on products advertised by Karin Novilda on Instagram. The influence of Attractiveness, Expertise, and Trustworthiness on consumer Purchase​​ intentions on products advertised by Karin Novilda on Instagram is shown by the​​ Adjusted R Square value​​ of 0.346. That is, Attractiveness, Expertise, and Trustworthiness can predict 34.6% changes in consumer purchase intentions on products advertised by Karin Novilda on Instagram. In contrast, the rest (65.4%) of consuming Purchase intentions on products advertised by Karin Novilda on Instagram are influenced by other variables not included in this research model.

• T-Test

 The T-test is used to test the effect of Attractiveness, Expertise, and Trustworthiness on Purchase intentions. The t-test will be obtained information regarding the direction of the influence of Attractiveness, Expertise, and Trustworthiness on Purchase intentions indicated by the value of the regression coefficient (beta).

• Accept Ho: if probability (p) > 0.05, which means Attractiveness, Expertise, and Trustworthiness do not have an influence significant on purchase intentions.

• Accept Ha: if probability (p)​​ ≤​​ 0.05, which means Attractiveness, Expertise, and Trustworthiness significantly affect purchase intentions.

T-test results on each of the research variables can be interpreted as follows:

• The influence​​ of Attractiveness on Purchase Intentions

 T-test results on Attractiveness variables obtained regression​​ coefficient values (beta) 0.178 with probability (p) 0.194. Based on the provisions of the t-test where the probability value (p) > 0.05 concluded that​​ H1​​ rejected, which means, Attractiveness does not have a significant influence on purchase intentions on products advertised by Karin Novilda on Instagram. This analysis implies that the higher Attractiveness of Karin Novilda as an ad star cannot significantly increase consumer Purchase intentions on products advertised by Karin Novilda on Instagram.

• The influence of Expertise on Purchase Intentions

 T-test results on the Expertise variable obtained regression coefficient (beta) 0.044 with a probability (p) 0.780. Based on the provisions of the t-test where the probability value (p) > 0.05 concluded that​​ H2​​ rejected, which means, Expertise has no significant influence on purchase intentions on products advertised by Karin Novilda on Instagram. The results of this analysis imply that the higher Expertise of Karin Novilda as an advertising star cannot significantly​​ increase consumer Purchase intentions on products advertised by Karin Novilda on Instagram.

• The influence of Trustworthiness on Purchase Intentions

 T-test results on the Trustworthiness variable obtained regression coefficient (beta) 0.415 with a probability (p) 0.003. Based on the provisions of the t-test where the probability value (p) ≤ 0.05 concluded​​ that​​ H3​​ is accepted, which means, Trustworthiness has a significant influence on Purchase intentions on products advertised by Karin Novilda on Instagram. This analysis implies that the higher level of honesty Karin Novilda as an ad star can significantly increase Purchase intentions or consumer confidence in the products advertised by Karin Novilda on Instagram.

• ​​ Discussion

 This research was conducted to examine how Karin Novilda's credibility as an endorser in advertising on Instagram. This study's endorser credibility was measured based on Karin Novilda's Attractiveness, Expertise, and Trustworthiness as an ad star. This study was also conducted to examine​​ how Karin Novilda's credibility as an ad star stimulates increased consumer Purchase intentions.

 The results showed that Karin Novilda's Attractiveness as an advertisement star did not have a significant effect. The regression analysis results also provide information that the Attractiveness of Karin Novilda as an advertisement star does not significantly​​ influence Purchase intention (beta = 0.178) and (p = 0.194). This study shows that Karin Novilda, as an advertising star, is less able to influence buyers in Purchasing intentions, even though Karin Novilda is a beautiful and classy celebrity.

 Karin Novilda's skill level as an advertisement star does not have a significant effect. The results of the regression analysis also provided information that Karin Novilda's Expertise as an advertising star does not significantly influence Purchase intention (b = 0.044) and (p = 0.780). Which means that Karin Novilda's Expertise was unable to control the increase in consumer purchase value even though Karin Novilda​​ was an advertising star who has a lot of experience and good quality skills.

 This study also provides information that Karin Novilda as an advertising star has a positive and significant effect on Purchase intention (b = 0.415) and (p = 0.003). The​​ results of this study provide evidence that in delivering advertisements, Karin Novilda is an artist who can be relied on because she conveys the content of advertising messages sincerely and honestly. Besides, the information contained in an advertisement​​ with an ad star, Karin Novilda can be trusted. Some of these things cause people to​​ believe in Karin Novilda as an advertisement star and have the intention to buy the products advertised by Karin Novilda.

 The results of previous research conducted by Gupta et al. (2015) where the endorser's credibility had a significant influence (86.7%; this study 34.6%) in influencing consumer purchase intentions. Sertoglu (2014) obtained results (p = 0.21; 0.08; 0.12), and Hasan and Jamil's research (2014) showed Attractiveness and Expertise (p = 0.12; 0.22), which means the research conducted by Sertoglu, Hasan and Jamil showed a positive and significant result. The difference between the results of this study and the analysis conducted by Gupta et al. (2015), Sertoglu (2014), Hasan and Jamil (2014) lie in the influence of the credibility of each endorser. In the study of Gupta et al. (2015), only the variable of Expertise did not have a significant effect on Purchase intention (Std. Beta (r) = 0.03 using the CFA model). Whereas in this study, not only the Expertise variable had insignificant results but also the Attractiveness variable. The difference between the results of this study and those of Gupta et al., Sertoglu and Hasan and Jamil is possible because the subjects and objects of the study are different,​​ this is what makes it possible to distinguish the results of this study from those of Gupta et al., Sertoglu and Hasan and Jamil.

Chapter V

Closing

 In this fifth chapter, the writer will conclude from the​​ research results that have been done. Furthermore, the authors make implications and suggestions aimed at interested parties.

• ​​ Conclusion

 Based on the results of the research that has been done. The results of the analysis of the influence of credibility​​ (Attractiveness, Expertise, and Trustworthiness) on Purchase intentions, can be concluded as follows:

• Attractiveness, Expertise, and Trustworthiness were significantly able to predict 34.6% of changes in consumer purchase intentions on products advertised​​ by Karin Novilda on Instagram.

• Attractiveness does not significantly affect the results of the t-test, which shows a probability (p) of 0.194 of changes in consumer purchase intentions of products advertised by Karin Novilda on Instagram.

• Expertise does not significantly affect the results of the t-test, which shows a probability (p) of 0.780 of changes in consumer purchase intentions of products advertised by Karin Novilda on Instagram.

• Trustworthiness has significantly able to predict according to the results of the t-test, which shows a probability (p) of ​​ 0.003 of changes in consumer purchase intentions on products advertised by Karin Novilda on Instagram.

• ​​ Managerial Implications

 To be successful in fierce business competition, every company must carry​​ out various ways or strategies to increase consumer purchase intentions of the products or services being sold. There are many ways that companies can do to improve consumer purchase intentions, one of which is through advertising programs.​​ Advertising is​​ one way to communicate the company's product or service brand to the public. There are many benefits that companies can get by running advertising programs such as: providing information on new products or services offered, reminding consumers of the company's products or services, persuading people to buy the company's products or services, and so on.

 The importance of advertising programs for the success of the company and consumers requires the company to create attractive and informative advertisements. One of the ways that many companies do to streamline marketing programs is by using an endorser. Companies usually use well-known people (such as artists or celebrities) with a high level of credibility as commercials for company products or services. The credibility of a celebrity can be seen from the group of Attractiveness, Expertise, and Trustworthiness it has. The use of artists or celebrities as commercials is done to make consumers more interested in the products or services. Another reason companies use well-known celebrities is that celebrities can build a company's brand image, which will arouse consumer purchase intentions.

 This study was conducted to examine the effect of Attractiveness, Expertise, and Trustworthiness on consumer Purchase intentions. This research aims to see all people in Surakarta who belong to the millennial generation (aged between 17 and 30 years) and have seen advertisements on Instagram with advertising stars Karin Novilda. The results of research conducted on 120 people provide evidence that Karin Novilda's Attractiveness and Expertise as an influencer is not sufficiently capable of delivering stimuli to increase consumer purchase value. At the same time, Karin Novilda's Trustworthiness shows significant results in increasing consumer purchase value. Based on this, companies need to design advertising programs, especially in determining and using well-known advertising stars to help improve brand image and consumer Purchase intentions.

• ​​ Research Limitations

 By the results of the data analysis and the discussions, the limitations of the study are:

• There was no specific type of product endorsed by Karin Novilda in this study.

• The majority of respondents in this study were men or women aged 18-20 years and were students. If​​ the coverage of respondents in this study is broader, the study results will better reflect the conditions in which celebrity endorsement can influence a person's purchase intention.

• Suggestions

 Based on the results of the research that has been done​​ and the managerial implications above. The authors formulate suggestions for interested parties, including the following:

• ​​ For Further Researcher

• Use specific types of endorsement products such as clothing, makeup, skincare, or others.

• Spread the​​ questionnaire more widely and evenly so that the results are also even, not just skewing to one.

• ​​ For Brands and Influencers

• The use of celebrities as advertisements must pay attention to the level of conformity between the celebrity's credibility and the​​ advertised product.​​ Celebrities with the appropriate Expertise in the product or service being promoted will go a long way in shaping and enhancing the personality of the brand. For example, in sports product advertisements using an athlete, cosmetic products using a beauty influencer, or in food products using a famous chef. The distribution of information in ads by celebrities must be done in an honest and more communicative and interactive style; this is done to increase consumer confidence and attract consumer attention to the products or services being sold.

• An attractive product or service advertisement must have a clear storyline or concept because a useful and stunning ad setting will contribute to consumers' desire to see ads from beginning to end.