The Effect Of Financial Capability, Technology Usage And Prior Bank Experience Into Millenial Online Shopping Behavior

. The rapid development of technology has made all aspects of human life easier. Online shop is a form of change brought about by technological developments. With an online shop, humans can shop wherever and whenever. This convenience, of course, must be followed by sufficient financial capabilities because the ease of shopping that is not followed by good financial capabilities will cause new problems, especially financial problems. With a quantitative approach, researchers want to find out whether Financial Capability, Technology usage, and prior bank experience influence a person's online shopping behavior. The sample in this study were residents of Jakarta and Tangerang aged between 19-34 years. Data collection was carried out by distributing questionnaires through the Google Form. Meanwhile, to analyze, researchers used Smart-PLS 3.0 software. The results of the study concluded that financial capability has a positive effect on online shops, while technology usage and prior bank experience have no effect on online shops.


I. INTRODUCTION
Technological developments make international access easier to reach, this very rapid development also brings many changes in human habits and lifestyles.One of them is a person's buying habits.Online shopping is now more favored by most people because it is more easily accessible from anywhere and anytime.Consumers can search for more information and can compare prices and products more easily, more options and conventional.Shopping online offers more satisfaction to save consumers' time [1].Communication technology has been proven to have an impact on lifestyle and the environment, and also has an impact on individual consumption [2].According to [3], generation Z does have different purchasing behavior compared to other generations.He found that gen Z is a consumptive buyer who feels the need to use his money if he feels he needs the products.Consumptive behavior occurs when individuals have a high desire to buy something without thinking about their needs [4].This phenomenon is supported by the development of delivery services that make online shopping easier.Not only shipping, diverse and increasingly easy payment methods also support this kind newly behavior.Consumptive purchasing behavior without being followed by low financial capability will cause financial problems.Like the practice of impulse shopping, individuals with low incomes will experience financial problems [5].In a survey conducted by [6], it was proven that 25.8% of online shops encourage consumptive behavior in students, 74.2% are environmental and are also supported by the pandemic which causes students to access more social media including e-commerce because learning activities are carried out online.
Financial capability includes an individual's knowledge of finance, the ability to manage their finances and income and make decisions related to finance.Financial capability also refers to individual ability to act(a person's knowledge, abilities, behavior, habits, motivation, self-confidence, and self-efficacy), and opportunity to act (a person's awareness of the financial products they need to manage their finances) [7].This concept refers to making good financial decisions, understanding how to control spending and income, and identifying goods and services needed [8].The concepts of financial capability and financial literacy are often misinterpreted.[4]states financial literacy is financial knowledge possessed by a person while in his opinion, [9]states that financial capability focuses on financial behavior and applies financial http://ijstm.inarah.co.id 1126 knowledge in action.In their research, [2] stated that students with high financial literacy are wiser in consumption activities.However, financial literacy can have a negative impact on consumptive behavior because this is also influenced by various circumstances [10].Based on the explanation above, the researcher developed a research framework and hypotheses.
Following is the research framework and hypotheses:

II. METHODS
This study wasconducted to determine whether financial capability has an influence on online consumptive behavior with logistic regression.Logistic regression is a statistical analysis method that describes the relationship between qualitative dependent variables having two or more categories with one or more independent variables on an interval scale.This study uses the SEM Partial Least Squares model with the aim of testing predictive relationships between constructs, to see the correlation or influence between these constructs [11].The assessment of the PLS SEM model is carried out by considering the results of the outer model and inner model [11].The population of this study is divided into two, namely, infinite population and finite population.The population of this study is the Indonesian population who are in the age range of 19-34 years.The sample of this study is Indonesian residents in the Jakarta and Tangerang areas who are in the age range of 19-34 years who have shopped online.
Data collection was carried out by distributing questionnaires using random sampling methods to residents of Tangerang and Jakarta areas who have an age range of 19-34 years who have shopped online to measure how their financial capability affects current decision making in online shopping and their consumptive behavior.The respondents chose one answer of the five Likert Scale questions ranging from strongly disagrees to strongly agrees.The primary data was further analyze using the structural equation modeling (SEM) method with SmartPLS 3.0 software.The following research model is proposed by the researcher: In this study, researchers have identified the answers to question indicators given to 116 respondents by finding the average or mean value of respondents' answers for each measurement using an interval scale.The calculation of descriptive analysis uses an interval scale of average values using the following formula: Class interval = (highest value -lowest value)/number of classes = 5-1/5 = 0.8.Based on the calculation of the interval scale above, it was decided that the interval between classes in this study was 0.8.http://ijstm.inarah.co.id 1127 This study comes up with the analysis of inner and outer model as follows:

III. RESULT AND DISCUSSION
The distribution of questionnaires conducted received 116 respondents.However, after going through the screening process, there were only 110 respondents who passed the screening, namely respondents with an age range between 19-34 years, domiciled in the Jakarta or Tangerang area, and had shopped through an online shop.The study conducted the validity testing which consist of the convergent and discriminant validity.The result of reliability testing and discriminat validity were explained in Table 1 and Table 2 From the results of data processing in the table above, it can be concluded from the results of the loading factor that is above 0.7 and the AVE value above 0.5, so that all indicators of each variable in the table above can be declared valid.Discriminant validity can be said to be good if the root of AVE in the construct is higher than the correlation of the construct with other latent variables, while in cross loading testing must show a higher indicator value of each construct compared to indicators in other constructs [12].From the table 3, the results of composite reliability and cronbach alpha show good results with the results above the criteria of ≥ 0.7.This shows that the indicators in the questionnaire have a fairly high consistency.R-Squared is a number that indicates the magnitude of the combination of independent variables together affecting the value of the dependent variable.This value is used to assess how much influence a particular independent latent variable has on the dependent latent variable.Table 4 show the results of the R 2 value test: Tabel 4. The value of R 2 Table 4 show that online shop behavior variable has the coefficient of determination is 0.695.This result shows that as many as 69.5% of indicators in online shopping behavior are influenced by financial capability, prior bank experience, and technology usage while the rest are influenced by other variables outside the study.In this case, the online shop variable falls into the category of moderate coefficient values.The F-squared test is performed to assess how much influence the independent variable has on the dependent variable.Here are the results of the F 2 test: Tabel 5.The result of F The result of F 2 testing shows that only financial capability that has strong effect into online behavior.Meanwhile, two other variables namely prior bank experience and technology usage do not have any effect into online shopping behavior.
The Q-Square test results is explained in the following table: Tabel 6.The result of R 2 testing Endogeneous Variable Q 2 Predict Online Shop Behavior 0.481 Table 6 shows that Q-Square has value above 0 (Q 2 >0) which means that endogenous variables have Predictive Relevance.This hypothesis test is used to see if there is a significant influence between the independent variable and the dependent variable.Decision making in this hypothesis test is based on [12] where p-values < 0.05 mean that the independent variable has a significant effect on the dependent variable.Here are the results of hypothesis tests that have been carried out by researchers: Tabel 7. The result of Hypotheses Testing The result of hypothesis testing shows that Consumer Financial Capability has a positive effect on Online Shopping Behavior since the value of the path coefficient (Original sample) has a value of 0.923, then p-values H1 has a value of 0.000, and t-statistic of 10.586.Based on this value, it can be concluded that the hypothesis is significant because the p-values < 0.005 and t-statistics are more than 1.96 and mean that the hypothesis is accepted.The result of hypothesis testing also shows that the use of technology has a negative influence on online shopping behavior because the value of the path coefficient (original sample) is -0.116.Based on the data obtained, p-values have a value of 0.419 and t-statistics of 0.810 so it can be concluded that the hypothesis is not significant because the value of t-statistics is less than 1.96 which means the hypothesis is rejected because the hypothesis proves insignificant that Technology Usage is negatively related to online shopping behavior.The other hypothesis testing also show that prior bank experience has no influence to online shopping behaviors with the value of the path coefficient (original sample) is -0.046.Based on the data obtained, the p-values of 0.669 and t-statistics of 0.428 do not qualify because the p-values are more than 0.005 and t-statistics are less than 1.96, this means that the hypothesis is rejected because the test results prove that the hypothesis is not significant that prior bank experience affects online shopping behavior.

IV. CONCLUSION
Based on the results of data processing, it can be concluded as follows that financial Capability has a positive relationship with the Online Shop.The test results using the PLS-SEM method, obtained a p-value of 0.000 or less than 0.005 and a t-statistic of more than 1.96 or 10.586 which indicates that there is a significant influence.This means that a person's decision to shop online for respondents aged 19-34 years is influenced by their financial capability.Technology Usage has a negative relationship with online shops.The test results using the PLS-SEM method showed the results of p-values of 0.669 and t-statistics of 0.428.This value means that the indicator is not significant because the p-values > 0.05 and t-statistics < 1.96.
Based on these results, the author concludes that technology usage or the use of technology in Jakarta and Tangerang area residents aged between 19-34 years has no effect on their online shopping behavior.Prior Bank Experience has a negative relationship with online shops.Based on the test results using the PLS-SEM method, a p-value of 0.419 and a t-statistics value of 0.428 were obtained, which means this value is not significant because the p-values > 0.05 and t-statistics values < 1.96.Based on these results, the author concludes that prior bank experience or experience in using banks of Jakarta and Tangerang residents aged 19-34 years does not affect their behavior in online shopping. V.

Fig 1 .
Fig 1. Proposed Research Framework Based on the research framework, the study proposed the following hypothesis: H1: Consumer financial capability positively affects Online Shopping Behavior H2: The technology usage affects Online Shopping Behavior.H3: Banking experience influences Online Shopping Behavior.

Table 1 .
. Result of Reliability Testing Main Test

Table 2 .
Result of Discriminant Validity Testing Main Test (Outer Loading) Based on table 2 above, each indicator in the research variable has the largest cross loading value when compared to the cross loading value in other indicators.Based on these data, it can be concluded that the indicators used have met good discriminant validity.

Table 3 .
Result of Composite Reliability Testing 2testing