THE EFFECT PERCEIVED OF SERVICE QUALITY, PERCEIVED OF WEBSITE QUALITY, AND PERCEIVED OF WEBSITE REPUTATION ON PURCHASE INTENTION: THE MEDIATION ROLE OF TRUST IN ONLINE SHOPPING

: This study aims to determine the effect of perceived of service quality, perceived of website quality, and perceived of website reputation on purchase intention through the mediating role of trust in online shopping. This research uses a quantitative method with Smart PLS software. A variable measurement scale is carried out using a likert scale. The sample in this study were university students and the general public with an age range of 17-56 years who had used the website of the e-commerce application. Dissemination of the questionnaire using an online questionnaire link. A total of 103 tabulations of questionnaires were collected and analyzed successfully. The object of analysis in this study is a website or online shopping service. The analytical method used is the validity and reliability test, coefficient of determination test (adjusted r-square), and hypothesis testing. The results of this study prove that the perceived of service quality, perceived of website quality, perceived of website reputation has an effect on trust and purchase intention. Trust has a mediating role in the bond between perceived of service quality, perceived of website quality, and perceived of website reputation on purchase intention. the company is to provide quality service to customers, achieve high market share, and increase profits (Dapas et al ., 2019). influence


Introduction
Information technology in the era of globalization as it is today has developed very rapidly. The internet is one of the human works produced along with the development of information technology. The facilities and services provided by the internet can make it easier for users to access various information from all over the world without knowing any boundaries because they are connected to a global computer network. The internet can also create new business opportunities and become a medium for marketing and trade (Cahyati & Mandala, 2017). Success in the company is to provide quality service to customers, achieve high market share, and increase profits (Dapas et al., 2019).
Website reputation and quality are extrinsic and intrinsic factors that can influence consumers' online purchasing decisions. These two factors can have a direct effect on online shopping attitudes and perceptions of risk accepted by consumers which in turn can lead to purchasing intentions for the product. In this study, the perception of information quality refers to consumers' overall assessment and evaluation of information quality, assessed from Fatta et al. (2016) has run a meta-analysis on users' perceived web quality, concluded that it is relevant effect easy to use, usability, and enjoyment, and suggested that it can encourage website use in online shopping. (Pawlasova & Klezl, 2017) defines the term as "user perception of the design of the website on which the group buying experience takes place". Lee et al. (2016) proposed a conceptual website quality whose type reflects the four dimensions of website quality: design, compliance/reliability, security, privacy, and trustworthiness, and customer service.The results of the study concluded that all measures of website quality have a relevant effect on purchase intention (Qalati et al., 2021). Thus, the following hypothesis is proposed:

Perceived of Website Reputation
A perception of a reputation is a consumer's perception of whether the retail store will be honest, care about its customers, and can fulfill its promises. This is how much customers trust it that the company as a seller must have honesty and care about its customers. Kim & Peterson (2017) meta-analysis of online trust has it was decided that risk awareness, security, privacy, information quality, system design quality, system quality, and reputation visibility were important antecedents of online trust. As a result, reputation often this is considered important in customer trust and long-term customer relationships. Long-term investment of resources, effort, and attention to building customer relationships is key for a reputable company. Consumers tend to favor companies with good reputations in e-commerce because they perceive less risk and not sure and know where to turn for help (Qalati et al., 2021). Therefore, the following hypothesis is proposed:

Trust
Online trust is the most important element of a business strategy because it reduces perceived risk and can create positive word of mouth. They state that during online shopping consumers, as trustors, find themselves in a risky situation where they use the internet as a tool to communicate their needs to e-vendors and share personal information about themselves. He chose the payment method. He expects the website to be a reliable vehicle for transactions and vendors to behave honestly and professionally when fulfilling purchase requests (Bauman & Bachmann, 2017). Khan et al. (2015) stated that there is no specific definition related to trust, it is a word used in many similar places in psychology, sociology, management, and others. These authors define trust as the binding force in online shopping between buyer and seller transactions. This term consists of 3 main elements (predictability, reliability, and fairness) and is considered an economic calculation in which values are explored through the differences between relationships, maintenance, and creation with the actual cost of serving them (Yuen et al., 2018).

H7: Trust Has A Significant Effect On Purchase Intention
The Mediating Role of Trust Jeon et al. (2017) investigated the mediating role of online trust and the utilitarian value between the perceived interactivity of websites and online travel communities. The analysis, found that online trust plays an important mediating role (Jeon et al., 2017). Based on e-

H7
commerce and customer repurchase intentions, direct and indirect relationships were also found between perceived site quality, perceived reputation, perceived ability to fulfill orders, trustworthiness, and repurchase intention (Qalati et al., 2021).
Based on the description of previous research, the conceptual framework for the formation of hypotheses can be explained in Figure 1.

Research Method
Respondents in this study were students and the general public with an age range of 17-56 years. The criteria for the selected respondents are women and men who use the website of the e-commerce application. The distribution of this research questionnaire uses an online questionnaire link. The total tabulation of the questionnaires collected was 110 and the questionnaires that could be processed were 103. The number of questionnaires that could not be processed was because the respondents filling out the questionnaires were not filled out properly or seriously, so the researchers felt that they were not objective if they were entered or inputted.
Respondents who have filled out the complete questionnaire are divided by gender, age, and domicile qualifications. Respondents by gender with the results were more dominated by women (52,4%) with a total of 54, while men (47,6%) with a total of 49 people. Respondents based on age, the results were dominated by ages 17-26 years (85,4%) with a total of 88 people, ages 27-36 years (7,8%) with a total of 8 people, while 37-46 years (3,9%) with the number of 4 people, while 47-56 years (2,9%) with the number of 3 people. Respondents based on domicile were predominantly from Yogyakarta and its surroundings, but some of the respondents came from outside the province. The measurement in this study uses a Likert scale, there are scores on five levels of answer choices with a score of "5" representing strongly agree and a score of "1" representing strongly disagree. Respondents were asked to answer questions that have been provided including perceived of service quality, perceived of website quality, perceived of website reputation, trust, and purchase intention.

Results and Discussion Factor Analyis a. Instrument Quality Test
Test the quality of the instrument in this study using Smart PLS. The validity test uses the quality of the instrument to measure. What has been done previously is that all indicators show that they are feasible to be used as research instruments, so it is necessary to test the validity of the questions. An indicator is said to be valid if the significance value of each question item is < 0,05. For this reason, the items that will be used for this research are question items that have a significance value of < 0,05. Based on the validity test on each question item, it shows that all indicators used to measure the variables of perceived of service quality, perceived of website quality, perceived of website reputation, trust, and purchase intention in this study have a significant value of less than 5% or below 0,05. So that all indicators that use this research are considered feasible and valid as measuring data which can then be analyzed. Discriminant validity is implemented to ensure that each latent variable is different from one another and all cross-loading values in this study have passed the discriminant validity test which shows a value above 0,7. Based on the table 1 above, it can be explained that the composite reliability value > the average variance extracted or the reliability value based on a significant test of 0,05 means that the items mentioned above are valid. While the high and low reliability empirically states that a number is called the reliability coefficient value. High reliability indicates that the value is close to 1.
Testing the reliability of the instrument by using the cronbach's alpha formula. because this research instrument is in the form of a stratified scale questionnaire. If the alpha value > 0,7 it means that the reliability is sufficient (sufficient reliability), while if the alpha > 0,8 this suggests that all items are reliable and all tests consistently have strong reliability. Based on the explanation above, it can be described for each variable as follows. Perceived of service 1uality use 3 question items and the value of the cronbach's alpha coefficient in this variable is 0,783, which results from state that the item or variable is valid and reliable. Perceived of website quality used 5 question items and the value of the cronbach's alpha coefficient in this variable is 0,966, which results from the state that the item or variable is valid and reliable.
Perceived of website reputation use 3 question items and the value of the cronbach's alpha coefficient in this variable is 0,944, which means that the item or variable is valid and reliable. Trust use 4 question items and the value of the cronbach's alpha coefficient in this variable is 0,900, which results from the state that the item or variable is valid and reliable. Last, purchase intention use 2 question items and the value of the cronbach's alpha coefficient in this variable is 0,891, which results from the state that the item or variable is valid and reliable.

b. Data Analysis Text
The coefficient determinant test (R-Square) is used to measure how far the model explains the dependent variable. The following are the results of the coefficient determinant test in table 2. Based on the table 2 above, it can be seen that the result for trust is 0,663, which means that the influence of the independent variables on perceived of service quality, perceived of website quality, and perceived of website reputation on consumer trust in online shopping is 66,3% and the remaining 33.7% is influenced by other variables which were not present in this study. Furthermore, based on the table above, it can be seen that the results for purchase intention are obtained by the number 0,750, which means that the influence of the independent variables on perceived of service quality, perceived of website quality, and perceived of website reputation on consumer purchase intentions in online shopping is 75% and the remaining 25% is influenced by other variables that are not included in this research.

c. Hypothesis Test
This study uses multiple regression analysis methods to test the hypothesis. The results of hypothesis testing can be seen in table 3 & 4. Table 3. Path Coefficient Table 4

. Hypothesis Test Result
Based on the table 3 and 4 above, several hypotheses were obtained. The first hypothesis shows that perceived of service quality has a positive effect on trust. Based on the table show the results of hypothesis testing with a coefficient value of 0,381 with a significance value of 0,000, which means the value is less than 0,05. Trustworthy, safe when consumers make transactions, as well as an attractive appearance, able to build consumer confidence so that consumer trust can be fulfilled properly. The second hypothesis show that perceived of service quality has a positive effect on purchase intentions. Based on the table shows the results of hypothesis testing with a coefficient value of 0,168 with a significance value of 0,031 which means the value is less than 0,05. Based on the test results it can be said that service quality has an important role in building consumer purchase intentions which are built on the latest information, and easy to understand the available uses and the services provided when shopping. These results provide evidence that consumers who are satisfied with the quality of online shopping services will increase online purchase intentions. The third hypothesis show that perceived of website quality has a positive effect on trust. Based on the table shows the results of hypothesis testing with a coefficient value of 0,249 with a significance value of 0,006, which means the value is less than 0,05. Based on the results of testing the hypothesis it can be concluded that website quality presented always displays the latest information, reliable information, is safe when consumers transact and has an attractive appearance, able to build consumer confidence, so that consumer confidence can be fulfilled properly. The fourth hypothesis show that perceived of website quality has a positive effect on purchase intention. Based on the table shows the results of hypothesis testing with a coefficient value of 0,315 with a significance value of 0.000, which means the value is less than 0,05. Based on the results of testing the hypothesis it can be concluded that website quality in this study has a role in the process of forming consumer buying intentions that are awakened by the latest information. Easy to understand use, reliable information, safe when transacting, and attractive appearance.
The fifth hypothesis show that perceived of website reputation has a positive effect on trust. Based on the table shows the results of hypothesis testing with a coefficient value of 0,316 with a significance value of 0,000, which means the value is less than 0,05. Based on the results of testing the hypothesis it can be concluded that the website reputation presented always displays the latest information, this website is well known to many people, and this website has a different point of view both to consumers and an attractive appearance, able to build consumer confidence, so that consumer confidence can be fulfilled properly. The sixth hypothesis show that perceived of website reputation has a positive effect on purchase intention. Based on the table shows the results of hypothesis testing with a coefficient value of 0,251 with a significance value of 0,005, which means the value is less than 0,05. Based on the results of testing these hypotheses can be concluded that website reputation in this study has a role in the process of forming consumer buying intentions which are built on the latest information, this website has widely known, and this website has a good view of the consumer. The last hypothesis show that trust has a positive effect on purchase intention. Based on the table shows the results of hypothesis testing with a coefficient value of 0,266 with a significance value of 0,003, which means the value is less than 0,05. The following are the results of the mediation test through the indirect effect test. The indirect effect test is the test of the trust as a mediation on the effect perceived of website reputation, perceived of website quality, and perceived of service quality on purchase tntention. The following is the result of the indirect effect perceived of website reputation on trust on purchase intention. In the table 5 above, it can be seen that the estimation results obtained have an indirect effect with several p-value (0,013). The indirect effect perceived of website quality on trust on purchase intention. In the table 5 above, it can be seen that the estimation results obtained have an indirect effect with several p-value (0,049). While the results of the indirect effect perceived of service quality on trust on purchase intentions. In the table 5 above, the estimated indirect effect of p-value (0,019) can be obtained. Based on the results of this study, it can be said that perceived of service quality, perceived of website quality, perceived of website reputation, and trust as mediating factors are influencing factors in purchase intentions.

Conclusion
This research still has some weaknesses which must be improved in future research.
1. Many factors can influence consumers in making online purchase intention through websites. However, the researcher only took the influence from perceived of service quality, perceived of website quality, and perceived of website reputation on purchase intention. This was chosen because there are variables that have not been studied much, especially perceived trust in online shopping. So for further research, it is expected to add variables outside of the variables that have been studied in order to obtain more varied results that can influence purchase intension and consumer confidence. 2. The objects in this study were taken by respondents in general, namely women and men with a vulnerable age of 17-56 years. For further research, it is recommended to expand the object of research so that it can expand the scope of research from various segments. 3. This study has many implications that can be helpful in various fields. From a theoretical point of view, this research helps in establishing the idea that service quality and websites act as modern opinion leaders working in the digital environment to influence e-commerce users and provide information. In addition, this study also presents the idea that source credibility can be considered as a mediator between perceived of service quality, perceived of website quality, perceived of website reputation, trust, and purchase intention. This study helps identify that this service quality is an integral part of the digital communications we see and use. 4. This study has a number of samples collected and successfully analyzed as many as 103 respondents who were taken by purposive sampling method. This research was also conducted by Mudiantono & Zatalini (2019) and Qalati et al. (2021). For further researchers, it is expected to add new variables or the topic is the same but the intended subject or target is different. As well as for further research, it is necessary to be able to test the difference between service quality and information quality in order to produce more detailed and broad research.