THE INFLUENCE OF TRUST AND INFORMATION QUALITY ON ONLINE PURCHASE DECISION IN THE SHOPEE APPLICATION (A CASE STUDY ON PT SRI BOGOR’S EMPLOYEE)

Article Info ABSTRACT Article history: Received May 8, 2021 Revised June 20, 2021 Accepted July 19, 2021 This study aims to determine how trust and information quality influence online purchasing decisions for Shopee application users. This research is a quantitative study with trust, information quality, and purchase decisions as variables. The population of this research was employees of PT SRI Bogor, who had at least used the Shopee application once. The sample used in this study consisted of 90 respondents who were taken using the accidental sampling technique. Data collection was carried out using questionnaires, in which each respondent was given 25 statements. The data were then analyzed using regression analysis techniques, classical assumption tests, and hypothesis testing. The results of this study indicate that: (1) Trust has a positive effect on purchasing decisions with a value of 28%; (2) Quality of information has a positive effect on purchasing decisions with a value of 37.4%; (3) Trust and quality of information simultaneously influence purchasing decisions with a value of 43.4%, while the remaining 56.6% is influenced by other variables not examined.


INTRODUCTION
In recent years, technological development is more sophisticated and continues to be more advanced than past several years. The development can be observed in various aspects starting from transportation, communication, trade, and et cetera. Hence, the people's lifestyle also changes and shifts from conventional to automatic because of technological development. The most conspicuous from the advancement is a gadget and the tendency of people to do activities in cyberspace such as buying through the internet or online shopping. Parallel with technological development, internet development also evolves rapidly, which very helpful for ordinary people in online shopping. Observing the statistics data in APJII sites (Asosiasi Penyedia Jasa Internet Indonesia) reveals that the internet user keeps on growing each year in Indonesia. A survey conducted by APJII in 2018 discloses Indonesian internet users reach up to 171,17 million users or 64,8% of the country's total population, and the number increased 10,12% from the previous year in 2017. With the increase of internet users yearly in Indonesia, it can encourage online shopping users' enormous potential, and thus, plenty of e-commerces are established in this country; one of them is Shopee. Shopee is one of the e-commerce existed in Indonesia. It's headquarter in Singapore, the company was established in 2015, and the company has a powerful influence to make online shopping wider in Indonesia. Based on the above e) Accuracy, Information comes from data or measurement results and recording of the facts. Thus, accuracy in measuring and recording facts will determine the accuracy of the data and the value of the information generated. f) Clarity, user can present information in various forms of text, tables, graphs, charts, and others. However, whatever the form chosen on the web, the critical thing simplifies the user to understand its meaning. Therefore, in addition to the presentation form, it must be correct. The user's ability to understand it must also be considered. g) Flexibility, good information is the one that can be easy to change its presentation conforming to the situation and requirement at hand. h) Punctuality, good information presented at the right time when it is needed. A late coming Information will become a worthless piece of information. (For instance, in decision making). Meanwhile, according to Sutabri (2012: 33), the quality of information depends on three aspects, e.g., a) Accurate, Information must be free from errors and misleading. Accurate also means that information must reflect its intention. b) Timelines, information coming to the receiver must not be late. Obsolete information is a worthless piece of information because it is a foundation in decision making in which if the decision is late, it will make a fatal impact on the organization. c) Relevance, the information is helpful for the users. The relevance of the information for every person is different.
For example, conveying information about the causes of damage in production machines to the company accountant is undoubtedly less relevant. It will be more relevant if it is addressed to the company's technical experts and vice versa.
McLeod (Susanto, 2013: 38) conveys that a piece of qualified information must have the following features: a) Accurate means that information must reflect the actual condition. The testing accuracy will be done by two people or more. If the test produces the same result, the data will be stated accurately. b) Punctuality means the information must exist when needed, not tomorrow or a few hours from now. c) Relevant means the information given must correspond to the requirements of the individual at every level within an organization. d) Complete means that information must be given in whole or complete.

RESEARCH METHODS Research Type
In this study, the researchers apply a quantitative approach because it wants to find the relation between trust (X1) and information quality (X2) toward purchase decision (Y). The research method used is a survey method. The primary data in this research is obtained in the field research by spreading questionnaires to the respondent.

Population and Sample
The population in this study is the employees in PT SRI Bogor, particularly the plant one tire assembly section. It consisted of 300 employees who had used/transacted using Shopee application by minimal one time, and sample in this research using Newman theory (Silaen, 2018:90) that is using 30% of the total population if the population is under 1000, the sample used in this research is 30% x 300 = 90 people respondent.

Sample Collection Techniques
The type of technique used is Accidental Sampling, which is a sampling technique based on chance, that is, anyone who happens to meet a researcher who has used the Shopee application at least once.

Data Analysis Techniques
In this study, the researchers apply SPSS version 24 software as the data processor for the questionnaire. The results of processing questionnaire data through SPSS version 24.0 software will produce a frequency distribution and percentage of questions for each indicator/statement.

1) Validity test
The validity test is applied to determine the level of obstacle or error from the questionnaire spread by the researchers. A valid instrument means that the instrument can be used to obtain data (measure) what should be measured.

2) Reliability test
The reliability test is the tool to measure a questionnaire that indicates a variable or a construct (Ghozali, 2011: 47). A questionnaire is reliable or reasonable if the respondent's answer to the questions is stable or consistent from time to time.

3) Classic Assumption test a) Normality Test
It is used to test whether the residual value generated from regression is distributed equally normal or not. b) Multicollinearity Test The multicollinearity test aims to find out the perfect or near-perfect correlation between variables on the regression model. A good regression model should not occur correlation between variables. c) Heteroscedasticity Test d) The heteroscedasticity test aims to test whether the regression model created equality of variance from the residual of one observation to the other observations. If the residual variance from one observation to the other observation is fixed, it is called homoscedasticity, and if it is different, it is called heteroskedasticity.

4) Statistics Test a) Regression Analysis test
The simple regression analysis was used to determine each first independent variable, e.g., trust (X1) and information quality (X2) toward the dependent variable purchase decision (Y). Two-predictors regression analysis was applied to find out the influence of independent variable simultaneously toward dependent variable. b) Correlation Test (R) According to Silaen and Widiyono (2013: 222), "Correlation analysis using statistics test was intended to measure the degree of correlation from two different variables, independent and dependent variables." Regression analysis was used to find out the relation between trust, information quality, and purchasing decisions.

c) T-test
The T-test was used to test whether independent variables significantly influence dependent variables by comparing to T table with T count obtained based on 5% (a = 0.05) significance level. d) F test The F test was used to find out the significant influence of independent variables simultaneously toward independent variables. e) Coefficient of Determination Analysis (R2) This coefficient of determination explains the influence of the variable X percentage on the fluctuations in the variable Y value, while the remaining difference from 100% is the influence caused by other factors not examined.

RESULTS AND ANALYSIS Validity Test
The validity test from the three variables concludes that the t-count value on all items used was more than 0.213 value or more significant than the t-table value. Thus, all items statements in all variables are valid.

1) Reliability Test
The deduction of reliability test on the three variables, e.g., trust (X1), information quality (X2), and purchasing decision (Y), the result of Cronbach's alpha calculation is > 0.60. Based on this, it can be deduced that all items in all variables are reliable.

2) Classic Assumption Test 1) Normality Test
The normality test result shows a significant value on the unstandardized residual of all dependent and independent variables that will be tested is as much as 0.200; it is more significant than the alpha 0.05 value, meaning that all residual variables are distributed normally.

2) Multikolonierity Test
The multicollinearity test shows the Tolerance value for trust variable (X1) and information quality (X2) is 0.727; it is greater than 0.10. Meanwhile, VIF value for trust (X1) and information quality (X2) variable is 1,375 < 10,00. It can be concluded that multicollinearity did not occur in the regression model.

3) Heteroskedesity Test
The heteroskedasticity test shows that significance (sig.) for trust (X1) is 0.95. Meanwhile, significance value (sig.) for information quality variable (X2) is 0,548. Because significance value (sig.) Both variables above are more significant than 0.05, and it can be deduced that it was never heteroskedastic.

4) Statistic Test a) Simple Linear Regression
The research result shows that the trust variable (X1) positively influences toward purchase decision variable (Y), and the Information quality variable (X2) also positively influences on purchasing decision variable (Y).

b) Two-predictors Linear Regression
The research result shows coefficient trust value by 0.296, which positively influences the purchasing decision, and information quality also shows coefficient value by 0.328, which positively influences purchasing decision.

c) Coefficient Determination
The research result shows a coefficient determination (R 2 ) value as much as 0.434, which means trust variable (X1) and information quality variable (X2) influence purchasing decisions as much as 0.434 x 100% = 43.4%; meanwhile, the rest of 56.6% was influenced by other variables.

d) T-test (partial)
This research result shows that for the trust variable, the T-count 5.848 > T

CONCLUSION
The research results on trust and information quality toward online purchasing decisions on the Shopee application (a case study on PT Sri Bogor's employees), namely: 1. There is a positive and significant effect of Trust on Online Purchase Decisions in the Shopee Application on PT SRI Bogor's Employees stated that the correlation effect is 0.529, and the influence of determination is 28%. 2. There is a positive and significant influence on the information quality toward online purchasing decisions on the Shopee application. For PT SRI Bogor's employee has accepted the correlation effect by 0.611, and the influence of determination is 37.4%. 3. There is a positive and significant influence of Trust and Information Quality Simultaneously on Online Purchase Decisions in the Shopee Application by PT SRI Bogor's employees, it is stated accepted the correlation effect amount of 0.659 and the influence of determination is as much as 43.4%.

Suggestion
Suggestion to e-commerce Shopee company, it is better to increase the security in the transaction and ensure again that the seller meets the applicable terms and conditions before selling products to users. Furthermore, the company provides up-to-date, complete, transparent, and exciting information about products, prices, locations, and promotions in the Shopee application. The user's trust level increases and the Shopee application's information quality becomes easy for users to make purchasing decisions.