Graduate Project | Course Paper

1

The relations between attributes of a third-party payment and its reuse

intention: Alipay

BY

Ye Zhiyuan

1430001110

Applied Economics

Chen Jiaqi

1430010002

Applied Economics

Submitted to

Dr. Wu Minglu

A Final Year Project Submitted to the Division of

Business and Management in Partial Fulfilment of

the Graduation Requirements for the Degree of

Bachelor of Business Administration (Honours)

Beijing Normal University – Hong Kong Baptist University

United International College

May 2017

2

Acknowledge

The authors of this final year project would like to sincerely express the greatest

gratitude to supervisor Dr. WU Minglu and assistant professor Dr. Thomas CHAN, both

of who currently work for United International College. Only after they have provided

their valuable assistance and flexible suggestions, can we successfully accomplish the

proposal and report. Furthermore, we have to mention that during the composition

period, some classmates, HE Yunping & OU Yangpian shared empirical computer

techniques with us when we came to them for help. We also appreciate those 30

volunteers who participated in our mock interview. They strongly helped in improving

the quality of questionnaire and interview procedure. We believe there are still some

other individuals who helped us, but unfortunately we cannot come up with all these

names at the time, since the space of the page is limited. What most importantly matters

is that the authors intend to send the most sophisticated wishes to the kind helpers. Good

luck to you guys!

Signature:

May 3rd, 2018

3

Table of contents

1.1 Statement of the problem(s) ………………………………………………………………… 4

1.2 Literature review ……………………………………………………………………………….. 4

1.2.1 Perceived risks …………………………………………………………………………….. 5

1.2.2 Economic benefits ………………………………………………………………………… 5

1.2.3 Convenience …………………………..……………………………………………………. 6

1.2.4 Ease of use …………………………………………………………………………………… 6

1.2.5 Consumer satisfaction …………………………………………………………………… 7

1.2.6 Reuse intention …………………………………………………………………………….. 7

1.2.7 Simultaneous equation model ………………………………………………………… 8

1.3 Objective(s) of the study …………………………………………………………………….. 8

1.4 Statement of hypotheses …………………………..…………………………………………. 8

2. Method ………………………………………………………………………………………………….. 9

3. Results …………………………………………………………………………………………………. 13

4. Discussion and implication …………………………..………………………………………… 22

5. Conclusion …………………………………………………………………………………………… 24

Appendix I: Research instrument related to proposed model ……………………….. 25

Appendix II: t-statistic & F-statistic reference …………………………………………….. 27

Reference …………………………..……………………………………………………………………. 28

4

1. Introduction

1.1 Statement of the problem(s)

Recently, when we, my partner and I, walk around the campus of United

International College and Beijing Normal University, we realize students always pay

for the bills with their smart phone, using a third-party payment, either Wechat or Alipay.

This phenomenon inspires our interest in what exactly stimulates virtually all students

to use electronic payment repeatedly. What are the factors that push students to use

electronic payment and what largely increase their reuse intention? There must be some

effective motivation which somehow keep students as loyal customers. After doing

sufficient online research, we reveal all of the academic journals we have found

focusing on a subject group rather than college students. Therefore, the aim of the

project is to conduct a research mainly concentrating on college students in Zhuhai.

1.2 Literature review

With the fast development of e-commerce in China, the third-party online payment

platform also develops in a high rate, a field which belongs to the emergence industry

(Hu, 2008). Established in 2004, Alipay now has become China’s predominant third-

party online payment platform. Third-party online payment services are well known for

escrow services, in which money is held by a third party on behalf of a supervisor

(Dahlberg, Guo, & Ondrus, 2015). Basically, Alipay provides an escrow service

enabling consumers to reflect their satisfaction with the goods they purchase online

before Alipay distributes money to the seller (Choi & Sun, 2016). In this process, Alipay

served as an authorized third party that provides guaranteed intermediation. In the

context of weak consumer protection regulations and laws, this unique payment service

is critical and necessary (Liu, 2015). In addition, Alipay also provides many more

applications and entertainment for users to engage in informative community. Alipay’s

successful operation would result from different features, which may affect the

satisfaction of users and finally are positively related to reuse intention (Odom, Kumar

& Saunders, 2002).

5

1.2.1 Perceived risks

According to the Ministry of Industry and Information Technology of China (MIIT,

2014), by January 2014 there were 0.84 billion mobile internet users in China, and only

25.1 percent of the Chinese mobile phone users use mobile payment, a percentage

which is much lower than the expected. Previous studies on third-party payment

acceptance mainly investigate the using motivations behind using m-payment, such as

ease of use, usefulness and convenience, but they lack attention of customer’s concern

suchlike perceived risk. (Kim, Mirusmonov & Lee, 2010). Considering the knowledge

gap, Yang (et al., 2015) aims to research how perceived risks are derived from various

uncertainties.

Forsythe and Shi (2003) define perceived risk is the subjective expectation of

possible losses of a customer when making decisions of online shopping. Perceived risk

of third-party payment refers to the extent of consumer’s perception of the possible

losses including financial loss, the violation of privacy, dissatisfaction with

performance, psychological discomfort, and wasting time (Yang et al., 2015). The

transfer of money between accounts through third-party would raise concern about

financial information, such as accounts and passwords being stolen and the subsequent

risk of financial loss. In mobile payment process, much more user information is

required, such as phone number, national identification number, consumption location,

shopping record etc., a situation which is also the major concern of the users. Moreover,

online payment usually draws support from wireless communication technologies that

potentially provide financial information to another industry (Kim, Mirusmonov & Lee,

2010). In addition, QR code scanning contains malicious software, which would be

used to illegally acquire private information.

1.2.2 Economic benefits

Consumers’ perceived risk and perceived profit are the two most fundamental

aspects affecting consumer decisions, including perceived negative and positive utility

(Peter & Tarpey, 1975). According to Treiblmaier, Pinterits and Arne Floh (2008), the

research showed that the economic factor that generates benefits for users has a

6

significant influence on users’ satisfaction towards online payments. Alipay and

TianHong Asset Management Company jointly launched an innovative product named

YuEBao on June 13, 2013, involving both payment and financial investment. This

financial product achieved rapid expansion. By the end of June, YuEBao received 2.5

million users and the transaction scale had reached RMB 6.6 billion (Xia & Hou, 2016).

Until today, the number of people using online financial products has grown, reflecting

a huge demand for financing in China. Some users of YuEBao choose not to be covered

by traditional financial institutions and turn to online financial instruments because of

the higher interest rate and possible money withdrawals at any time. The research

results of Choi and Sun (2016) supported that interests obtained from YuEBao account

increase the user’s satisfaction, thereby pushing patrons to reuse Alipay.

1.2.3 Convenience

According to Huang, Liu, and Wu (2017), consumers accept and reuse the

application for greater convenience, saving time and reducing cost. Convenience

yielded from third-party payment system lies in the fact that people can handle

transaction in any time and place where network is accessible (Yu et al. 2002).

Nowadays public services payments are available to be performed via the third-party

Alipay, mainly covering water charges, electricity and gas utilities, which were used to

be paid at banks or in certain locations. Moreover, Alipay launched joint services with

Didi Cab services, involving taxi hailing services, which offers the convenience to

passengers at a lower price than the traditional alternative does (Guo & Bouwman,

2016). Online shopping and online payment bring a large scale of convenience that

intensifies satisfaction to customers (David & Richard 2000).

1.2.4 Ease of use

Previous research indicated that perceived usefulness and perceived ease of use are

two determinants which cause people to accept or reject an application (Fred, 1989). In

Shin’s research (2010), it approved that perceived ease of use affects usage intention of

mobile payment systems. Ease of use reflects users’ operational experience with respect

to mobile payment. There is a negative impact of complexity on use intention,

7

indicaitng that ease of use creates positive utility on third-party payment (Xia & Hou,

2016). The higher degree of the ease of use, the higher the users’ adoption toward the

application; therefore, the more likely they would reuse the product. Consumers who

are familiar with and accustomed to the operation system more intend to acknowledge

the application.

1.2.5 Consumer satisfaction

The element satisfaction is an essential attribute of successful long-term business

relationships with customers (Balasubramanian et al., 2003). Consumer satisfaction

refers to cognitive and affective state of fulfillment in the consequence of the exchange

(Kim et al., 2009). After comparing the original expectation and the product’s actual

performance, the users develop a satisfaction level based on their confirmation level

and the expectation on which that confirmation was based. Based on previous studies

(Bhattacherjee, 2001), satisfaction was conceptualized as an affective state representing

the consumer’s emotional reaction to the entire e-commerce transaction through the

selling entity on the Internet. Thus, customer satisfaction is a customer’s subjective

judgment, deriving from observations of product performance. A higher consumer’s

perceived performance than the original expectation would lead to a relatively higher

level of satisfaction.

1.2.6 Reuse intention

Repurchase process is distinct from pre-purchase because the consumer has

previous experience to evaluate the level of satisfaction, which would largely affect

future purchase decisions and customer loyalty (Kim et al., 2009). Satisfaction acts as

an intermediate role of reuse intention, and several previous studies have shown the

positive effects of customer satisfaction on reuse intention (Choi & Sun, 2016).

Expectation-confirmation theory (ECT) is widely used to study consumer satisfaction

and repurchase intention and behavior (Kim et al., 2009). The underlying logic is: firstly,

customers generate an expectation of a specific product or service; secondly, after

consuming, they form perception of the performance of the product or service; thirdly,

a satisfaction level was established based on the conformation level; finally, customers

8

form a reuse intention based on the satisfaction level (Bhattacherjee, Anol &

Premkumar, 2004)

1.2.7 Simultaneous equation model

Empirical research is required to provide details on how the features of Alipay

affect customer satisfaction and how the satisfaction influences customers reuse

intention.

In economic theory, the relationships among multiple variables are examined by

simultaneous equation models. In simultaneous equation models, a dependent variable

of one equation is the independent variable of another equation, and equations are

simultaneously interrelated (Wang, Lee, & Chuang, 2016). According to Matzkin

(2008), the relationship between features of third party payment system, satisfaction

and reuse intention are interdependent and resulted in an endogenous problem. This

interactive relationship among variables imply the existence of a simultaneous

relationship and the single liner equation would fail to consider the interdependencies

effects could yield inconsistent findings (Hsing, 2016). The simultaneous equation

model is a better approach for investigating the interdependency and examining the

relationships among multiple variables (Wang, Lee, & Chuang, 2016).

1.3 Objective(s) of the study

In this project, an in-depth analysis will be taken with hypothesis to investigate the

relations between satisfaction level linked with Alipay adoption and the features of

Alipay, including perceived risks, economic benefit, convenience and ease of use.

Finally, we draw a conclusion on the issue of users’ reuse intention through the factor

of satisfaction level. Typically, this project mainly focuses on college students in Zhuhai

and Guangzhou. We hoped that this project would provide the present or potential

online vendors with certain amount of insight in order to help those vendors amplify

online shoppers’ purchase intention.

1.4 Statement of hypotheses

Hypothesis 1: Perceived risks negatively influence consumer satisfaction.

Hypothesis 2: Economic benefits students attained positively influence consumer

9

Economic

benefits

Perceived

risks

DV 1:

Consumer

satisfaction

Convenience

DV 2:

Reuse

intention

Monthly

income

Ease of use

IV

satisfaction.

Hypothesis 3: Convenience the application brings to students positively influences

consumer satisfaction.

Hypothesis 4: Ease of use positively influences consumer satisfaction.

Hypothesis 5: Consumer satisfaction positively influences reuse intention.

2. Method

After carefully reviewing the literature of different authors and following the

guideline, we choose multiple linear regression, simultaneous equation model &

hypothesis test as our cardinal method. The following phases were conscientiously

progressed: determining variables, articulating the interrelations, conceiving research

instrument, collecting data and interpreting data, nominating implication and limitation.

Guo & Bouwman (2016) denoted that interview is an optimal approach to collect

raw data. If we distribute online questionnaires, prospective participants might not take

them seriously. Since it is particularly hard for us to monitor the participants online, it

can be our concerned that those participants casually answer questions without correct

comprehension. In this case, the trustworthiness is likely to be much lower. Therefore,

after the questionnaire was mature, we walked around the Zhuhai campus of United

H 5

International College, Beijing Normal University and Sun Yat-sen University, taking a

hard copy of questionnaire with us. The paper was not directly displayed to students;

we randomly selected students who were passers-by around three campuses and

conducted a roughly 3-minute interview with them. After they answered those questions

listed, we accordingly recorded their responses.

According to Huang et al. (2017), the questionnaire should be considerably revised

and proofread, after a pre-test is held. It is suggested that we are supposed to invite a

few (roughly 30) college students to our mock interview. They are required to try to

answer our questionnaire during our mock interview, and when the work is done, those

participants at the first time are encouraged to give us feedback and flexible suggestion,

including what kind of questions should be involved/removed, what the appropriate

sequence of questions is and how long the interview lasts for. Meanwhile, a

11

is a factor that possibly performs an effect on our dependent variable; however, since it

is not an attribute of Alipay, we mark it as a moderating variable.

Research instrument is the soul of our project. In order to maintain the academic

creditability, we paraphrase ideas and questions from instruments provided by other

scholars. For most questions included in literature, the potential respondents are the

general population, so we need to modify those questions, to make them worthwhile

questions for college students to answer.

Several questionnaires shown in literature include skimming questions (Xie & Lin,

2014; Xia & Hou, 2016; Zhou, 2014; Davis & Michigan, 1989). After the debate

between me and my partner, we agree our own questionnaire starts with skimming

section, which is to make sure the participants are strongly familiar with Alipay and

acquaint how to operate it. The skimming part will easily indicate those unqualified

participants, and the interview will come to an end. In doing so, the skimming questions

help us intensify the efficiency of data collection. Not only does it save us time, but also

it suggests us smoothly look for the next interviewee. Section 2 is for collection of

demographic information. Although in the instrument of Xie & Lin (2014), collection

of personal information is placed in the first section, we reckon our questionnaire

should start with skimming questions followed by section of demographic information,

as this organization makes participative students more comfortable and then they prefer

to cooperate with us. Individual but not clandestine information will be measured,

including gender, age, year of study and monthly income.

For the upcoming sections, 6 sections are left behind. In other words, for the entire

questionnaire, there are 8 sections in total, within 6 sections of which each variable is

measured. Since our variables are somewhat qualitative, we need to quantify them with

Likert 5-point scale, ranking from strongly disagree to strongly agree. The complete

questionnaire is attached to appendix.

When all the required data were successfully received, we inspected the Excel

worksheet. Only data of students who chose “Yes” for question 1 & 3, and provided a

number larger than 3 for question 2, were accepted; others were deprived. Actually,

12

exact 300 observations were sustained.

Following our supervisor’s advice, first we generated respective linear regression

models, and then utilized simultaneous equation model. In order to clarify the confusion,

we define variables with notations.

PERCEIVED RISKS = per

ECONOMIC BENEFITS = eco

CONVENIENCE = con

EASE OF USE = eas

CONSUMER SATISFACTION = sat

REUSE INTENTION = int

MONTHLY INCOME = logarithm10(inc)

① sat = β0 + β1per + β2eco + β3con +β4eas + u

We ran the regression, and let SPSS display descriptive statistics. From F-test, we

recognized overall significance of four independent variables. From t–test, we realized

whether individual independent variable was statistically significant. For each

hypothesis, we either accept the null hypothesis or produce an alternative one.

Then comes the most pivotal portion of method section, the simultaneous equation

model. We assume that there is another relation between consumer satisfaction and

reuse intention. Instrumental variable should be introduced. Following Hsing’s idea

(2016), we noted an instrumental variable should be a predictor that is able to reflect

the similar property of our dependent variable consumer satisfaction; perhaps it can be

“impression” on a product. However, since our data collection is finished, and

instrumental variable is not included in the questionnaire, the measurement of

instrumental variable is missing. In this case, we have to deprive the instrumental

variable and still include 𝑠𝑎𝑡

 in equation ②. Fred (1989) announced that monthly

income also influenced reuse intention. Monthly income is observable since we asked

for information of personal income during an interview. Then we presume monthly

income is another endogenous variable. There are premises:

Covariance[𝑠𝑎𝑡

, 𝑢] = 0

13

Covariance[𝑠𝑎𝑡

, int] ≠ 0

Note: sat = 𝑠𝑎𝑡

 + 𝑢

For structural equation ①, although monthly income has something to do with

consumer satisfaction, it is included in the error term, but not directly included as an

explanatory variable. Analyzing the model created by Wang, Lee & Chuang (2016), we

need to calculate 𝑠𝑎𝑡

 and utilize 2-stage least square to run a reduced form equation

②.

② int = 0 + 1log10(inc) + 2𝑠𝑎𝑡

 + v

Because sat is very likely to interact with the error term v, we try 𝑠𝑎𝑡

 to eliminate

the adverse effect. The combination of structural equation ① and reduced form

equation ② becomes our simultaneous equation model.

Instrumental variable is quite helpful to structural equation that is moderately

identified. To the over-identified equation, although the instrumental variable might

produce an estimated parameter, the outcome is not dependable. The reason is that

instrumental variable does not reflect the properties of former predictor. In practice, for

the first stage, scholars usually calculate the coefficients and other statistical indicators

with OLS; for the latter stage, researchers generate a fitted value, such as sat

 in our

study. If every endogenous explanatory variable and instrumental variable are selected

from previous predictors, the instrumental variable will be representative. That is the

instrumental variable will be highly correlated to the endogenous variable that it stands

for.

3. Results

First, let’s have a brief view at skimming section. Among 328 interviewees, 98.48%

of persons have tried Alipay. 98.78% of individuals have utilized this application for

more than 3 months. Liu (2015) denotes that users are very likely to acquaint with the

basic operations and rudimentary functions. Also, 89.94% of people explicitly believe

they are currently making the best of the electronic platform.

52% of the participants are female, while the rest ones are male. No one chooses

to conceal his/her gender. Since gender might exert an impact on REUSE INTENTION,

14

156

144

0

20

40

60

80

100

120

140

160

180

200

G E N D E R

Female Male

we expected that the percentage of the genders could be balanced. In this way, the

impact can be eliminable. Moreover, gender is only a dummy variable and is not the

focused variable of this project, we are delighted to discover the outcome is consistent

with our expectation.

Figure I: Number of female and male interview takers

Figure II: Percentage of months during which individuals have used Alipay

Figure III: Percentage of personal opinions of whether they are familiar with the

application

15

Then let’s have a look at demographic information. Most of the college students

answering our questions are at the age of 22. Meanwhile, about 66.16% of participants

are year-4 students, so the age distribution corresponds to the year-of-study distribution.

The correspondence fits in our anticipation. Since the monthly expenditure of UIC

students is relatively higher, we forecast that the average monthly income of college

students from parents might be 3000 yuan. However, the figure shows over 50% of

students express their monthly income is below 2500 yuan. Yet, college students

acquiring the monthly income at comparatively low level still have incentives to use

Alipay. Since Alipay is popularized, almost every merchant located at campus,

including school canteen, coffee shop, supermarket, would place the QR code at the

cash counter, so that consumers are able to settle the payment easily with their tiny

mobile devices. Therefore, a massive amount of students prefer to pay with Alipay

under any circumstances, regardless of the sum of the bills. (Xu, Zheng & Zhou, 2011)

Figure IV: Percentage of people at various ages

Figure V: Percentage of people at different levels of income

Figure VI: Percentage of college students from year 1 to year 4

At the next step, using SPSS, we receive descriptive statistics of both demographic

information and seven variables.

Model reliability and validity were monitored by a confirmatory factor analysis.

Table I displays the indicators of Cronbach test. The confident reliability interval should

be ranged from 0.8 to 0.9 (Huang et al, 2017). In our research, most of the values of

composite reliability are all around 0.8 (except for Cronbach α of perceived risks).

These outcomes indicate the good reliability of the academic scales. The average

variance extracted (AVE) for every construct is precisely beyond 0.500: good

convergent validities!

Table I:

Items

Loading

AVE

CR

Perceived risks Cronbach α = 0.603

Perceived risks 1

0.599

0.637

0.581

Perceived risks 2

0.543

Perceived risks 3

0.567

Economic benefits Cronbach α = 0.799

Economic benefits 1

0.838

0.634

0.787

Economic benefits 2

0.697

Economic benefits 3

0.801

Convenience Cronbach α = 0.779

Convenience 1

0.797

0.823

0.698

Convenience 2

0.743

Convenience 3

0.707

17

Ease of use Cronbach α = 0.801

Ease of use 1

0.801

0.736

0.772

Ease of use 2

0.793

Ease of use 3

0.706

Consumer Satisfaction Cronbach α = 0.762

Consumer Satisfaction 1

0.799

0.637

0.781

Consumer Satisfaction 2

0.743

Reuse intention Cronbach α = 0.735

Reuse intention 1

0.767

0.811

0.689

Reuse intention 2

0.704

Reuse intention 3

0.722

Table II:

Descriptive Statistics

N

Minimum

Maximum

Mean

Std. Deviation

AGE

300

18

24

21.44

1.400

INCOME

300

1000

30000

3630.75

3541.360

RISK

300

1.00

5.00

3.6200

0.58791

BENEFITS

300

1.67

5.00

3.5400

0.63722

CONVENIENCE

300

1.67

5.00

4.0844

0.55361

EASE

300

2.00

5.00

3.9167

0.58542

SATISFACTION

300

2.00

5.00

3.7700

0.58587

INTENTION

300

2.00

5.00

4.1767

0.62665

Valid N (listwise)

300

Something draws our attention is that the standard deviation of monthly income is

particularly large. The range of monthly income is from 1000 yuan to 30000 yuan,

while the arithmetic mean is 3630.75 yuan, which is larger than our expectation, 3000

yuan. The statistics somehow illustrate college students of UIC, Sun Yat-sen University

(Zhuhai campus) and Beijing Normal University (Zhuhai campus) enjoy a satisfactory

living standard.

Using perceived risks, economic benefits, convenience, ease of use and reuse

intention as predictors, and setting consumer satisfaction as dependent variable, we get

the first multiple linear regression. Please keep in mind for the rest of the report,

variables will be denoted with notations listed in method section.

18

Table III:

Correlations

sat

per

eco

con

eas

Pearson

Correlation

SATISFACTION

1.000

0.051

0.388

0.459

0.576

RISK

0.051

1.000

0.087

0.179

0.144

BENEFITS

0.388

0.087

1.000

0.365

0.292

CONVENIENCE

0.459

0.179

0.365

1.000

0.460

EASE

0.576

0.144

0.292

0.460

1.000

Sig. (1-tailed)

SATISFACTION

–

0.188

0.000

RISK

0.188

–

0.066

0.001

0.006

BENEFITS

0.000

0.066

–

0.000

CONVENIENCE

0.000

0.001

0.000

–

0.000

EASE

0.000

0.006

0.000

–

N

SATISFACTION

300

RISK

300

BENEFITS

300

CONVENIENCE

300

EASE

300

Table III tells the level of correlations is low (most of the figures ＜0.2). We ran an

additional model (predictors: perceived risks, economic benefits, convenience;

dependent variable: ease of use) and get a R2 of 0.438. Variance inflation factor = 1/(1

– R2) = 1.779. It demonstrates the level of collinearity is low. Normally, when VIF ≥

10, the multi-collinearity exists!

Table IV:

19

Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

0.661a

0.437

0.430

0.45698

a. Predictors: (Constant), EASE, RISK, BENEFIT, CONVENIENCE

Table V:

ANOVAb

Model

Sum of Squares

df

Mean Square

F

Sig.

1

Regression

47.905

4

11.976

57.350

0.000a

Residual

61.604

295

0.209

Total

109.509

299

a. Predictors: (Constant), EASE, RISK, BENEFIT, CONVENIENCE

Table VI:

Coefficientsa

Model

Unstandardized

Coefficients

Standardized

Coefficients

t

Sig.

B

Std. Error

Beta

1

(Constant)

0.711

0.262

2.711

0.007

RISK

-0.081

0.046

-0.079

-1.781

0.076

BENEFIT

0.200

0.046

0.210

4.406

0.000

CONVENIENCE

0.228

0.056

0.211

4.083

0.000

EASE

0.439

0.051

0.429

8.556

0.000

a. Dependent Variable: SATISFACTION

Let us check the statistics listed in the tables above. The equation 1 can be re-written

with the given coefficients.

① sat = 0.711 – 0.081per + 0.2eco + 0.228con + 0.439eas

Adjusted R square indicates that the predictors of regression/equation ① are able

to explain 43.0% of the variation of consumer satisfaction (300 observations fit in this

regression model well). We can conclude that null hypotheses 2, 3, 4 referred in

20

literature review section are accepted (hypothesis 5 will be discussed later). Since t-

value of perceived risks is small, the proposed negative relation between risks and

satisfaction might not exist. Null hypothesis 1 is unaccepted. At table IV, F-statistic

implicates the overall significance of the predictors. Since the number of our

observations is exactly 300, the degree of freedom should be 295. After checking the

table of critical values, with four restrictions, we confirm the critical value at 5%

significance level is 2.37. As 57.35 is substantially larger than 2.37, we know that

generally speaking the four predictors are statistically significant and contribute great

explanation to the regression and to the dependent variable. For testing respective

significance of predictors, we implement t-test. For 1-tailed t-test, at 10% significance

level, the critical value is 1.282; at 0.5% significance level, the critical value is 2.576.

Compare if t-statistic is bigger than the absolute value of certain critical values (see

appendix). We draw a conclusion that economic benefits, convenience and ease of use

are individually, statistically more significant than perceived risks.

For the next phase, we regress equation ②, with 2-stage least square.

Table VII:

Model Description

Type of Variable

Equation 2

INTENTION

dependent

log10(inc)

predictor

SATISFACTION

predictor

𝑠𝑎𝑡



instrumental

MOD_2

Table VIII:

Model Summary

Equation 2

Multiple R

0.534

R Square

0.286

Adjusted R Square

0.281

Std. Error of the Estimate

0.624

21

Table IX:

ANOVA

Sum of Squares

df

Mean Square

F

Sig.

Equation 2

Regression

46.035

2

23.018

59.146

0.000

Residual

115.194

296

0.389

Total

161.229

298

Table X:

Coefficients

Unstandardized

Coefficients

Beta

t

Sig.

B

Std. Error

Equation 2

(Constant)

1.323

0.649

2.039

0.042

log10(inc)

-0.256

0.173

-0.086

-1.481

0.140

𝑠𝑎𝑡



0.988

0.091

0.953

10.876

0.000

According to the outcome, the complete equation ② should be:

int = 1.323 – 0.256log10(inc) + 0.988𝑠𝑎𝑡

 + v

Although the adjusted R2 of equations ② is not as high as that of equation ①, but

this model is also acceptable, since F-statistics is large enough. Predictors and

instrument of equation ② explain 28.1% of variation of reuse intention. F-statistic is

a pivotal indicator, indicating that the joint significance of log10(inc) and 𝑠𝑎𝑡

 is at

high level. Respectively, t-value of log10(inc) is small, so log10(inc) cannot be a

satisfactory predictor; t-value of 𝑠𝑎𝑡

 is quite massive, and thus 𝑠𝑎𝑡

 serves as an

effective instrumental variable. 𝑠𝑎𝑡

 is statistically significant, so it has a necessary

impact on reuse intention. Since the coefficient is positive, null hypothesis 5 is

accepted!

22

4. Discussion and implication

Table XI:

Path To

Path From

Hypotheses

Results

Consumer

Satisfaction

Perceived risks

H1: Perceived risks negatively

influence consumer satisfaction

Unaccepted

Economic

benefits

H2: Economic benefits students

attained positively influence

consumer satisfaction

Accepted

Convenience

H3: Convenience the application

brings to students positively

influences consumer satisfaction

Accepted

Ease of use

H4: Ease of use positively

influences consumer satisfaction

Accepted

Although Alipay now has become China’s leading third-party online payment

platform, its performance cannot be guaranteed to retain. In order to ensure sustainable

performance, Alipay needs to have a deeper understanding of how users’ satisfaction is

developed and how it affects ongoing reuse intention.

By browsing previous research, we selected four core features of Alipay as

determinants: perceived risk, economic benefits, convenience and ease of use. Three of

the four proposed features expect for perceived risk, worked well and were being

statistically significant, which deserve great attention. We noticed that in the reliability

test, Cronbach α of perceived risks is outside the confident reliability interval. Therefore,

in the later regression teat, perceived risk is not statistically significant. Particularly, the

magnitudes of influence induced by ease of use of Alipay are suggested by figures to

be the most significant factors positively affecting satisfaction.

According to the statistically significant figure, the feature convenience (H3) and

especially ease of use (H4) would be considered as the core competence or necessary

condition for customer satisfaction and so for reuse intention. Thus, strategic priority

should be given to these two perspectives to enhance the overall performance. For

Alipay, it could be better to launch more joint services with other online and offline

platform for users to engage in informative community activities. Besides, interface and

structure should be optimized to sustain its long-term reuse intention.

23

The perceived risks (H1) of an online payment system, which directly link to users’

financial security, are naturally being the first consideration before people decide to use.

However, in this research, this feature is statistically insignificance. Since our research

objective is all college student, the reason for insignificance of perceived risk could be

college students are lack of security awareness of asset. Nevertheless, the result of

statistically insignificance suggested that still, the managerial group of Alipay should

expand corresponding emphasis on keeping security control and reduce people’s

expectation of risks. Moreover, a reliable and stable market environment is vital for a

high-growth economy, the government should introduce relevant policies to standardize

the operation of the market.

The second determinant economic benefits (H2) may be considered as a sufficient

supplement to increase user’s satisfaction for sustainable management. As

Bhattacherjee (2001) concluded, this partial fulfillment comes from the unique

character of Chinese culture. Even if Chinese people emphasize social relationship

(“Guān xì”), it is based on its roots of mutual economic benefits, not just social

friendliness. Thus, Alipay should make effort to enhance the functioning of economy

factors for the role of customer satisfaction as a mediator for reuse intention.

Table XII:

Path To

Path From

Hypotheses

Result

Reuse

Intention

Consumer

Satisfaction

H5: Consumer satisfaction

positively influences reuse

intention.

Accepted

According to the figure of simultaneous equation model, t-value of log10(inc) is

small, so log10(inc) cannot be a satisfactory predictor. The reason could be the data of

income is somehow difficult to obtain, because people are not so willing to disclose

their real income. The figure also indicates that customer satisfaction plays a crucial

role on reuse intention. Thus, it indicates that Alipay need to make these four strategic

variables more intense to ensure its long-term business success. Moreover, due to the

complexity of sustainable operation in a dynamic ecosystem, actual practice is a

difficulty. Alipay should pay great effort to collaborate with other system, such as

offline banks, high-tech companies and government department.

24

Limitations exit in our project. Firstly, there are limited number of sample, 300.

This research targeted at Guangdong undergraduate student, while 300 samples may

not be representative enough. The balance of gender still needs to be guaranteed. If

imbalance takes place during the future interview, the researchers might measure

whether dummy variable gender has something necessary to do with satisfaction and

reuse intention. Secondly, since the R square in this project is approximately 0.43, it

suggested that there are still potential factors affecting users’ satisfaction and reuse

intention, while this project cannot cover all of them. Lastly, we mainly conducted

cross-sectional study, which people deal business in a changing world and user behavior

is dynamic. Therefore, a longitudinal research based on time serious would provide

more insights boosting satisfaction and reuse intention level to vendors.

5. Conclusion

Retaining loyal customers and facilitating their continuous use are essential to the

success of mobile payment service. By conceiving multiple linear regression and

simultaneous equation model, we input four core variables that somewhat cover the

dominant features of Alipay. This research identifies the connection between the

attributes of the third-party payment and the reuse intention of users and determines if

the connection is reverse, and how strong the connection is. Hypothesis test was well

completed, plus four original hypotheses are defended. Except perceived risks, others

all show positive influence on reuse intention. The result denotes that the research &

development team of Alipay shall make extra efforts to diminish the perceived risks

and sustain (or even enlarge) the level of economic benefits, convenience and ease of

use. In doing so, Alipay will become more powerful, further popularize and attract

infinite users.

25

Appendix I: Research instrument related to proposed model

Section 1: Skimming

Adapted from work of Xie & Lin (2014) and Xia & Zhou (2016)

1. Have you ever used Alipay?

A. Yes B. No

2. For how long have you used this application? Please enter the number of months.

3. Are you currently understood with the basic operation and usual functions of

Alipay? A. Yes B. No C. Not sure

Section 2: Demographic information

Adapted from Xie & Lin (2014)

4. Your gender.

A. Male B. Female C. Prefer not to disclose

5. Your age. Please enter an integer.

6. Year of study.

A. Freshman B. Sophomore C. Junior D. Senior

7. Your monthly income. Please enter an approximate integer.

Section 3: Measurement of perceived risks

Adapted from Yang et al. (2014) and Xia & Hou (2016)

8. It is my concerned that my personal information inputted into my Alipay account

will be shoplifted by another institute for criminal use.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

9. I’m worried all the time that economic losses happen if my account is stolen.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

10. I believe Alipay Inc. has implemented sufficient issues to protect my account.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

Section 4: Measurement of economic benefits

Adapted from Choi & Sun (2016)

11. Alipay offers certain amount of discounts.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

12. I have spent fewer fees charged when using Alipay than paying with cash or bank

cards.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

13. Unlike the financial instruments provided by commercial banks, those instruments

provided by Alipay are much more beneficial.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

Section 5: Measurement of convenience

Adjusted from Choi & Sun (2016)

14. Alipay breaks the confine of time and space.

26

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

15. When I purchase online tickets, suchlike train tickets, theatre tickets and tickets

for entry of theme parks or scenic spots, paying with Alipay brings convenience.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

16. When something emergent happens, I can obtain support from the department of

customer services.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

Section 6: Measurement of ease of use

Adapted from two separated articles of Zhou (2014)

17. It only took me a little time to figure out the operation of Alipay.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

18. I have never got in trouble with the operation.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

19. The application provides overt guideline for users on how to make the best of

diversified functions.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

Section 7: Measurement of consumer satisfaction

Revised from Guo & Bouwman (2016)

20. I have never complained about any problem of Alipay.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

21. My general experience and impression of this digital wallet is satisfactory.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

Section 8: Measurement of reuse intention

Modified from Xie & Lin (2014); Xia & Hou (2016); Zhou (2014)

22. Among cash, Alipay and bank cards, I will select Alipay regardless of any

circumstance

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

23. I like to recommend Alipay to other students.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

24. I will continue to use Alipay whenever I need to pay in the next 6 months.

A. Strongly agree B. Agree C. Neutral D. Disagree

E. Strongly disagree

Appendix II: t-statistic & F-statistic reference

Equation ① sat = 0.711 – 0.081per + 0.2eco + 0.228con + 0.439eas

F (4,295) = 2.37 (5% significance)

F = 57.35 ＞＞ 2.37

2-tailed t-test

Critical t-value = 1.282 (20% sig.)

= 1.645 (10% sig.)

= 1.960 (5% sig.)

= 2.326 (2% sig.)

= 2.576 (1% sig.)

(β1

) |𝑡| = 1.781 ＞ 1.645

(β2

) t = 4.406 ＞ 2.576

(β3

) t = 4.083 ＞ 2.576

(β4

) t = 8.556 ＞＞ 2.576

Equation ② int = 1.323 – 0.256log10(inc) + 0.988𝑠𝑎𝑡

 + v

F (2,297) = 3.00 (5% significance)

F = 59.146 ＞＞ 3.00

2-tailed t-test

Critical t-value = 1.282 (20% sig.)

= 1.645 (10% sig.)

= 1.960 (5% sig.)

= 2.326 (2% sig.)

= 2.576 (1% sig.)

(1) |𝑡| = 1.481 ＞ 1.282

(2) t = 10.876 ＞ 2.576

28

Reference

Balasubramanian, Sridhar, Prabhudev Konana and Nirup M Menon, (2003), Customer

Satisfaction in Virtual Environments: A Study of Online Investing, Management

Science, 49(7), 871-889.

Bhattacherjee, Anol & Premkumar. G, (2004), Understanding Changes in Belief and

Attitude Toward information Technology Usage: A Theoretical Model and

Longitudinal Test,” MIS Quarterly, 28(2), 229-254.

Choi, Y., & Sun, L. (2016). Reuse intention of third-party online payments: A focus on

the sustainable factors of alipay. Sustainability, 8(2), 147.

doi:10.3390/su8020147

Dahlberg, T., Guo, J., & Ondrus, J. (2015). A critical review of mobile payment

research. Electronic Commerce Research and Applications, 14(5), 265-284.

doi:10.1016/j.elerap.2015.07.006

Forsythe, S.M. and Shi, B. (2003), “Consumer patronage and risk perceptions in

internet shopping”, Journal of Business Research, Vol. 56 No. 11, pp. 867-875.

Fred, D. D., (1989), Perceived usefulness, perceived ease of use, and user acceptance

of information technology. The Management Information Systems Research

Center, 13(3), doi: 10.2307/249008.

Guo, J., & Bouwman, H. (2016). An ecosystem view on third party mobile payment

providers: A case study of alipay wallet. Info, 18(5), 56-78. doi:10.1108/info-01-

2016-0003

Hsing, Y. (2016). Determinants of the ZAR/USD exchange rate and policy

implications: A simultaneous-equation model. Cogent Economics & Finance,

4(1)http://dx.doi.org/10.1080/23322039.2016.1151131 Retrieved from

https://search.proquest.com/docview/1862241625?accountid=38789

Hu, Q. (2008). Chinese third-party payment risk analysis and countermeasures,

logistics management.

Huang, L., Liu, L., & Wu, J. (2017). Consumer acceptance of mobile payment across

time: Antecedents and moderating role of diffusion stages. Industrial

29

Management & Data Systems, 117(8), 1761.

Huang, X., Dai, X., & Liang, W. (2014). BulaPay: A novel web service based third–

party payment system for e-commerce. Electronic Commerce Research, 14(4),

611. doi:10.1007/s10660-014-9172-1

José Liébana-Cabanillas, F., Sánchez-Fernández, J., & Muñoz-Leiva, F. (2014). Role

of gender on acceptance of mobile payment. Industrial Management & Data

Systems, 114(2), 220-240. doi:10.1108/IMDS-03-2013-0137

Kim, C., Mirusmonov, M. and Lee, I. (2010), “An empirical examination of factors

influencing the intention to use mobile payment”, Computers in Human

Behavior, Vol. 26 No. 3, pp. 310-322.

Kim, Dan & Ferrin, Donald & Raghav Rao, H. (2009). Trust and Satisfaction, Two

Stepping Stones for Successful E-Commerce Relationships: A Longitudinal

Exploration. Information Systems Research. 20. 237-257, doi:

10.1287/isre.1080.0188

Liu, Y. (2015). Consumer protection in mobile payments in China: A critical analysis

of Alipay’s service agreement. Computer Law & Security Review, 31(5), 679-

688. doi:10.1016/j.clsr.2015.05.009

Lloyd, A. D., Antonioletti, M., & Sloan, T. M. (2016). Able but not willing? exploring

divides in digital versus physical payment use in china. Information Technology

& People, 29(2), 250-279. doi:10.1108/ITP-10-2014-0243

Matzkin, R. (2008). Identification in Nonparametric Simultaneous Equations Models.

Econometrica, 76(5), 945-978. Retrieved from

http://www.jstor.org/stable/40056491

MIIT (2014), Monthly Economy Development Report of Telecommunication Industry

of China (January 2014), Ministry of Industry and Information Technology of

China, Beijing.

Odom, M. D., Kumar, A., & Saunders, L. (2002). Web assurance seals: How and why

they influence consumers’ decisions. Journal of Information Systems, 16(2), 231-

250.

30

Peter, J.P, Tarpey, L. (1975) A comparative analysis of three consumer decision

strategies, Diffusion of innovation. Free Press, New York

Shin, D.H. (2010), “Modeling the interaction of users and mobile payment system:

conceptual framework”, International Journal of Human-Computer Interaction,

Vol. 26 No. 10, pp. 917-940.

Treiblmaier, H., Pinterits, A., & Floh, A. (2008). Success factors of internet payment

systems. International Journal Of Electronic Business, 6(4), 369-385.

doi:10.1504/IJEB.2008.020675

Wang, M., Cho, S., & Denton, T. (2017). The impact of personalization and

compatibility with past experience on e-banking usage. International Journal of

Bank Marketing, 35(1), 45-55. doi:10.1108/IJBM-04-2015-0046

Wang, M., Lee, M., & Chuang, J. (2016). Relations among audit committee

establishment, information transparency and earnings quality: Evidence from

simultaneous equation models. Quality and Quantity, 50(6), 2417-2431.

http://dx.doi.org/10.1007/s11135-015-0269-y Retrieved from

https://search.proquest.com/docview/1827074944?accountid=38789

Xia, H., & Hou, Z. (2016). Consumer use intention of online financial products: The

yuebao example. Financial Innovation, 2(1), 1-12. doi:10.1186/s40854-016-

0041-x

Xu, P., Zheng, S., & Zhou, S. (2011). Family and Western-Style Fast Food: Influences

on Chinese College Students’ Dairy Consumption. Journal Of Food Products

Marketing, 17(1), 1-24. doi:10.1080/10454446.2011.532370

Yang, Y., Liu, Y., Li, H., & Yu, B. (2015). Understanding perceived risks in mobile

payment acceptance. Industrial Management & Data Systems, 115(2), 253-269.

doi:10.1108/IMDS-08-2014-0243

Yu, H. C., His, K. H., & Kuo, P. J. (2002). Electronic payment systems: An analysis

and comparison of types. Technology in Society, 24(3), 331-347.

Zhong, J., & Nieminen, M. (2015). Resource-based co-innovation through platform

ecosystem: Experiences of mobile payment innovation in china. Journal of

Strategy and Management, 8(3), 283.

Zhou, T. (2014). An empirical examination of initial trust in mobile payment. Wireless

Personal Communications, 77(2), 1519-1531. doi:10.1007/s11277-013-1596-8

Zhou, T. (2014). Understanding the determinants of mobile payment continuance

usage. Industrial Management & Data Systems, 114(6), 936-948.

doi:10.1108/IMDS-02-2014-0068