ANALYZING BEHAVIORAL INTENTION OF E-WALLET USAGE IN FOOD AND BEVERAGE AT MICRO

: The purpose of this research is to find out the performance expectation, effort expectation, social influence, perceived risk, and perceived cost on behavioral intention for E-wallet usage by MSMEs in Manado. This research is conducted by using quantitative method using questionnaires as data collection method. Owners/the person in charge of the micro, small and medium enterprises in Manado use e-wallet that can be informants about these issues. Researchers determine the minimum sample to be 100 respondents. The requirements in selecting the samples, Owners/people in charge of MSME that use e-wallet in transactions. In this study, the primary data collected is data obtained by submitting questionnaire. The result of this research showed that performance expectation has a negative and significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City, effort expectation and social influence have a positive and significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City, perceived risk and perceived cost have a negative and not significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City, and performance expectation, effort expectation, social influence, perceived risk, and perceived cost, simultaneously has a positive and significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City. It is recommended that MSME owner or management must consider the importance of variables in this research, as well as its importance in order to optimize the business activities. MSME owner also should consider focusing on the variables that have a significant influence in usage of e-wallet to improve the business activities.


INTRODUCTION
The development of transactions between people goes through a lot of small and big changes. Prehistoric times we know are bartering and then developing into precious metals such as gold and silver. Began to form infrastructure and a regulatory system so that there could be banknotes that were recognized by many people, and the most recent development was electronic money. This development goes through thousands of years and the millions of technologies and beliefs that were created so that we can benefit from it today. In the past, transactions could be limited between places and the number of transactions. But now the transaction can be anywhere, anytime, and in large quantities. Electronic money can be divided into two, chips and servers. Both have their own advantages and disadvantages. Chip electronic money has a shape like an ATM card, has a small balance limit, the card can be used by other people and can be used offline is the biggest advantage for example is Tapcash, Flaz, Brizzi which is different from server-based electronic money or better known as e wallets that have a form of application (software) the balance that is stored can be more than chip-based, the biggest drawback is that it requires an internet connection to be able to use it.
Convenience is the main reason consumers prefer to use digital wallets over cash, which is 68%. This is because they don't need to carry a lot of money and debit/credit cards and don't have to worry about money and change when paying. The reason for using the promo is only 23%. That way, based on the same survey, consumers will continue to transact with digital wallets even though various promos that have been in effect have been removed. Consumers also use digital wallets for security reasons, amounting to 9%, such as the existence of transaction records and the ease of blocking accounts if a cell phone is lost. For the end user, the main benefits include greater flexibility when seeking alternative sources of funds and/or payment instruments that can be used to make payments at merchants. For the merchant, QRIS is more practical because only one standardized QR code is required in order to accept payments from various sources of funds and/or payment instruments. For the industry, QRIS supports interconnectedness and interoperability, thus minimizing fragmentation and enhancing efficiency.

THEORETICAL FRAMEWORK Entrepreneurship
Entrepreneurship is an activity to assess risks and opportunities to make the right decisions so that they can generate profits. Entrepreneurship also come from the idea or innovation from the people this argument also similar with Kurratko and Hodgetts (1992). The competitive behaviors that drive the market process, alternatively phrased as the introduction of new economic activity that leads to change in the marketplace (Davidsson, 2016). Entrepreneurship is the process of creating something new with value by devoting the necessary time and effort, assuming the accompanying financial, psychic, and social risks, and receiving the resulting rewards of monetary and personal satisfaction (Hisrich, Peters and Sheperd, 2010). Entrepreneurship is a trait that is found more in smaller and younger enterprises than in larger and older ones. This is due to the conditions favoring its development is more likely to be present in smaller and younger enterprises (Stevenson and Gumpert, 1985). Shane and Venkataraman (2000) Entrepreneurship is defined as an activity that involves the discovery, evaluation, and exploitation of opportunities to introduce new goods and services, ways of organizing, markets, processes, and raw materials through organizing efforts that previously had not existed.

E-Wallet
A digital wallet, also known as e-wallet, refers to electronic services that store payment instrument data, including card-based payment instruments and electronic money, which can store funds to make payments. e-Wallet Restrictions, Using Funds in e-Wallets funds contained in an e-Wallet can only be used for the following payments; shopping transactions paying utility bills. Funds contained in an e-Wallet are non-transferable to another e-Wallet. An e-wallet is a type of electronic card which is used for transactions made online through a computer or a smartphone. Its utility is the same as a credit or debit card. An e-Wallet needs to be linked with the individual's bank account to make payments E-wallet is a type of pre-paid account in which a user can store his/her money for any future online transaction. An E-wallet is protected with a password. With the help of an Ewallet, one can make payments for groceries, online purchases, and flight tickets, among others (The Economic Times, 2020).

229
Jurnal EMBA Vol.10 No.2 April 2022, Hal. 227-238 Previous Research Megadewandanu, Suyoto, and Pranowo (2016) identified the behavior and user acceptance factors of mobile wallet technology. Online survey was conducted among 372 respondents to test hypothesis based on UTAUT2 model. Respondents consisted of 61.29% of male and 38.71% of female with age proportion was dominated by age group of 20's of 78.76%. In addition, 50.81% of respondents never used mobile wallet before and 49.19% of respondents have ever used mobile wallet. Data obtained were confirmed using confirmatory factor analysis and analyzed using structural equation model. The study found that habit was the factor that most strongly affected individual behavioral intention to use mobile wallet in Indonesia, followed by social influence, effort expectancy and hedonic motivation. Sulaeman and Ninglasari (2020) examined the behavioral intention of the Muslim crowd funders to use the Zakat-based crowdfunding platform model by adopting the Unified Theory of Acceptance and Use of Technology (UTAUT) Model. The online platform, as an intermediary between Muslim crowd funders and MSMEs, provides financing services for MSMEs to fight any situation/pandemic like COVID-19 nowadays. This study used the primary data collected by using the online survey questionnaires, and then the analysis is conducted using partial least squares (PLS) regression. The empirical study shows that all the variables except for facilitating conditions have a significant positive effect on Muslim crowd funders' intention to use the Zakat-based crowdfunding platform model. Puteri and Wijayangka (2020) determined how acceptance of Gopay and OVO E-Wallet technology in MSMEs in Bandung uses the Unified Theory of Acceptance and Use of Technology model that has variable performance expectation, effort expectation, social influence, perceived risk, perceived cost and behavioral intention. The method used in this research is descriptive and causality methods with a quantitative approach, and using a hypothesis testing technique that is the T test. The sampling method used was purposive sampling with a total of 100 respondents. Based on the results of data processing using SmartPLS obtained performance expectation, effort expectation, social influence and perceived cost significantly influence behavioral intention of 97.5%. Perceived risk is known to have no effect on this study, this can be caused by differences in the type of technology, the monetary value of the technology users use and also the reputation of the community.

Conceptual Framework
Conceptual framework explains about the relation between the variables in this research:

Source: Literature Review
From this conceptual framework, this study aims to find out usage of using e wallets on micro small and medium enterprises.

Research Approach
This research is conducted by using quantitative method, which is a type of research that is `explaining phenomena by collecting numerical data that are analyzed using mathematically based methods or statistics (Creswell and Poth, 2003). According to Saunders, Lewis, and Thornhill (2009) used as a synonym for any data collection technique (such as a questionnaire) or data analysis procedure (such as graphs or statistics) that generates or uses numerical data.

Population and Sample
Population refers to the entire group of people, events, or object of interest that the researcher seeks to investigate (Sekaran & Bougie, 2010). According to Sugiyono (2018) Population is a generalization of the area consisting of objects or subjects that have certain qualities and characteristics determined by researchers to study and then draw conclusions. Owners/the person in charge of the micro, small and medium enterprises in Manado use e-wallet that can be informants about these issues. Based on the results of the sample count, the number 96.04 is obtained for the minimum number of samples. Researchers determine the minimum sample to be 100 respondents. The requirements in selecting the samples, Owners/people in charge of MSME that use e-wallet in transactions.

Data Collection Technique
In this study the information was collected using a questionnaire. The design of the questionnaire is influenced by several factors, including the type of research being conducted and the way in which it will be conducted (Proctor, 2000). Designing the questionnaire involved identifying the specific information needed to accomplish this study's objective, the correct phrasing of the questions, the sequence in which the questions would be arranged, and the layout that will best serve the research objectives (Zikmund, 2003). The questionnaire was designed using structured questions only. In other words, the questions had a pre-specified set of response alternatives and response format (Malhotra, 2004). The scaling process follows the guidelines by Hair et al (2009).

Operational Definition of Research Variable
Operational Definition of Research Variable will explain the definition of the variable from this research. Degree to which the consumer of mobile services believes that he or she may be exposed to certain types of financial, social, psychological, physical or time risk (Zhang, Zhu, & Liu, 2012)  Worries  Security Abrahão et al. (2016) Perceived Cost The initial, subscription, transaction and communication costs to which the consumer believes he or she will be submitted to in the future. It also includes the consumer's ability to buy a mobile device that is compatible with the mobile payment service (Shafinah et al., 2013)  Expensive  Make a lot of effort  Time Abrahão et al. (2016) Behavioral Intention Effective use by the consumer of a future product or service (Ventakest et al., 2003)  Access  helpful Abrahão et al. (2016) 231 Jurnal EMBA Vol.10 No.2 April 2022, Hal. 227-238

Result
This is the demographics statistics of respondents which is determined by dividing the discussion into several part: respondents' gender, respondents' age, and respondents' duration in marketplace platform. The total sample distribution based on several characteristics. Based on gender, there's a total 70 male people or 70% of the total sample is a male. On the other hand, there's a total 30 female people or 30% of the total sample is a female. Based on group of age, there's a total 66 people or 66% of the total sample is a group of people over the age of 30 years old, total 21 people or 21% of the total sample is a group of people between 25-30 years old, total 10 people or 10% of the total sample is a group of people between 20-25 years old, and 3 people or 3% of the total sample is a group of people under the age of 20 years old. Based on the duration of doing business, there's total 72 people or 72% of the total sample is a group of people doing business more than 1 year, 10 people of 10% of the total sample is a group of people doing business between 6 months to 1 year, and 18 people or 18% of the total sample is a group of people doing business between 3 months to 6 months.

Validity Test Result
Validity of a test means a test or an instrument is measuring accurately what it is supposed to measure. In SPSS, an instrument is considered accurate if it passed the 0.300 in corrected item-total correlation. Based on the data processed, it is concluded that every instrument in this research is valid, based on the value which passed 0.300.

Reliability Test Result
This part shows the reliability test results. Reliability is an analytical tool used by the researchers to test the stability and consistency of the research results (Sekaran & Bougie, 2010). Reliability less than 0.600 is poor, those in the 0.700 range, acceptable, and those over 0.800 are considered good. The table above shows that the Cronbach's Alpha 0.731 > 0.600, it is indicated that all research instrument indicator of variable is reliable.

Normality Test
Normality test is a test intended to test whether the regression model of the influence of recruitment, selection, and training and development on employee performance has a normal distribution or not. Regression model is good if the data distribution is normal or near normal. If the data is spread around the diagonal line and follow the direction of the diagonal line, then the regression fulfills the normality assumption.

Figure 3. Normality Test Result
Source: Data Processed Based on the figure above we can conclude that the data is normally distributed based on the graph in which is close to the normality line and distributed following the normality line.

Multicollinearity Test
Multicollinearity is an often-encountered statistical phenomenon in which two or more independent variables in a multiple regression model are highly correlated (Sekaran and Bougie, 2010). Jurnal EMBA Vol.10 No.2 April 2022, Hal. 227-238 Multicollinearity test used to test whether there is any correlation between independent variable in the regression model. The table above shows the calculation of multicollinearity, it can be known through the VIF and tolerance. If the value of VIF and tolerance is <10 then the regression model is free from multicollinearity. Based on the result in table 4.4, the symptoms of multicollinearity do not occur, because the VIF value of performance expectation, effort expectation, social influence, perceived risk, and perceived cost are 3.192, 2.619, 4.045, 1.370, and 3.307. It is concluded that there's no connection between the independent variables in this research. So, the assumption that there is no multicollinearity is met.

Heteroscedasticity Test
Heteroscedasticity test is a test to determine whether the error terms are normally distributed or not. The variance of the model must be different for all observations. Heteroscedasticity met when the pattern of data is well established.

Figure 4. Heteroscedasticity Test Result
Source: Data Processed A good regression model is a model that is free from the problem of heteroscedasticity. Based on the figure above, we can see that there is a spread of data of this regression analysis. This concluded that the independent variables which are performance expectation, effort expectation, social influence, perceived risk, and perceived cost are free from heteroscedasticity.

Multiple Linear Regression Analysis
As pointed out in the previous section that the objective of this study to know the influence of performance expectation (X1), effort expectation (X2), social influence (X3), perceived risk (X4), perceived cost (X5) on behavior intention (Y) and to test the hypothesis using linear regression methods. This study analyzes the regression of performance expectation (X1), effort expectation (X2), social influence (X3), perceived risk (X4), perceived cost (X5) on behavior intention (Y), with the help of a computer program package of SPSS version 25.0. The result of data processed is shown on  influence on behavior intention. One scale or one unit increase in performance expectation will decrease behavior intention by 0.049. 3. Coefficient value of 0.034 means that if the variable in this research, effort expectation, has a positive influence on behavior intention. One scale or one unit increase in effort expectation will increase behavior intention by 0.034. 4. Coefficient value of 0.040 means that if the variable in this research, social influence, has a positive influence on behavior intention. One scale or one unit increase in social influence will increase behavior intention by 0.040. 5. Coefficient value of -0.011 means that if the variable in this research, perceived risk, has a negative influence on behavior intention. One scale or one unit increase in perceived risk will decrease behavior intention by 0.011. 6. Coefficient value of -0.012 means that if the variable in this research, perceived cots, has a negative influence on behavior intention. One scale or one unit increase in perceived cost will decrease behavior intention by 0.012.

Coefficient Correlation (R) and Coefficient Determination (R2)
The correlation coefficient is used to measure the impact of performance expectation, effort expectation, social influence, perceived risk, and perceived cost on behavior intention. The correlation coefficient values can be seen in the table below: coefficient determination (R2) of 0.133 shows that the linear relationship in this model can explain the behavior intention (Y) for 13.3% while the rest 86.7% is explained by other factors not discussed in this research. Since independent variables used in this research is more than two variables, then adjusted R2 is preferably used. In this case adjusted R2 is 0.087. It means 8.7% variation of dependent variable can be explained by five independent variables, while 91.3% is explained by other causes.

Partial Test (T-Test)
Partial test of the hypothesis in this research is conducted by using t-test. It is used to test the impact of performance expectation (X1), effort expectation (X2), social influence (X3), perceived risk (X4), and perceived cost (X5), on behavior intention (Y) at α = 0.05 or compare the probability of the confidence level 95% of the partial coefficient (R) so that it can be seen the influence of the independent variables individually. Since there is a level of significance 5%, then the t-table will be t100; 0.05 = 1.660. Based on the calculations in the previous table, the interpretation as follows: 1. T-count for performance expectation (X1) is -3.027 at the level of significance of 0.003, which is lower than 0.05. It means that performance expectation (X1) has a significant partial influence on behavior intention (Y). 2. T-count for effort expectation (X2) is 3.173 at the level of significance of 0.002, which is lower than 0.05. It means that effort expectation (X2) has a significant partial influence on behavior intention (Y). 3. T-count for social influence (X3) is 2.118 at the level of significance of 0.037, which is lower than 0.05. It means that social influence (X3) has a significant partial influence on behavior intention (Y). 4. T-count for perceived risk (X4) is -1.650 at the level of significance of 0.102, which is higher than 0.05. It means that perceived risk (X4) doesn't have a significant partial influence on behavior intention (Y). 5. T-count for perceived cost (X5) is -0.973 at the level of significance of 0.333, which is higher than 0.05. It means that perceived cost (X5) doesn't have a significant partial influence on behavior intention (Y).

Simultaneous Test (F-Test)
Simultaneous test is conducted to determine the impact of recruitment, selection and training and development on employee performance. Simultaneous test is conducted using F-test with significance level 0.05 (5%). lower than 0.050. In conclusion, the independent variables simultaneously influence dependent variable significantly.

Discussion
Performance expectation has a negative significant influence on behavior intention of usage of e-wallet. This result is aligned with the previous research from Anwar and Rikumahu (2020) that Performance Expectation did not have a positive influence on Behavior Intention to Adopt. On the other hand, the result of this in contradicted with the result of the research from Antareza, Saefuloh, and Gunawan (2021) that stated that performance expectation has a positive significant influence on behavior intention to use technology, considering that business owner showed positive influence toward the usage of technology caused by the benefits and usefulness of the technology in order to improve their business activities. Furthermore, the result of this research is also not aligned with the previous research from Megadewandanu, Suyoto, and Pranowo (2016) that explained that performance does not have significant affect toward behavioral intention because the mobile wallet is still at introduction stage of product life cycle, which most of user are early adopter.
Effort expectation has a positive significant influence on behavior intention of usage of e-wallet. This result is aligned with the previous research from Puteri and Wijayangka (2020), in which is concluded that effort expectation is significantly influencing behavioral intention because the SMEs owner felt that it is easier to use Go-Pay and OVO (e-wallet) in doing transaction with the customers. This research is also aligned with the previous research from Sulaeman and Ninglasari (2020) that concluded that effort expectancy and behavioral intention shows a significant and positive relationship. On the other hand, the previous research from Sultan and Ramdhan (2016), stated that effort expectancy did not significantly influence the behavior intention, which resulted from the respondent who's did not consider whether it is difficult or easy to use Instagram.
Social influence has a positive significant influence on behavior intention of usage of e-wallet. This result is aligned with the previous research from Puteri and Wijayangka (2020), in which is concluded that social influence is significantly influencing behavioral intention because the SMEs owner explained that they felt that their usage of Go-Pay and OVO (e-wallet) is mainly influenced by their environment such as family and friends. This research is also aligned with the previous research from Sulaeman and Ninglasari (2020) that concluded that social influence and behavioral intention shows a significant and positive relationship. On the other hand, the previous research from Purnamaningsih, Erhan, & Rizzkalla (2018) concluded that social influence does not have an influence on behavioral intention. This previous research indicated that the respondent, which as businesses, like most business will drive the adoption of technology based on logical factor such as benefit, time, and cost. Meanwhile, social influence main driving forces are peer pressure that would not be a factor affecting a business bottom line.
Perceived risk has a negative and not significant influence on behavior intention of usage of e-wallet. This research apparently is aligned with the previous research from Puteri and Wijayangka (2020), in which is concluded that perceived risk is not significantly influencing behavioral intention because the SMEs owner felt that it is safe and convenient to use Go-Pay and OVO (e-wallet), but it is explained that it might not be significantly influencing behavioral intention regarding the personal data collected and storage online. The result of this research is also supported by the previous research from Sienatra (2020) that concluded that perceived risk doesn't have a significant influence on continuance intention. The result of this research is not aligned with the previous research from Achiriani and Hasbi (2021) that concluded that perceived risk has a significant influence on behavioral intention in using e-wallet Dana in Indonesia. Moreover, the previous research from Mustika & Rachmawati (2018) stated that perceived risk has a positive influence on behavioral intention. This caused by the assumption of risk like data theft or technical failure does not influence the usage of e-Tollpass.
Perceived cost has a negative and not significant influence on behavior intention of usage of e-wallet. This research is contradicted with the previous research from Puteri and Wijayangka (2020), in which is concluded that perceived cost is significantly influencing behavioral intention because the usage of Go-Pay and OVO (ewallet) might be using some capital and supporting equipment such as smartphone and internet data quota. This research is not aligned with the previous research from Achiriani and Hasbi (2021) that concluded that perceived cost has a significant influence on behavioral intention in using e-wallet Dana in Indonesia. Moreover, the previous research from Mustika and Rachmawati (2018) stated that perceived cost has a positive influence on behavioral intention. This is because the respondents feel that the cost needed to use e-Tollpass is not a negative factor influencing their decision of using the product.
Based on the result of this research, the F-test result showed that simultaneously all the independent variables significantly influence the behavior intention. This result is consistent with the previous research by 236 Jurnal EMBA Vol.10 No.2 April 2022, Hal. 227-238 Antareza, Saefuloh, and Gunawan (2021) that concluded that performance expectancy, effort expectancy, and social influence has an influence on intention to use e-wallet for business transactions. This result also aligned with the previous research by Puteri and Wijayangka (2020), in which is concluded that performance expectation, effort expectation, social influence, perceived risk, and perceived cost is significantly influencing behavioral intention in using e-wallet in Small-Medium Enterprises (SMEs).

Conclusion
Based on the result and data analysis, connected with research problem and hypothesis, there are several findings that can be concluded in this research, which are: 1. Performance expectation has a negative and significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City. 2. Effort expectation has a positive and significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City. 3. Social influence has a positive and significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City. 4. Perceived risk has a negative and not significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City. 5. Perceived cost has a negative and not significant influence on behavior intention to use e-wallet in food and beverage MSME in Manado City. 6. Performance expectation, effort expectation, social influence, perceived risk, and perceived cost, simultaneously impact on behavior intention to use e-wallet in food and beverage MSME in Manado City.

Recommendation
There are some recommendations that can be recommended based on the overall research of this research, which are: 1. The MSME owner or management must consider the importance of variables in this research, as well as its importance in order to optimize the business activities. 2. The MSME owner should consider focusing on the variables that have a significant influence in usage of ewallet to improve the business activities. 3. For future researcher, to study the inter-variables relationship similar to this research an improving it with another research object or maybe other related variables for better understanding the scope of this subject or this research.