Factors Influencing Educational Institutions Interest in Implementing
Online Payment Systems
Introduction :
The development of financial technology (fintech) in Indonesia is currently growing
very rapidly. This development is characterized by the increasing number of start-up
companies that have sprung up in this field. This is indicated by Bank Indonesia data as of May
2020 which shows that there are a total of 53 fintech companies that have been officially
registered, and there are also 161 fintech companies in the lending category that have been
registered with the Financial Services Authority as of March 2020. This is in line with
regulations that have been issued by the government in order to support the financial
technology ecosystem, such as Bank Indonesia Regulation No. 20/6/PBI/2018 on Electronic
Money and OJK Regulation No. 13/POJK.02/2018 on Digital Financial Innovation in the
Financial Services Sector.
In general, fintech services are divided into several classifications such as payment
services, capital raising, insurance, loans & deposits, investment management, and market
provisioning. Data as of 2019 shows that the payment and lending classifications are the 2
highest classifications that control the number of fintech companies in Indonesia (Daily Social
2019). This is also supported by the fact that there is one fintech company that became the first
start-up unicon in Indonesia, namely OVO, which is one of the fintech companies in the
payment classification. Some examples of other companies engaged in the payment
classification are GoPay from Gojek, OVO, Telkomsel's T-Cash which later changed to
LinkAja, and Dana.
Along with the increasing needs of society, the use of payment fintech services is also
projected to continue to increase with a very broad market potential, and can be implemented
in various categories. However, the development of fintech services is not without obstacles.
According to Bank Indonesia, there are still gaps in the fintech phenomenon in Indonesia,
namely in terms of infrastructure, regulation, and also the behavior of the community itself.
The education category is one category that has not been widely entered by payment
start-ups in Indonesia. Based on the results of PT Infra Digital Nusantara (PT IDN)'s internal
survey in 2018 of 109 educational institutions in the Jakarta, Bogor, Depok and Tangerang
areas, it shows that payment services for the education category are currently generally divided
into 3, namely cash payments, bank transfers and then manual reporting and checking, and
finally payment through virtual accounts. Conditions in the field show that the cash payment
system is still the main choice compared to other payment systems, which indicates that the
payment system in this category still tends to be conventional and has not moved towards
digital (Figure 1).
In addition, the payment system through transfers has also begun to be implemented in
several institutions. However, this method has drawbacks because the educational institution
still needs a manual reconciliation process so that the total payment data is not real time.
Another solution is to implement a payment system through a virtual account. This solution
can be said to be effective in terms of convenience for both parents and educational institutions,
but the implementation cost of creating a virtual account is quite high, so not all institutions
have the ability to implement it.
In some other countries, there are several online education payment services offered to
the market. For example, in the United States, there is a service called ParentPay, which offers
online education payments, and has collaborated with local banks and payment gateways
(Parentpay 2020). This service also has a mobile application to make it easier for parents to
make payments. Not only that, some services that have similar offerings include PaySchools
and MySchoolBucks. With the development of these services in other countries, there is
potential that similar services can also be offered in the Indonesian market, of course with
adjustments to the culture, regulations, and education level of the Indonesian people.
On this basis, PT IDN has been offering its services since 2018. PT IDN is a start-up
company in the field of financial technology, especially in the payment classification, and
focuses on the education payment category. PT IDN provides a platform for parents/students
to pay various education bills online through various payment options, and the education
institution gets the data in real time. PT IDN has a vision to change the behavior of Indonesian
society in the field of conventional payments to be completely paperless. However, one of the
challenges faced by the company is to build the trust of the public, as the main consumers, to
switch from cash payment systems to online payment fintech service systems, both the public
as parents of students, and the public as educational institutions.
Based on PT IDN's internal interviews with 1,000 parents in the Greater Jakarta area in
2018, 83% of them indicated that they are accustomed to making payments for PLN, PDAM
and other multifinance bills online, and 78% of them expressed interest in doing the same for
their children's school education payments (Yusra 2018). Therefore, PT IDN believes that
parents have the desire to utilize online services for their children's tuition payments, in other
words, there is a demand for online education payments. However, the acceptance of
technology from educational institutions is still a problem faced by PT IDN, even though
educational institutions are expected to be a provider or supply to the desire for online
education payments.
Financial Technology
According to Bank Indonesia (2019), financial technology is the use of a technology in
a financial system that results in new products, services, technologies, and/or business models,
and has an impact on monetary stability, financial system stability, and/or the efficiency,
smoothness, security, and reliability of a payment system. Bank Indonesia has classified the
types of financial technology services into:
1. Crowdfunding and Peer-to-Peer Lending, in this first classification, the service
offered is a means to raise or borrow money online. The following classifications of
fintech services are supervised by the Financial Services Authority (OJK),
specifically with the issuance of OJK Regulation number 77/POJK.01/2016 on
Information Technology-based Money Lending and Borrowing Services (OJK
2016).
2. Market Aggregator, the services offered in this second classification are in the form
of a portal that collects financial information or data and display it to the user, in
other words, it is also called a financial products comparison service.
3. Risk and Investment Management, this third classification is often referred to as an
online financial planner, as it offers financial planning services digitally.
4. Payment, Settlement, and Clearing, services in this last classification include e-
wallet and payment gateway services. This payment system aims to facilitate and
accelerate the payment or transaction process that is done entirely online. All
services in this classification fall under the supervision of Bank Indonesia.
Technology Adoption Model
There are several models of technology adoption, namely Theory of Reasoned Action
(TRA), Theory of Planned Behavior (TPB), Technology Acceptance Model (TAM), and
Unified Theory of Acceptance and Use of Technology (UTAUT). TRA has 2 main variables,
namely attitude and subjective norms, both of which will affect the behavioral intention
variable. TPB also has these 2 main variables, with the addition of 1 perceived behavioral
control variable. TAM has 3 main variables, namely perceived usefulness, perceived ease of
use, and attitude toward the system. Meanwhile, UTAUT has 4 main variables, namely
performance expectancy, effort expectancy, social influence, and facilitating conditions (Lai
2017).
The results of Lai's (2017) research comparing the four methods above to examine the
adoption of e-payments in Malaysia, show that TAM is the most suitable model to use for
research that aims to determine the acceptance of a new technology because TAM aims to find
the main determinants in the acceptance of a technology system, and can explain the behavior
of the general user population. In addition, TAM can also help predict whether a system will
be accepted or not accepted by users. Other models, namely TRA and TPB, are suitable for
technology products that already exist in the market and people generally already have certain
perceptions of these products. An example of a case suitable for using this model is research
on the acceptance of credit card products. While the UTAUT model is suitable for more in-
depth research, because there are variables of age, gender, and previous experience of
respondents that will also be studied.
TAM has been widely used in various information technology adoption studies. This is
because TAM has several advantages such as: TAM can explain the intention and behavior of
using technology consistently, TAM is a theory that is strong enough to be able to predict the
use of information technology, various studies using TAM have been carried out which
indicates that TAM has proven quality and statistically reliable (Olushola and Abiola 2017).
Based on the explanation above, the method chosen for this research is TAM, based on the fact
that this method is suitable for finding the main factors in the acceptance of a technology system
and can explain the behavior of the user population towards the use of a real system.
According to Davis (1980), the Technology Acceptance Model or TAM itself is a
model specifically created regarding a user's acceptance of a computer system or technology.
TAM aims to demonstrate a prototype system to potential users and then measure how deeply
motivated they are to use the system. TAM has 5 main constructs, namely perceived usefulness,
perceived ease of use, attitude toward using, behavioral intention to use, and actual system
use. According to Davis et al. (1989), there are two main concepts that are believed in user
acceptance, namely perceived ease of use and perceived usefulness. These two concepts will
then form an attitude toward using the system, where this variable is the main determinant of
whether a user or user will use or even reject the computer system offered. (Figure 2).
Perceived Ease of Use
According to Davis (1980), perceived ease of use is defined as the extent of the level
of effort that a prospective user must go through in using a computer system. This means that
the less effort required in adopting a computer system, the more this will affect the level of
acceptance by potential users. This is also strongly influenced by external factors. According
to Manalu (2019), perceived ease of use is one of the main determinants of a user's willingness
to use a computer system.
The main external factors that usually influence are social factors, cultural factors, and
political factors (Surendran 2012). Social factors include the language used, level of
knowledge, and supporting conditions/facilities. Meanwhile, political factors are usually the
effects of using technology in a political field. Davis' theory also states that the indicators used
to measure perceived ease of use are easy to learn, flexible, can control work, and easy to use.
In addition, based on the results of research by Murdiyanti et al. (2016), which examines the
acceptance of an archive preservation web technology at the organizational level, and also
Venkatesh and Davis (2000), mention that perceived ease of use can be examined through
indicators that are easy to learn and easy to use. Meanwhile, in Ciptaningsih's research (2011)
which examines the acceptance of a technology on e-archives, the perceived ease of use can be
examined through indicators of easy to learn and easy to use.
airline commerce, mentioned that one of the indicators of this construct is that it is
flexible to use.
Venkatesh and Davis (2000) reveal that perceived ease of use can affect perceived
usefulness, or perceived usefulness is defined. This is interpreted because the easier a system
is to use, the more useful it is in completing a job.
Perceived Usefulness of Technology
According to Davis in Surendran (2012), perceived usefulness is defined as a user's
perspective that a technology system can improve his job performance. Similar to the perceived
ease of use factor, the perceived technology usability factor is also strongly influenced by
external factors. Davis also conceptualizes that the perceived usefulness of technology can be
measured through indicators such as improving job performance, making work easier and
overall the technology or computer system used is felt to be beneficial to its users. This is also
supported by the research of Murdiyanti et al. (2016) and also further research from Venkatesh
and Davis (2000) which states that indicators that can explain the construct of perceived
usefulness of technology include making work easier, and reducing the costs required to
complete tasks. Meanwhile, the results of Fatmawati's (2015) research on technology
acceptance of a library information system explain that one of the indicators that can explain
this construct is a reduction in the time required to complete a particular task.
Davis said that the perceived usefulness of technology can affect the intention to use
technology. Several previous studies have also examined the adoption of a new system in the
education category. Maharoesman and Wiratmadja's (2016) research, for example, examines
user behavior in the adoption of internet banking as a means of paying for education at public
universities in Indonesia. The results showed that there are several significant factors that
influence students' interest in using internet banking as a means of paying for education,
namely factors of perceived usefulness, perceived enjoyment, and subjective norm. The results
of this study are also supported by the results of research by Jan and Contretas (2013) which
aims to determine the factors that influence technology acceptance for academic and
administrative systems at the university level. The study shows that the variables perceived
usefulness and subjective norms have a positive effect on attitude toward technology. Other
research has also been conducted by Mishra and Singh (2014) which aims to analyze the
important factors that influence the adoption of Internet Banking Services in India. The results
showed that trust, security, computer self-efficacy, perceived risk, perceived ease of use, and
perceived usefulness are the main driving factors that influence the adoption of internet banking
services in India.
Technology Acceptance Attitude
Attitude toward using Technology or attitude towards technology acceptance is defined
as the extent to which an individual has a pleasant or unpleasant attitude and assessment of an
object, which in this case is technology (Gunda 2014). This attitude can be defined as attitudes
and feelings, both positive and negative towards a technology. In TAM theory, this attitude has
an influence on the intention to use technology. This is also supported by Mahdi's research
(2018) which states that the attitude variable has a significant direct effect on intention or
intention to use a technology, and the results of Annilda's research (2017) which suggests that
the user's attitude in using a technology will affect the interest or intention to use the
technology.
According to Davis (1980), the construct of technology acceptance attitude can be
explained by indicators of the level of pleasure of a user in using a particular system. This is
also in line with the statement of Murdiyanti et al. (2016), where the results of their research
also state that this construct can be measured through indicators of the level of trust of a user
in using the system.
Intention to Use Technology
Behavioral intention to use or also called technology use intention is the degree of
possibility of a person in using an object, which in this case is a technology (Surendran 2012).
This construct can be explained through indicators of a user's interest in using a particular
system (Venkatesh and Davis 2000), and also indicators of the motivation of a user in
telling/recommending the system to other users and potential users (Murdiyanti 2016).
Furthermore, the intention to use a technology, in the TAM construct, will affect the
actual use of the system. This is supported by the results of research by Ladeinde (2011) which
aims to examine the relationship between perceived usefulness, perceived ease of use, and
intention to use simulation technology on members of the Project Management Institute. This
study shows that there is a strong correlation between the usefulness or benefits that will be felt
and the ease of use of technology on the intention to use simulation technology. Then this is
also supported by the results of Wibela's research (2018) which states that the intention to use
technology has a positive and significant effect on the actual use of services in his research on
digital financial inclusion services.
Actual System Usage
Actual system use, also called actual system use, refers to the direct use of the system
by an individual to complete their work (Davis 1980). Davis also said that actual system use
can be measured through the frequency and duration of time spent in using a technology. This
is in accordance with the resultsnresearch by Murdiyanti et al. (2016) which says that the actual
use of the system can be explained through indicators of regular use of the system by users in
a certain frequency and also the satisfaction felt by users in using a system. In addition, this is
also supported by research by Fawad et al. (2018) regarding the acceptance of a new technology
in a business environment using the TAM model and Employee Readiness for e-Business
(EREB) which says that perceived ease of use and perceived usefulness will affect actual
system use.
Operational Definition of Variables
1. Age: the age of the respondent when filling out the questionnaire.
2. Gender: the gender of the respondent when filling out the questionnaire.
3. Position/Title in the Organization: the position the respondent has in the
educational organization where he/she works.
4. Organization classification: the type of organization where the respondent
works. This variable is measured using 2 statements that reflect the variable,
namely the classification of organizational levels, and the ownership status of
the organization.
5. Perceived ease of use: a situation where respondents who will use the system
believe that the system is not complicated and easy to operate. This variable
is measured using 4 statements that reflect these variables.
6. Perceived technology usefulness: the state in which respondents who will
use the system believe that the system can help the related business processes.
This variable is measured using 5 statements that reflect these variables.
7. Technology acceptance attitude: a respondent's attitude towards the system
to be used. This variable is measured using 4 statements that reflect these
variables.
8. Intention to use technology: a situation where respondents want to
implement the system and use it in the organization. This variable is measured
using 6 statements that reflect these variables.
9. Actual system use: the situation in the field where respondents use or do not
use a system in the organization.
Data Analysis
1. Descriptive Analysis
This analysis is a tabulation of data in tabular form so as to facilitate interpretation and
analysis. This method is used to analyze the demographics of the respondents of this study and
also analyze the profile of the organization to be studied, which includes the classification of
organizational levels, and organizational ownership status.
2. SEM (PLS-SEM)
The analysis in this study will use PLS-SEM assisted by the SmartPLS program. SEM
is a method for predicting constructs in models that have many factors and relationships. While
the PLS-SEM method is considered appropriate for use in this study because PLS-SEM can be
used for complex structural models and has many indicators / constructs, has a small sample
size, and does not require normally distributed data (Hair et al. 2019).
According to Monecke and Leisch (2012), PLS SEM consists of three components,
namely the structural model, measurement model and weighting scheme. The weighting
scheme is a special feature of PLS SEM and is not found in SEM.
covariance-based. In the structural model, also known as the inner model, all latent
variables are connected to one another according to the theory of substance. Latent variables
are divided into two, namely exogenous and endogenous variables. In the measurement model,
also known as the outer model, all manifest variables or indicators will be connected to their
respective latent variables. In the PLS framework, one manifest variable can only be connected
to one latent variable. All manifest variables associated with one latent variable are referred to
as a 'block'. Thus, each latent variable has a block of manifest variables. A block must contain
at least one indicator. The way a block is connected to a latent variable can be reflective, that
is, the manifest variable acts as an indicator that is influenced by the same concept; or
formative, that is, the indicator shapes or causes changes in the latent variable (Wijanto 2008).
The PLS algorithm aims to estimate the value of all latent variables using an iteration
procedure. In PLS, there are 2 main evaluations, namely outer model evaluation or reflective
model measurement, and inner model evaluation or structural model measurement which aims
to assess the effect of one variable on other variables.
In the reflective model measurement, the evaluation is carried out by means of validity
and reliability tests. The validity test aims to test whether each question item is valid and can
reflect the measured attributes, while the reliability test is carried out on these valid question
items to test whether the item is reliable. There are two types of validity in PLS SEM, namely
convergent validity and discriminant validity. Convergent validity means that a set of indicators
represents one latent variable. This value can be illustrated through the average variance
extracted (AVE) value. The AVE value must have a minimum value of 0.5, which means that
one latent variable is able to explain more than half of the variance of its indicators on average.
Meanwhile, discriminant validity means that two conceptually different concepts must show
significant differences. If interpreted statistically, the AVE value of each latent variable must
be greater than the highest r2 value with other latent variable values. The second criterion for
discriminant validity is that the loading value for each indicator is expected to be higher than
its respective cross-loading value. In other words, if an indicator has a higher correlation with
other latent variables than with the latent variable itself, then the fit of the model should be
reconsidered. For reliability, the Cronbach's Alpha value can be used, where this value reflects
the reliability of all indicators in the model. The minimum value is 0.7 while the ideal value is
0.8 or 0.9. 2In measuring the structural model, the value criterion that can be used is the R
value of the endogenous latent variable, where this value is categorized as substantial if it has
a value above 0.67.
