Integrating Quality Features into Technology Acceptance Model for Examining the Acceptance of Mobile Payment Go-Pay

. Go-Pay is the innovation in Financial Technology offered in terms of payment efficiency. The main objective of this study was to bring further insights into consumer intentions to use mobile payments Go-Pay to helps improving service quality and increasing user satisfaction of the Go-Jek user. Mobile payment is in a developing stage and consumers are still reluctant to use mobile payment. Data was collected from Go-Jek users who have used Go-Pay as mobile payment systems in Bandung city. A survey study was conducted with the SEM analysis of 200 participants with four out of six hypotheses are supported. Research results demonstrated that there were significant relationships between service quality and intention to use, while there was no significant relationship between perceived ease of use and perceived usefulness to intention to use mobile payment Go-Pay. Integrating Quality Features into Technology Acceptance Model for Examining the Acceptance of Mobile Payment Go-Pay. The


INTRODUCTION
Information technology has become an inseparable part of daily life. Along with the increase in human activity, we need support tools to facilitate and provide a sense of comfort for people in carrying out daily activities such as technological developments in financial. The meaning of FinTech is innovation in financial services. FinTech helps people to access financial products and financial literacy easier. Therefore, the mobile payment technology is here as a result of this important development of information technology. Mobile payment which refers to the use of a mobile device to provide financial information, communications and transactions to customers from anywhere and at any time [1]. However, a low acceptance rate of mobile payment services by customers is an important concern of banking organization and financial company worldwide. Therefore, it takes understanding from consumers to understand mobile payment acceptance through examining technological factors that influence user's intention to use mobile payments [2].
Cash payment transactions are replaced by the non-cash payment system based on server and chip by mobile phone. FinTech's growth was also driven by the launching of the grand design efforts to increase the non-cash payments or often referred to as toward a Less Cash Society by Bank Indonesia as the holder of the Indonesian payment system regulatory authority. Less Cash Society is defined as a culture or trend that develops in the community in making payment transactions through non-cash payment media [3]. The mobile payment method, primarily payments using a smartphone, significantly increases payment efficiency compared to conventional payment methods [4].
Online mobile payment can save time and reduce transaction costs. FinTech begins to change human lifestyles in various aspects. One of them is like a transaction activity using online transportation. Online transportation companies see consumer needs to support efficiency and ease in payments, therefore the company provides mobile payment to facilitate consumers to pay.
For example Go-Jek initiated Go-Pay as server-based digital wallet to provide electronic money transaction system [5]. Go-Pay has collaborated with banks in Indonesia, and consumers can easily pay for top-up via ATM, internet banking, mobile banking and also Go-Jek driver [6]. Go-Pay acceptance in the society is based on quality offered by Go-Jek as service provider, therefore measurements are needed to find out the success of an information system. Through information technology systems, business processes can be carried out easily, quickly, effectively, and efficiently.
[5] [6] The Technology Acceptance Model (TAM) determines a user's acceptance and usage of new technology. Davis (1985) formulated TAM to understand the user's acceptability of information systems and to formulate the conceptual framework of information technology adoption by adapting the theory of reasoned action (TRA) [7].
To assess whether Go-Pay mobile payment technology is accepted or not, it can be measured through the Technology Acceptance Model (TAM) development model with the concept of quality. Quality Technology Acceptance Model (QTAM) focuses on quality and integrates the technology acceptance model and the service quality framework [8]. The only known model that uses service quality variable is the DeLone and McLean Information System Success Model [8]. They define service quality as overall support from the service provider to the users. They imply that service quality becomes important as users now become customers. According to DeLone and McLean, service quality is important for information system success as poor user support will lead to lost customers and sales [9]. If consumer perceptions are filled with high quality services that are useful and easy to use, they will be more likely to use the service and more customers will be willing to use mobile payment [10].

Figure1. Research Model
To measure Perceived Usefulness, the Chin and Todd's items (1995) were used and scale items from Davis (1989) was used for Perceived Ease of Use . E-Service quality is measured with items from Barnes and Noble (1997). To measure intention to use, scale items were adapted from DeLone & McLane (1992).

METHOD Population and Sample
A survey was used to study the relationships hypothesized in this research. The respondents of this study were consumers of Go-Jek who used mobile payment Go-Pay in Bandung. Based on BPS Indonesia, Bandung City has 3.701.501 people [11], and based on previous research Go-Jek

E-Service Quality (SEQ)
-  (2018), the age range of users is between 18 and 30 years. The sampling technique in this study was purposive sampling or sampling techniques with certain considerations. Purposive sampling is the selection of a group of subjects based on certain characteristics that are considered to have a close connection with the characteristics of the population that have been known previously. In other words, the sample units contacted are adjusted to certain criteria that are applied based on the research objectives or research problems [13]. The respondents of this study were Go-Jek users who have made payments with Go-Pay.

Responsiveness
Analysis methods and data using Structural Equation Modeling (SEM), the sample size must meet the minimum sample size for the application of the SEM model. According to Hair, Aderson, Tatham and Black in Kusnendi [14] the minimum sample size for SEM analysis is 100 to 200. In line with some of the opinions above, Joreskog and Sorbom (1988) in Wijanto (2008) state that the relationship between the number of variables and the minimum sample size in a structural model can be seen in table 1 below.

Measurement
This study used a semantic differential measurement scale or semantic difference scale. According to Umar (2008), scale is used to measure the meaning of an object or concept for respondents. This scale contains elements of evaluation (such as, good-bad and honest-not honest) and potential elements (active-passive and slow-fast). This scale is used because of interval data. The respondent must describe and support the statement.

Structural Equation Modeling
Analysis of the data used in this study is a statistical test of the structural equation model (structural equation model -SEM). The use of structural equation models is based on the condition of free (exogenous) and bound (endogenous) latent variables. In principle SEM is a combination of factor analysis and path analysis. The purpose is to confirm or test empirically and simultaneously measurement models and structural models that are built on the basis of theoretical studies. Thus, one of the advantages of SEM compared to the regression method and other multivariate methods is the application of SEM procedures simultaneously to a hybrid / full SEM model (a combination of measurement models and structural models).

RESULTS AND DISCUSSION
Of the 200 completed questionnaires, 183 data were used because 17 respondents did not use Go-Pay. Table 3 presents he demographic profile of the respondents. The gender distribution of the study was 63 % female and 37 % male, respectively. The majority of the respondents were young adults between 15 -35 years old.
Structural Equation Modeling (SEM) data analysis techniques was used to test the hypothesis. Before testing the hypothesis, a model assumption test was performed to see whether the model could be analyzed with SEM.

Model Assumption Test Results
There are several assumptions that must be fulfilled in the Structural Equation Model (SEM), including normality, outlier data, and multicollinearity. The results of testing the assumptions are presented as follows.

Data Normality
Normality test needs to be done both for single data normality and multivariate normality in which several variables are used in the final analysis (Ferdinad, Augusty, 2014: 62). Data normality was performed using a critical ratio skew value of ± 2.58 at a significance level of 0.01. The results of testing these assumptions are presented in table 4 as follows, Based on table 4, From the results of the normality test using Amos, the multivariate data were normally distributed based on critical ratio skew value > 2.58.

Match Test Results
Match test is carried out to check the level of compatibility between the data and the model, the validity and reliability of the measurement model, and the significance of the coefficients of the structural model. Matching test is carried out through several stages, namely the suitability of the whole model, the suitability of the measurement model, and the compatibility of the structural model.

Overall Model Fit
The suitability test for all models was conducted to evaluate the degree of compatibility or Goodness of Fit (GOF) between the data and the model in general. Based on the data processing results using AMOS, the obtained test results are as follows: The suitability test of the model above produces a probability value ( Based on this information, although not all Goodness-of-Fit shows the Fit model, it can still be said that the overall model is fit because according to Maholtra (2010: 733) the minimum size that can be used is absolute (such as GFI, AGFI), one size which is absolutely bad (such as RMSR, SRMR, RMSEA) and one comparative measure (such as NFI, NNFI, CFI, TLI, RNI). Based on the statement, each size is represented so that it can be said that the model is fit.
The magnitude of each latent variable in direct effect (Standardize effect) is presented in the table 6 as follows. The results of the significance test of the estimated path coefficients in the model after trimming are all significant at the 5% error rate or the P value of each path coefficient has a value <0.05. It can be said that the model is the best fit model in explaining the phenomenon of research, therefore the model will be used as an empirical basis for answering the proposed research problem. Hypothesis 1 H1.0 = Service Quality (SEQ) has no significant effect on Perceived Ease of Use (PEOU) H1.1 = Service Quality (SEQ) has a significant effect on Perceived Ease of Use (PEOU) The value of the relationship between SEQ and PEOU is 0.22 with a probability <0.05 (p = 0.0019) meaning H1.0 is rejected and H1.1 is accepted.

Hypothesis 2
H2.0 = Service Quality (SEQ) has no effect on Perceive Usefulness (PU) H2.1 = Service Quality (SEQ) has a significant effect on Perceived Usefulness (PU) The value of the relationship between SEQ and PEOU is 0.10 with a probability <0.05 (p = 0.042) means that H2.0 is rejected and H2.1 is accepted.

Hypothesis 3
H3.0 = Service Quality (SEQ) has no significant effect on Intention to use (Int) H3.1 = Service Quality (SEQ) has a significant effect on Intention to use (Int) The value of the relationship between SEQ with Int is equal to 0.18 with a probability <0.05 (p = 0.0199) means that H3.0 is rejected and H3.1 is accepted.  The concept of perceived ease of use shows the degree to which a person believes that the use of information systems is easy and does not require the user's effort to be able to use it. This concept includes the clarity of the purpose of the use of information systems and ease of use of the system for the purpose in accordance with the wishes of users (Davis et al., 1989). This concept provides the understanding that if the information system is easy to use, then users will tend to use it. Therefore, in developing an information system, it is necessary to consider the perceived usefulness and perceived ease of use of the user of the information system.

CONCLUSION
The main objective of this study was to bring further insights into consumer intentions to use mobile payments Go-Pay offered in terms of payment efficiency. Mobile payment is in a developing stage and consumers are still reluctant to use mobile payment.
In order to retain the customers, Go-Jek must try to achieve a high level of E-service quality. As a result, customers will enjoy the benefits of the mobile payment technology, and it would indirectly benefit Go-Jek as well. From this study, it can be concluded that at the intention to use is influenced by the service quality offered. This concept provides the understanding that if the information system is easy to use, then the user will tend to use the information system.
Based on statistical results, high service quality will give consumers a positive attitude through their perceptions about the usefulness and ease of use. With high service quality, consumers will see that the service is more valuable and easier to use. Through this perception their positive attitude will be built which in turn will lead to positive intentions.
For further research, it can be done in a different context to validate more concrete models such as the role of trust or risk can be added to the model to fully explain the more efficient use of mobile payments.