ANALYSIS OF PERCEIVED VALUE AND TRUST OF SATISFACTION AND ITS IMPACT ON THE LOYALTY OF ONLINE SHOPPING CONSUMERS IN TOKOPEDIA

The number of Internet Users is increasing and the competitors of Online Business are making Online Shopping business tight, but the increase in Internet users is not in line with the increase in Online Shopping users. Data obtained from the Association of Indonesian Internet Service Providers (APJII, 2020), the reason why Internet users never change their goods or services online 6.7% is worried that the goods will not arrive. And data from the National Consumer Protection Agency (BPKN) for 2018-2020 there were complaints with problems regarding phishing and accounts via OTP. Tokopedia is an E-marketplace Platform that carries Online Purchase of Goods, with Tokopedia, consumers can easily fulfill their daily needs. The development of Tokopedia is very convincing that many consumers are making products. The more consumers who make repeated purchases, the more growth will be. However, there will be repeated every consumer who feels loyal to what is offered and has a customer. Loyalty must have several components because Loyalty is built on the basis of several components. So that this research discusses the Perceived Value and Trust in Satisfaction and its impact on Consumer Loyalty. This study refers to previous research as a conceptual basis. The research sample was taken from 256 respondents who were taken from the Tokopedia application using Google Docs in the form of a questionnaire survey. From the results of this study, consumer loyalty is very hampered by the Perceived Value, trust and satisfaction as a mediation.


INTRODUCTION
The current technological developments make Indonesia one of the countries that has an appeal in the eyes of other developing countries. Because the total population and ethnic composition in Indonesia are very diverse, Indonesia is very attractive. In this era, each individual can easily obtain information as well as become a resource through the internet network. This is in line with the data released by https://www.internetworldstats.com/top20.html, namely Indonesia is the 4th (fourth) position in the world's largest Internet users in 2020 with an internet user value of 171,260,000 with a population of 273,523,615 population. From this data, it can be stated that technological developments from 2000 to 2020 are very significant. Along with current technological developments, the sale and purchase of goods or services has progressed very rapidly. One of the conveniences in buying and selling transactions is Electronic Trading or what is currently known is E-Commerce. However, these technological advances have not been accepted by all consumers. Data obtained from the Association of Indonesian Internet Service Providers (APJII 2020), the reason why Internet users never shop for goods or services Online 6.7% is worried that the goods will not arrive. And data from the National Consumer Protection Agency (BPKN) for 2018-2020 there were complaints with problems regarding phishing and account abuse through OTP. Meanwhile, according to the Indonesian Consumers Foundation, 6.3% of complaints throughout 2019 came from e-commerce issues. Of the total e-commerce complaints, most 28.2% of the public complained that ordered goods were not accepted. In addition, 15.3% of online consumers complain of ordered goods not according to specifications and a difficult refund system (http://ylki.or.id/). One of the E-Commerce platforms that carries Online Purchase of Goods is Tokopedia. Statistical results obtained from https://www.statista.com/, show the reasons Indonesian online consumers dislike the Tokopedia online platform in 2018. In that year, 21.7% of respondents who were surveyed answered that one of the main reasons for being dissatisfied with this service because of the time it takes to deliver the goods.
Previous studies have analyzed a lot about online shopping behavior, such as: Lin andSun (2009), Luo et al. (2012), Ludin and Cheng (2014), Hamdan and Yuliantini (2021). To win the competition in the eyes of consumers, a good corporate image is needed because consumers have an important role in all businesses. Consumers are intangible assets. Therefore, to get the top ranking in the hearts of consumers, business people carry out various activities so that the company is known and accepted by consumers so that consumers can feel loyal. Loyalty is defined as a continuous repurchase of the same brand or in other words the action of someone who buys a brand and only pays attention to certain brands (Sondakh, 2014). Griffin (2005: 5) suggests that consumers are said to be loyal or loyal if consumers show regular buying behavior or there are conditions that require consumers to buy at least twice in a certain time interval.
In a relationship the most important is Perceived Value, Trust, Satisfaction and Loyalty. According to Kotler and Armstrong (2014: 150) Satisfaction is a feeling of pleasure or disappointment that arises after comparing the performance (result) of the product thought against the expected performance (or outcome). Beginning with consumer satisfaction and trust in the products or services offered so as to make consumers loyal.
The results of research conducted by Hsin Hsin Chang et al., (2010) on consumer loyalty, namely loyalty is built on the basis of high value perceptions. In addition this result found that customers with a high perceived value have a stronger relationship between satisfaction and customer loyalty than customers with a low perceived value. Other factors that affect consumer loyalty are trust and satisfaction. Roshana Gul (2014) states that trust is another factor that affects customer loyalty. Brand trust is a promise between the brand and the consumer that will fulfill the needs of the consumers according to their wishes. Research conducted by Jennifer H. Gao (2019) states that satisfaction tends to arise after consumers build a certain level of trust in online shopping. This study also explains that there is a significant and positive correlation found for trust in online shopping and consumer satisfaction with online shopping.
Empirical review and theoretical review are used in this research. Previous studies related to research to be conducted as an empirical review, while theoretical reviews were conducted to find out the theories and concepts developed from each source related to the research model.

PREVIOUS RESEARCH
The literature study in this study uses three references from previous studies. The first study used research conducted by Hsin Hsin Chang et al., (2010) the relationship between Service Quality, Consumer Value Perception, Customer Satisfaction and Consumer Loyalty. about Perceptions of Value on Consumer Loyalty, namely the existence of a significant positive effect of Value Perceptions on Online Consumer loyalty. In addition this study found that customers with a high perceived value have a stronger relationship between satisfaction and customer loyalty than customers with a low perceived value. In the post-purchase stage, rational factorssuch as customer perceived valueplay important roles because they can strengthen the relationship between satisfaction and loyalty. The definions of Perceived Value is the difference between all benefits felt by consumers and costs incurred by consumers and A product can be said to have great value if it can meet customer needs, wants and demands (Kotler and Keller (2012: 147).
The second study uses research by Roshana Gul (2014. Trust is another factor that affects customer loyalty. Brand trust is a promise between a brand and consumers that the needs of consumers will be fulfilled according to their wishes. According to Gul (2014) trust is a positive psychological state that occurs in certain relationships, in other words, customer trust is a positive psychological state that occurs between customers and certain products or services. trust is built between parties who do not know each other both in the interaction and transaction process (McKnight, Kacmar, and Choudry (in Bachmann & Zaheer, 2006).
The third research uses research. Research conducted by Jennifer H. Gao (2019) states that satisfaction tends to arise after consumers build a certain level of trust in online shopping. This study also explains that there is a significant and positive correlation found for trust in online shopping and consumer satisfaction with online shopping. Satisfaction is a feeling of pleasure or disappointment that arises after comparing the performance (results) of the product that is thought of against the expected performance (or results) Kotler (2014: 150). Loyalty is the attitude and commitment that is held to buy or support back a preferred product or service in the future Kotler and Keller (2009: 138). Research conducted by Hamdan (2018) reveals that satisfaction has a positive and significant effect on customer loyalty.

RESEARCH METHODS
The type of research used is explanatory research with a quantitative approach, such as the explanation of According to Sugiyono (2014), the explanatory research method is a research method that intends to explain the position of the variables under study and the influence between one variable and another. As for this research, the analysis built can be seen in the image below: Available Online: https://dinastipub.org/DIJMS Page 599 .

Figure 1. Research method
The sampling technique uses non probability sampling. By using purposive sampling technique is the taking of sample members from the population that limit the specific criteria of someone who provides information according to the needs of the author. The sample used in this study were consumers from Tokopedia in DKI Jakarta who had the following criteria: This research instrument is a questionnaire which is arranged based on the indicators of the research variables. These indicators are described in detail in the question points. The questionnaire used in this study used an Online questionnaire using Google Docs which was measured by a Likers Scale. Respondents were asked to fill in each item by selecting one of the five available options. The scoring and measurement of alternative answers use a Likert scale which has five alternative answers.
In this study the authors used the assistance of the SEM (Structural Equation Modeling) Method with Partial Least Square (PLS) on the grounds that Perceptions of Value, Trust, Satisfaction and Consumer Loyalty Tokopedia are considered as variables that cannot be measured directly or are called latent variables. And what can be measured is based on the indicators (manifest variable), and together it involves the level of measurement error.
This study has 7 hypotheses. To analyze the models and hypotheses proposed, researchers analyzed using the Structural Equation Modeling (SEM) method. H1 = Perceived Value has a positive and significant direct effect on Customer Satisfaction H2 = Trust has a positive and significant direct effect on consumer satisfaction H3 = Perceived Value has a positive and significant direct effect on Consumer Loyalty H4 = Trust has a positive and significant direct effect on Consumer Loyalty H5 = Satisfaction has a direct, positive and significant effect on Customer Loyalty H6 = Perceived Value has an indirect, positive and significant effect through Satisfaction on Consumer Loyalty H7 = Trust has a positive and significant effect through Satisfaction on Consumer Loyalty.

FINDINGS AND DISCUSSION
In this chapter, we will discuss the stages of processing data collection using SmartPLS version 3.3.2 software. This study has 7 hypotheses. To analyze the models and hypotheses proposed, researchers analyzed using the Structural Equation Modeling (SEM) method. Based on Figure 3, it is known that there are 4 (four) variables in this study where the Perceived Value and Trust is the independent variable. Meanwhile, Customer Satisfaction is a mediator variable and Loyalty is the dependent variable. In addition, in each variable there are indicators that represent these variables in accordance with the theoretical basis. At this stage, the outer model analysis will be carried out using the data quality test in SEM. To produce data with good quality, a validity test that measures the suitability of each indicator is needed and a reliability test that measures the certainty of the research data.

Descriptive Research Respondents
The questionnaires distributed from 26 September 2020 to 3 October 2020 to 256 research respondents. There were 14 respondents aged> 45 years, 61 respondents 36 -45 years old, 128 respondents 26 -35 years old and 53 respondents aged 17-25 years old. Furthermore, at the beginning of the questionnaire there are several entries that indicate the personal data of the research respondents. The following is a description of the characteristics of the respondents in this study. Based on table 4.1.1, it is known that out of 256 respondents the number of workers as teachers was 2%, employees (private / ASN) were 73%, art workers were 1%, students were 5%, professionals (consultants / doctors etc.) as much as 2%, entrepreneurship as much as 6% and others as much as 11%. Based on table 4.1.2, it is known that out of 256 respondents, 39% of them are in West Jakarta, Central Jakarta is 6%, South Jakarta is 41%, East Jakarta is 10%, North Jakarta is 4%.

Validity Test
The validity test for the reflective indicators uses the correlation between the item scores and the construct scores. The following are the results of the data validity test in this study.

Source: PLS output version 3.3.2, Data Processing
Based on Figure 3, all variables in this study have valid data. This is because the AVE value is above the requirement of 0.50 (Ghozali, 2014: 42).

Reliability Test
Reliability test shows an accuracy, consistency and accuracy of a measuring instrument in making measurements (Ghozali, 2014: 44). If a study has reliable data, then the research data has been tested for reliability and consistency of results. Reliability test in PLS  Based on table 4.3, it is known that all variables in this study are reliable. This is because the resulting value meets the criteria, namely ≥ 0.70.

Test Results of the Coefficient of Determination (R2)
The evaluation model for the inner model can be evaluated by looking at the variance described, namely the R square value in the research construct: Based on Table 4.4, it can be seen that the R-Square value for the Loyalty construct is 0.723. This means that the model has a goodness-fit model level. This also means that the variable Customer Loyalty can be explained by the three variables in the model, namely Perceived Value, Trust and Satisfaction of 72%, while the remaining 28% is explained by other variables not in this study.

Result of Path Analysis Test or Path Coefficients
To analyze the direction of the relationship between positive or negative variables which can be explained by looking at the value of the path analysis below: Based on Table 4.5, it can be seen that the value of each relationship between variables is at a value> 0 to 1, which means that the relationship is positive. Therefore it can be concluded:

Significance Test Results or t-Statistics
To see the significance results of the parameter coefficient, it can be calculated from the valid variable dimensions. Researchers want to know that there are positive or negative and significant or insignificant effects based on the calculation of P values which must be below 0.05 and the t statistic is greater than 1.96 (Ghozali, 2014: 65).

Specific Indirect Effects Test Results
To see the significance results on the moderator variable. Researchers want to know whether there is an indirect or indirect effect from the T-value.

CONCLUSION AND SUGESTION
Based on the conclusions obtained in this study. Here are some suggestions: a) In the Perceived Value with Result (PV5) dimension the outer loading value is 0.829. This shows that customers are very happy with the services provided by Tokopedia because they are very good to consumers. Therefore, it is hoped that Tokopedia can maintain and / or improve again so that consumers can be more loyal in shopping at Tokopedia. The trust dimension (TR2) shows an Outer Loading value of 0.827. This shows that consumers rely heavily on Tokopedia to shop to meet their needs. Therefore, it is recommended that Tokopedia should further improve the ease of features on the website so that consumers can be more loyal and trust shopping on Tokopedia. The satisfaction dimension (ST5) shows an Outer Loading value of 0.859. This shows that consumers are satisfied because the Order tracking system on Tokopedia is very easy to track. Therefore, satisfaction is the main key to consumer loyalty. b) Tokopedia as an e-commerce market leader must be able to maintain through the achievement of market performance and profit performance, be led by leaders who have strong management and are able to create a good work culture and must pay attention to customer online experience from the start of product search, purchase and product delivery. A superior organization can create competitive advance through the creation of product leadership, operational excellence and customer intimacy so that Tokopedia can achieve a sustainable competitive advance among its competitors. c) Tokopedia must also evaluate merchants who cooperate on the application website so that all merchants in Tokopedia are guaranteed the quality of the products sold to match what consumers expect. d) Tokopedia must also improve customer data security and improve product delivery services to match the time expected by consumers so that consumer confidence can increase in shopping at Tokopedia.
For further researchers who are interested in aspects of marketing strategy and online shopping. This research can be used as a reference for conducting research in similar fields. For further research, several things can be considered, including the following: a) The research was conducted by adding variables which are factors that can affect online consumer loyalty. b) Comparing several research objects such as other online buying and selling sites so that they are expected to provide better results. c) Research is carried out more specifically in selecting research objects and research locations such as Fashion, Electronics, locations in certain cities so that the research results can be more targeted.