p–ISSN: 2723 - 6609 e-ISSN: 2745-5254
Vol. 5, No. 7 July 2024 http://jist.publikasiindonesia.id/
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3437
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-
Satisfaction as Mediation
Gilang Pratama Hafidz
1
*, Muhamad Khaidir Ali
2
Universitas Esa Unggul, Indonesia
Email:
1*
2
*Correspondence
ABSTRACT
Keywords: customer e-
loyalty, customer e-
satisfaction, customer
perceived value, e-service
quality (e-sq).
In the e-commerce sector, attaining customer loyalty is a
formidable task that necessitates particular focus from firms
to outdo their contenders. This study sought to
comprehensively comprehend the impact of e-service
quality (e-SQ), encompassing seven utilitarian and two
hedonic aspects, on customer e-loyalty in the marketplace.
Both e-satisfaction and perceived value were examined as
moderators and mediators, respectively, to provide a more
empirical perspective. The study distributed online
questionnaires to 265 participants between the ages of 18
and 60 who made at least two purchases from the top five
Indonesian marketplaces within the last three months and
reside in Greater Jakarta (Jabodetabek). The collected data
underwent analysis using Structural Equation Modeling -
Partial Least Square (SEM-PLS). The findings indicated that
e-S) has the potential to enhance customer e-satisfaction.
Similarly, e-SQ can enhance customer loyalty both directly
and indirectly by increasing e-satisfaction. Additionally,
customer satisfaction with the electronic platform has a
positive impact on e-loyalty. Notably, perceived value does
not play a moderating role in the influence of both factors.
This study indicates that businesses operating in the e-
commerce sector could enhance e-SQ factors like
enjoyment, system availability, fulfilment, and efficiency to
maximize e-satisfaction and e-loyalty.
Introduction
The rapid progress of technology in Indonesia has made the competition in the e-
commerce industry also more intense and competitive. This can be observed in
Indonesia's 2017, where approximately 30 million individuals shopped online out of a
total population of 260 million as well as a plethora of electronic marketplaces, including
Tokopedia, Shopee, Bukalapak, Lazada, Blibli, etc., emerged to facilitate trade
transactions without direct interaction (Das, Tamhane, Vatterott, Wibowo, & Wintels,
2018).
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3438
Based on Iprice.co.id (2022), Shopee, Tokopedia, Bukalapak, Lazada and Blibli
became the five largest marketplace players in Indonesia in the fourth quarter of 2020.
Surprisingly, Indonesian online shoppers are not very loyal to one specific marketplace,
as evidenced by the significant changes in the number of visits by loyal customers. For
instance, during Q4 of 2020, Shopee emerged as the marketplace with the largest number
of yearly visitors, amounting to 129,320,800, with Tokopedia following closely behind
with 114,655,600 visitors. In Q4 2021, Shopee dropped to second place with an annual
visitor count of 138,776,700, while Tokopedia claimed the top spot with 157,443,300
visitors. This indicates that maintaining customer loyalty is crucial in the e-commerce
industry, both online and in traditional markets (Kim, Wang, & Roh, 2021).
In the e-commerce sector, developing customer loyalty online poses a difficult task
that demands that online businesspeople pay particular attention to move ahead of their
rivals (Ghali, 2021). Companies only obtain profits when customers have made an
average purchase of more than four times. However, how the company produces customer
e-loyalty itself remains a problematic and complex phenomenon. This is due to the
difficulty of switching to similar competitors, which is usually one click away.
Companies also need a lengthy process to build customer e-loyalty. (Kotler, Armstrong,
Harris, & He, 2020) stated that e-satisfaction is very crucial, and it is key for companies
to build profitable and sustainable relationships. Satisfied customers not only make repeat
purchases but also spread positive word-of-mouth and cultivate e-loyalty by buying
similar items. Moreover, for both established and emerging e-commerce industries, e-
service quality (e-SQ) plays a vital role in e-loyalty formation (Eryiğit & Fan, 2021).
However, online shoppers report concerns, anxiety, and risk perception, which puts
pressure on companies to enhance e-service quality (e-SQ) by improving its functionality
(utilitarian) and making it a more satisfying user experience (hedonic) (Shatnawi, 2019).
Therefore, business people must understand how consumers interpret e-SQ, and the
consequences companies receive (Parasuraman, Zeithaml, & Malhotra, 2005).
Numerous researches have examined the connections between e-SQ, customer e-
satisfaction, e-loyalty, and perceived value. This research showed that e-SQ is related to
customer e-satisfaction (Rodríguez, Villarreal, Valiño, & Blozis, 2020) and influenced by
their e-loyalty. According to preliminary research by (Brusch, Schwarz, & Schmitt,
2019), customer e-satisfaction relates to e-loyalty. It is also an e-SQ mediation against e-
loyalty. Customer perceived value strengthens their e-satisfaction and e-loyalty.
However, there are inconsistencies regarding the analysts’ views on the relationship
between both factors. Some preliminary research stated that there is no direct correlation
between e-SQ and customer e-loyalty, as opposed to several others. Furthermore, the
researchers who examined the factors that cause customer e-loyalty and customer e-
satisfaction usually concentrated only on utilitarian characteristics while ignoring hedonic
characteristics. The purchase decision process involves both rational and emotional
factors (Shatnawi, 2019). As a result, researchers have criticized the absence of hedonic
constructs from previous studies on e-service quality.
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3439
The most crucial act, which frequently presents obstacles for a business, is
ascertaining the value of its services to customers (Parasuraman et al., 2005). According
to preliminary studies, consumers' perceptions of value vary based on individual tastes,
level of service knowledge, purchasing power, and capability. Therefore, companies must
be able to understand how perceived value encourages buyers to feel the benefits that are
offered and accepted by them (Amalia & Nurlinda, 2022). Additionally, over the past few
years, the strategies used to determine consumer perceived value have been updated,
making it fascinating to continue researching (El-Adly & Eid, 2016). Moreover, the role
of customer-perceived value moderation in the relationship between e-satisfaction and e-
loyalty is added to this research model. This analysis is also centred on the Indonesian
marketplace, which makes this research unique.
This research aims to enhance empirical understanding of the impact of e-SQ while
accounting for relevant knowledge gaps. Our study investigates seven utilitarian and two
hedonic dimensions of customer e-loyalty in the marketplace. To better comprehend their
effects, e-satisfaction and perceived values will be evaluated as mediating and moderating
factors, respectively. This gives stakeholders new insights needed to boost
competitiveness, especially through aspects of e-SQ.
E-Commerce
The rapid expansion of technology, particularly the internet, has transformed the
process of buying, selling, and marketing products in present times. E-commerce,
according to DeLone & McLean (2004), denotes the use of the Internet as a medium for
engaging buyers and sellers in conducting and processing business transactions and
exchanging products or services for money. Defined e-commerce as the use of
information and communication technology for distributing products and services to
consumers through supplier organizations and conducting financial and non-financial
transactions. This definition includes the buying and selling of products or services via
the Internet along with the entire process (Nisar & Prabhakar, 2017). This is according to
Laudon & Laudon's (2018) research, which defines e-commerce as the utilization of the
internet, networks, and other digital information to exchange goods and services. E-
commerce consists of multiple categories, specifically business-to-consumer (B2C),
business-to-business (B2B), and consumer-to-consumer (C2C) transactions conducted by
companies, individual buyers, businesses, and fellow consumers, respectively. Therefore,
based on the above explanations, this research defined e-commerce as a comprehensive
transaction process used by organizations to sell products or services to consumers
through information and communication technology.
Research Methods
Research Design
The causality research, which determines the relationship between two or more
variables, was used to design this research (Rahma & Syah, 2023). It comprises
exogenous, mediator, moderator, and endogenous variables in the form of e-SQ, customer
e-satisfaction, perceived value, and e-loyalty, respectively.
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3440
Variable Measurement
In recent years, the proper specification of the model has become a significant
concern in e-SQ research, with the belief that e-SQ may be more effectively
conceptualised in a formative as opposed to reflective manner. Unlike reflective
measurement models, formative measurement models establish the causal flow from
indicators to latent constructs.
According to (Theodosiou, Katsikea, Samiee, & Makri, 2019), there are
fundamental factors to consider here. Firstly, it is logical to anticipate that every e-SQ
dimension causes the overall e-SQ, not the other way around. For instance, users'
appraisals of website quality concerning contact, privacy or system availability aid in
forming perceptions of overall e-SQ. Secondly, e-SQ dimensions cannot be switched. For
instance, privacy encompasses particular features within the construct's conceptual field
that cannot be substituted by other facets like system accessibility. Moreover, the
divergent e-SQ gauges do not necessarily fluctuate in correlation with other associated
variables. Lastly, no data indicate that different e-SQ facets have identical antecedents
and outcomes. Thus, this indicates that the conceptualization of e-SQ should utilize
reflective indicators for first-order constructs and formative indicators for second-order
constructs. This suggests that Type II dimensions, reflective-formative, may present a
more suitable approach. Ultimately, this allows the formative model to provide a more
conceptually deeper understanding of e-SQ in the e-commerce industry.
The measurement of e-service quality (e-SQ) variables is based on the research of
Parasuraman et al. (2005) and is utilitarian consisting of seven dimensions: efficiency (8
statements), fulfilment (7 statements), privacy (4 statements) and system availability (4
statements), responsiveness (5 statements), compensation (3 statements) and contact (3
statements) combined with the research of (Theodosiou et al., 2019) He-donistic consists
of two dimensions: enjoyment (4 statements) and virtual emotion (3 statements) which
total 40 statements operationalized as reflective – formative type. Furthermore, the related
size of the customer variable perceived value, e-satisfaction, and e-loyalty adopted from
Chang et al. (2009) consists of 3, 4, and 6 statements operationalized as reflective,
respectively. Therefore, the total measurements used in this research amounted to 53
statements.
Population and Sample
The population of this research is all Indonesian marketplaces (Sho-
pee/Tokopedia/Bukalapak/Lazada/Blibli) infinite number of users, while the sample used
a non-probability sampling method. In this study, we used a purposive sampling
technique. The criteria for respondent selection were those within the age range of 18 to
60 who have made purchase activities at least two times in the last three months and
domiciled in the Greater Jakarta area. The number of samples was determined based on
(Sarstedt, Hair Jr, Cheah, Becker, & Ringle, 2019) the minimum sample needed, which
is 5 times the statements. Therefore, this research requires 265 respondents (53 statements
x 5), with a proportion of 20% from each marketplace.
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3441
Data collection was implemented using a non-probability method in the form of
judgment or purposive sampling and predetermined according to criteria and quality. The
data collection utilizes Google Form as a questionnaire tool distributed twice online with
a Likert scale of 1 – 7, where 1, 2, 3, 4, 5, 6, and 7 de-note strongly disagree, disagree,
disagree enough, neutral, agree, and strongly agree, respectively. As there are more
options on it, there is a higher chance that the response will be related to the respondent's
experience (Joshi et al., 2015). Therefore, this research used primary data derived from
the results of the distributing questionnaires after processing.
Data Analysis Methods
This research first conducted a pretest test on 30 samples with validity testing using
statistical software. It aims to determine the extent of the instrument’s accuracy used in
measuring the variables. Kaiser Meyer Olkin (KMO) and Measure of Sampling Adequacy
(MSA) were used to test the questionnaire validity with values >0.50 for each variable
(Napitupulu, Kadar, & Jati, 2017). The MSA value must be >0.50 for each indicator,
hence, when there is a statement indicator not capable of exceeding that value, and then
the indicator must be omitted in further analysis. In addition, reliability testing was also
carried out to know the level of consistency and trust of measuring instruments using
Cronbach's Alpha method with the general pro-vision of an α value of 0.6 – 0.7 (Ursachi
et al., 2015).
Structural Equation Modeling (SEM), specifically the Partial Least Squares (PLS-
SEM) technique, was utilized as the statistical software in this study. PLS-SEM is a
variance-based method for estimating structural equations that includes measurement and
structural components to maximize the described variance of the endogenous latent
variables. In essence, PLS-SEM is appropriate for research studies aimed at developing
theories. Therefore, the Disjoint Two-stage Approach is suitable when there are higher-
order models known as Hierarchical Component Models (HCMs) Type II: Reflective-
Formative (Sarstedt et al., 2019).
A reflective outer model test stage is carried out, determining the loading factor,
composite reliability, convergent validity, and discriminant validity. Meanwhile, the
formative measurement models are evaluated by analyzing the collinearity, significance,
and relevance of outer weight. This is followed by determining the inner model testing
using one-tailed, collinearity, R-square, path coefficients, and indirect effects. R-Square
guidelines used are 0.75, 0.50, and 0.25, indicating strong, moderate, and weak. In
addition, a bootstrapping procedure was carried out in which the entire original sample
was used to resample at a t-value of 1.65 and a confidence level of 95% for significance
testing to determine the influence between variables.
Results and Discussion
A pretest was conducted to evaluate the validity and reliability of the study. Validity
testing requires KMO and MSA values above 0.50. Our calculations revealed that all 53
statement indicators in this study met these requirements, with both the KMO test (0.580
- 0.907) and the MSA test (0.549 - 0.933) satisfying these criteria. Moreover, the
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3442
assessment of reliability can be measured through Cronbach's alpha with a range of 0.6
to 0.7 considered acceptable (Ursachi et al., 2015). After processing the data, it was
discovered that all indicators of the questionnaire statement exceeded the value limit set
beforehand (0.751-0.964), indicating its reliability and usefulness for future analysis.
Descriptive analysis of respondent demographic information reveals that the sample
of 265 respondents is comprised of 81 men (30.6%) and 184 women (69.4%). The sample
represents each marketplace (Shopee, Tokopedia, Bukalapak, Lazada, and Blibli), with
53 respondents (20%) from each. The respondents are domiciled in Jakarta (25.3%),
Bogor (22.3%), Depok (15.1%), Tangerang (14.7%), and Bekasi (22.6%). The majority
of respondents are aged between 18 and 21 years, which accounts for 120 respondents
(45%) who are believed to be the core marketplace market in the future, followed by 22-
39 years old as many as 101 respondents (38.1%) who are believed to be active users of
the current marketplace, and 40-60 years old as many as 44 respondents (16.6%) who are
believed to be the group that spends the most money in the marketplace. Unexpectedly,
119 respondents (44.9%) have been using the marketplace for over three years, with the
highest frequency of shopping being 8-12 times in the last three months, as reported by
88 respondents (33.2%). Additionally, accessories & fashion and care & beauty were
equally popular product categories, with 68 respondents each (25.7%), while the majority
of respondents spent between Rp5,000,0000 to Rp9,999,999 per month, amounting to 112
respondents (42.3%).
Table 1
Descriptive Analysis of Respondents
Items
N
Categories
Frequency
Percentage
Respondent Demographics
Gender
265
Male
81
30.6%
Female
184
69.4%
Age
18-21 years old
120
45.3%
22-39 years old
101
38.1%
40-60 years old
44
16.6%
Domicil
e
Jakarta
67
25.3%
Bogor
59
22.3%
Depok
40
15.1%
Tangerang
39
14.7%
Bekasi
60
22.6%
Monthly
Expense
s
< IDR 2,999,9999
30
11.3%
IDR 3,000,000 - IDR
4,999,999
101
38.1%
IDR 5,000,000 – IDR
9,999,999
112
42.3%
> IDR 10,000,000
22
8.3%
Respondent Usage
Marketp
laces
Used
265
Shopee
53
20%
Tokopedia
53
20%
Bukalapak
53
20%
Lazada
53
20%
Blibli
53
20%
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3443
Marketp
lace
Usage
Period
<1 Year
55
20.8%
1-2 Years
91
34.3%
≥3 Years
119
44.9%
Online
Shoppin
g
Frequen
cy
2 times
55
20.8%
3-7 Times
74
27.9%
8-12 Times
88
33.2%
>12 Times
48
18.1%
Product
Categor
y
Accessories and Fashion
68
25.7%
Electronic
41
15.5%
Food and Drink
23
8.7%
Household Goods
65
24.5%
Care and Beauty
68
25.7%
Furthermore, a cross-tabulation analysis was conducted on 265 respondents'
demographics and usage patterns to identify specific characteristics. The data reveals that
the majority of marketplace users are in the 18-21 age range, with 12.5% selecting
Tokopedia, 11.3% opting for Blibli and 8.3% choosing Shopee. Bukalapak's users consist
predominantly of those between 22-39 years old (14.3%), while Laza-da's user base is
evenly distributed between the 18-21 and 22-39 age groups at 7.5%. Secondly, regarding
product categories by gender, it can be noted that the care & beauty category (20.8%) and
household supplies (20.4%) are popular amongst female marketplace users, whereas male
marketplace users tend to shop online for products in the accessories & fashion category
(10.9%) and electronics (7.2%). Finally, according to the length of marketplace usage,
the shopping frequency indicates that the majority of marketplace users are individuals
who have been utilising the marketplace for over 3 years, purchasing online between 8-
12 times within the past 3 months (18.5%). This is followed by individuals who have
been using the marketplace for 1-2 years, making online purchases 8-12 times within the
past 3 months (12.5%).
Assessing Reflective Measurement Models
(Sarstedt et al., 2019) State that the initial step in evaluating the reflective
measurement model is to assess the loading of the indicators. This involves analyzing the
loading factor values that exceed 0.70. Results from data analysis in this study indicated
that EFF1, EFF2, EFF3, EFF6, FUL4, FUL5, and SYS4 indicators do not meet the
standards and must be eliminated. Conversely, the remaining indicators have outer
loading values (0.719-0.886) greater than 0.70.
The second step utilized internal consistency reliability to assess the consistency of
outcomes among indicators, gauging the latent construct through composite reliability
and Cronbach's alpha value of >0.6 - 0.7 (Hair & Alamer, 2022). The data processing
results showed that all constructs had a composite reliability value from 0.796 to 0.932
and a Cronbach's alpha value from 0.625 - 0.912, satisfying the specified requirements.
The third step, convergent validity, measures if indicators are positively correlated
with other indicators of the same construct. It is evaluated through Average Variance
Extracted (AVE) analysis at values exceeding 0.50. The data processing analysis
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3444
indicates that the reflective indicators had AVE values ranging from 0.560 to 0.778,
which is above the threshold of 0.50. As a result, the convergent validity of the reflective
indicators is valid.
Table 2
Loading Factor, AVE, Composite Reliability, and Cronbach's alpha
No.
Items
Convergent Validity
Internal Consistency Reliability
Loading
Factor
AVE
Composite
Reliability
Standardized
Cronbanch's α
EFF4
0.746
0.612
0.863
0.788
EFF5
0.830
EFF7
0.813
EFF8
0.736
FUL1
0.799
0.605
0.885
0.837
FUL2
0.779
FUL3
0.747
FUL6
0.787
FUL7
0.777
SYS1
0.761
0.609
0.824
0.681
SYS2
0.786
SYS3
0.794
PRI1
0.839
0.726
0.888
0.811
PRI2
0.879
PRI3
0.836
RES1
0.833
0.652
0.903
0.866
RES2
0.838
RES3
0.856
RES4
0.734
RES5
0.770
COM1
0.880
0.778
0.913
0.857
COM2
0.879
COM3
0.886
CON1
0.848
0.713
0.882
0.799
CON2
0.847
CON3
0.839
ENJ1
0.765
0.560
0.836
0.741
ENJ2
0.719
ENJ3
0.726
ENJ4
0.782
VE1
0.750
0.566
0.796
0.625
VE2
0.739
VE3
0.766
e-SAT1
0.824
0.714
0.882
0.799
e-SAT2
0.841
e-SAT3
0.868
e-LOY1
0.837
0.695
0.932
0.912
e-LOY2
0.824
e-LOY3
0.858
e-LOY4
0.871
e-LOY5
0.840
e-LOY6
0.769
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3445
No.
Items
Convergent Validity
Internal Consistency Reliability
Loading
Factor
AVE
Composite
Reliability
Standardized
Cronbanch's α
CPV1
0.774
0.644
0.879
0.817
CPV2
0.808
CPV3
0.774
CPV4
0.853
Finally, a construct is considered to have discriminant validity if it captures unique
phenomena not identified by other constructs in the model. It indicates the extent to which
a concept is distinct from others based on empirical standards. Henseler et al. (2015)
describe three methods for validating the discriminant, with the first based on the Fornell-
Larcker Criterion that determines the reference square root of each AVE construct, which
should be greater than its highest correlation with other constructs. It is crucial to use
these methods to ensure the validity of the discriminant. The second method is Cross
Loading, which sets a reference value of >0.70.
Table 3
Fornell-Larcker Criterion
CO
M
CO
N
LO
Y
SAT
CP
V
EFF
ENJO
Y
FUL
L
PRI
RES
SYS
VE
CO
M
0.88
2
CO
N
0.63
2
0.84
5
LOY
0.48
3
0.57
6
0.83
4
SAT
0.53
5
0.64
2
0.76
0
0.84
5
CPV
0.48
6
0.63
3
0.75
8
0.74
0
0.80
3
EFF
0.58
6
0.59
4
0.69
7
0.74
0
0.62
8
0.78
2
ENJ
0.48
8
0.60
5
0.71
6
0.72
0
0.66
3
0.69
0
0.748
FUL
L
0.59
7
0.61
3
0.66
1
0.66
5
0.67
9
0.66
9
0.588
0.77
8
PRI
0.53
5
0.57
7
0.57
8
0.65
7
0.55
7
0.73
2
0.587
0.55
3
0.85
2
RES
0.70
3
0.64
3
0.67
7
0.67
9
0.60
7
0.73
3
0.681
0.73
8
0.63
3
0.80
8
SYS
0.41
9
0.44
1
0.67
5
0.68
4
0.59
2
0.69
8
0.627
0.60
2
0.59
9
0.62
3
0.78
0
VE
0.63
7
0.63
0
0.56
0
0.64
9
0.59
7
0.66
7
0.675
0.55
4
0.61
8
0.64
6
0.54
2
0.75
2
Table 4
Cross Loading
CO
M
CO
N
LO
Y
SA
T
CP
V
EF
F
ENJ
OY
FU
LL
PR
I
RE
S
SY
S
VE
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3446
COM
1
0.88
0
0.5
21
0.4
33
0.4
47
0.4
03
0.5
40
0.421
0.52
4
0.4
94
0.6
38
0.3
47
0.5
61
COM
2
0.87
9
0.5
64
0.4
04
0.4
55
0.3
92
0.4
82
0.428
0.50
0
0.4
46
0.6
06
0.3
39
0.5
35
COM
3
0.88
6
0.5
86
0.4
40
0.5
11
0.4
85
0.5
27
0.441
0.55
2
0.4
75
0.6
15
0.4
17
0.5
86
CON
1
0.55
2
0.8
48
0.4
92
0.5
15
0.5
14
0.5
13
0.491
0.54
6
0.4
81
0.5
62
0.3
72
0.4
91
CON
2
0.47
1
0.8
47
0.5
07
0.5
54
0.5
53
0.5
14
0.531
0.50
5
0.5
13
0.5
37
0.4
23
0.5
52
CON
3
0.58
3
0.8
39
0.4
59
0.5
58
0.5
35
0.4
78
0.510
0.50
4
0.4
67
0.5
31
0.3
21
0.5
53
CPV
1
0.43
7
0.4
86
0.5
55
0.5
96
0.7
74
0.4
97
0.558
0.48
5
0.4
97
0.5
04
0.4
44
0.5
79
CPV
2
0.33
0
0.4
88
0.6
82
0.6
21
0.8
08
0.5
07
0.538
0.53
0
0.4
50
0.4
72
0.5
17
0.4
55
CPV
3
0.28
6
0.4
74
0.5
02
0.5
26
0.7
74
0.3
82
0.455
0.53
4
0.3
00
0.3
63
0.3
86
0.3
67
CPV
4
0.49
6
0.5
78
0.6
66
0.6
23
0.8
53
0.6
04
0.571
0.62
7
0.5
20
0.5
90
0.5
33
0.5
11
EFF4
0.44
1
0.5
25
0.5
97
0.6
12
0.5
68
0.7
46
0.594
0.53
7
0.5
67
0.5
56
0.5
24
0.4
87
EFF5
0.53
7
0.4
70
0.5
55
0.5
73
0.5
28
0.8
30
0.543
0.56
7
0.5
87
0.6
27
0.5
72
0.5
93
EFF7
0.45
6
0.4
22
0.5
39
0.5
96
0.4
74
0.8
13
0.516
0.53
4
0.6
00
0.5
72
0.6
35
0.5
28
EFF8
0.39
5
0.4
33
0.4
79
0.5
27
0.3
75
0.7
36
0.496
0.44
6
0.5
29
0.5
36
0.4
41
0.4
76
ENJ1
0.33
4
0.4
33
0.5
43
0.5
48
0.4
62
0.5
12
0.765
0.42
2
0.4
15
0.5
07
0.5
45
0.4
36
ENJ2
0.33
2
0.4
24
0.4
31
0.4
32
0.4
42
0.3
99
0.719
0.37
2
0.3
72
0.4
29
0.3
08
0.5
11
ENJ3
0.56
0
0.5
23
0.4
81
0.4
99
0.4
80
0.5
75
0.726
0.49
3
0.5
28
0.6
04
0.4
01
0.6
84
ENJ4
0.26
9
0.4
42
0.6
49
0.6
39
0.5
81
0.5
61
0.782
0.46
8
0.4
43
0.5
00
0.5
71
0.4
31
FUL1
0.50
0
0.4
65
0.4
78
0.5
00
0.4
93
0.5
35
0.415
0.79
9
0.3
94
0.6
30
0.4
41
0.4
54
FUL2
0.44
1
0.4
52
0.4
59
0.4
66
0.4
97
0.4
70
0.420
0.77
9
0.3
89
0.5
35
0.4
67
0.3
95
FUL3
0.50
9
0.4
13
0.4
91
0.4
37
0.4
37
0.4
78
0.429
0.74
7
0.4
36
0.6
01
0.4
05
0.4
19
FUL6
0.49
2
0.5
62
0.5
69
0.5
90
0.6
16
0.5
80
0.511
0.78
7
0.4
89
0.5
72
0.5
05
0.5
10
FUL7
0.38
7
0.4
75
0.5
54
0.5
67
0.5
71
0.5
26
0.496
0.77
7
0.4
32
0.5
38
0.5
08
0.3
70
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3447
e-
LOY
1
0.30
9
0.3
97
0.8
37
0.5
93
0.5
91
0.5
14
0.525
0.53
8
0.4
12
0.5
15
0.5
71
0.3
33
e-
LOY
2
0.35
9
0.4
55
0.8
24
0.6
31
0.6
06
0.5
97
0.594
0.55
8
0.4
81
0.5
37
0.5
66
0.4
35
e-
LOY
3
0.48
7
0.5
06
0.8
58
0.6
43
0.6
24
0.5
87
0.613
0.57
8
0.4
51
0.6
24
0.5
60
0.5
00
e-
LOY
4
0.40
3
0.4
37
0.8
71
0.6
14
0.6
75
0.5
93
0.591
0.61
6
0.4
87
0.5
86
0.6
10
0.4
16
e-
LOY
5
0.35
3
0.5
17
0.8
40
0.6
68
0.6
30
0.6
02
0.634
0.50
1
0.5
27
0.5
65
0.5
55
0.5
04
e-
LOY
6
0.50
2
0.5
66
0.7
69
0.6
51
0.6
62
0.5
92
0.620
0.50
9
0.5
31
0.5
55
0.5
12
0.6
08
PRI1
0.45
2
0.5
24
0.4
48
0.5
61
0.5
06
0.6
28
0.484
0.45
2
0.8
39
0.5
24
0.4
67
0.5
26
PRI2
0.48
2
0.5
24
0.5
28
0.5
75
0.4
75
0.6
14
0.503
0.43
6
0.8
79
0.5
46
0.5
30
0.5
66
PRI3
0.43
2
0.4
25
0.4
99
0.5
42
0.4
44
0.6
28
0.513
0.52
7
0.8
36
0.5
49
0.5
34
0.4
85
RES1
0.55
5
0.5
18
0.5
68
0.5
77
0.5
39
0.6
45
0.556
0.60
8
0.6
13
0.8
33
0.5
57
0.5
50
RES2
0.52
5
0.5
02
0.5
22
0.5
67
0.4
97
0.5
78
0.576
0.57
9
0.4
98
0.8
38
0.5
25
0.5
02
RES3
0.56
8
0.5
46
0.6
40
0.6
47
0.5
43
0.6
49
0.600
0.67
6
0.5
40
0.8
56
0.6
00
0.5
46
RES4
0.51
6
0.4
60
0.4
76
0.4
49
0.4
40
0.5
37
0.515
0.54
0
0.4
37
0.7
34
0.4
14
0.4
78
RES5
0.69
0
0.5
74
0.5
08
0.4
70
0.4
16
0.5
39
0.495
0.56
3
0.4
54
0.7
70
0.3
84
0.5
32
e-
SAT1
0.36
7
0.5
24
0.6
26
0.8
24
0.6
03
0.6
24
0.634
0.56
1
0.5
36
0.5
46
0.5
81
0.5
32
e-
SAT2
0.51
0
0.5
56
0.6
30
0.8
41
0.6
21
0.5
92
0.578
0.56
9
0.5
32
0.5
71
0.5
56
0.5
25
e-
SAT3
0.48
0
0.5
48
0.6
69
0.8
68
0.6
51
0.6
59
0.612
0.55
6
0.5
95
0.6
02
0.5
95
0.5
87
SYS1
0.24
2
0.3
66
0.5
89
0.5
79
0.5
65
0.4
53
0.518
0.47
7
0.3
94
0.4
48
0.7
61
0.3
45
SYS2
0.37
3
0.3
13
0.4
54
0.4
76
0.3
78
0.5
85
0.473
0.44
3
0.5
18
0.5
06
0.7
86
0.4
36
SYS3
0.37
8
0.3
47
0.5
20
0.5
32
0.4
21
0.6
09
0.470
0.48
3
0.5
04
0.5
08
0.7
94
0.4
97
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3448
VE1
0.33
8
0.4
99
0.5
03
0.5
68
0.4
71
0.5
51
0.580
0.44
5
0.5
72
0.5
37
0.5
41
0.7
50
VE2
0.51
8
0.4
54
0.3
39
0.4
16
0.3
93
0.4
12
0.458
0.36
9
0.3
56
0.4
07
0.2
90
0.7
39
VE3
0.62
4
0.4
61
0.3
90
0.4
52
0.4
72
0.5
20
0.462
0.42
3
0.4
26
0.4
91
0.3
45
0.7
66
The third is the Heterotrait-Monotrait Ratio (HTMT), with a reference value of
<0.90. The results of the menu data processing analysis show that all constructs meet the
good validity requirements of the three approaches, except for the HTMT perceived value
against e-satisfaction (0.911), efficiency against e-satisfaction (0.929), enjoyment against
e-satisfaction (0.929), system availability against e-satisfaction (0.917), privacy against
efficiency (0.915), system availability against efficiency (0.955), virtual emotion against
efficiency (0.931), and virtual emotion against enjoyment (0.991).
Table 5 HTMT
CO
M
CO
N
LO
Y
SA
T
CP
V
EFF
ENJO
Y
FUL
L
PRI
RE
S
SY
S
V
E
CO
M
CO
N
0.76
5
LO
Y
0.54
5
0.67
4
SAT
0.64
5
0.80
4
0.89
0
CPV
0.57
4
0.78
0
0.86
7
0.91
1
EFF
0.71
1
0.74
6
0.81
9
0.92
9
0.76
6
ENJ
0.62
5
0.78
8
0.85
2
0.91
7
0.83
8
0.89
0
FUL
L
0.70
5
0.74
4
0.74
9
0.80
4
0.81
0
0.81
6
0.736
PRI
0.64
1
0.71
6
0.67
1
0.81
5
0.67
7
0.91
5
0.756
0.66
8
RES
0.82
1
0.77
5
0.75
6
0.80
7
0.70
9
0.88
4
0.847
0.86
5
0.75
2
SYS
0.55
2
0.59
2
0.84
5
0.91
7
0.77
0
0.95
5
0.850
0.78
8
0.81
2
0.80
1
VE
0.89
0
0.88
2
0.71
9
0.89
5
0.82
2
0.93
1
0.991
0.75
1
0.83
8
0.86
1
0.79
9
Assessing Formative Measurement Models
In this study, the dimensions of e-SQ (efficiency, fulfilment, privacy, system
availability, responsiveness, compensation, contact, enjoyment, and virtual emotion) are
second-order constructs with a reflective-formative type. Each first-order construct is
reflective in its relationship to the e-SQ indicators, while the dimensions of e-SQ are
formative. According to (Hair & Alamer, 2022), a high correlation is not expected
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3449
between indicators in formative measurement models. This test is carried out based on
the Variance Inflation Factor (VIF) value, provided it is <5.00, and when it exceeds the
limit, there is a problem with collinearity. The results of the data processing analysis
concluded that the collinearity issue was unfounded because the VIF value (2.338 –
3.750) is by the applicable reference.
Table 6
Formative Measurement Model Evaluation
Variable
Indicator
VIF
Outer
Weights
(Outer
Loadings)
T
Statistics
P Values
E-Service
Quality
(E-SQ)
Efficiency
(EFF)
3.644
0.200 (0.877)
2.554
0.005
Fulfillment
(FUL)
2.619
0.185 (0.808)
2.682
0.004
System
Availability
(SYS)
2.338
0.251 (0.828)
4.754
0.000
Privacy (PRI)
2.454
0.060 (0.754)
0.950
0.171
Responsiveness
(RES)
3.750
0.054 (0.827)
0.562
0.287
Compensation
(COM)
2.506
-0.041 (0.622)
0.665
0.253
Contact (CON)
2.350
0.163 (0.744)
2.169
0.015
Enjoyment
(ENJ)
2.705
0.319 (0.876)
4.997
0.000
Virtual
emotions (VE)
2.647
0.003 (0.739)
0.046
0.482
Furthermore, the required criteria to evaluate formative indicators is conducted by
examining the outer weight and loading using bootstrapping to determine their
significance. Therefore, the indicator is maintained when the outer weight is significant.
The indicator should be kept assuming it is not statistically significant at the outer weight
but at the loading value ≥0.50. Conversely, when the indicator value on the outer loading
is <0.5, the model is considered for deletion. The data processing results show that
formative indicators have passed the existing provisions. The data processing results
show that efficiency, fulfilment, system availability, contact, and enjoyment have
significant outer weight values (T statistics of 2.169 – 4.997 > 1.65 with a P value of
0.000 – 0.015 < 0.05), while privacy, responsiveness, compensation, and virtual emotion
has an insignificant outer weight value (T statistic of 0.046 – 0.950 < 1.65 with P value
0.171 – 0.482 > 0.05). However, the outer loading value (0.622 – 0.827) ≥ 0.50 so that
the indicators are maintained. Thus, formative indicators have passed the existing
provisions.
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3450
Assessing Structural Models
Collinearity needs to be investigated to make sure it isn't skewed the regression
findings before evaluating the structural linkages. This process, similar to evaluating
formative measurement models, determines the VIF values using the exogenous
constructs' latent variable scores. The VIF value (1.00 - 4.05) was < 5.00, meeting the
specified criteria based on the data processing outcomes.
The guidelines of 0.75 (strong), 0.50 (moderate), and 0.25 (weak) indicate that the
coefficient of determination (R2) measures the combined effect of exogenous latent
variables on endogenous. The data processing results reveal that the e-SQ variable
effectively explains customer e-satisfaction by 69.30%. This means that 30.70% of other
variables not analyzed in this research explain customer e-satisfaction. In addition, the e-
SQ, customer e-satisfaction, and customer perceived value variable is rated R2 by 71.10%
to explain the e-loyalty, while the remaining 28.90% explained others, not in this research.
This study utilized a bootstrapping procedure to assess the significance of the path
coefficient in testing existing hypotheses, with consideration given to notable influences.
When the T-value surpasses 1.65 with a 95% confidence level, and the P-value remains
below 0.05, the original sample is deemed appropriate to determine the relationship
direction.
Mediation Analysis
A mediation analysis was performed to evaluate the impact of customer e-
satisfaction on the relationship between e-SQ and e-loyalty. The findings showed a
significant indirect effect of e-SQ on customer e-loyalty through e-satisfaction (H4:
Original sample = 0.119, T statistic = 1.680 > 1.65, P value = 0.047 < 0.05). The effect
of total e-SQ on customer e-loyalty was also significant (Original sample = 0.547, T
statistic = 7.102 > 1.65, P value = 0.000 < 0.05), with the inclusion of mediators. The
impact of e-SQ on customer e-loyalty is noteworthy (Original sample = 0.456, T
statistical = 4.645 > 1.65, P value = 0.000 < 0.05). The study indicates that the
contribution of supplementary customer e-satisfaction towards the link between e-SQ and
e-loyalty influences e-loyalty indirectly. Hence, according to the results from the
mediation analysis test, H4 is confirmed.
Table 7
Mediation Analysis Results
Total effect
(e-SQ -> Customer e-
Loyalty)
Direct effect
(e-SQ -> Customer e-
Loyalty)
Indirect Effects of e-SQ on
Customer e-Loyalty
Coefficient
P-Value
Coefficient
P-Value
Coefficient
T-
Value
P-
Values
0.574
0.000
0.456
0.000
H4
0.119
1.680
0.047
The initial finding of this study indicates that e-SQ has a favourable effect on
customer e-satisfaction. It implies that the e-satisfaction of marketplace customers will
increase alongside the enhancement in the provided e-SQ. Customers are content with
their purchase from the marketplace owing to the hedonic aspect of e-SQ, which is
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3451
demonstrated in the form of the enjoyment dimension and is capable of offering a
comfortable shopping experience. This sense of comfort is derived by customers from the
system's availability dimension. They perceive that the marketplace functions properly
and runs smoothly. Customers experience efficiency since they can access the
marketplace quickly, specifically by loading the page. From the cross-tabulation analysis,
it is evident that respondents aged 18-21 years dominate, with 120 respondents (45.3%)
selecting the Tokopedia marketplace (12.8%), Blibli (11.3%), Shopee (8.3%), Laza-da
(7.5%) and Bukalapak (5.7%). The advanced technology proficiency of the demographic
group indicates a deeper comprehension of the marketplace's service quality. This is
consistent with prior research, indicating that e-SQ positively affects e-satisfaction
highlighting the importance of the hedonic aspect, which may enhance the probability of
new users revisiting a website.
This study demonstrates that e-SQ has a beneficial effect on customer e-loyalty,
specifically, the better the e-SQ offered by the marketplace, the greater the increase in e-
loyalty amongst its customers. The marketplace is preferred by customers due to its
fulfilment aspect, where it fulfils delivery promises and is capable of delivering goods
within an acceptable timeframe as deemed by customers. In addition, the contact aspect
of e-SQ, specifically the marketplace, offers a company helpline for customers as well as
an online customer service representative to report any issues that may arise. The cross-
tabulation analysis results reveal that the largest proportion of female marketplace users
purchase care & beauty items (20.8%) and household supplies (20.4%) online, whilst the
majority of male marketplace users buy accessories & fashion items (10.9%) and
electronics (7.2%) online. These findings suggest gender differences in online shopping
preferences and, consequently, varying delivery requirements. This statement is
supported by prior studies that demonstrated a positive correlation between e-SQ and
customer e-loyalty.
Furthermore, the third research finding indicates that customer satisfaction with
online experiences has a favourable effect on e-loyalty. This implies that enhanced
customer e-satisfaction leads to increased e-loyalty. Customers feel it is a wise choice to
shop at the marketplace so that even if customers have to buy again at the marketplace,
they will still be satisfied. In effect, this makes the marketplace the best place for them to
shop. This is reinforced by previous research that found customer e-satisfaction is the
basis of the intention to remain loyal to online consumers (Brusch et al., 2019).
Another important finding of this study is that customer e-satisfaction acts as a
mediator for e-SQ in terms of customer e-loyalty, which is acceptable. This implies that
an improvement in the e-satisfaction of customers by the marketplace will result in an
increase in the e-loyalty of marketplace users due to better provision of e-SQ. Customers
enjoy browsing in the marketplace as it is organized efficiently and provides thorough
information, leading to customer satisfaction. As a result, they often recommend the
marketplace to others seeking online shopping advice and choose it as their preferred
online shopping destination. The findings of this investigation are corroborated by prior
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3452
scholarship that discovered e-satisfaction impacts user conduct and e-SQ evaluation
influences e-satisfaction, subsequently impacting e-loyalty.
Finally, the results showed that customer-perceived value does not strengthen the
relationship between e-satisfaction and e-loyalty. These results are not in line with the
research conducted by (Kim et al., 2021) that customer perceived value is moderation
between e-satisfaction and e-loyalty. This is because the majority of respondents have
had sufficient shopping experience 8-12 times in the last 3 months (33.2%) consisting of
usage duration of more than 3 years (18.5%), 1-2 years (12.5%), and under 1 year (2.3%)
This makes process makes customers satisfied with the e-SQ provided by the
marketplace, which ultimately results in the inability of variables of customer perceived
value to strengthen the relationship between e-satisfaction and e-loyalty. Therefore,
although customer perceived value has increased, it does not raise their e-satisfaction,
which impacts e-loyalty. In other words, satisfied customers will potentially increase e-
loyalty, even when there is not a perceived value variable that acts as moderation. This is
reinforced by the results of the second hypothesis that e-SQ directly affects customers' e-
loyalty.
Conclusion
This study analyzes how e-SQ affects customer e-satisfaction as a mediator, and in
turn, customer e-loyalty. Additionally, the moderating role of customer-perceived value
is considered. Bagozzi's (1992) appraisal-emotional reactions-behavior framework
guides this research, which focuses on marketplace customers in Indonesia. Therefore, it
can be concluded from the research findings that higher levels of e-SQ, particularly
relating to enjoyment, system availability, fulfillment, and efficiency, provided by the
marketplace, correspond to increased e-satisfaction and e-loyalty among its customers.
The study consisted mainly of respondents aged between 18 and 21 (45.3%), suggesting
that they have a greater level of technological proficiency and are therefore more familiar
with marketplace service quality. In addition, it appears that male and female respondents
exhibit differing preferences for online shopping. Males are inclined towards accessories
& fashion (10.9%) as well as electronics (7.2%), while females display dominance in care
& beauty (20.8%) as well as household supplies (20.4%). These variances suggest unique
requirements for product delivery.
Then, higher e-satisfaction among marketplace customers leads to increased
customer e-loyalty. As the marketplace improves e-SQ through the e-satisfaction of its
customers, their e-loyalty will also increase. Customer e-satisfaction acts as a mediator
between e-SQ and customer e-loyalty. This means that as the marketplace improves e-SQ
through the e-satisfaction of its customers, their e-loyalty will also increase. However,
the results demonstrate that even if there are no moderating variables in the form of
customer perceived value, satisfied customers can enhance e-customer loyalty.
This research has limitations that should be addressed in future investigations.
Specifically, the study only included participants residing in the Greater Jakarta area as
online marketplace customers. Therefore, the findings may not be generalizable to other
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3453
populations. Furthermore, the correlation between e-loyalty and geography may be a
potential weakness. Those with limited geographic mobility may feel compelled to
remain loyal to online purchasing, due to the convenience of this option in light of their
location, compared to those with greater mobility. Further research is needed to include
users from diverse areas. Additionally, this study has restricted its determination of e-
loyalty to the variables of e-SQ, customer e-satisfaction, and perceived value. Further
research is recommended as several factors related to customer e-loyalty remain to be
investigated, including e-trust and e-commerce innovation. It is anticipated that more
comprehensive research will provide an in-depth understanding of the subject.
Gilang Pratama Hafidz, Muhamad Khaidir Ali
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3454
Bibliography
Brusch, Ines, Schwarz, Benjamin, & Schmitt, Rebecca. (2019). David versus goliath-
service quality factors for niche providers in online retailing. Journal of Retailing
and Consumer Services, 50, 266–276.
Das, Kaushik, Tamhane, Toshan, Vatterott, Ben, Wibowo, Phillia, & Wintels, Simon.
(2018). The digital archipelago.
Eryiğit, Canan, & Fan, Yanrong. (2021). The Effects of Convenience and Risk on E-
Loyalty through the Mediating Role of E-Service Quality: A Comparison for China
and Turkey. Journal of International Consumer Marketing, 33(5), 613–626.
Ghali, Zohra. (2021). Motives of customers-loyalty towards e-banking services: a study
in Saudi Arabia. Journal of Decision Systems, 30(2–3), 172–193.
Hair, Joseph, & Alamer, Abdullah. (2022). Partial Least Squares Structural Equation
Modeling (PLS-SEM) in second language and education research: Guidelines using
an applied example. Research Methods in Applied Linguistics, 1(3), 100027.
Kim, Yaeri, Wang, Qiran, & Roh, Taewoo. (2021). Do information and service quality
affect perceived privacy protection, satisfaction, and loyalty? Evidence from a
Chinese O2O-based mobile shopping application. Telematics and Informatics, 56,
101483.
Kotler, P., Armstrong, G., Harris, L. C., & He, H. (2020). Principles of Marketing. Eighth
Eur. Pearson Education Limited. Harlow, UK.
Napitupulu, Darmawan, Kadar, J. Abdel, & Jati, R. Kartika. (2017). Validity testing of
technology acceptance model based on factor analysis approach. Indonesian
Journal of Electrical Engineering and Computer Science, 5(3), 697–704.
Nisar, Tahir M., & Prabhakar, Guru. (2017). What factors determine e-satisfaction and
consumer spending in e-commerce retailing? Journal of Retailing and Consumer
Services, 39, 135–144.
Parasuraman, Ananthanarayanan, Zeithaml, Valarie A., & Malhotra, Arvind. (2005). ES-
QUAL: A multiple-item scale for assessing electronic service quality. Journal of
Service Research, 7(3), 213–233.
Rahma, Asya Zahratur, & Syah, Tantri Yanuar Rahmat. (2023). The Relationship
Between Service Quality, Patient Satisfaction, Behavioral Intentions And Prices In
Outpatient Healthcare In The Hospital Industry. Jurnal Ekonomi, 12(02), 232–242.
Rodríguez, Pablo Gutiérrez, Villarreal, Ricardo, Valiño, Pedro Cuesta, & Blozis, Shelley.
(2020). A PLS-SEM approach to understanding E-SQ, e-satisfaction and e-loyalty
for fashion e-retailers in Spain. Journal of Retailing and Consumer Services, 57,
102201.
Impact of E-Service Quality
in Customer E-Loyalty of Marketplace with the Role of E-Satisfaction As Mediation
Jurnal Indonesia Sosial Teknologi, Vol. 5, No. 7, July 2024 3455
Sarstedt, Marko, Hair Jr, Joseph F., Cheah, Jun Hwa, Becker, Jan Michael, & Ringle,
Christian M. (2019). How to specify, estimate, and validate higher-order constructs
in PLS-SEM. Australasian Marketing Journal, 27(3), 197–211.
Shatnawi, Tamather. (2019). Development of an integrated conceptual framework for
electronic hedonic service quality (e-HSQ): An exploratory study. Journal of
Internet Commerce, 18(4), 395–427.
Theodosiou, Marios, Katsikea, Evangelia, Samiee, Saeed, & Makri, Katerina. (2019). A
comparison of formative versus reflective approaches for the measurement of
electronic service quality. Journal of Interactive Marketing, 47(1), 53–67.
