Jurnal Manajemen Bisnis Vol. 9, No.2 (2022) September, Page.

263-273
e-ISSN: 2621-1971|p-ISSN: 2088-7086

CONSUMER SATISFACTION AND ONLINE PURCHASE

DECISIONS: EXAMINING THE PRICE AND PRODUCT
ANALYSIS
Rezki Fani1*, Sri Wahyuni Nur2, Adnan Hasan3
1*,2,3
Institut Agama Islam Negeri Parepare, Jl. Amal Bhakti No.8, Bukit Harapan, Kec. Soreang, Kota Parepare,
Sulawesi Selatan 91131
ARTICLE INFO ABSTRACT
This research is a quantitative descriptive study with a path analysis
method that aims to determine the influence of price and product
assessments on online purchasing decisions and their impact on
consumer satisfaction. The population in the study were college students
living in Maros County and were consumers of online shopping
platforms. The total sample of 98 people was obtained by sampling method
using purposive sampling technique. Based on the results of hypothesis
testing using the smartPLS application version 3.0, it shows that of the
seven hypotheses proposed, only the price variable has no direct effect on
consumer satisfaction. While other hypotheses have a positive and
significant influence both directly and through intervention variables
Correspondence Email :
[email protected]
ABSTRAK
Penelitian ini merupakan studi deskriptif kuantitatif dengan metode
analisis jalur yang bertujuan untuk mengetahui pengaruh penilaian
Keywords: harga dan produk terhadap keputusan pembelian online dan dampaknya
Price, Product Review,Online Purchase terhadap kepuasan konsumen. Populasi dalam penelitian ini adalah
Decision, Consumer Satisfaction mahasiswa yang tinggal di Kabupaten Maros dan merupakan konsumen
platform belanja online. Jumlah sampel sebanyak 98 orang diperoleh
dengan metode pengambilan sampel menggunakan teknik purposive
sampling. Berdasarkan hasil pengujian hipotesis menggunakan aplikasi
smartPLS versi 3.0, menunjukkan bahwa dari tujuh hipotesis yang
diajukan, hanya variabel harga yang tidak berpengaruh langsung
terhadap kepuasan konsumen. Sementara hipotesis lainnya memiliki
pengaruh positif dan signifikan baik secara langsung maupun melalui
variabel intervensi

INTRODUCTION
The buying and selling culture has changed significantly since the 2000s. This is marked
by changes in people's lifestyles which tend to be more practical than before. These changes can
be found in several people's lifestyles such as food tastes, fashion, and shopping. In the midst of
busy people, especially urban communities, purchases that do not require a physical visit are
also an option that is starting to be looked at. This change in transaction activity contributed to
the growth of e-commerce in Indonesia. This development has even attracted the government's
attention with the publication of the XIV Economic Policy Package on e-commerce (Humas
Kementerian Koordinator Bidang Perekonomian, 2016). This policy package was issued after
seeing a significant increase in terms of e-commerce transactions. In addition, along with the
increase in people's income, changes also occur in the buying and selling culture that leads to
an increase in online purchases. Bank Indonesia (BI) estimates that the total e-commerce
transactions in 2022 will reach Rp 530 trillion (KONTAN.CO.ID, 2021). This figure is certainly
based on the trend of increasing public transactions on online shopping platforms.

Fani, Nur & Hasan 263

 Social Research and Monitoring Sociab, Chamber of Commerce and Industry (KADIN),
Kemkominfo, Accenture published 2015 data regarding the number of internet users in
Indonesia. It was found that 77% of people use the internet in searching for product information
and online shopping (Harahap, 2018). This data is likely to surge amid the development of
today's use of the internet which is not only to find information about news, but also to search
for groceries. There are more than ten widely used online shopping sites, including Lazada,
Tokopedia, Bukalapak, etc.
The migration of people's spending methods to digital-based is a consequence of the
development of technology and information. This includes social changes that can simply be
interpreted as a form of accepted lifestyle, either due to changes in geographic, cultural,
demographic, ideological, or new discoveries in society (Pratama, 2022). The more modern the
civilization of a society, the simpler, faster and easier its life process will be. In addition, the
increasing number of smartphone users also plays an important role in increasing online
shopping transactions. Some online shopping applications can be easily downloaded via mobile
phones. The simplification of this life process will also affect consumer behavior in the process
of selecting the products they want.
The rapid growth of online shopping using online shopping platforms has certainly
increased business competition in this sector. Not a few online shopping platforms are unable
to compete and have decreased transactions. This is triggered by many factors. Consumers each
have their own subjectivity and are increasingly selective in choosing what platform they use to
shop online. Often we will access
More than 2 (two) online shopping sites then compare them before making a purchase
decision. The purchasing decision process consists of five phases, namely problem
identification, information search, evaluation of alternatives, purchase decisions and subsequent
buying behavior (Kotler, 2009). In online purchases, consumers will pay more attention to their
products. Consumers will see how many sales figures for the product, how the assessment of
people who have bought it, whether they are safe to buy the product, whether it is easy to
communicate with the seller or not, and various considerations before making a purchase
decision. In addition, after making a purchase, consumers will evaluate the results of their
decisions. If the evaluation results are good, consumers will feel satisfied and the probability
level to repurchase or recommend the online store will certainly be greater.
The current global situation also plays a major role in the development of online
purchasing transactions, especially with the Covid-19 pandemic situation that threatens the
entire world, causing many changes in people's lifestyles. The COVID-19 pandemic requires
governments around the world to issue policies to carry out activities at home, including
activities from home or work from home. This makes the level of public visits to shopping
centers decrease and makes the level of visits to online shopping platforms increase. A survey
conducted by YouGov on “COVID-19 impact on Indonesian consumer behavior', online
shopping activities increased by 20% and food orders through delivery services by 14% (Barrus,
2020). The COVID-19 pandemic is certainly an opportunity for online stores and online
shopping platforms to further improve service and production.
Online purchasing decisions are very different from offline purchasing decisions.
Consumers will not see directly the product they want to buy. Consumers will only see visually
and descriptively things related to the product. This is a distinctive feature of online shopping
that requires sellers to explain their products in detail and as attractively as possible. This online
purchase decision is also influenced by various factors, such as the assessment of other

Consumer Satisfaction and Online Purchase Decisions,… 264

consumers who have made purchase transactions, security, and return services if the goods
received are defective and do not match the product description.
Online purchasing decisions are based on (Hardiawan, 2013) : a) Efficient search (speed of
time, ease of use, and effort of ease of search), b) Value (competitive price and quality),
Interaction (information, security, waiting time and navigation)
The increasingly massive use of technology and information makes online shopping
decisions more selective. Prospective buyers will easily make a comparison process in which
online store he will decide on a purchase. For this reason, online platforms and online sellers
must synergize and provide the best quality in displaying the products to be sold. The online
platform provides an attractive, fast and easy website. Meanwhile, online sellers or shops that
work with online platforms provide an explanation of the quality of the products they sell on
online shopping platforms.
Price is basically the main criterion for consumers to make a purchase. In the marketing
mix, price is considered as a quality feature of a product. Price is the amount of value, usually
in the form of money, that is needed to add to a combination of goods or services, with the
addition of some goods (Philip Kotler, 2012). Price is also a monetary unit or other measurement,
including other goods or services that are exchanged for ownership or use of an item or service
(Tjiptono, 2012). Price is often very decisive for the purchase decision. Online purchases through
online shopping platforms have an advantage in terms of the price of a product. We often
encounter this mechanism, such as payments with the Cash On Delivery (COD) system, Pay
Later (a payment mechanism in installments), this of course, in addition to providing many
choices for consumers, it can also make consumers adjust to their financial conditions. This
system is a form of adjustment of the online shopping platform to the lifestyle and financial
condition of the community.
Experience, evaluation, and consumer opinion are instruments of product assessment (S.
Bae., 2011). This can influence potential buyers to recognize the product as trusted and reliable
when making purchasing decisions. Product reviews are always available in every e-commerce
for consumers who have purchased certain products. Product reviews on the marketplace can
include consumer ratings for purchased products. Consumers can also upload product photos
and give stars or ratings in the product rating column.
Product review is an assessment along with consumer reviews of certain products after
finalizing the order. Rating is used as a measure of their shopping experience in online stores.
Product assessment is very useful because it provides a reference for potential buyers before
making a purchase decision. This is because in online sales, potential customers cannot touch,
experiment, or see in person. Consumers usually make reviews from manufacturers as a
reference. This is so that mistakes in buying products or wrong orders can be avoided, so the
simple way is to start by doing research on products through product assessments (Debora,
2016).
Several studies have been carried out regarding online shopping. Lifestyle, ease of use,
and quality of information have a positive impact on purchasing decisions, but trust does not
(Wardoyo & Andini, 2017). Meanwhile (Nadia Ramadhanty & Fikriyah, 2020) explains that the
survey results show that lifestyle, price, time and cost efficiency, activity groups, quality of
information, and promotions influence the decision to buy hijab in online stores, trust and
convenience in buying scarves online. does not affect shopping decisions in online stores. Both
studies have some similarities and differences in results. In contrast, ease of use shows different
results.

Fani, Nur & Hasan 265

 The hypotheses we propose in this study are:
H1: Price has a positive effect on purchasing decisions
H2: Product appraisal has a positive effect on purchasing decisions
H3: Price has a positive effect on consumer satisfaction
H4: Product assessment has a positive effect on consumer satisfaction
H5: Purchase decisions have a positive effect on consumer satisfaction
H6: Price has a positive effect on consumer satisfaction through decisions purchase
H7: Product assessment has a positive effect on customer satisfaction through buying
decision

RESEARCH METHOD
This research is a quantitative research. namely research methods based on the philosophy
of positivism, which are used to examine certain populations or samples, data acquisition using
research tools, quantitative or statistical data analysis aimed at testing hypotheses that have been
proposed (Sugiyono, 2017).
This research was conducted in Maros Regency. The population in this study are students
who live in Maros Regency with an unknown number. The sample was selected using a non-
probability method and purposive sampling technique with the criteria of students who
understand how to buy online, have made online purchase transactions on online buying and
selling platforms at least five times in 2022. Closed online questionnaires have been prepared on
Google Form and links are distributed to all social media applications, such as Whatsapp, and
Instagram, Facebook. The survey was conducted from January to April 2022. In total, 156 people
responded to the survey. After eliminating data from respondents who did not match the set
criteria and eliminating inconsistencies, 98 eligible responses were used in this analysis. Data
was collected from questionnaires distributed through social media, WhatsApp, and Google
Form. The population of this research is students who live in Maros Regency who have made
online purchase transactions on online purchasing platforms. The path analysis technique in this
study is to use the path analysis model as follows:

Price (X1)

Purchase Consumer
Decisions (Y) Satisfaction (Z)

Product Review
(X2)

Figure 1. Research Model

RESULTS AND DISCUSSION

Respondent Description

Consumer Satisfaction and Online Purchase Decisions,… 266

 Students who live in Maros Regency who shop online on an online shopping platform are
the sample. Description of the sample characteristics of 98 people. This description of the
respondent's identity is intended to provide an explanation of the respondent's own condition.

Figure 2. Gender

Gender
Figure 2 explains that the female gender has
more numbers than the male gender, namely
68% or 67 respondents for women compared to
31 32% or 31 respondents for men. This condition
Male
shows that college students shop online more
Female
67 than college students.

Source: Processed Data 2022

Figure 3. Age of Respondents

Age

Based on Figure 3, it explains that the

18 23
percentage of online shopping made by
students for the age range of 17 to 20 years is 23
people, ages 21 to 24 years is 57 people, and the
age range is 25 years and over as many as 18
people. This condition shows that students aged
57
21-24 are more likely to shop online.

17-20 21-24 25 above

Source: Processed Data 2022

Data Statistical Analysis

The first step in analyzing using Partial Least Square is testing the outer model. The
method used for analysis with the outer model is the PLS algorithm. In this study, the outer
model was tested using convergent validity, discriminant validity, reliability, and formative
construct tests. The PLS application used is the SmartPLS version 3.0 application.

Convergent Validity Test

Some of the criteria used to measure convergent validity are looking at the values of the
outer loading and average extracted variance. The first stage in testing convergent validity is
outer loading. The indicator is said to be valid if the value of the outer loading indicator is more
than 0.7 (Ghozali, Imam, 2015).

Fani, Nur & Hasan 267

 Figure 4: Output Program Smart PLS, 2022

Table 1. Outer Loading

Price Product Review Purchasing Decision Customer Satisfaction
(P) (PS) (PD) (CC)
Pr1 0,717
Pr2 0,895
Pr3 0,849
PR1 0,773
PR2 0,897
PR3 0,907
PD1 0,881
PD2 0,873
PD3 0,880
PD4 0,886
PD5 0,779
CC1 0,811
CC2 0,787
CC3 0,833
CC4 0,862
CC5 0,812
Source: Processed Data 2022

Consumer Satisfaction and Online Purchase Decisions,… 268

 Seeing the value of outer loading in table 1, all indicators have a value of > 0.7, this informs
that all indicators are valid to measure the construct.

Discrimination Validity
The model for measuring validity and reliability, the coefficient of model determination
and the path coefficient for the equation model, can be seen in the following table:

Table 2. Discriminat Validity

Purchase Consumer
Price Product Review
Desicion Satisfaction
(H) (PP)
(KP) (KK)
X1.1 0,717 0,551 0,577 0,369
X1.2 0,895 0,625 0,762 0,519
X1.3 0,849 0,734 0,644 0,600
X2.1 0,607 0,773 0,506 0,682
X2.2 0,665 0,897 0,762 0,556
X2.3 0,728 0,907 0,704 0,518
Y1 0,631 0,635 0,881 0,628
Y2 0,732 0,662 0,873 0,523
Y3 0,711 0,656 0,870 0,549
Y4 0,778 0,780 0,886 0,629
Y5 0,597 0,549 0,779 0,633
Z1 0,495 0,549 0,421 0,811
Z2 0,561 0,653 0,682 0,787
Z3 0,494 0,546 0,462 0,833
Z4 0,469 0,510 0,579 0,862
Z5 0,469 0,495 0,630 0,812
Source: Processed Data 2022

The results of the cross loading calculation in table 2, it is known that the loading value
of each indicator on the construct of the cross loading value. This means that the construct or
latent variable has better discriminant validity than the indicators in the other blocks.
Another method to measure discriminant validity is to measure the value of the square
root of average variance extracted (AVE). The recommended value for a good model is 0.5 or
higher. The next test is the composite reliability of the indicator block that measures the
construct. If the composite reliability value is above 0.60, then the construct is said to be reliable.
In addition, it can also be seen by looking at the reliability of the construct or latent variable
which is measured by looking at the Cronbach's alpha value of the indicator block that measures
the construct. If the value of Cronbach's alpha is above 0.7, then the construct is said to be
reliable. To illustrate the value of the construct results of each variable, namely price, product
assessment, purchase decisions and consumer satisfaction with each variable and indicator, the
loading value table for the research variable constructs resulting from running the Smart PLS
program is shown in the following table:

Fani, Nur & Hasan 269

Table 3. Construct Reliability, Validity and Average Variance Extracted (AVE)
Composite Average Variance
Variabel Cronbach’s alpha
Reliability Extracted (AVE)
Price 0,760 0,863 0,678
Product Review 0,823 0,895 0,742
Purchase Decision 0,910 0,933 0,737
Consumer
0,880 0,912 0,675
Satisfaction
Source: Processed Data 2022

Based on table 3, it shows that the Average Variance Extracted (AVE) of each variable,
namely price and product assessment, purchasing decisions and consumer satisfaction has a
construct > 0.50 meaning all constructs are reliable. Thus it can be stated that each variable has
high discriminant validity.
Meanwhile, it can be seen in table 4.4 above that the composite reliability value of each
variable shows the construct value > 0.60. This result informs that each variable has met the
composite reliability standard so that it can be concluded that all variables have a high level of
reliability. Furthermore, in the table Cronbach's alpha each variable shows a construct value >
0.70 so this result shows that the variables in the study have met the requirements of Cronbach's
alpha value, which means that all variables have high reliability values. In conclusion, the
indicators in this study have high discriminant validity in shaping their respective variables.

Inner Model Analysis

Coefficient of Determination (R2)
Based on the results of data processing carried out with the help of the SmartPLS 3.0
program, the R Square value is as follows:

Table 4. Koefisien Determinasi (R2)

R Square
Purchase Decision 0,702
Consumer Satisfaction 0,529
Sumber : Output Smart PLS 3.0, 2022

Based on the value of R Square, the value of the purchasing decision variable is 0.702. This
value indicates that the percentage value of purchasing decisions is 70.2%, which means that the
variable price and product assessment has an effect of 70.2% on purchasing decisions. While the
value of R Square for the variable consumer satisfaction is 0.529. These results explain that the
percentage value of customer satisfaction is 52.9%. Which means that the variables of price,
product assessment and purchasing decisions have an effect of 52.9% on consumer satisfaction.

Hypothesis test
After assessing the inner model, the next step is to evaluate the relationship between latent
constructs based on the hypothesis that has been proposed in this study. Hypothesis testing in
this study was carried out by comparing the T-Statistics value and the P-Values value. The
hypothesis is accepted if T-Statistics > 1.97 and P-Values < 0.05. The following are the results of
the direct effect Path Coefficients:

Consumer Satisfaction and Online Purchase Decisions,… 270

Table 5. Path Coefficients (Direct Effect):
Parameter P
(Direct Effect) TStatistics Result
Coefficient Values
H1 Price(X1)→Purchase
0,527 5,346 0,000 Accepted
Decision(Y)
H2 Product Review(X2)
0,360 3,640 0,000 Accepted
→Purchase Decision(Y)
H3 Price(X1) → Consumer
-0,007 0,049 0,961 Rejected
Satisfaction(Z)
H4 Product Rating(X2) →
Consumer 0,360 2,504 0,013 Accepted
Satisfaction(Z)
H5 Purchase Decision(Y) →
Consumer 0,419 3,037 0,003 Accepted
Satisfaction(Z)
Sumber : Output Smart PLS 3.0, 2022

Based on the table of direct influence, of the five hypotheses that have a direct effect there
is 1 (one) hypothesis that is rejected, namely H3 because the value of T Statistics < 1.96 and P-
Values > 0.05 while the other 4 (four) hypotheses are accepted because the T-value is Statistics
> 1.96 P-Values < 0.05. The following is an explanation of the hypothesis:

H1: Price has a positive effect on purchasing decisions

The t-statistic value of the price on purchasing decisions is 5.346 with a parameter
coefficient value of 0.537. the t-statistic value is more than the t-table value (5.346 > 1.96) with a
significance value of 0.000. Thus, the price hypothesis has a positive effect on purchasing
decisions is accepted. This informs that consumer purchasing decisions for online goods are
influenced quite high by the price factor.

H2: Product appraisal has a positive effect on purchasing decisions

The t-statistic value of product assessment on purchasing decisions is 3.640 with a
parameter coefficient of 0.360. This t-statistic value is more than the t-table value (3,640 > 1.96)
with a significance value of 0.000. Thus, the product appraisal hypothesis has a positive effect
on the purchase decision is accepted. This informs that consumer purchasing decisions for
online goods are influenced quite high by product assessment factors. Information about
product quality, speed of delivery, conformity with consumer expectations can be seen from the
product assessment and used as a reference before making the purchase process

H3: Price has a positive effect on consumer satisfaction

The t-statistic value of product assessment on purchasing decisions is 0.049 with a
parameter coefficient value of -0.007. this t-statistic value is more than the t-table value (0.049
<1.96) with a significance value of 0.961. Thus, the price hypothesis has a positive effect on
consumer satisfaction is rejected. This informs that consumer satisfaction with online goods is

Fani, Nur & Hasan 271

not influenced by the price factor. Often consumers feel there is a discrepancy in the prices
offered by online stores with consumer expectations for these products.

H4: Product appraisal has a positive effect on consumer satisfaction

The t-statistic value of product assessment on consumer satisfaction is 2,504 with a
parameter coefficient value of 0.360. This t-statistic value is more than the t-table value (2,504 >
1.96) with a significance value of 0.013. Thus, the product assessment hypothesis has a positive
effect on consumer satisfaction is accepted. This informs that consumer satisfaction with online
goods is positively influenced by product assessment factors. Consumers after making a
purchase will feel satisfied after they previously saw the assessment of the product.

H5: purchasing decisions have a positive effect on consumer satisfaction

The t-statistic value of purchasing decisions on consumer satisfaction is 3,037 with a
parameter coefficient of 0.419. This t-statistic value is more than the t-table value (3.037 > 1.96)
with a significance value of 0.003. Thus, the purchase decision hypothesis has a positive effect
on consumer satisfaction is accepted. This informs that consumer satisfaction with online goods
is positively influenced by purchasing decision factors.

Tabel 6. Path coefficients

H6 Price(X1)→Purchase
Decision(Y) → Consumer 0,221 2,469 0,015 Accepted
Satisfaction(Z)
H7 Product
Review(X2)→Purchase
0,151 2,469 0,014 Accepted
Decision(Y) → Consumer
Satisfaction(Z)
Sumber : Output Smart PLS 3.0, 2022

Based on the table of indirect influence, it shows that the two hypotheses that have an
indirect effect are accepted because the T-Statistics value > 1.96 P-Values < 0.05. The following
is an explanation of the hypothesis:

H6: Price affects consumer satisfaction through decisions purchase

The t-statistic value of price on consumer satisfaction through purchasing decisions is
2,469 with a parameter coefficient value of 0.221. This t-statistic value is more than the t-table
value (2,469 > 1.96) with a significance value of 0.015. Thus, the Price Hypothesis affects
consumer satisfaction through the purchase decision is accepted. This informs that consumer
satisfaction is positively influenced by price through purchasing decisions.

H7: Product assessment has an effect on consumer satisfaction through buying decision
The t-statistic value of product assessment on consumer satisfaction through purchasing
decisions is 2,469 with a parameter coefficient of 0.151. The t-statistic value is more than the t-
table value (2,469 > 1.96) with a significance value of 0.014. Thus, the product assessment
hypothesis affects consumer satisfaction through the purchase decision is accepted. This informs
that consumer satisfaction is positively influenced by product evaluation through purchasing
decisions.

Consumer Satisfaction and Online Purchase Decisions,… 272

CONCLUSION
Based on the results of hypothesis testing, it can be concluded several things, namely: 1)
Price has a positive effect on purchasing decisions, but has a negative effect on consumer
satisfaction, 2) Product assessment has a positive effect on purchasing decisions and consumer
satisfaction, 3) Price through purchasing decisions has a positive effect on consumer satisfaction
and 4) Product assessment through purchasing decisions has a positive effect on consumer
satisfaction.

REFERENCE
Barrus, H. (2020). Tips Startup Hadapi Krisis Covid-19. Industry.Co.Id.
Debora, Y. (2016). Berapa Besar Pengaruh Ulasan Pembeli Saat Berbelanja Online? Tirto.Id.
https://tirto.id/berapa-besar-pengaruh-ulasan-pembeli-saat-berbelanja-online-b7Gm
Ghozali, Imam, H. L. (2015). Konsep, Teknik, Aplikasi Menggunakan Smart PLS 3.0 Untuk Penelitian
Empiris. BP Undip.
Ghozali, I. (2016). Aplikasi Analisis Multivariate dengan Program SPSS (8th ed.). Universitas
Diponegoro.
Harahap, D. A. (2018). Perilaku Belanja Online Di Indonesia: Studi Kasus. JRMSI - Jurnal Riset
Manajemen Sains Indonesia, 9(2), 193–213. https://doi.org/10.21009/jrmsi.009.2.02
Hardiawan, A. C. (2013). Pengaruh Kepercayaan, Kemudahan, danKualitas Informasi Terhadap
Keputusan Pembelian Secara Online. Universitas Diponegoro.
Humas Kementerian Koordinator Bidang Perekonomian. (2016). Paket Kebijakan Ekonomi XIV:
Peta Jalan E-Commerce!
KONTAN.CO.ID. (2021). Transaksi e-commerce meningkat pesat, infrastruktur logistik harus siap
mengikuti.
Kotler, P. (2009). Manajemen Pemasaran (1st ed.). PT. Prehallindo.
Nadia Ramadhanty, R., & Fikriyah, K. (2020). DETERMINAN KEPUTUSAN PEMBELIAN
JILBAB MELALUI ONLINE SHOP. 3, 40–55.
Philip Kotler, G. A. (2012). Principles of Marketing (14th ed.). Pearson.
Pratama, C. D. (2022). 4 Teori Perubahan Sosial dan Contohnya. Kompas.
S. Bae., & T. L. (2011). Product type and consumers’ perception of onlineconsumerreviews.
Electronic Markets, 21(4), 255–266.
Sugiyono. (2017). Metode Penelitian Kuantitatif, Kualitatif, dan R&D. Alfabeta.
Tjiptono, F. (2012). Strategi Pemasaran (Kedua). Andy.
Wardoyo, W., & Andini, I. (2017). Faktor-Faktor Yang Berpengaruh Terhadap Keputusan
Pembelian Secara Online Pada Mahasiswa Universitas Gunadarma. Jurnal Manajemen
Dayasaing, 19(1), 12. https://doi.org/10.23917/dayasaing.v19i1.5101

Fani, Nur & Hasan 273