THE CREATION OF INTENTION TO BUY THROUGH ONLINE SOCIAL TIES AND SOCIAL MEDIA ADVERTISEMENT

Objective: The growing number of social media users makes many companies want to get subscribers through their social media activities. There are several choices of strategies that can be done by the company to be able to take the huge market. Research Design & Methods: This study compares the effect of advertising activity on social media and user group development that has a strong tie to intention to buy. Findings: This research found that the social media advertisements has a better impact to intention to buy compared to social ties. Implications & Recommendations: This study was conducted on social media users on land Transportation Company that already has a group in social media, Facebook, and does advertising activities on it. Contribution & Value Added: The results of this study provide theoretical implications in the verification of the previous research results and managerial implications that can help the company in conducting marketing activities through social media.


INTRODUCTION
The development of the internet is now growing rapidly with the number of users who increase every year. In March, 2017 the number of internet users around the world reached about 3.7 billion, or approximately 49.7% of the total world population. The percentage of internet users' growth from 2000 to 2017 was 936%, so that the average number of users was growing by 117% each year. The largest number of Internet users are in Asia with a percentage of 50.1% of all internet users worldwide (Internet World Stats, 2017).
The growth of internet users can not be separated from the number of social media users around the world which are increasing every year. By 2017, the number of active users of social media is 2.8 billion or 37% of the world's population. Of the total active social media users, 2.5 billion of them use a smartphone. Facebook is the most accessed social media, and then the second is Youtube, Qzone from China, Instagram, Tumblr, Twitter and Linkedin (Statista, 2017).
Indonesia has the 4th largest social media user worldwide with a growth of 34% over the previous year. The number of social media users in Indonesia reached 106 million users or 40% of the total population of Indonesia. Of all those social media users, there are 92 million or 87% of total active social media users in Indonesia who use smartphone to access social media (www.wearesocial.com, 2017). The growing number of outstanding social media users makes producers interested in promoting their products in social media. It can be seen in the amount of advertising revenue up at the end of 2016 amounted to 6.8 billion dollars with the number of advertisers about 2 million, and is expected to increase by 59% from the previous year. It also can be found in social media of Instagram which has experienced an increase in the number of active advertisers from around 200 thousand in 2016 to about 1 million active advertisers in 2017, with earnings of $595 million (www.wordstream.com, 2017).
Researches on advertising in social media have resulted in several findings that can help companies in conducting marketing activities through advertisements on social media. Advertisements on social media can increase customer confidence in the product by increasing perceived value after seeing the promotion of the company ( Van-Tien Dao et al., 2014). Advertisements on social media that are appealing and appropriate to social media users will give rise to a desire to spread viral positive information about the advertised product (Petrescu et al., 2015).
Social ties are also very influential on the users' desire to perform activities of electronic word of mouth (e-WOM) which is very helpful in the spread of product positive information. The source of referral information and strong social tie will have a greater impact than those with weaker social tie (Steffes & Burgee, 2009).
From some previous research on social ties in social media, some say that advertising messages delivered by parties who have higher social ties, will improve the attitude of the message recipient to the brand. The same study also explains that advertising coming from companies that advertise on Facebook, will improve attitudes toward the brand and also increase purchase intention (Yang, 2012). Therefore, the study concludes that advertisements submitted by companies that advertise on Facebook have a better impact than advertisements delivered by parties who have high social ties, such as close friends. This is in contrast to previous research that found that advertising messages delivered by people with high social ties have a greater impact on the recipients than the advertisements submitted by producers (Brown & Reingen, 1987). Another contradictory study said that researchers conducted field experiments on Facebook and found that information and product recommendations provided by friends who have strong ties to consumers are considered to have a high diagnostic level (J.-J. Wang et al., 2018). Other studies say that the strength of social ties and social influence has a positive effect impact on purchase intention. Not only social ties exist between consumers and online friends who do business on social media platforms, but also social influence from comments and opinions of other consumers or family members affect purchase intention (Zhang & Wang, 2019). The gap in the difference between the results of the studies relating to the large ratio of advertising influences conveyed by those who have high social ties compared to the advertisement submitted by the producers is what this study will fill.

Social Media Advertisements
Social media has become a very attractive medium for companies to do promotion because of the ever-increasing number of users and nearly one-third of the world's population. Facebook users prefer to respond to information provided through the brand community rather than advertisements on Facebook. They give more confidence to the brand community compared to advertising on social media (Chi, 2011). It becomes a question of whether advertisements on social media without or with the influence of the tie power in social media will have an effect on online trust.
Advertisements on both paid and unpaid social media as earned from e-WOMs of social media users has a good impact on overall brand purchases (Thornhill et al., 2017). So companies can combine the power of paid and non-paid advertisements on social media to get maximum performance.
Carmichael & Cleave (2012) advertised on Facebook and monitored two advertisements to capture the user-advert interaction and its effectiveness for small businesses. For the amount a small business would invest in such social media advertising, they found that this form of advertising could substantially improve the visibility of such businesses and easily direct consumer traffic to the advertised business page. H1: Advertisements on social media give more effect on users e-trust compared to online social tie

Online Social Tie
Online social ties become an important factor for consumers in making decisions based on online marketing activities. Online users are more confident of an information source that has a strong tie in understanding and evaluating product quality before making a decision (J.-C. Wang & Chang, 2013).
Online social ties become the basis of user generated content (UGC) which is the core of social media. The stronger the online social tie, the stronger the UGC's influence in giving a positive impact on products shared by social media users (Zeng & Wei, 2013).
H2: Online social power gives more effects on e-trust users compared to advertising on social media.

e-Trust
With the growing number of transactions taking place in e-commerce, especially in social media, electronic trust (e-trust) becomes an important thing for social media users. The e-trust development mechanism which consists of certification by a third party, reputation and return policy will increase the trust of online sellers. The e-trust development effort undertaken by the company is the most effective mechanism of trust development . The implications of e-trust built by this company are a direct influence on perceived usefulness and intention to buy (Mansour et al., 2014).
Another thing that is no less important is the fast response to customer information questions according to their wishes, which ultimately leads to satisfaction and development e-trust (L. Wang et al., 2015).
H3: E-trust users on online companies affect the perceived usefulness of users.
H4: E-trust users on online companies have an effect on users' intention to buy.

Perceived Usefulness
Social media users who want to make an online purchase will have a perception of the usefulness or benefits of purchasing the brand. Perceived usefulness has more influence on purchase intentions than perceived ease of use. Factors in perceived usefulness consist of informative, fast, efficient and easy (Kucukusta et al., 2015).
Perceived usefulness needs to be emphasized in advertising on social media by companies to differentiate with competitors. In addition, perceived usefulness can also be developed from product quality, completeness of information and security (Sohn, 2017).
H5: Perceived usefulness of the users in the company affects the users intention to buy.

Intention to Buy
Purchase intention or intention to buy is a kind of decision-making that studies the reason to buy a particular brand by consumer (Shah et al., 2012). Purchase intention may be changed under the influence of price or perceived quality and value. In addition, consumers are affected by internal or external motivations during the buying process. Customers always think that purchase with a low cost, simple packaging and little known product is a high risk since the quality of these products is not trustable (Gogoi, 2013).

Research Model
From an intensive study of literature using a variety of good reference studies, an empirical research model is produced as presented in the following Figure 1. Model illustrates that advertising in social media and online social ties has the potential to develop online trust. The impact will affect the perceived usefulness of social media users of the advertised product and intention to buy against the product.

Research Object
The object of this research is the users of social media of Facebook incorporated in the group managed by the bus company of Bejeu. The group has about 86,000 members consisting of customers, users, Bejeu bus lovers who actively share, communicate and conduct activities in a community called Black Bus Community (BBC).

Method of Collecting Data
Determination of the number of samples for Structural Equation Modeling (SEM) analysis between 100 until 200 so that it can be processed through SEM, used formula in Equation 1. This formula is suitable for use when the parameter measured is the value of the proportion. n is the number of samples, Z is the standard normal number and E is the level of error in estimating, if the confidence level is 95% and the estimation error is not more than 8%, so we get 150 research samples. This study used primary data obtained directly from the respondents with a target of 150 members of the group taken as a sample. The data were collected by survey method using questionnaires which contained questionnaires submitted directly to respondents.
The questionnaire was administered online using the Google Forms facility and key chains were provided to 150 members who meet the survey requirements. From the results of Google Forms, it obtained valid data of 122 samples.

Data Processing and Analysis
Data entered through Google Forms is then converted into SPSS software before doing further analysis. Next, they will do some testing using AMOS software to obtain the results of the tests required.

FINDING The Description of Research Variables
The description of research variables is obtained by calculating the index value. The results are presented in the following Table 1. From the calculation of the index value for the research variables, it was obtained that the variables studied were in the average category.

Confirmatory Test
Confirmatory tests were conducted to confirm whether the variables used were the correct measuring instruments or not. Confirmatory testing requires that the standardized value should be greater than 0.5 (show in Table 2). From the results of tests conducted, the value could be met on each measured indicator so that it could be concluded that these indicators were the right measuring tool.

Normality Test
Research data is normal if the value of CR Multivariate is in the range ± 2.58. CR Multivariate in this study was 2.578 which was still in the range specified. Therefore, it can be said that the data of this study was normally distributed

Outliers test
Outliers test has a close relation with the normality test. When the assumption of normality cannot be proven, it also indicates an outlier's problem. Outlier is examined from Mahalanobis distance <value of Chi Square

Multicollinearity and Singularity
Multicollinearity and Singularity was conducted to determine the closeness of the relationship between two exogenous variables i.e. facebook line and online social tie with respect to determinant of sample covariance matrix value which was greater than zero.
The testing of this model yielded a determinant of sample covariance matrix of 2542988.505 which was more than zero. This means that there is no multicollinearity problem between the facebook line variables and the online social tie and both show the singularity for each variable.

Residual Assumption
A study of residuals resulting from the research models testing is needed to determine whether the research model modification is necessary or not. The residual assumption was assessed by observing the value of standardized residuals which was greater than 2.58. Testing of this model yielded two standardized residual values which were greater than 2.58. However, the amount did not exceed 5% of the total standardized residual value which was greater than 2.58. Therefore, it can be concluded that there is no residual problem and the model does not require modification.

Construct Reliability and Variance Extracted
Construct reliability and variance extracted calculations are presented in the following Table 3. The values set for reliability construct are > 0.7 and variance extracted > 0.5. Referring to the above calculation results it can be concluded that the assumption of reliability construct and variance extracted can be met for each research variable.

Model Feasibility
Feasibility analysis of the model is required to justify that the model has been developed as a fit model. Model feasibility testing is performed by examining in the following

Hypothesis Testing
Hypothesis testing became the final step undertaken in this study. The following Table 5 is the results of the hypothesis testing. The results of hypothesis testing conducted in this study resulted in probability values which were smaller than 0.05. It indicated that the five hypotheses tested could be accepted and verified statistically. For H1 and H2, as compared to the power of influence between advertisement in social media and online social tie, therefore, if it was seen from standard value of estimate of both variables, it was found that advertisement in social media have bigger influence compared to online social tie toward etrust. Thus, from the results of the comparison, then H2 was rejected.

CONCLUSION
This research found that the social media advertisements has a better impact to purchase intention compared to social ties. The theoretical implication of this study is that the results of previous studies which suggest that advertising messages conveyed by people who have high social ties have a greater effect than those advertised by producers Brown & Reingen (1987) are not proven. This study is in line with previous research that explains that advertising on facebook has a greater impact than advertising delivered by people who have strong social ties. This research also contributes in adding antecedent which has been proven to have an effect on the development of e-trust for companies in social media, i.e. advertising in social media and online social tie.
The use of social media to conduct marketing activities becomes increasingly important and should be done for the company. This can be seen in the results of research that advertising in social media and corporate brand group development has a considerable influence in the development of e-trust for the company. By having a high e-trust from customers, the company will be able to make its customers have a good perceived usefulness and ultimately can increase intention to buy the company's products.
In other words, with high e-trust, the company will be able to convert the customer's trust into intention to buy which will have a good impact for the company.