THE EFFECT OF ONLINE SERVICE QUALITY ON BERRYBENKA E-COMMERCE CUSTOMER SATISFACTION

This study aims to see the effect of online service quality which consists of dimensions (ease of use, web design, responsiveness, customization and assurance) on customer satisfaction in Berrybenka e-commerce. This study used primary data obtained by distributing questionnaires to 110 respondents who were selected using a purposive sampling method. The data analysis method used is multiple linear regression method with the help of SPSS version 20. This study shows that ease of use has a significant effect on Berrybenka e-commerce customer satisfaction, web design has a significant effect on Berrybenka e-commerce customer satisfaction, responsiveness has a significant effect on customer satisfaction. Berrybenka ecommerce customer, customization has a significant effect on Berrybenka e-commerce customer satisfaction and assurance has a significant effect on Berrybenka e-commerce customer satisfaction. Simultaneously, ease of use, web design, responsiveness, customization and assurance have a significant effect on Berrybenka's e-commerce customer satisfaction.

behavior by customers towards the company because it can create a sense of comfort and security for its customers. Guarantees affect customer satisfaction, because if customers feel safe and comfortable in transactions, it will create a sense of satisfaction that cannot be measured.

conceptual framework
Based on the explanation above, the following conceptual framework can be developed: H1 H2 H3 H4 H5 H6 Image 1 conceptual framework

Research Hypothesis
Based on the problem formulation and conceptual framework, the hypotheses in this study are: H1: Ease of Useeffect on customer satisfaction in Berrybenka e-commerce. H2: WebDesigneffect on customer satisfaction in Berrybenka e-commerce. H3: Responsiveness influences customer satisfaction in Berrybenka e-commerce. H4: Customization affects customer satisfaction in Berrybenka e-commerce. H5: Assurance has an effect on customer satisfaction in Berrybenka e-commerce. H6: Ease of Use, Web Design, Responsiveness, Customization, and Assurance has an effect on customer satisfaction in Berrybenka e-commerce.

RESEARCH METHOD 3.1. Subjects and Research Locations
In accordance with the chosen research title, the writer will use the research subject, namely consumers who make purchases at the Berrybenka store through the official website. Meanwhile, the research location was carried out on the people of Medan City who had made purchases at the Berrybenka store through the official website.

Population and Sample
In this study, the researcher determined the sample with the following criteria: people who had purchased products sold on the official Berrybenka website and made purchases more than once. As forthe number of samples for this study amounted to 110 people.

Classic assumption test
The classical assumption test consists of several stages, namely, normality test, heteroscedasticity test and multicollinearity test.

Data analysis method
The data analysis method in this study is to use multiple linear regression, to see the effect of online service quality on Berrybenka e-commerce customer satisfaction using the SPSS 20.0 program. The regression equation can be formulated as follows: Multiple Linear Regression Analysis Y = α + β1X1 + β2X2+ β3X3+ β4X4 + β5X5+ e Where:

Validity Testing Results
Validity testing is used to measure the validity of a questionnaire. The results of validity testing can be seen in Table 1.  Table 1 it can be seen that the calculated r value is greater than 0.187, so it can be concluded that each statement item in the questionnaire is declared valid.

Reliability Test Results
The reliability test is an index that shows the extent to which the measuring device is reliable. For further explanation, see Table 2.  Table 2, it can be seen that the value of Cronbach's alpha from the variable ease of use (X1) is 0.704> 0.60, web design (X2) is 0.677> 0.60, responsiveness (X3) is 0.757> 0.60, customization (X4) of 0.691> 0.60, assurance (X5) of 0.636> 0.60 and customer satisfaction (Y) of 0.606> 0.60 so that all the question variables above are reliable.

Classical Assumption Testing Results
Data Normality Test Results Data normality testing was carried out to find out whether the data obtained through distributing questionnaires was normally distributed or not. The results of the normality test can be seen in the image below:

Figure 3 Graph of normality probability plot test results
Based on the results of the SPSS output in Figure 2, it can be seen that the data is spread around the diagonal line, so it can be concluded that the data in the regression model of this study are normally distributed. Furthermore, testing the normality of the data using statistical analysis through the Kolmogorov Smirnov test (KS) with statistical tools using the SPSS program, can be seen in Table 3. Based on Table 3 it can be seen that the Asymp. Sig (2-tailed) of 0.996 is above the standard significance of 0.05. According to Ghozali (2018: 167) states that the terms of the Kolmogorov Smirnov test (KS) if the significant value is > 0.05 then the data distribution is normal. Therefore, it can be concluded that with a confidence level of 95% this data is normally distributed.

Heteroscedasticity Test Results
The heteroscedasticity test aims to test whether in the regression model there is an inequality of variance from the residuals of one observation to another (Ghazali, 2018: 137). The results of the heteroscedasticity test can be seen in Figure 4.

Figure 4
Heteroscedasticity Test Based on Figure 4, it can be seen that the dots spread evenly inside, above and below the number 0 on the Y axis. In addition, these dots do not form a certain pattern, it can be concluded that there were no symptoms of heteroscedasticity in this study.

Multicollinearity Testing Results
The multicollinearity test aims to test whether the regression model found a correlation between the independent (independent) variables. A good regression model does not have a correlation between the independent variables (Ghozali, 2018: 107). The results of the multicollinearity prerequisite test can be seen in Table 4.  Table 4, it can be seen that the VIF value for each research variable is as follows:

THE EFFECT OF ONLINE SERVICE QUALITY ON BERRYBENKA E-COMMERCE CUSTOMER SATISFACTION Elliza Sabila, Nurainun
1. VIF valuefor the variable ease of use of 1.944 <10 and a tolerance value of 0.514 > 0.10 so that the ease of use is stated not to have symptoms of multicollinearity. 2. VIF valuefor the web design variable it is 2.244 < 10 and the tolerance value is 0.446 > 0.10 so that the web design is declared to have no symptoms of multicollinearity. 3. VIF valuefor the responsiveness variable of 1.683 <10 and a tolerance value of 0.594 > 0.10 so that responsiveness is declared to have no symptoms of multicollinearity. 4. VIF valuefor the customization variable it is 1.915 <10 and the tolerance value are 0.522> 0.10 so that customization is declared to have no symptoms of multicollinearity. 5. VIF valuefor the assurance variable it is 1.021 <10 and the tolerance value are 0.980> 0.10 so that assurance is declared to have no symptoms of multicollinearity.

Data analysis method 4.6. Multiple Linear Regression
This study aims to determine the effect of online service quality on customer satisfaction at Berrybenka stores through the official website. Therefore, multiple linear regression analysis is used to analyze the effect. The results of multiple linear analysis with the SPSS version 20 program were obtained as follows:  Table 5 above, it can be seen that the coefficients of each variable and constant that are observed in this study are the ease-of-use coefficient (X1) 0.233 web design coefficient (X2) 0.217 responsiveness coefficient (X3) 0.227 customization coefficient (X4) 0.237 coefficient assurance (X5) 0.131 and constant 1.177. This means that the independent variables affect customer satisfaction (Y). The equation can be written as follows: Y =1.177 + 0.233X1 + 0.217X2 + 0.227X3 + 0.237X4 + 0.131X5 +e From this equation it can be explained that the effect of e-service quality on customer satisfaction at the Berrybenka store through the official website. The results of the analysis show that: a. The constant value (α) is 1.177, this indicates that if there is an influence (increase/decrease) in the values of the variables X1, X2, X3, X4 and X5 considered constant, the value of customer satisfaction at the Berrybenka store via the official website is 1.177. b. The first coefficient is Ease of use (X1) of 0.233, this states that if there is an increase in Ease of use (X1) of 1, it will increase Purchase Decision (Y) by 23.3%. c. The second coefficient is Web design (X2) of 0.217, which means that every increase in Web design (X2) is 1, it will increase customer satisfaction (Y) by 21.7%. d. The third coefficient is Responsiveness (X3) of 0.227, which means that for every increase in Responsiveness (X3) of 1, it will increase customer satisfaction (Y) by 22.7%. e. The fourth coefficient is Customization (X4) of 0.237, which means that for every increase in customization (X4) of 1, it will increase customer satisfaction (Y) by 23.7%. f. The fifth coefficient is Assurance (X5) of 0.131, which means that every increase in assurance (X5) is 1, it will increase customer satisfaction (y) by 13.1%.

CONCLUSIONS AND SUGGESTIONS 5.1. CONCLUSION
Based on the results of the analysis that has been carried out in testing the effect of online service quality which consists of (ease of use, web design, responsiveness, customization and assurance) on Berrybenka e-commerce customer satisfaction, it can be concluded that: 1. Ease of usesignificant effect on customer satisfaction. Ease of use can make it easier for users to carry out their activities when shopping on the official Berrybenka website. Customers who frequently shop on the official Berrybenka website will have no difficulty using the official Berrybenka website. This shows that the easier the site is to use, the more satisfied customers will be. 2. Web designsignificant effect on customer satisfaction. Web design provides information easily to users. This can please the customer and satisfy his needs. An attractive website will make it easier for customers to find the products they need. This shows that the better the website that is owned, the more satisfied customers will be using the website. 3. Responsivenesssignificant effect on customer satisfaction. Responsiveness can be measured by how fast it provides services to customers. Customers who feel their complaints are being addressed properly will feel their needs are met. This shows that the more responsive the official Berrybenka website is, the more problems faced by customers will be resolved soon. So, satisfaction in using the official website will also increase. 4. Customizationsignificant effect on customer satisfaction. Berrybenka's official website can narrow down the choices for customers, minimizing the time customers spend going through all the products to find what they want. Berrybenka's official website can determine customer needs, which can narrow the time it takes customers to find the products offered. This shows that the easier it is for customers to find the product they are looking for, the more satisfied customers are to make transactions on the official Berrybenka website. 5. assurancesignificant effect on customer satisfaction. Because it can foster trust behavior by customers towards the official Berrybenka website because it can create a sense of comfort and security for customers. This shows that the more secure the transaction, the more satisfied the customer will be.

5.2.SUGGESTIONS
From the research that the authors have conducted, basically the quality of online services has been going well, but there are still a number of things that need to be improved. Therefore, the author gives suggestions: 1. On the websiteofficial Berrybenka, in terms of user convenience such as website errors and inaccessibility which make customers no longer want to visit the official Berrybenka website. This is of particular concern to the official Berrybenka website to provide adequate web hosting servers so that devices and websites don't have errors when used.