THE INFLUENCE OF TRUST, CONVENIENCE AND RISK ON THE DECISION OF THE PEOPLE OF SOUTH BANJARMASIN TO USE THE SHOPEE PAYLATER PAYMENT METHOD ON THE SHOPEE MARKETPLACE

This research aims to analyze the influence of trust, convenience, and risk on the decision of the people of South Banjarmasin to use the Shopee Paylater payment method. This research is field research with a quantitative approach. The population in this research is the people of South Banjarmasin who have used the Shopee Paylater feature. The sample in this study consisted of 100 respondents. Data collection used a questionnaire with multiple linear regression analysis. The results of the F test show that trust, convenience, and risk simultaneously influence the decision of the people of South Banjarmasin to use Shopee Paylater payment method. This is different from the results of the T test, where only convenience has a partial effect on people’s decisions, while trust and risk do not have a partial effect on people’s decisions in using the Shopee Paylater payment method.


INTRODUCTION
Indonesia is a developing country where the people are very open to new technology and products in the world.This in addition to the number of interest users in Indonesia increasing very rapidly from year to year.With technological advances running rapidly, one of them is the internet.The internet was introduced to society in 1969, until now the internet has become a necessity for individuals and organtizations (Astria dan Wahdiniwaty 2021).
The development of the internet has led to new things happening in society, namely purchasing or shopping for goods online.Online shopping can make it easier for buyers to get the products they need, buyers don't need to go to the store for a long time and can be an effective decision for buyers because online shopping is more helpful than offline shopping which is related to traffic jams, limited parking spaces, crowded places, and time is limited.This is a big challenge and opportunity for E-commerce companies to develop well in Indonesia.
According to latest report for the second quarter of 2020, PETA E-commerce iPrice, the first position was taken by Shopee, where visitors to the Shopee website reached 93.44 million per month.Meanwhile, Tokopedia only earns 86.103 million per month.Shopee can occupy first position, influence by several things, such as TV, advertisements that appear on website/social media and Youtube (Indrawati dan Shaina 2021).Shopee is an E-commerce company that is an online shopping platform by focusing on the mobile platform, making it easier for customers to find the items they need.The payment methods provided by Shopee are very varied, such as Shopeepay, Shopee Paylater, bank transfer, COD, and so on.However, there is something interesting about the Shopee payment method, namely the paylater feature .Shopee published Shopee Paylater feature in 2019 as an instant loan solution of up to Rp750,000 with the aim of making it easy for users to make payments within 1.2 to 3 months without needing a credit card which is subject to interest of 2.95%.The convenience provided by Shopee Paylater made people interested in using it in 2019, so that the number reached 102,971 people with 81,423 active borrowers in various cities in Indonesia (Na'im, Solikah, dan Mawftiq 2023).
Shopee paylater has several risks to users, such as excessive consumer behavior, unnoticed costs, disruption of financial management, risks to Bank Indonesia checking arising from arrears and identity hacking (Iskandar 2019).Consumer trust greatly influences online purchasing decisions because it can help consumers overcome perceptions of uncertainty and risk in making purchases (Lestari dan Iriani 2018).Another risk is the costs imposed on consumers.These costs are an interest rate of 2.95% for all loan terms, a penalty fee of 5% per month and a handling fee or administration fee of 1% per transaction (Noorridha, Aulia, dan Syifa 2023).
The easier access to electronic transactions in e-commerce must also be accompanied by strict consumer data security factors to prevent data misuse or account hacking by third parties.Based on the facts that occured, the presence of the paylater feature has opened up a new opportunity for irresponsible parties to hack accounts.
Apart from the problems above, the differences in the results of previous studies also form the background for this research.A study from (Putri dan Iriani 2020), has found that there is a partial positive influence of the trust variable and the convenience variable on purchasing decisions using Shopee Paylater online loans.Meanwhile, a study conducted by (Dwiyanti 2022), has found that consumer trust variable partially also has a positive and significant effect on purchasing decisions, but the convenience variable partially has no effect on purchasing decisions using the Spaylater online loan method.

RESEARCH METHODS
This study is field research with the quantitative aprroach.This study examined the influence of the hypothesised variables.The variables in this research consist of trust, convenience, risk as the independent variable, and purchasing decisions as the dependent variable.For the trust variable, the indicators consist of the seller's sincerity in serving consumers, the seller's ability to guarantee satisfaction and security during transactions, the seller's integrity in fulfilling agreements made with consumers, and the consumer's willingness to accept negative consequences that may occur (Kotler & Keller, 2016).The indicators in the convenience variable consist of the level of understanding in use, operating system, level of clarity in operation, and level of flexibility (Tampubolon 2022).The risk variable has indicators in the form of finance, performance, psychology, and social (Ariani 2017).While the indicators contained in the purchasing decision variable are suitabiliy to needs, benefits, product accuracy with consumer expectations, and repeat purchases (Thomson, 2018).
The population in this study (Sugiyono, 2014) are the people of South Banjarmasin who have used the Shopee Paylater feature.This research has used non-probability sampling technique in determining the sampel, and purposive sampling method in taking the sampel, with the criteria being that the people of South Banjarmasin are aged 20 -29 years and have an income.The reason for selecting this sample is based on research results from a Special Poll Report for the First Semester of 2021 Edition entitled "Indonesia E-Commerce Trends 2021", which states that the largest market share of Shopee users are those aged 20 -29 year (47%) (Satriawan dan Saputra 2023).The number of samples obtained was 100.
The technique of data analysis uses several methods consisting of research instrument tests, namely tools that can be used to collect, process, and evaluate data from respondents using the same measurement pattern (Siregar 2017).The technique of data analysis in this research used: Firstly, classical assumption test which consists of normality test,

Validity and Reliability Test Results
The result of the validity test on the instruments of the purchasing decision (Y), trust (X1), convenience (X2), and risk (X3), each of which consists of 4 questions, are declared to have passed the validity test because they have fulfilled the requirements by obtaining an rcount value greater than r-table .While the results of the reliability test show that the Cronbach's Alpha value for the purchasing decision variable (Y) is 0.689, for the trust variable (X1) it is 0.582, for the convenience variable it is 0.655, and for the risk variable it is 0.682.All values are greater than 0.6 which is the reliability standard.
Based on this results, it can be stated that the instruments for variables of purchasing decision, trust, convenience, and risk, have passed the reliability test because the Cronbach's Alpha value obtained is greater than the reliability coefficient value.

Classical Assumption Test Results
The normality test aims to test whether the data used in the research is normally distributed or not (Matondang dan Nasution 2021).The normality test in this study used the Kolmogorov-Smirnov and P-Plot.

Figure 1. Normality Test Results
The figure above shows that the dots are collected and tend to follow a diagonal line.Thus the data is normally distributed and the normality assumption is met.

Multicollinearity Test Results
The multicollinearity test is carried out to test whether there is a correlation between the independent variables in the regression model (Matondang dan Nasution 2021).The decision making criteria in this test are based on the tolerance value and Variance Inflation Factor (VIF) value.If the tolerance value is greater than 0.10 and the VIF value is smaller than 10,00 then there is no multicollinearity in the regression model.Conversely, if the tolerance value is smaller than 0.10 and the VIF value is greater than 10.00 then multicollinearity occurs in the regression model.

Table 2. Multicollinearity Test Results
The results of the multicollinearity test above show that the tolerance value for the trust variable is 0.737 which is greater than 0.10 and the VIF value is 1.357 which is less than 10.00.The tolerance value for the convenience variable is 0.749 which is greater than 0.10 and the VIF value is 1.335 which is less than 10.00.The tolerance value for the risk variable is 0.964 which is greater than 0.10 and the VIF value is 1.038 which is less than 10.00.So, the conclusion is that the regression model in this study did not have multicollinearity.

Heteroscedasticity Test Results
The heteroscedasticity test aims to test whether there is an inequality of variance from the residuals of one observation to another in the regression model.If the variance from the residual of one observation to another is constant then it is called homoscedasticity, but if the variance from the residual of one observation to another is different then it is called heteroscedasticity.A good regression model is if homoscedasticity occurs or heteroscedasticity does not occur.If the the test result is above the significance level (r > 0.05), it means that heteroscedasticity does not occur.Conversely, if the test results are below the significance level (r < 0.05), it means heteroscedasticity occurs (Matondang dan Nasution 2021).

Heteroscedasticity Test Results
Based on the Glejser test, it was found that the significance value of the trust variable was 0.933, the convenience variable was 0.189, and the risk variable was 0.757.Because the significance level of all variables is above 0.05, in accordance with the basis for decision making in the Glejser test, it can be concluded that heteroscedasticity does not occur in this regression model.

Multiple Linear Regression Analysis
Multiple linear regression analysis is an analytical tool for forecasting the influence value between two or more independent variables on one dependent variable in order to prove whether there is a functional or causal relationship between two or more independent variables on one dependent variable (Yuliza dan Yeneti 2022) Based on the table above, a multiple linear regression equation can be prepared between the independent variable and the dependent variable by entering the multiple linear regression coefficient into the form of the multiple linear regression equation as follows: Y = 6.891 + 0.080X1 + 0.488X2 + 0.031X2 + e The explanation for the multiple linear regression equation is: 1.A constant value of 6.891 is interpreted as if all independent variables are considered constant or have not changed.So, the purchase decision using Shopee Paylater is 6.891.A positive constant value means that if all independent variables increase or have an effect in units, the dependent variable also increases.2. The regression coefficient value for the trust variable (X1) is 0.80, meaning that the trust variable has a positive influence on purchasing decisions, or in other words, every time there is an increase in the trust variable by one unit, the purchase decision using the Shopee Paylater payment method by the people of South Banjarmasin will increase by 0.80 with the assumption that the other variables are fixed or constant.3. The regression coefficient value for the convenience variable (X2) is 0.488, meaning that the convenience variable has a positive influence on purchasing decisions, or in other words, every time there is an increase in the convenience variable by one unit, the purchase decision using the Shopee Paylater payment method by the people of South Banjarmasin will increase by 0.488 with the assumption that the other variables are fixed or constant.4. The risk variable (X3) regression coefficient value is 0.031, meaning that the risk variable has a positive influence on purchasing decisions, or in other words, every time there is an increase in the risk variable by one unit, the purchase decision using the Shopee Paylater payment method by the people of South Banjarmasin will increase by 0.031 with the assumption that the other variables are fixed or constant

Hypothesis Test Results
Hypothesis testing aims to prove the hypothesis proposed in the research, whether it is accepted or rejected (Wardani 2020).Hypothesis testing in this research consists of the T-test to determine the influence of the variable trust (X1), convenience (X2), and risk (X3) on people's purchasing decisions partially, and the F-test to determine the influence of the variable trust (X1), convenience (X2), and risk (X3) on people's purchasing decisions simultaneously.

a. T-Test Results
This test is used to determine the effect of each independent variable on the dependent variable separately (Akbar 2023).

T-Test Results
The table above shows that: Firstly, hypothesis test of the trust variable on purchasing decisions, the coefficient value is positive with t-count value of 0.894 which is smaller than the t-table value of 1.98 with a significance level of 0.374 which is greater than 0.05.This shows that the trust variable does not influence the decision of the people of South Banjarmasin to use Shopee Paylater payment method.Secondly,hypothesis test of the convenience variable on purchasing decisions, the coefficient value is positive with t-count value of 5.083 which is greater than the t-table value of 1.98 with a significance level of 0.000 which is less than 0.05.This shows that the convenience variable has a positive and significant effect on the decision of the people of South Banjarmasin to use Shopee Paylater payment method.Thirdly, hypothesis test of the risk variable on purchasing decisions, the coefficient value is positive with t-count value of 0.526 which is smaller than the t-table value of 1.98 with a significance level of 0.600 which is greater than 0.05.This shows that the risk variable does not influence the decision of the people of South Banjarmasin to use Shopee Paylater payment method.

b. F-Test Results
This test is carried out to see whether the independent variables together have a significant effect on the dependent variable (Akbar 2023).

Table 6. F-Test Results
From the table above it can be seen that F-count value of 14.331 which is greater than the F-table value of 2.70 with a significance level of 0.000 which is smaller than 0.05.This means that the variables of trust, convenience, and risk have a significant effect simultaneously on the purchasing decisions of the people of South Banjarmasin in using Shopee Paylater payment method.

c. Coefficient of Determination Test
The coefficient of determination (R 2 ) aimsto measure how far the model's ability is to explain dependent variations (Ayuningtiyas dan Gunawan 2018).The value is between 0 and 1 (Haqiqi, Susanti, dan Ferawati 2020).If the value is small then the ability of the independent variables to explain variations in the dependent variable is very limited (Purbowo, dkk. 2023).A value close to 1 means that the independent variables provide almost all the information needed to predict the variation in the dependent variable (Hariyati 2023).

Table 7. Coefficient of Determinaton Test
Based on the table above it is known that the coefficient value in column R is 0.556, and the determinant value in the adjusted R square column is 0.288 or 28.8%.This shows that 28.8% of data changes in purchasing decision variables are influenced by the variables of trust, convenience, and risk.Meanwhile, the remaining 71.2% is influenced by the other variables or factors, apart from these three variables.
et al multicollinearity test, and heteroscedasticity test.Secondly, multiple linear regression analysis which consists of regression equation, correlation coefficient, and determination coefficients (adjusted R2).4) Hypothesis test consists of a T-test (partial test) and F-test (simultaneous test).

Table 1 . Normality Test Results of Kolmogorov P
THE SHOPEE MARKETPLACE ... -Plot test results show that the data points tend to follow the normal line and are collected around it.Meanwhile, from the results of normality test, the Asymp.sig value of 0.200 is greater than the significance level of 0.05.

Table 4 . Multiple Linear Regression Analysis
THE INFLUENCE OF TRUST, CONVENIENCE AND RISK ON THE DECISION OF THE PEOPLE OF SOUTH BANJARMASIN TO USE THE SHOPEE PAYLATER PAYMENT METHODON THE SHOPEE MARKETPLACE ...
doi.org/10.55681/economina.v2i11.1002THEINFLUENCEOF TRUST, CONVENIENCE AND RISK ON THE DECISION OF THE PEOPLE OF SOUTH BANJARMASIN TO USE THE SHOPEE PAYLATER PAYMENT METHODON THE SHOPEE MARKETPLACE ...
THE INFLUENCE OF TRUST, CONVENIENCE AND RISK ON THE DECISION OF THE PEOPLE OF SOUTH BANJARMASIN TO USE THE SHOPEE PAYLATER PAYMENT METHODON THE SHOPEE MARKETPLACE ...