The Influence of Service Quality on Customer Satisfaction and Its Impact on Repurchase Interests in Telkomcell Products in Timor Leste

Telkomcel's customers started to complain and lost their trust and their customers got worse and worse. In order to take back the customer's trust, Telkomcel implemented two main strategies, namely the Push Strategy and the Pull Strategy in its marketing activities. This study aims to provide an overview of Specific Service Quality in three dimensions, namely Tangible, Empathy and Assurance, Customer Satisfaction and Repurchase Intention at Telkomcel Timor Leste, and to determine the effect of service quality on Repurchase Intention with customer satisfaction interventions at Telkomcel East Timor. Based on the results of distributing questionnaires to 200 people at several universities in Timor Leste such as UNTL (Timor Loro Sae National University), UNPAZ (Universidade Da Paz), UNITAL (Universidade de Oriental) and DIT (Dili Institute of Business) ISC (Institut Unggul Cristal) 200 students, lecturers and administrative staff on campus randomly using a Likert scale and non-probability sampling techniques and purposive sampling. SEM (Structural Equation Modeling) data analysis technique, with the help of SmartPLS 3.0 software.


INTRODUCTION
Telkomcel as one of the operators that competes in the Timor-Leste telecommunications market, coupled with the Pandemic, of course requires Telkomcel to change its marketing strategy.In 2020, Telkomcel will still focus on the needs of existing customers and will also continue to try to increase the customer base (Telkomcel, 2020 and2021).However, from the other side, Telkomcel is also facing a pandemic, and the Timor Leste government has launched a new regulation regarding frequencies in Timor Leste, in 2021 which is a critical year, because there is a reduction in several frequencies because it is considered that there is no license yet.With this situation giving a significant impact on Telkomcel, customers start complaining and losing trust and customers are getting lower and lower, to take back the trust of customers, Telkomcel applies two main strategies, namely Push Strategy and Pull Strategy International Journal of Social Service and Research https://ijssr.ridwaninstitute.co.id/IJSSR Page 1884 results were significant, such as research conducted by (Hamzyah) that: (1) Tangible variables have a positive (6.757) and significant (0.000) effect on customer satisfaction or it can be said that tangible has a positive and significant effect on customer satisfaction.(2) The variable empathy has a positive (3.295) and significant (0.001) effect on customer satisfaction or it can be said that empathy has a positive and significant effect on customer satisfaction.(3) The variable that has the dominant influence on customer satisfaction is tangible (X2), this is obtained from the beta value based on standardized coefficients, namely 0.523 or 52.3% compared to other variables with the required significance.
However, the results of this research were carried out at one of the PT.PLN (Persero) in Enrekang Regency, the researcher wants to try to conduct research in the telecommunications sector, namely the Telkomcel Timorleste Company.Reliability of assurance (Assurance), empathy (Empathy), and tangibles (Tangibles).One of the strategies that Telkomcel can use to maintain its position in the Timor Leste telecommunications market, is because researchers observe market situations and conditions in Timor Leste.Therefore to find out how much influence the three dimensions, namely Tangible Empathy and Assurance have on (Customer Satisfaction) and (Repurchase Intention) Of all the dimensions of service quality, all have a positive effect on satisfaction.However, of all the variables, only empathy has a positive and significant effect on satisfaction.It can be interpreted that empathy in serving students such as a friendly attitude and respect for administrative staff in serving students will contribute to student satisfaction so that it increases significantly.Furthermore, student satisfaction which increases significantly for the attention and service provided by administrative staff will contribute significantly to student intentions to do word of mouth.
The desires of consumers are always changing.Entrepreneurs are required to conduct a survey to find out what kind of product strategy should be used to develop products that can be accepted by consumers.A new product development strategy must be made that is able to survive and compete with competitors' product types, so that the company can achieve its goals Based on his scientific study, Zungu & Mason (2019) said that product development is not only important for the success of the product but also is the key to determining the success of the company.The steps used in product development are searching for and finding development ideas from various sources after that think about the company's ability to carry out this development, if the company feels capable then predict customer demand as the key to the success of this development.Carry out the development process within the company (functional specifications) then carry out production of the product (product specifications).After that, review the design, carry out market tests, as well as market introductions, and finally carry out periodic evaluations for the product development process that is accepted by customers.(Layla Hafni, 2020) Of all the activities carried out by a company, it will ultimately lead to the value that will be given by customers regarding perceived satisfaction.The definition of consumer satisfaction varies widely and will provide a broader understanding, while some experts define customer satisfaction as follows: According to Kotler in (Ang et al., 2012), satisfaction is the level of feeling at which a person states the results of a comparison of product (service) performance) is accepted and expected.According to Chandra, Hafni, Chandra, et al. (2019) states that consumer satisfaction or dissatisfaction is the consumer's response to the evaluation of the perceived discrepancy/disconfirmation between previous expectations (or other performance norms) and the product's actual performance that is felt after its use (Stefani, 2020).

METHODS
This study used a cross-sectional approach to determine the timing and frequency of data collection and processing.When data is collected at one time, processed, evaluated, and conclusions drawn, this is called using a cross sectional technique (Indrawti, 2015: 18).The population in this study were students and lecturers and staff at 10 campuses such as UNTL, UNPAZ, UNDIL, UNITAL, UCTL, IOB, CRISTAL, ICS, DIT, and ISJB in Timor Leste.This population is estimated at 17,621 people.
(Indrawati 2015: 164), the sample is people from the population selected to participate in the research, namely from several campuses so far Telkom has divided student SimCards either to be observed, treated, or surveyed about what is being investigated.The method used in this study is a stratified random sampling approach, meaning that individuals from the population do not have the same opportunity to be selected as a sample, regardless of that the researcher chooses several students to be representative and the researcher uses a 5% margin of error whether they have a chance the same or not.Meanwhile, the sample strategy for this study was stratified random sampling.( Based on the results of calculations using the slovin formula, the minimum sample for this study is 201 respondents.Respondents to this study were people who had used Telkomcel products at least once, and to reduce errors in filling out this questionnaire, the researchers rounded up the number of respondents to 201 respondents.

Validity test
Questionnaire question items are measuring instruments (tools) that can test what the research objectives are, so all statements in a questionnaire are required to measure their validity.Validity indicates the extent to which a measuring instrument can measure something that has been determined, so that the greater the validity of a measuring instrument, the more likely the measuring instrument will fulfill its purpose or be more oriented to what it should measure (Indawati 2015: 146) .

Source: Primary data processed in 2022
From the data table 1 above, it can be seen that the overall value of the loading factor shows more than > 0.7 so that all of the research variables are declared valid and can be continued in other research analyses.The discriminant test results in table 2 present the results of the cross loading calculation, which shows that the cross loading value of each indicator in the service assurance variable is above the cross loading value of the other variables.All values above the threshold are 0.7 so that the research instrument is said to be discriminantly valid.The results of obtaining Customer Satisfaction values are greater than 0.7 so that the research instrument is said to be discriminantly valid.Whereas the Repurchase Intention variable indicator is above the cross loading value of the other variables.The results of the acquisition of the Repurchase Intention value above are greater than 0.7 so that the research instrument is said to be discriminantly valid.

Reliability Test
Reliability is defined as something related to the level of trust, reliability, consistency, or stability of the results of one measurement.Reliability is one of the characteristics that a good measuring instrument theoretically has a reliability coefficient between -1 to +1, but a coefficient that is less than zero (0) is meaningless because the interpretation of reliability always refers to a positive coefficient (Indawati , 2015:155 ).Based on the test results in table 3 the composite reliability value of the Assurance variable was 0.895 with a Cronbach's Alpha value of 0.843.On the variable Customer Satisfaction, a composite reliability value of 0.911 was obtained with a Cronbach's Alpha value of 0.870.on the Empathy variable, a composite reliability value of 0.880 was obtained with a Cronbach's Alpha value of 0.819.The Repurchase Intention Variable obtained a composite reliability value of 0.913 with a Cronbach's Alpha value of 0.872.the Tangible variable obtained a composite reliability value of 0.920 with a Cronbach's Alpha value of 0.883.

Table 1Composite Reliability and Cronbach's Alpha
So it can be interpreted that all the variables above are said to be reliable because the composite reliability is greater than 0.7 and Conbach's alpha is above 0.6, so the questionnaire can produce stable or constant measuring results.From table 4.2.3 above, all variables with reflective indicators have an AVE (Average Variance Extracted) value of more than 0.5.It can be said that all of these variables are valid.Perceived risk and perceived benefit variables do not use the AVE measure because they are formative indicator variables.

Discussion of Research Results Evaluation of the Measurement Model (Outer Model)
After testing the outer model that has fulfilled, then testing the inner model (structural model) is carried out.The inner model can be evaluated by looking at the r-square (indicator reliability) for the dependent construct and the t-statistical value of the path coefficient test.The higher the r-square value means the better the prediction model of the proposed research model.The path coefficients value indicates the level of significance in hypothesis testing.Indicator reliability aims to assess whether latent variable measurement indicators are reliable or not.You do this by evaluating the results of the outer loading of each indicator.A loading value above 0.7 indicates that the construct can explain more than 50% of the indicator variance (Wong KK, 2013;Sarstedt et al., 2017).Therefore, it refers to the rule of thumb if the value indicates an R-square model of 0.67 (Strong), 0.33 (Moderate) and 0.19 (Weak) (Chin, 1988) so this research model is classified as moderate.However, the results from Table 4.3.1 show that, the two variables such as Responsiveness, Reliability have a greater influence on the affected variables than the factors or dimensions derived from the model studied.So based on the table of F Square values above, the effect size that is small or negligible is the effect of X2 on Y1 and Y2.The large size effect is the effect of X1 on Y1, while the other effects are moderate.

Table 6. Fornell-Larcker Criteria
Table 6 Fornell-Larcker criteria shows the AVE root value of each construct or variable.The AVE roots are indicated by numbers marked in yellow in the table.The AVE root value is then compared with the model's R-square value.It is known that the R-square model has a value of 0.742.From table 4.17 it can be seen that the AVE root score for each variable is greater than the R-square value.Thus all variables are declared valid and can be used for further tests.Discriminant validity can also be known from the cross loading score or cross validity where the indicator value of a variable to the variable itself must be greater than the correlation of the indicator to other variables.The cross loading score is known from the calculation results using the PLS algorithm technique.Outer Loading Table 4.2.2Outer Loading calculation can be seen from the attachment From the Outer Loading calculation results it is known that the indicators of all variables are valid.Q-square analysis (Q2) and q2 effect size

Source: SmartPLS Processing Data Version 3
Based on table 7 above, it can be seen that the variability of the Y1 Customer Satisfaction construct is 0.283.Furthermore, the construct Repurchase Intention Y2 is 0.345.Therefore it refers to the rule of thumb chin (1998) if the value of Q 2 > 0 then the model has predictive relevance so from these results it shows that this model has predictive relevance.The Output Path Coefficient as shown in the picture above is to see the magnitude of the direct effect (DIRECT EFFECT) of each independent (exogenous) variable on the dependent (endogenous) variable.

Tangible and Customer Satisfaction
The magnitude of the parameter coefficient for the Tangible variable on the Customer Satisfaction Variable of 0.377 which means there is a Tangible positive influence on Customer Satisfaction Or it can be interpreted that the better the Tangible value then Customer Satisfaction will increase.Increase of one tangible unit will also increase Customer Satisfaction by 37.7%.Based on calculations using bootstrap or resampling, where the results of the Tangible estimation coefficient test for Customer Satisfaction bootstrap results are 0.371 with a t-count value of 4.791 greater than t-table (1.64) then the p-value is 0.000 <0.05 so Accept H1 or which means a direct influence Tangible to Customer Satisfaction statistically significant or significant .Judging from the previous research data by Caecilia Soares (2018) that: Tangible has a significant effect on Customer Satisfaction.Because the resulting t-count value of 9,004 is greater than t-table (1.64) with a significance level of 95% and a confidence level of 5%.Based on this analysis it can be concluded that the tangible variable as a dimension for strengthening service quality at Telkomcel Timor Leste, from these results shows that the reduction in frequency has an impact on customer trust decreasing from 95% to 37.7% meaning that the tangible being studied here is the Student Simcard part of (Mobile Network Operator (MNO) will have a direct impact on customer satisfaction, especially students /I in Timor Leste who use Telkomcel Student SimCards.

Empathy and Customer Satisfaction
The magnitude of the parameter coefficient for the Empathy variable on Customer Satisfaction of 0.103 which means there is no positive effect of Empathy towards Customer Satisfaction.Or it can be interpreted that the Empathy value is getting worse then Customer Satisfaction will decrease.Decrease one unit of Empathy will also reduce Customer Satisfaction by 10.03%.Based on calculations using bootstrap or resampling, where the test results estimate the Empathy coefficient of Customer Satisfaction bootstrap results are 0.103 with a t-count value of 1.488 smaller t-table (1.64) then the p value is 0.137 > 0.05 so Accept H0 is not significant or does not have a direct effect on Empathy on Customer Satisfaction is not significant or not statistically significant.The results of this statistic show that, meaning this SimCard student, telkomcel distributes it for free and with a data package of $ 25 per month and free calls to fellow student simcards, so the customer emapi will be higher and customer satisfaction will also increase automatically.

Assurance and Customer Satisfaction
International Journal of Social Service and Research https://ijssr.ridwaninstitute.co.id/

IJSSR Page 1892
The magnitude of the parameter coefficient for the Assurance variable on Customer Satisfaction of 0.296 which means there is a positive influence of Assurance on Customer Satisfaction Or it can be interpreted that the better the Assurance value then Customer satisfaction will increase.Increase of one unit of Assurance will also increase Customer Satisfaction by 29.6%.Based on calculations using bootstrap or resampling, where the results of the Assurance estimation coefficient test on Customer Satisfaction bootstrap results are 0.296 with a t-count value of 3.946 greater than t-table (1.64) then the p-value is 0.000 <0.05 so Accept H1 or which means the direct influence of Assurance to Customer Satisfaction statistically significant or significant .From these statistical results it proves that Assurance has a significant impact on Customer Satisfaction , which means that, in the process of distributing SimCadr students, Telkomcel does this formally, first making an application letter to the university, and the university answers based on a request from Telkomcel, and based on data -This student data that Telkomcel distributes this SimCard is guaranteed not to be published because there are already regulations that have been implemented by Telkomcel.

Tangible and Repurchase Intentions
The magnitude of the parameter coefficient for the Tangible variable on Repurchase Intention of 0.246, which means there is a positive influence on the Tangible variable on the Repurchase Intention variable .Or it can be interpreted that the better the Tangible value, the Repurchase Intention will increase.Increase of one tangible unit will also increase Repurchase Intention by 24.6%.Based on calculations using bootstrap or resampling, where the results of the Tangible estimation coefficient test for Repurchase Intention bootstrap results are 0.246 with a t-count value of 3.284 greater than t-table (1.64) then the p-value is 0.001 <0.05 so Accept H1 or which means a direct influence Tangible to Repurchase Intention statistically significant or significant .Based on this analysis it can be concluded that the tangible variable as a dimension for strengthening service quality at Telkomcel Timor Leste, from these results shows that the reduction in frequency has an impact on Repurchase Intention of 24.6% meaning that the tangible being studied here is the Simcard Student part of (Mobile Network Operator) MNO will have a direct impact on Repurchase Intention, especially students /I in Timor Leste who use Telkomcel Student SimCards.

Empathy and Repurchase Intention
The magnitude of the parameter coefficient for the Empathy variable on Repurchase Intention of 0.117 which means there is no positive influence on the Empathy variable on the Repurchase Intention variable .Or it can be interpreted that the value of Empathy is getting worse then Repurchase Intention will decrease.Decrease one unit of Empathy will also reduce Repurchase Intention by 11.7%.Based on calculations using bootstrap or resampling, where the test results estimate the Empathy coefficient of Repurchase Intention bootstrap results are 0.117 with a t-count value of 1.589 smaller t-table (1.64) then the p-value is 0.113 > 0.05 so Accept H0 is not significant or which means there is no direct influence of empathy to Repurchase Intention is not significant or not significant.The results of this statistic show that, it means that this SimCard student, Telkomcel distributes it for free and with a data package of $ 25 per month and free calls to fellow SIM Card Students, so the emapi of customers will be higher to request or request and the spread of information on the distribution of Simcar is also fast from students one to another student is also fast, so the demand for student SimCards will increase accordingly.

Assurance and Repurchase Intention
The magnitude of the parameter coefficient for the Assurance variable on Repurchase Intention is 0.203, which means that there is a positive influence on Assurance on Repurchase Intention.Or it can be interpreted that the better the Assurance value then Repurchase Intention will increase.Increase of one unit of Assurance will also increase Repurchase Intention by 20.3%.Based on calculations using bootstrap or resampling, where the results of the Assurance estimation coefficient test on Repurchase Intention bootstrap results are 0.203 with a t-count value of 2.767 greater than t-table (1.64) then the p-value is 0.006 <0.05 so Accept H1 or which means the direct influence of Assurance to Repurchase Intention statistically significant or significant .From the results of this statistic proves that Assurance has a significant effect on Repurchase Intention, which means. the process of distributing student SimCadr, this is Telkomcel doing it formally, first making an application letter to the University, and the university answered based on a request from Telkomcel, and based on the student data that Telkomcel did the distribution of this SimCard guaranteed it will not be published because it already exists regulations that have been implemented by Telkomcel.So from this process requests for student SimCards each period run well, and efficiently.

Customer Satisfaction and Repurchase Intention
International Journal of Social Service and Research, Carlos Soares Ribeiro, Diane Indiyati

IJSSR Page 1893
The magnitude of the parameter coefficient for the variable Customer Satisfaction on Repurchase Intention is 0.318, which means there is a positive influence on Customer Satisfaction on Customer Satisfaction.Or it can be interpreted that the better the value of Customer Satisfaction then Repurchase Intention will increase.An increase in one unit of Customer Satisfaction will also increase Repurchase Intention by 31.8%.Based on calculations using bootstrap or resampling, where the test results of the estimated Customer Satisfaction coefficient of Repurchase Intention bootstrap results are 0.318 with a tcount value of 4.839 greater than t-table (1.64) then the p-value is 0.000 <0.05 so Accept H1 or which means the direct effect of Customer Satisfaction on Repurchase Intention is statistically significant or significant .So based on statistical results that Customer Satisfaction has a very significant influence on Repurchase Intention, which means that.In this three-dimensional study, one dimension is not significant for Variavel Y1 and Y2, meaning that empathy does not have a direct impact on these two variables because what is examined here is a student SimCard with a tag of $25/month and free calls to fellow SimCard students, so customers always empathetic to telkomcel products.And two other dimensions have a significant influence on the variables Y1 and Y2, meaning that the tangible and assurance variables directly affect the two variables.When there is a reduction in frequency, it will have an impact on customer satisfaction and by itself customer interest in repurchasing Telkomcel products will also decrease.

CONCLUSION
Based on the results of the research and discussion that has answered the formulation of the problem regarding the influence of Service Quality in this study, researchers only take three dimensions of Servqual, namely Tangible, Assurance, and Empathy for Repurchase Intention with Customer Satisfaction as an intervening variable in Tekomcel products in Timor Leste, so it can be concluded as follows: (1) The results of the Tangible estimation coefficient test on Customer Satisfaction bootstrap results are 0.371 with a t-count value of 4.791 greater than t-table (1.64) then the p value is 0.000 <0.05 so Accept H1 or which means a direct influence Tangible to Customer Satisfaction statistically significant or significant .Based on this analysis it can be concluded that the tangible variable as a dimension to directly influence service quality at Telkomcel Timor Leste, from these results shows that the frequency reduction has an impact on trust of 37.7% meaning that the tangible being studied here is the Simcard Student part of (Mobile Network) Operator) MNO will have a direct impact on customer satisfaction, especially students /I in Timor Leste who use ) then the p-value is 0.113 > 0.05 so Accept H0 is not significant or which means there is no direct influence of empathy to Repurchase Intention is not significant or not significant .The results of this statistic show that, it means that this SimCard student, Telkomcel distributes it for free and with a data package of $ 25 per month and free calls to fellow SIM Card Students, so the emapi of customers will be higher to request or request and the spread of information on the distribution of Simcar is also fast from students one to another student is also fast, so the demand for student SimCards will increase accordingly.(6) Assurance estimation coefficient test results against Repurchase Intention bootstrap results are 0.203 with a t-count value of 2.767 greater than t-table (1.64) then the p-value is 0.006 <0.05 so Accept H1 or which means the direct influence of Assurance to Repurchase Intention statistically significant or significant .From the results of this statistic proves that Assurance has a significant effect on Repurchase Intention, which means. the process of distributing student SimCadr, this is Telkomcel doing it formally, first making an application letter to the University, and the university answered based on a request from Telkomcel, and based on the student data that Telkomcel did the distribution of this SimCard guaranteed it will not be published because it already exists regulations that have been implemented by Telkomcel.So from this process requests for student SimCards each period run well, and efficiently.( 7) the test results of the estimation coefficient of Customer Satisfaction against Repurchase Intention bootstrap results are 0.318 with a t-count value of 4.839 greater than t-table (1.64) then the p-value is 0.000 <0.05 so Accept H1 or which means the direct effect of Customer Satisfaction on Repurchase Intention is statistically significant or significant .So based on statistical results that Customer Satisfaction has a very significant influence on Repurchase Intention, which means that.In this three-dimensional study, one dimension is not significant for Variavel Y1 and Y2, meaning that empathy does not have a direct impact on these two variables because what is examined here is a student SimCard with a tag of $25/month and free calls to fellow SimCard students, so customers always empathetic to telkomcel products.And two other dimensions have a significant influence on the variables Y1 and Y2, meaning that the tangible and assurance variables directly affect the two variables.When there is a reduction in frequency, it will have an impact on customer satisfaction and by itself customer interest in repurchasing Telkomcel products will also decrease.
Figure 1PLS Initial Model or PLS StageBased on Figure4.3.1 above, there are 2 models in this structural model, namely: 1.The influence model of X1, X2, X3 on Y1.So Y1 as endogenous latent variable, while X1, X2, X3 as exogenous latent variable.2. The influence model of X1, X2, X3 and Y1 on Y2.So Y2 as endogenous latent variable, while X1, X2, X3 and Z as exogenous latent variables.The constructs or latent variables in this structural equation include: X1, X2, X3, Y1 and Y2.Each of these latent variables has indicators or manifest variables in it.Based on the PLS concept above, the results of the outer model analysis on your data are as follows: The results of the analysis are as follows:

Figure
Figure 2PLS Initial Model or PLS Stage Values Source: Data Processing SmartPLS Version 3 Average Variance Extracted International Journal of Social Service and Research https://ijssr.ridwaninstitute.co.id/IJSSR Page 1888 Journal of Social Service and Researchhttps://ijssr.ridwaninstitute.co.id/ the path coefficients between constructs is carried out to see the significance and strength of the relationship and also to test the hypothesis.Path coefficients values range from -1 to +1.The closer to the +1 value, the stronger the relationship between the two constructs.A relationship that is closer to -1 indicates that the relationship is negative(Sarstedt et al., 2017).The results of the analysis at the inner level are as follows: (The calculated T value of the loading factor and path coefficient).

Figure 3Calculated T
Figure 3Calculated T value from loading factor and path coefficientAnd if what is displayed is the p value of the loading factor and the path coefficient, it is as follows:

Table 4 . R-Square Values Source
: SmartPLS Processing Data Version 3 Based on table 4.3.1.It can be seen that the R-Square of the Customer Satisfaction variable can be explained by the service quality variables namely Tangible, Empathy, Assurance of 40.40% and 59.60% influenced by other variables not used in this study such as Responsiveness, Reliability and the model is classified as moderate.While the R-Square value of the Repurchase Intention variable is 49.20%.it shows that the Repurchase Intention variable can be explained by service quality variables namely Tangible, empathy, Assurance of 49.20% and 50.80% influenced by other variables that are not used such as Responsiveness, Reliability

Table 5 . F Square International Journal of Social Service and Research, Carlos
Soares Ribeiro, Diane Indiyati IJSSR Page 1889 Source: Taken from Outer Stage data

Empathy (X2) -> Repurchase Intention (Y2)
Telkomcel Student SimCards.(2) The test results of the estimation coefficient of Empathy towards Customer Satisfaction bootstrap results are 0.103 with a t-

International Journal of Social Service and Research https
://ijssr.ridwaninstitute.co.id/IJSSRPage 1894 count value of 1.488 smaller t-table (1.64) then the p value is 0.137 > 0.05 so Accept H0 is not significant or does not have a direct effect on Empathy on Customer Satisfaction is not significant or not statistically significant.The results of this statistic show that, meaning this SimCard student, telkomcel distributes it for free and with a data package of $ 25 per month and free calls to fellow student simcards, so the customer empathy will be higher and customer satisfaction will also increase automatically.(3)Assuranceestimationcoefficienttestresults against Customer Satisfaction bootstrap results are 0.296 with a t-count value of 3.946 greater than t-table (1.64) then the p-value is 0.000 <0.05 so Accept H1 or which means the direct influence of Assurance to Customer Satisfaction statistically significant or significant .From these statistical results it proves that Assurance has a significant impact on Customer Satisfaction , which means that, in the process of distributing SimCadr students, Telkomcel does this formally, first making an application letter to the university, and the university answers based on a request from Telkomcel, and based on data -this student data that Telkomcel distributes SimCards to students and lecturers at the university, this data is guaranteed not to be published because there are already regulations that have been implemented byTelkomcel.(4)Theresults of the Tangible estimation coefficient test on Repurchase Intention bootstrap results are 0.246 with a t-count value of 3.284 greater than t-table (1.64) then the p-value is 0.001 <0.05 so Accept H1 or which means the direct influence of Tangible Variables to Repurchase Intention statistically significant or significant .Based on this analysis it can be concluded that the tangible variable as a dimension for strengthening service quality at Telkomcel Timor Leste, from these results shows that the reduction in frequency has an impact on Repurchase Intention of 24.6% meaning that the tangible being studied here is the Simcard Student part of (Mobile Network Operator) MNO will have a direct impact on Repurchase Intention, especially students /I in Timor Leste who use the Telkomcel Student SimCard.(5) the test results of the estimation coefficient of Empathy towards Repurchase Intention bootstrap results are 0.117 with a tcount value of 1.589 smaller t-table(1.64