GENERATION Z REPURCHASE INTENTION IN INDONESIA E-COMMERCE: E-WOM MODERATION AND CUSTOMER TRUST MEDIATION

: This study aims to analyze the effect of e-service quality on customer trust and repurchase intention, customer trust on repurchase intention, mediation of customer trust in e-service quality on repurchase intention and the effect of e-service quality on moderated customer trust by e-WOM on e-commerce in Indonesia. This research is quantitative research by distributing questionnaires to respondents. The population in this study are Generation Z e-commerce customers in Indonesia. The sample selection method uses non-probability sampling with convenience sampling. The sample of this research is 203 respondents. Data analysis used the PLS Structural Equation Modeling (SEM) method. The results of this study indicate that e-service quality has a positive and significant effect on customer trust and repurchase intention, customer trust has a positive and significant effect on repurchase intention, e-service quality has a positive and significant effect on repurchase intention through customer trust and e-WOM is unable to moderate the effect of e-service quality on customer trust


Introduction
Indonesia will have a demographic bonus from 2030 to 2040.The demographic bonus is defined as a population of productive age greater than non-productive age.In 2030 the Indonesian population of productive age will be more than 60% of the total population (Kominfo 2020).One of these demographic bonuses is Generation Z. Based on the 2020 population census, generation Z occupies the largest portion in Indonesia, namely 27.94% (BPS 2022).Generation Z are individuals born between 1995 and 2012 (Barhate and Dirani 2022;Maioli 2017).Generation Z has different characteristics from previous generations, they were born when technology was widely used, and they did not know a world without the internet (Twenge 2010).The results of a survey conducted by the Alvara Research Center show that the generation that uses the internet the most in Indonesia is Generation Z, with a percentage of 97.7%.In line with the increasing number of Generation Z in Indonesia, the Internet development in Indonesia has also increased rapidly, evidenced by the number of Internet users in Indonesia, which has continued to increase over the past five years.Internet users in Indonesia have grown by 54.25% in the last five years.In January 2022, the number of internet users reached 204.7 million people, with an internet user penetration of 73.7% of the total population of Indonesia (Annur 2022).The increase also occurred in transactions through e-commerce.E-commerce transactions in Indonesia grew by 50.58% from 2020 to 2021.In 2021, the number of ecommerce transactions in Indonesia will reach 401 trillion rupiahs (Elena 2022).Generation Z transactions in e-commerce contributed 36% of all transactions in 2020 (Alinea 2020).Generation Z's contribution to e-commerce transactions in Indonesia will continue to increase due to demographic bonuses in Indonesia from 2030 to 2040.Competition in e-commerce is fierce.This can be seen from the number of visits to Indonesia's top five e-commerce sites, namely Tokopedia, Shopee, Bukalapak, Lazada and Blibli, which have fluctuated from 2018 to 2021.There has even tended to be a decrease in visits to the top five e-commerce sites in Indonesia, from 381.08 million in 2018 to 357.46 million in 2021 (Iprice 2021).The decrease in the number of visits to e-commerce is in contrast to the increase in internet users and an increase in e-commerce transactions in Indonesia.This indicates that customers still have low repurchase intention in the same e-commerce.Low customer repurchase intention is an indication of low customer loyalty.Based on survey results from Surveysensum, 42% of customers still frequently switch from one e-commerce to another (Kompas 2022).Repurchase intention in e-commerce is a customer intention that will cause repurchasing behaviour as described in the Theory of Planned Behavior (TPB).Intention shows how hard someone dares to try, intention shows how much effort someone plans to make, and intention relates to subsequent behaviour (Ajzen 1991).Customer repurchase intention in e-commerce is influenced by e-service quality.Rohwiyati and Praptiestrini (2019) explain that increasing e-service quality will increase customer repurchase intention.Lestari and Ellyawati (2019) and Hongdiyanto et al. (2020) also revealed that customer repurchase intention is significantly influenced by e-service quality.However, oleh Yunus, Fauzi, and Rini (2021) research shows different results and proves that e-service quality has no significant effect on customer repurchase intention.Then, Ikhsan and Lestari (2021) put forward different results, which explained that e-service quality has a negative effect on customer repurchase intention.E-service quality can also affect customer trust in e-commerce.Good e-service quality will increase customer trust in e-commerce (Pratiwi, Sunaryo, and Mugiono 2021).Tran and Vu (2019) also explained that customer trust in online shopping would increase if the e-service quality of e-commerce is considered reasonable by customers.Customer trust can also mediate the relationship between e-service quality and customer repurchase intention (Pratiwi et al. 2021;Tran and Vu 2019).Liu and Tang (2018) and Zhu, Kowatthanakul, and Satanasavapak (2020) explain that high customer trust in e-commerce is required to increase customer repurchase intention.However, research conducted by Ikhsan and Lestari (2021) and Ariyanti, Maihan, and Lestari (2021) revealed different results.It proved that customer trust did not significantly impact e-commerce customer repurchase intention.There are not many studies that use e-WOM as a moderating variable, so a more profound study is needed.E-WOM is expected to increase the effect of e-service quality on customer trust so that it will have an impact on increasing customer repurchase intention in e-commerce in Indonesia.This study aims to analyze the effect of e-service quality on customer trust and repurchase intention, customer trust on repurchase intention, mediation of customer trust in e-service quality on repurchase intention and the effect of e-service quality on customer trust moderated by e-WOM on e-commerce in Indonesia.Surabaya, October 28-29 th , 2023

Theory of Planned Behaviour
The Theory of Planned Behaviour focuses on a person's intention to perform a behaviour because the intention is an intermediate variable that causes behaviour from an attitude or other variables.The intention mediates the influence of various motivational factors that impact behaviour.Intention also shows how hard someone dares to try, intention shows how much effort someone plans to do, and intention is closely related to subsequent behaviour.Intentions can be expressed through the Theory of Planned Behaviour (TPB), namely beliefs or attitudes towards behaviour, subjective norms and behavioural control (Ajzen 1991).The formation of intention plays a distinctive role in directing action, namely connecting deep considerations that are believed and desired by someone with specific activities (Ajzen 1991).

Repurchase Intention
Repurchase intention is a customer's evaluation of repurchasing products or services from the same seller or not, taking into account the current situation and the possibility of future events (Hellier et al. 2003).According to Dharmmesta and Handoko (2000) repurchase intention is a transaction made for the same product or service and then will repurchase.The first step for ecommerce is how to encourage customers to make purchases on e-commerce, and then the next step is how to encourage customers to make repeat purchases (Novitasari et al. 2022).Repurchase intention is important in maintaining a company's image and success of a company.Pappas et al. (2014) explained three indicators of repurchasing requests in e-commerce: (1) Intention to continue shopping in e-commerce in the future.(2) Will continue shopping on ecommerce in the future.(3) Will regularly use e-commerce in the future.Parasuraman, Zeithaml, and Malhotra (2005) explain that e-service quality is the ability of a website to facilitate online shopping, effective and efficient transactions and delivery.This understanding explains the concept of e-service quality, from the pre-purchase process (ease of use, product information, ordering information, and personal data protection) to the postpurchase process (delivery and return policy).According to Ladhari (2010) the indicators used for measuring e-service quality are: (1) Reliability/fulfilment, one of the main dimensions in traditional service quality instruments, refers to service performance according to promises in a precise and timely manner.(2) Responsiveness is a response that refers to the willingness to help customers or users, responsiveness to customer questions and problems, and the existence of alternative communication channels provided by the website.(3) Ease of use/ability, leading to easy access to existing information.This is a crucial reason for customers to make virtual shopping decisions.Ease of website use is essential in e-service quality because the e-commerce ecosystem can make customers feel challenged to use e-commerce.(4) Privacy/security, leading to the protection of customer personal and financial information.The intended protection is measured as the website's ability to feel safe by customers.This indicator is very suitable because the potential risk of financial loss and fraud in virtual transactions is relatively high.Security is a crucial influence on the intention to revisit a website and make a purchase.

E-Service Quality
(5) Web design refers to online catalogues' aesthetic features, content and structure.Web design has an essential role in attracting and retaining visitors, and content that has an equally important role.( 6) Information quality refers to the availability and accuracy of information customers obtain when visiting the website.E-service quality can affect customer trust in e-commerce.Good e-service quality can increase customer trust in e-commerce (Pratiwi et al. 2021).Tran and Vu (2019) also explained that customer trust in online shopping would increase if the e-service quality of e-commerce is considered reasonable by customers.E-service quality can also affect customer repurchase intention in e-commerce.Rohwiyati and Praptiestrini (2019) explain that increasing e-service quality will increase customer repurchase intention.Lestari and Ellyawati (2019) and Hongdiyanto et al. (2020) also revealed that customer repurchase intention is significantly influenced by e-service quality.Hypothesis 1: E-service quality has a positive effect on customer trust Hypothesis 2: E-service quality has a positive effect on repurchase intention

Customer Trust
Customer trust is a person's desire to involve himself with others engaged in exchanges caused by someone having faith in different parties (Moorman, Deshpande, and Zaltman 1993).Customer trust leads to a person's belief that other people will behave based on their expectations and the hope that other people who have been chosen to be trusted will not act opportunistically by taking advantage of the situation (Kamtarin 2012).McKnight, Choudhury, and Kacmar (2002) explained that the indicators used to measure customer trust are: (1) Trusting Belief is how high a person believes and feels trust in other people in certain situations.Trusting Belief is the opinion of someone who believes (customers) in a trusted party (online shop seller) where the seller has characteristics that can provide benefits to customers.(2) Trusting Intention is done intentionally in which a person is willing to depend on other parties in certain situations; this happens individually and leads directly to other people.The element that forms trusting Intention is the willingness to depend, namely, the individual's willingness to depend on e-commerce.Liu and Tang (2018) and Zhu, Kowatthanakul, and Satanasavapak (2020) explain that high customer trust in e-commerce is required to increase customer repurchase intention.However, research conducted by Ikhsan and Lestari (2021) and Ariyanti, Maihan, and Lestari (2021) revealed different results.It proved that customer trust did not significantly impact e-commerce customer repurchase intention.Customer trust can also mediate the relationship between eservice quality and customer repurchase intention (Pratiwi et al. 2021;Tran and Vu 2019).Tran and Vu (2019) explain that e-service quality significantly influences repurchase intention through customer trust.Hypothesis 3: Customer trust has a positive effect on repurchase intention Hypothesis 4: E-service quality has a positive effect on repurchase intention through customer trust E-Word of Mouth (WOM) E-WOM is a sentence or discussion about a product, service or organization, whether positive or negative and can be accessed by the whole community online (Hennig-Thurau et al. 2004).According to Litvin, Goldsmith, and Pan (2008) E-WOM is the entire non-formal communication intended for customers through internet media relating to the use or specifications of products, both in the form of goods and services as well as sellers.Generally, word of mouth, or WOM, can be interpreted as a marketing communication tool incorporated into the promotion mix (Prahiawan et al. 2021).Marketing communication aims to build awareness and increase customer intention in a product or brand, which in turn can make customers take action to buy the brand.E-WOM is a positive or negative statement about a product or company created by a potential customer, which is available to many people and institutions via the Internet (Prahiawan et al. 2021).Surabaya, October 28-29 th , 2023 According to Lin, Wu, and Chen (2013), indicators for measuring e-WOM are (1) Quality of e-WOM, Quality of e-WOM Related to the quality of the content of the reviews read by customers and how easily customers can understand these reviews.(2) The number of e-WOM, The number of e-WOM is related to the number of comments posted and can represent product performance.The number of comments also describes the popularity of the product.(3) Sender's expertise is the sender of the message when making comments that are used to attract customers to make decisions.E-WOM that is positive and has good quantity and quality will strengthen the effect of e-service quality on trust with customers because, with the help of E-WOM, customers will quickly find out about the quality of online services provided by e-commerce.Therefore it will have an impact on increasing customer trust in e-commerce.Hypothesis 5: E-WOM moderates the effect of e-service quality on customer trust

Research Model
The research model is shown in Figure 1 below:

Research design
This research is quantitative research by distributing questionnaires to respondents.This research is cross-sectional research.The research was conducted in Indonesia from January to May 2022.

Population and Sample
The population in this study is the top 5 e-commerce customers in Indonesia (Tokopedia, Shopee, Bukalapak, Lazada and Blibli) who fall into the Z generation category (born 1995-2012).The sample selection method uses non-probability sampling with convenience sampling.Because the population size is uncertain, determine the number of samples using recommendations from Roesco.This explains that the number of samples is at least 10-20 times the variable under study (Sugiyono 2010).So, the minimum sample in this study with four variables is 4x20 = 80 respondents.Because this study tested models and hypotheses using SEM PLS, Hair et al. (2010) recommend a sample size of 100-200 respondents.So, in this study, 500 questionnaires were distributed throughout Indonesia following the weight of the population based on the 2020 population census on six major islands in Indonesia (56.10% in Java, 21.68% in Sumatera, 7.36% in Sulawesi, 6.15 % in Kalimantan, 5.57% in Bali-Nusa Tenggara and 3.17% in Papua-Maluku).Questionnaires that were filled out and met the criteria and continued for data processing in this study totalled 203.

Variable Measurement
Each variable is measured by distributing questionnaires to respondents using a 5-point Likert scale (where 1-5 indicates strongly disagree to agree with the statements submitted strongly).The variable of repurchase intention is measured

Data Analysis
The descriptive analysis used in this study is an analysis of the characteristics of the respondents and an analysis of the respondents' responses.Then a questionnaire feasibility test consisted of: (a) Validity test, a validity test conducted to measure the accuracy of research instruments or questionnaires.The questionnaire is valid if the questions or statements can reveal something that the questionnaire will measure.The accuracy value of the questionnaire can be measured using the correlation coefficient.The questionnaire is excellent and valid if the correlation coefficient is > 0.3 (Robinson 2018).(b) Reliability Test is a questionnaire test conducted to measure the consistency of respondents' answers.The reliability test was carried out with the Cronbach Alpha statistical test.The questionnaire is reliable if the Cronbach Alpha value is > 0.70 (Robinson 2018).After that, the PLS-SEM Model test was carried out by testing the outer model and inner model.The outer model consists of Outer loading, Composite reliability (CR) and Average Variance Extracted (AVE).While the Inner model consists of R 2 and Q 2 .Then a hypothesis test is carried out by looking at the P-values from bootstrapping on the PLS-SEM used as an alternative in determining whether the hypothesis is accepted or rejected.The criteria for the P value in testing the hypothesis are as follows: (a) P-values < α = 0.05, the hypothesis is accepted.This means that partially exogenous variables have a significant influence on endogenous variables.(b) P-values > α = 0.05, the hypothesis is rejected.This means that partially exogenous variables do not significantly affect endogenous variables.

Characteristics of Respondents
The characteristics of the respondents in this study are shown in Table 1.

Validity and Reliability Test Results
The results of testing the validity of each statement item and the reliability of each variable are shown in Table 2 below: Table 2 shows that all statement items in this study are valid because the correlation value of all items is greater than 0.30.The Cronbach's Alpha value for each variable in this study was greater than 0.70, meaning each variable was reliable.

PLS-SEM Model Test Results
The results of the PLS Structural Equation Modeling (SEM) test in this study are shown in Figure 2 and Table 3.Based on the PLS-SEM model test, it can be seen that (Table 3) each indicator in this study has a loading factor value greater than 0.60.so that all indicators can be said to be valid.The Composite Reliability value for each research variable is greater than 0.70.meaning that all variables in this study are reliable.The AVE value for each research variable is greater than 0.50.meaning that all variables in this study are valid.The R square value shows a moderate relationship between exogenous variables and endogenous variables.It can be explained that the R Square value of 0.728 or 72.8% of the customer trust variable can be explained by the e-service quality and e-WOM variables.While the remaining 27.2% is explained by other factors not examined by this study.While the test results on the repurchase intention variable obtained an R Square value of 0.652 or 65.2%, the Surabaya, October 28-29 th , 2023 e-service quality and customer trust variables can explain the repurchase intention variable.While the remaining 41.1% is explained by other factors not examined in this study.Then, the Q square value is above 0.00, meaning that this research model has a good observation value or the model has a predictive relevance value.

Hypothesis Test Results
The results of hypothesis testing in this study are shown in Table 4.

Discussion
Effect of E-Service Quality on Customer trust E-Service quality positively and significantly affects e-commerce customer trust in Indonesia.
That is, the better the e-service quality received by customers, the higher the trust of Generation Z customers in e-commerce in Indonesia.The results of this study support the results of research conducted by Pratiwi, Sunaryo, and Mugiono (2021) and Tran and Vu (2019), which explain that good e-service quality can increase customer trust.E-Service quality is the ability of e-commerce websites to facilitate online transactions, including effective and efficient delivery of goods purchased online (Parasuraman et al. 2005).Surabaya, October 28-29 th , 2023 E-Service quality will shape the attitude toward behaviour, which is an indicator of the Theory of Planned Behaviour (TPB) put forward by Ajzen (1991).In this study, attitude is represented by customer trust in e-commerce.E-Service quality that is received and felt by e-commerce customers will shape customer attitudes in the form of customer trust, which will determine further behaviour.From this study, it was found that e-service quality is an antecedent of attitude towards TPB.
Effect of E-Service Quality on Repurchase Intention E-Service quality has a positive and significant effect on the repurchase intention of Generation Z customers in e-commerce in Indonesia.The better the e-service quality, the higher the intention to repurchase Generation Z customers in e-commerce in Indonesia.The results of this study are in line with the results of research conducted by Rohwiyati and Praptiestrini (2019), Lestari and Ellyawati (2019) dan Hongdiyanto et al. (2020), which explains that e-service quality has a positive and significant influence on customer repurchase intention.However, the results of this study are not in line with the results of research conducted by.Yunus, Fauzi, and Rini (2021)prove that e-service quality does not significantly impact customer repurchase intention.
Repurchase intention is a customer's desire to re-transact a product or service because they are satisfied with the goods or services in previous transactions.Repurchase intention is an indicator of intention in the Theory of Planned Behaviour (TPB).From the results of this study, it was found that there was a direct effect of e-service quality on intentions represented by the repurchase intention of Generation Z customers in e-commerce.

The Effect of Customer Trust on Repurchase Intention
Customer trust has a positive and significant effect on the repurchase intention of Generation Z customers in e-commerce in Indonesia.That is, the higher customer trust, the higher the intention to repurchase Generation Z customers in e-commerce in Indonesia.The results of this study support the results of research conducted by Liu and Tang (2018) dan Zhu, Kowatthanakul, and Satanasavapak (2020), which prove that customer trust has a positive and significant influence on repurchase intention.However, the results of this study do not support the results of research conducted by Ikhsan and Lestari (2021) dan Ariyanti, Maihan, and Lestari (2021), which prove that customer trust has no significant effect on repurchase intention.
Customer trust is a positive attitude of customers towards a product, and these customers believe that the seller will meet customer expectations.In the Theory of Planned Behaviour (TPB), customer trust is a representation of attitude toward behaviour which will affect the intention represented by repurchase intention and ultimately impact customer behaviour in repurchasing.
Effect of E-Service Quality on Repurchase Intention through Customer Trust E-service quality significantly affects repurchase intention through Generation Z customer trust in e-commerce in Indonesia.That is, customer trust can mediate the effect of e-service quality on repurchase intention.Good e-service quality will increase customer trust so that customers who trust e-commerce will have a high repurchase intention.This study's results align with the results of research conducted by Pratiwi et al. (2021) dan Tran and Vu (2019), which explain that customer trust can mediate the effect of service quality on repurchase intention.
The results of this study support the theory put forward by Baron and Kenny (1986)regarding partial or partial mediation, that is if the effect of the independent variable on the dependent variable decreases but is not equal to zero by entering the mediating variable into the equation.So, customer trust in this study is a partial mediating variable in the relationship between e-Surabaya, October 28-29 th , 2023 service quality and repurchase intention because there is a significant direct effect between eservice quality and repurchase intention.
Effect of E-Service Quality on Customer Trust moderated by E-WOM E-WOM cannot moderate the effect of e-service quality on Generation Z customer trust in ecommerce in Indonesia.This is supported by statistical analysis of hypothesis testing in this study because E-WOM is an antecedent variable of customer trust.Therefore, the effect of E-WOM as a moderating variable is not significant.
In addition, the cause of E-WOM does not moderate the effect of e-service quality on customer trust because Generation Z is a generation that understands information technology and spends a lot of time using social media, so they read and see E-WOM a lot about e-commerce.Therefore E-WOM is an antecedent of customer trust, not a moderation.Positive e-WOM will have a direct effect on increasing customer trust.
In the Theory of Planned Behavior (TPB), there are indicators of subjective norms which can be interpreted as social influences that cause a person to take action or behave (Ajzen 1991).E-WOM in research can represent subjective norms on TPB.This study found that E-WOM influences customer trust, which represents an attitude towards TPB.That is, subjective norms influence the attitude of TPB.

Conclusions
Generation Z can use information technology in various activities, including online trading or e-commerce.From this study, for Generation Z customers in Indonesia, e-service quality is important in increasing customer trust and repurchase intention.Then, high customer trust will generate Generation Z customer repurchase intention in e-commerce in Indonesia.Customer trust can also partially mediate the effect of e-service quality on repurchase intention.This study also found that E-WOM is not a moderating variable between e-service quality and customer trust but functions as an antecedent of customer trust.Then, E-service quality is an antecedent of attitude on the Theory of Planned Behavior (TPB), which in this study is represented by customer trust.Then, E-WOM in this study represents a subjective norm on TPB that has a direct influence on the attitude of customer trust.

Figure
Figure 1.Research Model using three indicators proposed by Pappas et al. (2014): Intention to continue shopping in e-commerce in the future, will continue shopping in e-commerce in the future and will regularly use e-commerce in the future.The e-service quality variable is measured using six indicators proposed by Ladhari (2010): Reliability/fulfilment, Responsiveness, Ease of use/ability, Privacy/security, Web design and Information quality.The customer trust variable is measured using two indicators by McKnight, Choudhury, and Kacmar (2002): Trusting Belief and Trusting Intention.The e-WOM variable is measured using three indicators proposed by Lin, Wu, and Chen (2013): e-WOM quality, e-WOM quantity and Sender's expertise.