IDENTIFYING THE FACTORS AFFECTING THE

ADOPTION OF E-COMMERCE IN SRI LANKA:

INTERNET USERS' PERSPECTIVE

Koswatte Gedera Hashani Navarathna

179124U

Degree of Master of Business Administration in Information Technology

Department of Computer Science and Engineering

University of Moratuwa

Sri Lanka

June 2019
 IDENTIFYING THE FACTORS AFFECTING THE
ADOPTION OF E-COMMERCE IN SRI LANKA:
INTERNET USERS' PERSPECTIVE

Koswatte Gedera Hashani Navarathna

179124U

The dissertation was submitted to the Department of Computer Science and

Engineering of the University of Moratuwa in partial fulfilment of the requirement for
the Degree of Master of Business Administration in Information Technology

Department of Computer Science and Engineering

University of Moratuwa

Sri Lanka

June 2019
 DECLARATION

I declare that this is my own work and this thesis does not incorporate without
acknowledgement any material previously submitted for a Degree or Diploma in any
other University or institute of higher learning and to the best of my knowledge and
belief it does not contain any material previously published or written by another
person except where the acknowledgement is made in the text.

Also, I hereby grant to University of Moratuwa the non-exclusive right to reproduce

and distribute my thesis/dissertation, in whole or in part in print, electronic or other
medium. I retain the right to use this content in whole or part in future works (such as
articles or books).

……………………………….. ………………………….

K.G.H. Navarathna Date

(Signature of the candidate)

The above candidate has carried out research for the Master‟s thesis under my
supervision.

……………………………….. ………………………….

Dr. H.M.N. Dilum Bandara Date

Signature of the Supervisor

i
 COPYRIGHT STATEMENT

I hereby grant the University of Moratuwa the right to archive and to make available my thesis
or dissertation in whole or part in the University Libraries in all forms of media, subject to the
provisions of the current copyright act of Sri Lanka. I retain all proprietary rights, such as
patent rights. I also retain the right to use in future works (such as articles or books) all or part
of this thesis or dissertation.

-------------------------------

K.G.H. Navarathna

ii
 ABSTRACT

Electronic commerce brings world economic market into a common portal where buyers and
sellers could meet overcoming their economic and geographic barriers. Despite being one of
the earliest adopters of the Internet in the region and several initiatives by stakeholders to
promote e-commerce, e-commerce adaptation in Sri Lanka appears to be relatively low.
Maturity level of e-commerce adoption in Sri Lanka, as well as any obstacles and hindrances
preventing the widespread adoption are not clear. The objective of this research is to identify
barriers that prevent Internet users from wide-spread adoption of e-commerce in Sri Lanka.
We adopted a mix-method research methodology to identify Internet user related adoption
factors and to understand e-commerce merchants‟ view on those factors. First, a preliminary
survey was carried out by interviewing stakeholders to identify barriers affecting the
adoption of e-commerce. Next, a survey was used to identify customer adoption factors.
Survey data were analyzed using Structural Equation Modeling to identify key factors and
their relationships. Finally, interviews were conducted with e-commerce merchants to
identify their view on those factors and what actions had been taken to enhance the adoption
level. Factors such as affordability, knowledge and awareness, and facilities expected from
retailers were identified as the most significant factors contributing to the success of e-
commerce adoption in Sri Lanka. Government and legal factors, consumer perception, and
digital infrastructure are the least significant factors. Smoothening the delivery process,
introducing convenient return policies, and enhancing government involvement to promote
e-commerce are the key recommendations to enhance the e-commerce adoption.

Keywords: Internet users, e-commerce, e-commerce adoption, Structural Equation

Modeling

iii
 ACKNOWLEDGEMENT

Initially I would like to thank my thesis supervisor Dr. Dilum Bandara of CSE
department at University of Moratuwa. He was always available there to clear my
doubts and questions regarding the research or writing. He constantly allowed this to
be my own work, while guiding me to the correct direction.

I would also take this opportunity to thank the industry experts who took part in the
preliminary survey for this research project. Without their valuable input and ideas,
the questionnaire could not have been successfully prepared.

Apart from them, I would also like to thank the professionals who supported in
validating the results for this research writing.

I would also like to acknowledge word boutique team as the second reader of this
dissertation, for their valuable comments made on this dissertation.

Last but not least, my sincere gratitude goes to my parents and to my husband for
providing me constant support and encouragement throughout my postgraduate
studies and research. This achievement would not have been possible without them.

Thank you.

K.G. Hashani Navarathna

iv
 TABLE OF CONTENTS

DECLARATION .......................................................................................................... I
COPYRIGHT STATEMENT ...................................................................................... II
ABSTRACT ................................................................................................................ III
ACKNOWLEDGEMENT .......................................................................................... IV
TABLE OF CONTENTS ............................................................................................. V
LIST OF FIGURES .................................................................................................... VII
LIST OF TABLES .................................................................................................... VIII
LIST OF ABBREVIATIONS ..................................................................................... IX
1. INTRODUCTION ................................................................................................ 1
1.1. Background ................................................................................................... 1
1.2. Problem statement ......................................................................................... 2
1.3. Research objective ........................................................................................ 2
1.4. Research significance .................................................................................... 3
1.5. Outline ........................................................................................................... 3
2. LITERATURE REVIEW ..................................................................................... 4
2.1. Organizational-level challenges in e-commerce adoption ............................ 4
2.2. Factors affecting online transaction and shopping behavior ......................... 8
2.3. e-commerce adoption in developing countries ............................................. 9
2.4. Summary ..................................................................................................... 11
3. RESEARCH METHODOLOGY ........................................................................ 12
3.1. Research process ......................................................................................... 12
3.2. Conceptual framework ................................................................................ 12
3.3. Structural equation modeling ...................................................................... 17
3.4. Population and sample selection ................................................................. 18
3.5. Data collection ............................................................................................ 19
3.6. Summary ..................................................................................................... 19
4. DATA ANALYSIS AND DISCUSSION ........................................................... 22
4.1. Analysis of pre-survey interviews ............................................................... 22
4.2. Data preparation for analysis ...................................................................... 23
4.3. Demographic analysis ................................................................................. 24

v
 4.4. Questionnaire analysis ................................................................................ 30
4.5. Validity and reliability of the data .............................................................. 31
4.6. Relationship with factors affecting successful e-commerce adoption ........ 33
4.7. Factor analysis of different demographics .................................................. 36
4.8. Analysis of expert feedback ........................................................................ 38
4.9. Summary ..................................................................................................... 39
5. RECOMMENDATIONS AND CONCLUSION ................................................ 40
5.1. Conclusion .................................................................................................. 40
5.2. Research limitations .................................................................................... 42
5.3. Future work ................................................................................................. 42
REFERENCES........................................................................................................... 44
APPENDIX A: PRE- SURVEY QUESTIONNAIRE ................................................ 49
APPENDIX B: SURVEY QUESTIONNAIRE.......................................................... 51
APPENDIX C: HISTOGRAMS FOR LIKERT SCALE QUESTIONS..................... 57
APPENDIX D: FACTOR ANALYSIS BASED ON DEMOGRAPHGIC
CHARACTERISTICS ............................................................................................... 60

vi
 LIST OF FIGURES
Figure 3.1: Research approach. 13

Figure 3.2: Conceptual framework. 15

Figure 4.1: Gender distribution of the sample. 24

Figure 4.2: Age distribution of survey responders. 25

Figure 4.3: Educational qualification distribution of participants. 26

Figure 4.4: Employment status of the sample. 27

Figure 4.5: Summary of the smployment sector (no of samples = 378). 27

Figure 4.6: Income level distribution of the sample. 28

Figure 4.7: Online purchasing frequency distribution. 28

Figure 4.8: Estimated online transaction or purchases value distribution. 29

Figure 4.9: Estimated percentages of online purchases. 30

Figure 4.10: Structural model. 33

Figure 4.11: Estimated model with bootstrapping. 34

vii
 LIST OF TABLES

Table 3.1: Pre-identified factors. 14

Table 3.2: Definitions of constructs. 15

Table 3.3: Sub-factors and indicators used to measure the constructs. 20

Table 4.1: Summary of company profiles. 22

Table 4.2: Criteria for validity and reliability or smart PLS 3.0. 31

Table 4.3: Critical considerations regarding validity and reliability. 32

Table 4.4: Discriminant validity values of the constructs. 32

Table 4.5: Path coefficients. 34

Table 4.6: Corresponding weight and „p‟ value for each indicator. 36

viii
 LIST OF ABBREVIATIONS

AF Affordability
AVE Average Variance Extracted
CP Consumer Perception
BOI Board of Investment
CEO Chief Executive Officers
CIO Chief Executive Officers
DI Digital Infrastructure
DV Dependent variable
DW Durbin-Watson
e-commerce electronic commerce
e-economy electronic economy
e-readiness electronic readiness
e-marketing electronic marketing
ERI Electronic Readiness Index
FR Facilities expected from retailers
GCE General Certificate of Education
GDP Gross Domestic Product
GLS Government and Legal Support
GOF Goodness of Fit
ICT Information and Communication Technology
IT Information Technology
IV Independent Variable
KAF Knowledge and Awareness Factors
LSA Level of Success in e-commerce Adoption
LV Latent Variable
ME Micro Enterprises
NRI Network Readiness Index
PLS Partial Least Square
ROI Return on Investment
SEM Structural Equation Modeling
SME Small and Medium-sized Enterprises

ix
 1. INTRODUCTION

1.1. Background

Electronic commerce or e-commerce brings world‟s economic market at your

fingertips. It involves any commercial or business transactions which involve transfer
of e-money and information of buyers and sellers throughout the Internet. Schneider
and Perry (2000) stated that, e-commerce includes business activities such as serving
customers, associating with business partners and exchanging business documents
performed using the Internet. Broadly, e-commerce can be explained as a general
medium which allows performing any form of commercial transactions over the
Internet (Whitely, 1998). There is a potential ability in e-commerce to improve
efficiency and productivity in many sectors; therefore, has gained significant
attention across industries and states.
While it has become a part of life in some countries, it has not gained wide-
spread adoption in many others. Sri Lanka has several popular e-commerce sites like
Kapruka.com, MyDeal.lk, WoW.lk, and takas.lk. Based on the Central Bank 2017
annual report (Central Bank of Sri Lanka, 2017), even though there has been a rapid
growth in Sri Lankan e-commerce industry, still overall sales are comparatively low
when compared to similar countries (Kurnia, 2005). Informal information implies,
maintaining inventory is one of the key issues. Most e-commerce sites rely on
external partners to maintain inventory and this has led to several challenges such as
reliability and limited inventory. Another issue is the delivery of products to
customers, especially to customers outside Colombo.
Sri Lanka is still in a stage of transforming from the traditional transaction
methods into e-commerce. Thus, there is significant opportunity to increase the level
of adoption. Moreover, current level of e-commerce maturity in Sri Lanka and
factors hindering the adoption are unclear. According to the literature, most of the
studies were focused on organizational-level adoption of Sri Lanka and rest of the
world (Senarathna et al., 2011; Suriyapperuma et al., 2015; Grandon et al., 2004;
Choshin et al., 2015). A little amount of analysis seemed to exist on the Internet user
perception on adoption (Javadi et al., 2012; Makhitha, 2014). Therefore, it is

1
imperative to identify the challenges according to Internet users‟ perception, to reach
higher maturity levels.

1.2. Problem statement

34.66% of the total population in Sri Lanka, used either fixed broadband or
mobile broadband (Telecommunications Regulatory Commission of Sri Lanka,
2018). But as per the 2017 annual report of the Central Bank (Central Bank of Sri
Lanka, 2017), overall sales of most popular e-commerce sites in Sri Lanka are
comparatively low. Also, peer discussions and customer rating on e-commerce sites
reveal profit is only a part of the problem where the bigger problem seems to be
Internet users not using them frequently or not buying high-value items. In this
context, the problem to be addressed by this research can be formulated as:
What are the Internet user-centric factors affecting the adoption of e-commerce in
Sri Lanka?
This study focuses on possible challenges that prevent Internet users from
adapting to e-commerce (i.e., Internet user related factors) rather than the challenges
affect organizational level. This would be benefitted for e-commerce companies,
because both organizational perspective and consumer perspective goes hand in hand
in attracting, interacting, and retaining the customers. Therefore, knowing the
consumer perspective will be an advantage. According to the background search,
most of the past research work conducted in other countries over the world including
Asia, has studied organizational level factors but not in Sri Lankan consumer related
factor.

1.3. Research objective

The key objective of the research is to identify the Internet user related factors
that affect the e-commerce adoption in Sri Lanka. Apart from that, following are the
sub-objectives to be achieved in this study.
 To identify factors that affect the e-commerce adoption from related research
work and domain experts

2
  To examine the strength of association between the identified Internet user-
related factors and indicators used to measure e-commerce adoption in Sri
Lanka
 Solicit feedback from e-commerce business stakeholders regarding the
identified Internet user-related factors

1.4. Research significance

The outcome of this research would be identification of a set of challenges that

prevents Internet users from frequently using e-commerce in Sri Lanka. By looking
at the identified challenges and recommendations given, relevant business
stakeholders could consider about the actions that can be initiative to overcome those
identified challenges. This could benefit them in regularizing and smoothening their
e-commerce offerings and online processes.

1.5. Outline

The rest of the thesis is organized as follows; Chapter 2 presents the literature
review on identifying the factors affecting the e-commerce adoption along with the
justification of current research work. Chapter 3 contains a detailed description on
research methodology, where research approach, data collecting methods, and tools
are explained. Data analysis is presented in Chapter 4. Conclusion, research
limitations, and future work are discussed in Chapter 5.

3
 2. LITERATURE REVIEW

Section 2.1 presents the literature on organizational-level challenges in e-

commerce adoption while Section 2.2 focuses on factors affecting online transaction
and shopping behavior. e-commerce adoption in developing countries is described in
Section 2.3. Section 2.4 presents the summary of related work.

2.1. Organizational-level challenges in e-commerce adoption

Small and Medium-sized Enterprises (SMEs) gives a major contribution to the

national economy and Gross Domestic Product (GDP). These SMEs are often
recognized as economy growth engines. A research has been conducted by Choshin
and Ghaffari (2015) on determining the significant factors which affect the e-
commerce adoption in small and developing business firms. Main objective of this
study was to propose a framework to determine the effective factors for successful e-
commerce adoption in SMEs. Infrastructure, costs, customers‟ satisfaction, and
knowledge and awareness have used as four variables in constructing the model.
Based on a survey of 180 staff employees; Post Bank, Eastern and Western
Azerbaijan, authors concluded that customers‟ satisfaction, amount of costs,
infrastructure, and knowledge and awareness significantly affect the success of e-
commerce. Limitation of study was that, the data sample represented only a single
organization.
Senarathna et al. (2011) examined the relationship between the organizational
factors and e-commerce adoption in Sri Lanka using 200 SMEs in Colombo district.
Authors concluded that there is a positive correlation between the ability to adapt
quickly to changing conditions in SMEs and e-commerce adoption. Apart from that
managerial attitude regarding innovation has a positive relationship with e-commerce
adoption. However, no relationship between cultural characteristics and commerce
adoption in SMEs was found.
A research by the Central Province, Sri Lanka (2004) pointed out the
challenges faced by developing nations in e-commerce adoption. Even though the
large-scale funding agencies such as World Bank and Asian Development Bank were

4
willing to help, the outcomes were not so appreciable. Moreover, there were only a
few successful stories based on such funding. It suggested well-coordinated support
between government sectors and the private companies, as well as greater participant
of public are essential to formulate the e-economy system. e-readiness score of Sri
Lanka is 3.66 (Economist Intelligence Unit, 2003). This is far below the countries
like India and Thailand which have scores of over five. The factors influencing this
low rating are due to low teledensity (i.e., number of telephone connections for every
hundred individuals living within an area), outdated legislative framework, low level
of computer literacy, and lack of human resources in the field. To overcome these
deficiencies, authors suggested formulating favorable government policies including
legal framework, enhancing knowledge of computer literacy and provide adequate
infrastructure facilities through national GDP.
It has been accepted that Sri Lanka has the potential to adopt these concepts
due to importance of geographical location, availability of human and natural
resources. Majority of the economy is run by SMEs and Micro Enterprises (MEs) in
Sri Lanka. Those will be benefiting from adapting ICT capabilities in improving their
quality of services, as well as the overall economy. However, these researchers
focused on large-scale organizations. Little was explored about consumer perception;
hence, there is a need to explore the consumer- related factors.
Another research focusing on Board of Investment (BOI) companies in Sri
Lanka has shown that compared with neighbor countries, usage of Information
Technology (IT) by SME in Sri Lanka are yet at a lower level (Gunawardana, 2008).
The main purpose of this study was to determine the firm-based characteristics on e-
commerce adoption in SMEs in Sri Lanka. 30 SMEs in Colombo metropolitan area
registered under BOI were used as sample population. Simple linear regression was
preformed to identify the relationship between e-commerce adoption and firm-based
characteristics. Among tested firm-based characteristics, involvement of the
management has been identified as significant characteristic for e-commerce
adoption. Supplier service factor was a critical success factor for both involvement of
the management and adoption in IT. Moreover, e-marketing was another critical
success factor that determines the involvement of the management.

5
 There are several limitations of this study. Sample population was only
restricted to SMEs in Colombo metropolitan area. Also, sample size was
comparatively small, and research was narrowed down through Business-to-Business
activities. Hence, there is a lack of identifying the needs and behavior of Internet
consumers in e-commerce activities. Moreover, there was no consideration on
demographic factors. Therefore, there is a necessity on conducting a research with
more representatives analyzing the Internet consumer- related factors including
demographic consideration as well.
Suriyapperuma et al. (2015) investigated the Internet adoption impact on SME
performance in Sri Lanka by analyzing secondary data of 250 articles. Advantages of
Internet, complexity, business orientation, new work practice adoptability and ICT
costs were determined as direct influential factors on Sri Lankan SMEs and their
performances. However, due to the use secondary data, there can be a mismatch with
the current context. Therefore, there is still a need to understand Internet users‟
perspective in adopting e-commerce.
A study was carried by Kapurubandara et al. (2008) to understand and
determine the challenges faced by developing countries in e-commerce adoption. It
analyzed the internal and external challenges faced by SMEs and identified the
support required to overcome them. The studied area was limited to Colombo
district, as it has the highest density of companies where ICT is used. To have
richness and validity in the research, authors carried out preliminary pilot interviews,
a survey, and interviews with SMEs. It allowed for both qualitative and quantitative
methods of analysis with good observations. They have identified lack of required
skill in employees, security issues on Internet payment methods, cost spend on e-
commerce and Return on Investment (ROI) from e-commerce as internal barriers.
Absence of government coordination in legal and infrastructure facilities, social
norms were found to be external barriers. With the findings they highlighted the
necessity of providing support to SMEs if they are ready to successfully adopt ICT
and e-commerce.
Though this research is based on Sri Lankan context, their focus was to cover
enterprise level. They have not explored the consumer-related challenges. Also,
sample size was not adequate to represent SMEs in Sri Lanka.

6
 Grandon et al. (2004) focused on the contribution of top management of small
companies in Chile towards the adoption into e-commerce concept. Data was
collected through a survey questionnaire. A preliminary discriminant analysis was
conducted to determine the factors with their respective order of importance.
Organizational readiness, managerial productivity, external pressure, decision aids,
compatibility, and perceived usefulness were the found to be significant in
contributing to e-commerce adoption.
In this study there is no differentiation among the chosen executives. They
categorized them all as managers/owners. This might lead to confusion in results as
the knowledge about e-commerce is different from profession to profession. Since
the sample was limited to Bio-Bio region of Chile, there were no national-wide
representation of the entire population and sample size was small. Though the
objective of this research is achieved, there is no attempt to identify the contribution
to e-commerce on consumer perspective in a company. So, there is an emerging need
of identifying consumer related factors relevant to the profession.
Grandon and Pearson (2004) identified the determinant factors of strategic
value and e-commerce adoption as per the top managers in SME; Midwest region of
the United States (US). Two independent research streams are combined in this
study. Based on the previous research work, operational support, managerial
productivity and strategic decision aids were the prominent facts in perception of
strategic value while organizational readiness, external pressure, perceived ease of
use and perceived usefulness were the four prominent categories in e-commerce
adoption. Hypotheses were tested to find a link between these two streams. Sample
population was surveyed using an online survey. Confirmatory factor analysis and
Canonical analysis is used to test the model using collected data.
According to the findings, all three facts in perception of strategic value stream
was found to be significant and perceived usefulness, perceived ease of use,
compatibility and external pressure groups were identified as significant in e-
commerce adoption. Among them, the perceived usefulness, compatibility and
perceived ease of use pointed out to be most significant facts in e-commerce
adoption. Number of employees considered to this research is a major limitation.
They have used firms where number of employees varies between 10 and 200 as

7
their sample. In this context only five firms consisted of more than 200. Therefore,
this sample was biased for smaller firms.

2.2. Factors affecting online transaction and shopping behavior

Javadi et al. (2012) analyzed factors affecting online shopping behavior of

consumers in Iran. Using an email questionnaire 107 randomly selected consumers
of five major online stores in Iran were surveyed. Results indicated that financial risk
and non-delivery risk has negative impact on online shopping behavior while
subjective norms, domain specific innovativeness have positive effect on online
shopping behavior. Due to the time constraints, authors have not considered all
possible factors affecting on online shopping behavior. The population represents
only the consumers of electronic goods in Iran. Therefore, the results might not be
applicable for other countries and other consumer goods.
Online shopping seems to be very popular among the younger generation.
Makhitha (2014) identified factors that determine attitudes of Generation Y students
on online shopping in South African context. Population for this research was
Generation Y undergraduates and postgraduates at a main university of technology in
South Africa. The questionnaire was designed based on related work. The findings
stated that majority of students do not do online shopping even though they
frequently use Internet. Convenience, perceived risk and security, shopping intention,
ease of use and usefulness were the five factors identified based on the factor
analysis. Among them shopping intention, convenience, ease of use, and usefulness
factors had a significant positive impact on the attitude of online shoppers. Perceived
risk and security had a negative impact on the attitude of online shoppers.
Applicability of this research is limited because only undergraduate and post
graduate students were considered as the sample.
The growth of online transactions goes hand in hand with an increase of rate in
fraud. Today, most of the merchant activities take place through the Internet. In
Kenya though the frequency of fraud activities is low, with the increasing e-
commerce adoption, it is predicted to increase fraud related incidents. Based on this
fact, a study has been conducted by Kanyaru and Kyalo (2015) to find the challenges

8
connected with Internet merchant activities and best practices that can be used to
prevent such challenges. Secondary data collected from scholarly journals, books and
credible publications were used.
Authors concluded that lack of customer awareness about online market
security risks, increase of highly vulnerable mobile devices as the challenges in
Internet merchant activities. Strong security systems, incident management plans and
regular security assessments, continuous detection and conventional protection
measures were identified as the best practices that can be used to prevent such
challenges. Since this is based on secondary data, there can be reliability issues.
Lin and Lee (2005) identified the effect of organizational learning factors and
knowledge management processes on e-business adoption. Data was collected from
202 information system executives in Taiwanese firms using a survey. Structural
Equation Modeling (SEM) approach was used to test the research framework. The
results confirmed that all three factors; knowledge level, training available, and
technical expertise in organizational learning have a significant relationship with
successful e-business adoption. Both knowledge acquisition and knowledge
application in knowledge management processes has significant relationship with e-
business adoption. But there is no significant relationship between knowledge
sharing and e-business adoption. According to the authors, there were three major
restrictions in this study. The sample population may reflect the cultural differences
between Taiwan and other countries. Apart from that, this study has not considered
the all organizational learning factors. Also, factors related to technological
innovation can be changed over time. Another major limitation was the sample size.
The sample population only included large enterprises in Taiwan. Also, in many
cases large firms tend to be more mature in e-business adoption rather that the SMEs.
Therefore, the results do not imply the entire e-business adoption.

2.3. e-commerce adoption in developing countries

It is important to know the world‟s trends and adoption towards e-commerce

throughout the existing systems. Based on the Grounded theory, a research has been
conducted by Kabango and Asa (2015) to determine the factors that support e-

9
commerce development in developing countries. This study aims to determine the
association of attitudes and e-commerce adoption in developing countries. The
development of e-commerce is measured along the four facts; trust and loyalty,
accountability and awareness, quality and benefits, security and privacy. In
conclusion they stated that, trust, security, fraud and hacking, awareness and
perceived usefulness, accessibility, perceived quality and role of government has an
association with e-commerce adoption. However, authors do not mention the
population sampled and model adopted to analyze the data. Moreover, the
investigation did not take place in multiple cities; especially in rural areas where the
accuracy and richness could be enhanced for the results.
Bui et al. (2003) identified the importance of e-commerce in a nation. Authors
considered eight factors such as macro economy, digital infrastructure,
knowledgeable citizens, ability to invest, competitiveness, access to skilled
workforce, culture and cost of living and pricing. Key objectives of this research
were to identify the factors that contribute to increased e-readiness, develop
measures that can quantitatively measure e-readiness and provide a framework to
calculate e-readiness of a country across these eight factors. Based on possible
indicators for each factor, following formula was derived to calculate composite e-
readiness index for a given country:
(2.1)

Where e-readiness is the overall e-readiness value, i is country, j is each of the 52

measures, wij is relative weights assigned to the 52 measures, eij is individual score
for each measure on a scale of 1 to 10, and n is the total number of measures. Several
selected countries from East Asia, USA, and G7 were used to interpret the selected
methodology. According to the results, they have concluded that selected East Asian
countries need more improvement regarding the e-commerce area.
Even though they tested the concept among several countries, each country has
their own unique set of e-readiness characteristics. Therefore, these eight factors can
be different or there maybe additional factors than eight to be considered when
applying this methodology. Also, with the rapid change, collected data can be
outdated as well.

10
 e-readiness of a country refers to the capacity and state of preparedness to
participate in the electronic world. In other terms, it measures the quality of ICT
infrastructure in a country and the capacity of consumers, businesses and
governments to use that ICT to their benefit. Since the proposed research is aimed on
e-commerce adoption of e-commerce in Internet user level, there can be different
factors that should be considered in calculating the national e-readiness value (Center
for International Development, 2002).

2.4. Summary

This chapter presented a detailed discussion of past research work done

worldwide, based on identifying the factors affecting e-commerce adoption. Also, it
discussed about strengths and limitations on the literature done and the importance of
having a research work for consumer level as well. Related work on e-commerce
adoption mainly focused on identifying the organizational factors. We identified
twelve key factors through literature review that mainly contribute to a successful e-
commerce adoption. Even though there were several research work on other countries
identifying the factors affecting online shopping behavior of consumers, no such
study exists for Sri Lanka. Also, they did not cover aspects such as bank and other
financial transactions, health services. Moreover, consumer perspective can be
changed with time, country and developing technology. Therefore, this research is
conducted to fill that research gap. Apart from that, there were research work on
calculating the national e-readiness and identifying the factors affecting on that. But
national e-readiness does not represent the e-commerce adoption. Therefore, the
factors that affect national e-readiness may relevant or may not relevant to e-
commerce adoption. Hence, this research focused on identifying the consumer
perspective factors while putting the maximum effort to address the limitations found
in literature review.

11
 3. RESEARCH METHODOLOGY

Section 3.1 presents the research process. Section 3.2 explains the conceptual
framework used in the study. Section 3.3 explains the statistical technique used.
Section 3.4 presents the population and sample selection while Section 3.5 presents
data collection process. The summary is presented in Section 3.6.

3.1. Research process

Figure 3.1 illustrates the research approach used in the study. There are three
phases. In the first phase, a preliminary survey (see Appendix A) was carried out by
process of interviews with domain experts to find the factors that need to be
considered apart from the factors identified during the literature study. In the second
phase a detailed questionnaire (see Appendix B) was distributed among the potential
consumers of e-commerce. This questionnaire covered the purpose of this study, as
well as the indicators which has been gathered in the previous phase. In the third
phase a qualitative approach was used to validate the results by interviewing relevant
professionals to solicit expert feedback on the identified customer-related factors.

3.2. Conceptual framework

Based on the literature survey factors listed in Table 3.1 were identified as
related to the adoption of e-commerce. Ability to invest (spending capacity),
knowledgeable citizens, digital infrastructure, perceived usefulness, perceived ease
of use, product risks, non-delivery risk, convenient product return policy, attitude,
trust and loyalty, security and privacy and customer satisfaction were found as key
factors that adopt e-commerce adoption.
Consumer perspective on e-commerce changes with time and advancing
technology. Also, some of the pre-figured factors may not be relevant to the Sri
Lankan context. Therefore, a pre-survey was also conducted to identify the additional
factors that may need to be considered apart from ones identified from the literature
study. Several CEOs and marketing managers of popular e-commerce retailers were
interviewed during the pre-survey.

12
 Literature Pre Pre-identified
Survey Survey factors
affecting e-
commerce
Phase 01 adoption

Data analysis Survey

Phase 02

Consumer-Related Expert
Factors affecting e- feedback on
commerce adoption in results
Sri Lanka

Phase 03

Figure 3.1: Research approach.

Focused business area, e-commerce role in the business, e-commerce strategies
used to attract consumers and consumer performance in e-commerce were discussed
in the interviews. Apart from those, challenges in growing and promoting online
business, suggestions in overcoming those challenges and factors identified from the
literature study were also discussed as well. Not having proper delivery mechanisms,
lack of e-commerce exposure and knowledge of local merchants, lack of legislation
and rules for e-commerce, and lack of awareness of available legal framework for e-
commerce were pointed out as major challenges that they faced. Poor government
involvement such as not having international payment methods, poor support in IT
infrastructure and poor e-commerce promotion were also listed as challenges in
promoting e-commerce among consumers. Improvement in government involvement
in promoting e-commerce to public sector, enhancement in legislation and practices
on e-commerce and having rating systems and consumer reviews displayed on the e-
commerce web site were their suggestions to encourage consumers, as well as the
service providers.

13
 Table 3.1: Pre-identified factors.

Related Sources

[Kabango et al; 2015]

[Choshin et al; 20157

Grandon et al; 2004

[Javadi et al; 2014]

Lane et al; 2004

[Lin et al; 2005]

Bui et al; 2003
Factors Affecting e-commerce Adoption

Ability to Invest (Spending Capacity) X

Knowledgeable Citizens X X X X X X
Digital Infrastructure X X
Perceived Usefulness X
Perceived Ease of Use X X X
Perception about the quality of e-products and e- services X X
Non-Delivery Risk X
Convenient product return policy X
Attitude X
Trust and Loyalty X
Security and Privacy X X
Customer Satisfaction X X

After identifying the factors, the focus was to determine the association
between these factors and e-commerce adoption in Sri Lanka. However, these
variables were not directly measurable. Therefore, to develop a relationship between
these unobservable measures and e-commerce adoption, a framework was developed
using a set of observable measures.
These immeasurable variables were described as Latent Variables (LV) /
Constructs (Bentler and Bonnet, 1980). These hypothetical constructs are created by
the researcher to understand the research area. We identified seven LVs / constructs
based on the facts found in literature review and pre-survey. Those identified LVs
are listed in Table 3.2.

14
 Table 3.2: Definitions of constructs.
Construct / Latent Variables Definition
Digital infrastructure (DI) Degree to which a consumer believes that Digital
infrastructure would impact their e-commerce adoption
behavior
Government and Legal Support Degree to which a consumer believes that Government
(GLS) and Legal Factors would impact their e-commerce
adoption behavior

Affordability (AF) Degree to which a consumer believes that Affordability

would impact their e-commerce adoption behavior
Facilities expected from retailers Degree to which a person consumer believes that
(FR) Retailers would impact their e-commerce adoption
behavior.
Knowledge and Awareness (KAF) Degree to which a consumer believes that Knowledge
and Awareness would impact their e-commerce
adoption behavior
Consumer Perception (CP) Degree to which a consumer believes that perception
would impact their e-commerce adoption behavior
Level of Success in e-commerce Consumer ability to use e-commerce
Adoption (LSA)

Figure 3.2 shows the research framework generated using the constructs /
variables identified. Independent variables are represented in left while dependent
variables are represented in right.

Affordability

Government and
Legal Support

e-
Knowledge and
commerce
Awareness
adoption

Digital
Infrastructure

Facilities
expected from
Retailers

Consumer
Perception

Figure 3.2: Conceptual framework.

15
 Following hypothesis can be derived based on the research framework:
Let;
HA: Alternative Hypothesis
H0: Null Hypothesis
H1
H10. There is no significant relationship between the maturity of digital infrastructure
and level of e-commerce adoption.
H1A. There is a significant positive relationship between the maturity of digital
infrastructure and level of e-commerce adoption.

H2
H20. There is no significant relationship between the government and legal support
and level of e-commerce adoption.
H2A.There is a significant positive relationship between the government and legal
support and level of e-commerce adoption.

H3
H30. There is no significant relationship between the affordability and level of e-
commerce adoption.
H3A. There is a significant positive relationship between the affordability and level of
e-commerce adoption.

H4
H40. There is no significant relationship between the facilities expected from retailers
and level of e-commerce adoption.
H4A. There is a significant positive relationship between the facilities expected from
retailers and level of e-commerce adoption.

H5
H50. There is no significant relationship between the knowledge and awareness and
level of e-commerce adoption.
H5A. There is a significant positive relationship between the knowledge and
awareness and level of e-commerce adoption.

16
H6
H60. There is no significant relationship between the Consumer Perception and level
of e-commerce adoption.
H6A. There is a significant positive relationship between the Consumer Perception
and level of e-commerce adoption.

3.3. Structural equation modeling

The literature review showed that in such situation the concept of Structural
Equation Modeling (SEM) can be used. Structural Equation Modeling (SEM) is a
statistical method use to analysis relationship in between network of data. It is
currently used in various social science analyses. It evolved from the earlier methods
such as genetic path modeling. It includes composite analysis, factor analysis, path
analysis and various model analyses. It involves measurement of directly
immeasurable (unobserved constructs) latent variables using one or more observed
(indicator/manifest) variables. SEM assesses relationships between such latent
variables. SEM allows testing multiple constructs simultaneously. Also, it calculates
the strength of association between constructs. When association among variables is
tested, the associations are free of measurement error as it has been estimated and
removed. Therefore, reliability of measurement is assured (Bentler and Ullman,
2003, Chapter 23).
To example the concept of educational level cannot be directly measured,
instead college final year GPA, exam mark, and grades of subjects is used. Those are
observed variables. Hence, SEM provides the relationship in numerical value. This
value indicates the strength of the relationship among constructs. Ultimately it allows
identifying good indicators to predict latent variables.
There are many advantages in using SEM. Since it estimates and eliminates
error in data gathering, leaving only common variance, measurement error is
removed. Reliability of measurement is accounted for estimating and removing the
measurement error. Complex relationships would easily be analyzed with this
technique. It allows simultaneous examination of several multidimensional
constructs. Alternatively, it calculates the strength of association between constructs

17
(Bentler and Ullman, 2003, chapter 23). LVs are unobservable and not directly
measurable. Therefore, to estimate the LV s in the model, the concept of indicator
variables is adopted. Indictor variables are observable and empirically measurable.
Table 3.3 shows the constructs along with the sub-factors and indicators used to
measure the above constructs in the questionnaire (see Appendix B). Based on the
factors, collected from the literature and pre-survey, a detailed questionnaire was
distributed to collect sample data. Random responses were excluded using the
answers for question number 04 and 05 (see Appendix B).

3.4. Population and sample selection

The focused population represents the potential Internet users‟ in Sri Lanka.
According to the statistical data from the Telecommunications Regulatory
Commission of Sri Lanka (2018) by December 2018, there were 7,263,161 fixed
broadband, narrowband, and mobile broadband subscriptions. This represents
34.67% from the total population. Given this population, the sample size was
determined to be 384 sampels based on the Cochran‟s formula (Bartlett et al., 2001)
with a confidence interval of 0.05 and confidence level of 95%. This technique is
more appropriate for large populations (Bartlett et al., 2001; Glenn, 1992).
Convenience sampling (availability sampling) is a non-probability
sampling method. It depends on data collection from population members who are
conveniently available. Samples are selected because they are easiest to recruit for
the research. It is the most common of all sampling techniques. It is fast,
inexpensive, easy, and the subjects are readily available. Simplicity of sampling and
the ease of research. It is helpful for pilot studies, hypothesis generation, and data
collection can be facilitated in less time (Etikan et al., 2015). Therefore, we used
convenience sampling for the survey. Even though random sampling may cover all
the social groups in the sample population, it takes quite long time to collect data.
Also, the subjects may be unavailable due to the technical diiculties (i.e.,
geographical distance) and limited available time for data collection. Also, there are
cases where it is difficult to divide the population into strata (i.e., we can study on
equal number of males and feales eventhough sex distribution is not equal) (Showkat

18
et al., 2017). Therefore, to collect data within the given time frame, convenience
sampling was the most appropriate method.

3.5. Data collection

Data were collected using both survey and interview approach. Due to the time
constraints, Internet users who were readily accessible or available were targeted to
collect the data. For online data collection process, e-mail addresses were selected as
per the convenience.
The questionnaire was divided into three sections. The first section contained
questions related to the purpose of the study. In the second section some questions
were closed ended questions. There the respondents were able to make their response
on a 5-point Likert scale, differ from strongly disagree to strongly agree and some
questions were designed to rate the answer on 1 -10 scale according to the
respondent preference. Questions to collect demographic details of the respondents
were included into third section. Post research interviews were carried out with
information officers in popular organizations to validate the obtained results.

3.6. Summary

We proposed to use a mix method of both qualitative and quantitative methods

to identify the factors affecting e-commerce adoption in Sri Lanka. Primarily a
survey was carried out and later a set of interviews was done to validate the results.
Structural Equation modeling statistical techniques was proposed to analyze the
gathered data. Internet users‟ population of Sri Lanka was taken as the population
and the calculated sample size was 384. Convenience sampling method was used as
the sampling method. A detailed analysis of the methodology adopted for this
research was presented including conceptual framework and hypothesis used.

19
 Table 3.3: Sub-factors and indicators used to measure the constructs.

Indicators
Construct Sub-Factors (Survey
Questions)

Digital
Infrastructure Having Internet connectivity issues (i.e., speed, break downs) 1. Q.10(6)
(DI)
Poor safety and return policies 1. Q.10(9)
Poor security and privacy issues 2. Q.10(10)

Government Awareness on Sri Lankan laws regarding e-commerce 3. Q.11(1)

and Legal Satisfaction on government involvement with regards to e-
Factors 4. Q.11(4)
commerce
(GLF)
Trust in payment methods 5. Q.11(8)
Satisfaction on confidentiality maintained by the online service
6. Q.11(9)
providers in Sri Lanka
Importance of Price 1. Q.09(1)
Affordability
(AF) 2. Demogra
Income level
phy Q.06
Having Home delivery and multiple pick up options 1. Q.09(4)
Promotions 2. Q.09(2)
Low delivery fess 3. Q.09(5)
Facilities
expected
Delivery time 4. Q.09(6)
from retailers Having lack of choices and services 5. Q.10(1)
(FR)
Payment issues 6. Q.10(3)
Delivery Issues 7. Q.10(4)
Good Customer support 8. Q.10(8)
Degree of satisfaction with quality of products/services 9. Q.11(7)
Knowledge
and
Awareness
Lack of knowledge in using the technology 1. Q.10(2)
(KAF)
Time taken for checkout 1. Q.09(3)
Peer reviews 2. Q.09(8)
Consumer Brand and quality 3. Q.09(7)
Perception
(CP) Customer reviews 4. Q.09(9)
Availability of options 5. Q.09(10)
Lack of confidence on the product quality 6. Q.10(5)

20
 Cost associated with online transactions 7. Q.10(7)
Trust on Online Transaction methods 8. Q.11(3)
Satisfaction in user-friendliness of e-commerce web sites in Sri
9. Q.11(5)
Lanka
Satisfaction with the number, variations, and quality of payment
10. Q.11(6)
methods offered by e-commerce web sites in Sri Lanka
Level of Cost for of all online purchases over the last 12 months (in
1. Q.06
Success in e- rupees)
commerce
Adoption Number of online purchasing, transactions, and service websites
2. Q.03
(LSA) know and use

21
 4. DATA ANALYSIS AND DISCUSSION

Section 4.1 presents the analysis of pre-survey interviews. Data preparation for
analysis is explained in Section 4.2. Section 4.3 illustrates the demographic analysis.
Section 4.4 presents the questionnaire analysis. Section 4.5 describes the validity and
reliability analysis of the data while Section 4.6 explains the relationship between
results and e-commerce adoption. Section 4.7 illustrates relationship between results
and the demographic data. Post-interview analysis is presented under Section 4.8.
Chapter summary is presented in Section 4.9.

4.1. Analysis of pre-survey interviews

Some of the pre-figured factors in literature study may not be relevant to the
Sri Lankan context (see Table 3.1). Also, they can be outdated as consumer
perspective on e-commerce changes with time and advancing technology. Therefore,
pre-survey interviews were conducted to identify unique contributory factors in the
Sri Lankan context that may need to be considered apart from ones identified from
the literature study.
Two interviews were conducted to identify the unique factors that might affect
Sri Lankan Internet users‟ behavior. Table 4.1 presents the profile of the interview
participants. Challenges in growing and promoting online business, suggestions in
overcoming those challenges, and future e-commerce adoptions were discussed.

Table 4.1: Summary of company profiles.

Participant 1 Participant 2
Designation Marketing manager of an online CEO from a financial solution
retailing company providing company
Focused Leading retailers in home
business area appliances (i.e., TV, washing
machine), mobile phone and
tabs
Customer base Island-wide consumer base
(mostly online users in their
mid-30s)

22
 Because the retailer does not have any regional offices, deliveries are handled
by the central store located in Colombo. Therefore, handling delivery and returns
island-wide has become a serious matter for them. This increases consumer
frustration and loss of their trust in business. Lack of IT exposure and lack of IT
knowledge of their merchants in online transactions was another challenge. These
merchants still rely on paper documents rather than digital transactions. This has
lengthened the primary purchasing procedure and result in out of stock conditions
which makes buyers unhappy.
According to the domain expert, when it comes to Sri Lankan consumers, lack
of legislation and lack of awareness of available legal framework act as a strong
challenge. They have suggested enhancing the legislation and practices from internal
affairs to court on digital forensic evidence and IT crimes to encourage consumers,
as well as the service providers. Moreover, poor government support such as not
having international payment methods, poor IT infrastructure, and poor e-commerce
promotion framework were pointed out as challenges by both interviewees.
Hence, government involvement in promoting e-commerce to the public was
listed as an essential factor to consider in improving the e-commerce adoption of
consumers. Also, to give more transparency to the online business they suggested
having rating systems and consumer reviews displayed on the web site. In summary,
following factors were also found to be essential in promoting e-commerce adoption:
1. Knowledgeable citizens
2. Non-delivery risk
3. Attitude
4. Lack of legislation and rules for e-commerce
5. Lack of awareness of available legal framework for e-commerce
6. Poor government involvement

4.2. Data preparation for analysis

414 responses were collected between November 19, 2018 to January 20, 2019
using both online and paper-based questionnaire. 239 responses were collected from
online survey while the remaining 175 responses were collected from paper-based

23
survey. No missing values were recorded. Among them 12 responses were excluded
due to unreliability of the answers for question number four and five (see Appendix
B). In question four, those who have responded as “nothing” for “Which types of
goods/services do you typically purchase/consume online?” question, have continued
to complete the questionnaire. As those responses were unreliable, those cases were
excluded. In question five, those who have responded that they never perform any
kind online transaction, have continued to complete the questionnaire. Therefore,
those cases were also excluded and finally 402 responses were used for the data
analysis.

4.3. Demographic analysis

43% Male
57% Female

Figure 4.1: Gender distribution of the sample.

Figure 4.1 shows the gender distribution of the sample. Based on the statistical
data from Census of Population and Housing 1911-2012 (Census of Population and
Housing, 2012); Department of Census and Statistics, 51.59% of the Sri Lankan
population was male and 48.41% were female. Therefore, the sample population is
somewhat biased in terms of gender distribution.
According to a study by “AdWeek and “Washington Post” (Moses, 2013; Lu,
2014), men rely on online purchasing rather than women as they think it saves times
and cost whereas for women there are various reasons behind this behavior. Women
are mostly hedonic shoppers. Women want to know more about brand, how products
are going to make them feel and need more tactile selection rather than visual

24
selection. Thus, women like manual shopping. Alternatively, women tend to consider
both customer reviews and objective information. They behave in a more
comprehensive manner. Men tend to take objective information over subjective
information into consideration. So it has been shown that men favor mobile shopping
which save their time and cost, while women enjoy browsing products and catalogs,
and sharing information within their networks (Moses, 2013; Lu, 2014).

350

300
No of respondents

250

200

150

100

50

0
16-25 26-35 36-45 46-55 >55
Age group

Figure 4.2: Age distribution of survey responders.

As seen on Figure 4.2 majority (76.9%) of the survey participants were

between 26-35 years age category. According to the statistical data from Census of
Population and Housing 1911-2012 (Census of Population and Housing, 2012);
Department of Census and Statistics, majority (23.54%) of the population belong to
more than 55 years age category. But in the sample, majority is belonged to 26- 35
years age group. Therefore, the sample data is towards younger customers. The key
reason for this might be respondents whose age between 26-35 years was ready to
answer the questionnaire than the other generations. They may have more hands-on
experience in e-commerce. Most respondents over 46-years were reluctant to answer
the questions as they stated that they are not involved in online purchasing and
transaction services.
252 (62.7%) respondents were having bachelor‟s degrees. Only a few
responders were educated below GCE Advanced Level. This distribution is presented

25
in Figure 4.3. Therefore, the sample is biased along the educated population. To
overcome this, stratified sampling would be a better sampling method.

0.7% 1.5%
0.7%
3.7%
3%
6.7%

20.9%

62.7%

GCE OL GCE AL Diploma Bacheor

Master Professional PhD Other

Figure 4.3: Educational qualification distribution of participants.

Full-time employment was defined as being employed 35 hours or more per

week and who has permanent single occupation. Based on Sri Lanka Labor force
Survey 2015 (Sri Lanka Labour Force Survey, 2015), majority of the labor force
belonged to 26-35 years age category. Because most of the samples were from this
age category, 87.3% of the sample population was full-time employed (see Figure
4.4). Part-time employment was defined as occupying a job less than 35-hours or less
per week and who do not have permanent single occupation. 5.2% belong to this
category. Among the population 3.0% were housewives and 0.8% belong to the
retired category.

26
 3.0% 0.8%
3.7%

5.2%

87.3%

Employed_Full Employed_Part Unemployed

Housewife Retired

Figure 4.4: Employment status of the sample.

1.6% 2.4%

7.1%

5.6%
24.6%
7.1%

31%
20.6%

Health Engineering IT Legal

Finance Educational Administrtive Other

Figure 4.5: Summary of the employment sector (no of samples = 378).

378 respondents were employed at the time of data collection. Among them,
31.0% of respondents were belonging to IT sector. Health sector employees and
Engineering sector employees represented 24.6% and 20.6%, respectively. Research

27
students, tourism sector employees and armed forces employees were categorized
into the other category. These details are summarized in Figure 4.5.

5.22%

6.72%

11.94%
37.31%

16.42%

22.39%

< 25 000 25 000 - 49 999 50 000 - 74 999

75 000 - 99 999 100 000 - 149 000 > 150 000

Figure 4.6: Income level distribution of the sample.

Income level of the participants was categorized according to the World Bank
income classifications and household income and expenditure survey - 2016 (Fantom
et al., 2016; Household Income and Expenditure Survey, 2016). These were varied
between less than LKR 25,000 to more than LKR 150,000. Income level distribution
is represented in Figure 4.6. 37.3% of the sample population was having more than
LKR 150,000 monthly incomes.

200
180
160
140
No of respondents

120
100
80
60
40
20
0
Once or several Once or several Once or several Rarely
times a week times a month times a year

Figure 4.7: Online purchasing frequency distribution.

28
 Figure 4.7 presents the how frequently the participants perform online
purchasing or transactions. Majority of the sample population (47.3%) performed
online purchasing or transactions at least once a month. 26.1% performed at least one
transaction per year.

120

100
No of respondents

80

60

40

20

Estimated expenses per year (LKR)

Figure 4.8: Estimated online transaction or purchases value distribution.

Figure 4.8 shows the estimated value of transactions or purchasing performed

within a year. Among the 402 respondents, 28.4% respondents estimated between
LKR 5,001 – 15,000 value online purchases or transactions, where 27.6%
respondents estimated between LKR 15,001 – 50,000 values.
Because most of the responders belonged to health, IT, and engineering
professions the sample is bias. Therefore, the income-level distribution does not
represent the income distribution in Sri Lanka.
Estimated percentage of online expenses compared with total expenses related
to purchases is present in Figure 4.9. 67.9% of the sample population mentioned that
they spent less than 10% of their income for online purchases over the last 12
months. 5.9% spent over 50% of their income on online purchases.

29
 5.9%

26.1%

67.9%

<10 11-50 > 50

Figure 4.9: estimated percentages of online purchases.

4.4. Questionnaire analysis

Based on that analysis (see Appendix C), majority of the sample, have a neutral
opinion on awareness of Sri Lankan customer laws regarding e-commerce.
Significant sample stated that, they are not aware with the laws as well. Further,
analysis showed that, having knowledge and awareness on performing online
purchasing/transactions is an essential factor. Many of the respondents stated that
government involvement in promoting online business in poor. These statements are
consistent with feedback from pre-survey interviews. Major portion of the sample
was satisfied with the facilities provided by the retailers such as user-friendly web
environment, number of variations, and quality of payment methods, and secure
payment methods. However, most of the respondents were uncertain about the
quality of the products, customer support, and confidentiality maintained by the
service providers.

30
4.5. Validity and reliability of the data

Table 4.2: Criteria for validity and reliability.

Consideration Constructs Value
Cronbach Alpha Coefficient > 0.6
Composite Reliability > 0.7
Average Variance Extracted > 0.4
Significance Level (p value) (For 95% confident interval) < 0.05
Discriminant Validity <0.85

Quality of the research is improved with scientifically accepted methodology,

validity and reliability of gathered data. On the other hand, for better statistical
analysis, validity and reliability of data set must be above the agreed standards.
Cronbach alpha coefficient, composite reliability, and average variance extracted are
used to determine the validity and reliability of the gathered data. Desired thresholds
are represented in Table 4.2 (Fornell et al., 1981). In SEM, there are various methods
to test the validity and consistency of a data set. In smart PLS Composite Reliability
and Average Variance Extracted are used for this purpose. These criterion are the
most widely used methods. The accepted threshold for each test are listed in Table
4.2 (Fornell et al., 1981).
Based on literature, Cronbach alpha coefficient is the most used method to test
such parameter in Structured Equation Modeling (SEM). In Smart PLS 3.0 statistical
software composite reliability and average variance extracted are considered to be
superior compared with traditional Cronbach alpha coefficient. In Cronbach‟s alpha,
it is assumed that all the indicators used have the same importance on the construct.
But in general, some indicators will be more important while some are not.
Therefore, these differences are considered when you determine the composite
reliability of the construct (Hair et al., 2013).

31
 Table 4.3: Critical considerations regarding validity and reliability.
Construct
AF CP DI FR GLS KAF LSA
Criterion

Cronbach alpha coefficient 0.632 0.785 1.000 0.788 0.565 1.000 0.705
Composite reliability 0.724 0.841 1.000 0.834 0.592 1.000 0.831
Average variance extracted 0.573 0.401 1.000 0.402 0.340 1.000 0.631

Table 4.3 lists details of critical considerations regarding validity and

reliability. Cronbach alpha, composite reliability, and average variance extracted for
affordability (AF), knowledge and awareness (KAF), level of success adoption
(LSA), Consumer Perception (CP), facilities expected from retailers (FR) and digital
infrastructure (DI) were above the expected threshold level. In government and legal
construct Cronbach alpha (0.565), composite reliability (0.592) and average variance
extracted (0.340) were below the threshold. Therefore, out of seven, six constructs
were above the expected threshold.
Discriminant validity test was carried out to determine whether the constructs
used in the model are correlated or not. It checks the degree in which a construct is
distinct from other constructs (Hair et al., 2013). Table 4.4 shows the discriminant
validity values for the constructs used in the model.

Table 4.4: Discriminant validity values of the constructs.

AF CP DI FR GLS KAF LSA

AF 0.757
CP 0.649 0.624
DI 0.052 0.247 0.624
FR 0.664 0.830 0.333 0.608
GLS 0.181 0.350 0.004 0.307 0.583
KAF 0.049 0.209 0.361 0.320 0.039 0.298
LSA 0.422 0.424 0.031 0.419 0.238 0.169 0.794

As seen in Table 4.4, none of the constructs were correlated; hence, are distinct
from each other. Since the moderate values were < 0.85, those were taken into
account as well (Henseler et al., 2012). These metrics prove that the data set is

32
consistent and the constructs we used to measure the adoption are accurate except the
data set used in Government and Legal Analysis (GLS). Therefore, to maintain the
reliability and validity of the data set, those data were removed and the analysis was
proceed further.

4.6. Relationship with factors affecting successful e-commerce adoption

SMART PLS 3.0 Partial Least Square (PLS) based software is used for the data
analyzes. Compared to the covariance and structural analysis, PLS provides both
reflective and formative scales easily. Also, no prior distributional assumptions are
needed, and comparatively small sample size is acceptable in PLS.
Figure 4.10 shows the model used to analyze the dataset using Smart PLS.
Constructs are indicated in Blue while yellow squares represent the indicator
variables. To estimate the constructs, indicator variables are used.

Figure 4.10: Structural model.

The PLS regression algorithm was used with boot strapping method to estimate
the model. It estimated statistical significance of various PLS-SEM results including
path coefficients with p values, factor scores and indicators‟ weights/loadings. The

33
estimated model with bootstrapping, including corresponding p values in inner model
and outer weights and p values in outer model is showed in Figure 4.11. Inner model
represents the relationship between a construct and level of adoption (i.e., AF ->
LSA). Outer model represents the relationship between a construct and indicators
related to that. (i.e., AF -> AF1 & AF-> AF2).

Figure 4.11: Estimated model with bootstrapping.

Table 4.5: Path coefficients.

AF -> LSA 0.039

CP-> LSA 0.189
DI -> LSA 0.949
FR -> LSA 0.001
KAF -> LSA 0.000

Affordability (p=0.039), facilities expected from retailers (p=0.001), government and

legal (p=0.022), and knowledge and awareness (p=0.000) constructs have p values
less than 0.05 (see Table 4.5). Both Consumer Perception (p=0.189) and digital

34
infrastructure (p=0.949) were having p values below the threshold. Therefore, out of
the six hypotheses tested using the SEM technique, three were found to be
significant. Therefore, following three hypotheses were proven and shown to be
significantly affecting the level of successful e-commerce adoption:
 H3. There is a significant positive relationship between the
affordability and level of success in e-commerce adoption.
 H4. There is a significant positive relationship between the facilities
expected from retailers and level of success in e-commerce adoption.
 H5. There is a significant positive relationship between knowledge
and awareness and level of success in e-commerce adoption.
No significant positive relationship between the maturity of digital
infrastructure and Consumer Perception and level of e-commerce adoption
(hypotheses H1 and H6) were observed. Therefore, digital infrastructure and
Consumer Perception were not shown to be significantly affecting the level of
successful adoption of e-commerce.
Loading and weights indicate the strength of association between construct and
indicators. It should be more than 0.5 and the corresponding p should be less than
0.05 to achieve minimum corresponding association. Strength of the association is
increased with high loading and weight value. Table 4.6 shows the indicators
considered for each accepted hypothesis with corresponding weight and p value for
each indicator. Affordability construct consisted with two indicator variables. Among
those two indicators, both importance of price factor and income level was
considered as highly relevant indicators.
Corresponding loading/weights and p values for lack of knowledge in using
the technology were above the threshold. Hence, the lack of knowledge in using the
technology indicator in knowledge and awareness were considered as a relevant
indicator. Among the nine indicators used facilities expected from retailers,
promotions, having home delivery and multiple pick up options, low delivery fees,
delivery time, and delivery issues were considered as highly relevant indicators.

35
 Table 4.6 – Corresponding weight and p value for each indicator.

Construct Indicator Weight of p value for

indicator each indicator

Affordability Importance of Price 0.863 0.000

Income level 0.633 0.000
Facilities Having Home delivery and multiple 0.855 0.000
expected from pick up options
retailers
Promotions 0.707 0.000
Low delivery fess 0.866 0.000
Delivery time 0.856 0.000
Having lack of choices and services 0.455 0.000
Payment issues 0.466 0.000
Delivery Issues 0.528 0.000
Good Customer support 0.406 0.000
Degree of satisfaction with quality of 0.223 0.006
products/services
Knowledge & Lack of knowledge in using the 1.000 0.000
Awareness technology

This implies that most of the online consumers are not highly concerned on brand
and quality of the product/services that are advertised. Their purchasing manner is
highly depending on the price, benefits offered by retailers, and delivery options.
Lack of knowledge about the technology is a key technical barrier. Digital
infrastructure is not much of a concerned, as there are many means to access e-
commerce. Behavior norms are not highly considered as relevant to e-commerce
adoption.

4.7. Factor analysis of different demographics

Demographic characteristics such as gender, age group, educational level, and

employment sector were tested separately using multi-group analysis to determine
whether the behavior changes across demographics (see Appendix D). Multi-group
analysis determines whether pre-defined data groups have any significant
differences. If the determined p value in parametric test is smaller than 0.05 the data

36
groups are having significant differences (Henseler et al., 2012). Effect size of the
association is determined by the path coefficient value. Higher the value represents
higher the effect. Using this method, multiple data groups can be tested
simultaneously, and it allows doing a comparison between them as well.
With regards to gender, it was found that there is a significant (p = 0.012)
difference between male and female with respect to affordability. Females (0.379)
were more concern about affordability than males (0.031). Most of the time females
prefer to do bargaining (Gupta, 2015). Therefore, the price factor is significant in
their purchasing behavior where males are more concerned on the convenient and
easiness than the price (Gupta, 2015). Knowledge and awareness, as well as facilities
expected from the retailers constructs did not indicate any significant difference with
regards to gender.
While 26-35 years age group showed consistent results with findings obtained
in Section 4.3, other age groups (16-25, 36-45, 46-55, and above 55) did not
contained adequate data to calculate significance for each age group with good
reliability and validity. Therefore, we could not interpret the results based on the age
group.
As majority of the participants had Bachelor‟s and Master‟s Degrees we
analyze factors related to these sub-groups. Results showed that Bachelor‟s degree
holders and Master‟s degree holders have a significant (p = 0.021) difference based
on facilities expected from the retailers‟ factor. The effect of this fact was a higher
value to the Bachelor‟s degree category (0.298) where it was a lower value to the
Master‟s degree category (-0.250).
When it comes to the employment, majority of participants were from
engineering, IT, and health sectors. Other employment categories did not contain
adequate data to calculate significance difference for each sector. Engineering sector
employees (0.484) is significantly (p = 0.040) concerned about the affordability
factor than the IT sector employees (0.043). All three sectors are keen on facilities
provided by retailers. Health sector (0.689) has a significant higher effect with
regards to facilities provided by retailers than engineering (0.588) and IT sectors (-
.0039).

37
4.8. Analysis of expert feedback

Post-validating enables verifying the relevance and accuracy of the research

findings. This helps to make correct decisions in applying the results into action. The
literature had no proven evidence to show that the Internet user behavior in e-
commerce in Sri Lanka. Hence, post-interviewees were the best representatives we
had as they were experts in the domain. Therefore, to validate the results, a set of
post interviews were carried out with two professionals; two information officers of
reputed e-comerce retailing companies. According to them affordability, facilitates
expected from retailers, and having knowledge and awareness are the significant
factors which affect e-commerce adoption which are consistent with our findings.
Moreover, their discussion revealed that the Internet user whose monthly income is
more, do more online purchasing compared to the person whose monthly income is
less. Income plays a major role to purchase online products. Higher income people
prefer to purchase online more, as it gives them reliability and convenient.
Because we have considered potential Internet users as the population, we tried
to collect more responses from them. Generally younger generation is more updated
with the technological revolution than the old generation. Therefore, most of the
responses were between 26-36 years category. This category is more likely to
consider the price, rather than the quality of the product (Nanson, 2018).
This fact has been found to be true from current research works. Therefore,
importance of price and income level became significant than the behavioral norms.
Legal factors were not highlighted as significant in this study. According to
interviewees, consumers in this age category most of the time used to avoid dealing
with legal matters such as formal complaining to legislative institutes because they
were unlike to get into such legal matters as they considered them to be a waste of
time and money.
Delivery is the most common for any age category. In Sri Lanka, still e-
commerce-based companies are not wealthy enough to have their own delivery
mechanisms. Therefore, they must rely on external parties. These external parties are
not much reliable. Quality of the service is not assured either. Therefore, delivery is
the main concern of any age category in e-commerce purchasing.

38
 To improve e-commerce adoption in general, they suggested making aware of
the benefits in online purchasing. Also, address the issues from their side such as
delivery as well.

4.9. Summary

This chapter presented a detailed description about how the data were analyzed
and a detailed discussion on the results obtained through the analysis. Except data for
government and legal support, rest of the data for digital infrastructure, affordability,
facilities expected from retailers, knowledge and awareness, consumer perception,
and level of success in e-commerce adoption have found to be valid and reliable. Out
of the six constructs tested using SEM technique, three were found to be significantly
affecting the e-commerce adoption. Importance of price and income level,
promotions, having home delivery and multiple pick up options, low delivery fees,
delivery time, and delivery issues, lack of knowledge in using the technology
indicators were pointed to be highly relevant indicators. Results were also analyzed
based on demographic data and indicators were identified for each demographic
category. Finally, results were verified with post interviews conducted with several
experts.

39
 5. RECOMMENDATIONS AND CONCLUSION
Section 5.1 presents the summary of research with the conclusions drawn from
the data analysis. Research limitations are described under Section 5.2.
Recommendations for future studies are presented in Section 5.3.

5.1. Conclusion

Even though the Internet usage is rapidly increasing, still there is a lag in
widespread adoption of online purchasing/transactions compared to traditional
transaction methods. Therefore, the main objective of this study was to identify the
factors affecting the e-commerce adoption in Sri Lanka from Internet users‟
perspective. Without being confined to the literature survey, a preliminary survey
was conducted to identify additional factors. A comprehensive questionnaire was
then developed as the research instrument. It was shared with a small sample of
potential Internet users in Sri Lanka, and both Sinhala and English versions were
used to reduce potential language barriers affecting the collected responses. Then
data cleaning and re-validation processes was followed. Missing values were
removed to maintain the integrity of the dataset. Accuracy was improved by
excluding question answer mismatch. Finally, data were analyzed using structural
equation modeling technique. The aim was to increase the validity and reliability of
the data set with suitable values for Cronbach Alpha coefficient, Composite
Reliability, Average Variance Extracted, and Discriminant Validity. Based on the
analysis it was conclduded that a all constructs were valid and reliable, except for
government and legal construct.
Out of six hypotheses tested, affordability, facilities expected from retailers,
and knowledge and awareness were turned out to be significantly impacting the
adoption of e-commerce in Sri Lanka. Consumer behavior and digital infrastructure
were not found to have a significant impact on the e-commerce adoption among
Internet users. Importance of price and income level appeared to be strong indicators
in affordability construct while having home delivery and multiple pick up options,
low delivery fees, delivery time, and delivery issues were turned out to be the
strongest indicators in facilities expected from retailers construct. Having knowledge

40
and awareness on using the technology was the strongest indicator for knowledge
and awareness construct.
During the literature survey and pre-survey interviews several potential key
factors were highlighted that may contribute to the e-commerce adoption. Because
we were concerned about the Internet users‟ perspective, digital infrastructure factor
was not significant in our study. When people in developing countries get involved
with online purchasing, they concern more about factors like affordability and
facilities given by the retailers (Kabango & Asa, 2015). Because Sri Lanka goes into
developing country category (World Economic Situation and Prospects, 2018) this
fact is applicable to us as well. Therefore, even though literature highlighted the
consumer perception as a challenge, it did not apply for Sri Lankan Internet users‟ e-
commerce adoption.
Results were further analyzed based on demographic data and the following
key points were identified:
 Affordability was having a significant impact on the e-commerce adoption of
female population than the male population.
 Bachelors‟ degree holders were more concerned by the facilities given by the
retailers than the Masters‟ degree holders.
 Engineering sector is more concerned about affordability than the IT sector. All
three sectors are keen on facilities provided by retailers while health sector had a
higher effect with regards to facilities provided by retailers than engineering and
IT sectors.
The key conclusion of this results and analysis is that the population under 26-
35 age in years with higher educational qualifications and relatively higher income
are more likely to be the main e-commerce consumer base among Internet users in
Sri Lanka. Their online transactions and purchasing decisions are affected by price,
income level, having home delivery and multiple pick-up options, low delivery fees,
delivery time, delivery issues, and having knowledge and awareness on using the
technology.
Results were verified with a set of interviews carried out with domain experts.
According to their opinion, the results were acceptable, and some indicators were

41
biased based on age category. During the discussion, they suggested that
stakeholders should be looking more into smoothen their product delivery process
Such as decreasing the delivery time, having convenient delivery policies and having
multiple options or multiple centers to deliver island wide to reduce the delay.
Moreover, they stated that government should take initiatives to motivate people to
transfer into online transactions by reducing traditional transaction methods (e.g., bill
payments and bank transactions) as well.

5.2. Research limitations

Sample population was biased based on the gender, age category, and
profession. Sample population was restricted to urban geographic areas in Sri Lanka
as well. Education and economic wealth of the people explains people‟s
improvement in technology. Also, the living area which is more advanced in
facilities and technology is likely to have more involvement with online business
rather than the less or underdeveloped geographic area. Moreover, this study is only
focused on Internet users‟ perspective on e-commerce adoption. Non-Internet users
were not included to this study.
Spiritual and cultural norms were not analyzed in the study. There was a
possibility that, the prior perceptions may change with the real experience and time.
Other scenarios like factors from merchants, government, were not considered as
well.

5.3. Future work

To ensure the presence of key subgroups within the sample population, it is
recommended to use stratified random sampling in future studies. This method will
prevent the sample from being biased towards one or few social groups. Since this
study was limited to urban geographic areas in Sri Lanka, Sample with more
representatives might give a broader representation to get more significant results.
It is necessary to design a cross sectional descriptive study to explore the
matters of none e-commerce users. Their believes (e.g., hidden charges, privacy in
danger, compulsory registration, difficulty in reversing process if a mistake happens),

42
fears and factors which limit e-commerce behavior must be researched. This will
facilitate us to understand practical issues which they might face. By addressing
those aspects, e-commerce adoption can be improved.
Metrics to measure the impact of cultural beliefs would improve the
understanding of e-commerce adoption behavior in those aspects as well. Also,
questions to measure the percentage who are not involved with e-commerce would
give a representation of status of e-commerce usage in Sri Lanka.

43
 REFERENCES

Alaaraj, H., & Ibrahim, F. W. (2014). An Overview and Classification of E-

Readiness Assessment Models. International Journal of Scientific and Research
Publications, 4(12), 1-5.

Bentler, P. M, & Bonnet, D. C. (1980). Significance tests and goodness of fit in the
analysis of covariance structures. Psychological Bulletin, 88(3), 588-606.

Bartlett, J. E., Kotrlik, J. W., & Higgins, C. C. (2001). Organizational Research:

Determining Appropriate Sample Size in Survey Research. Information Technology,
Learning, and Performance Journal, 9(1), 43.

Bentler, P. M., & Ullman, J. B. (2003). Structural Equation Modeling. Handbook of

Psychology, 23, 607-634.

Bui, T. X., Sankaran, S., & Sebastian, I. M. (2003). A framework for measuring
national e-readiness. International Journal of Electronic Business, 1(1), 3-22.

Choshin, M., & Ghaffari, A. (2016). An investigation of the impact of effective

factors on the success of e-commerce in small- and medium-sized
companies. Computers in Human Behavior, 66, 67-74.

Central Bank of Sri Lanka. (2018). Economic, Price and Financial System Stability,
Outlook and Policies. Annual Report 2017, 1, 1-25.

Delpachithra, S., Howard, S., Lane, S.M., & Vyver, G.V. (2004). An electronic
commerce initiative in regional Sri Lanka; The vision for the central province
Electronic Commerce Portal, The Electronic Journal of Information Systems in
Developing Countries, 16(1), 1-18.

Department of Census and Statistics. (2012). Census of Population and Housing

(2012). Retrieved January 25, 2019, from http://www.statistics.gov.lk/PopHouSat/
CPH2011/Pages/ Activities /Reports/FinalReport/FinalReportE.pdf

44
Department of Census and Statistics. (2015.). Sri Lanka Labour Force Survey (2015).
Retrieved January 25, 2019, from http://www.statistics.gov.lk/samplesurvey/
LFS_Annual Report_2015.pdf

Department of Census and Statistics. (2016). Household Income and Expenditure

Survey - 2016. Retrieved January 25, 2019, from http://www.statistics.gov.lk/HIES/
HIES2016 /HIES2016_FinalReport.pdf

The Economist Intelligence Unit Limited, & IBM Corporation. (2003). The 2003 e-
readiness rankings. Retrieved January 25, 2019, from http:// graphics.eiu.com /files /
ad_pdfs /eReady_2003.pdf

Etikan, I., Musa, S., A., & Alkassim, R., S. (2015). Comparison of Convenience
Sampling and Purposive Sampling, American Journal of Theoretical and Applied
Statistics, 5(1), 1-4.

Fornell, C. G., & Bookstein, F. L. (1982). Two structural equation models: LISREL
and PLS applied to consumer exit-voice theory. Journal of Marketing Research, 19,
440–452.

Glenn, D. I. (1992). Determining Sample Size, Fact Sheet PEOD-6. Florida: Florida
Cooperative Extension Services. Universisty of Florida.

Grandon, E., & Pearson, J. M. (2004). Perceptions of Managers/Owners of Small and

Medium Sized Firms in Chile, Communications of the Association for Information
Systems, 13, 81-102.

Grandon E., & Pearson J. M. (2004). Electronic commerce adoption: an empirical

study of small and medium US businesses, Information & Management, 42, 197-216.

Gunawardena, K., D. (2005). An empirical study of potential challenges of

implementing E-commerce strategy; Special reference to BOI (Board of Investment)
companies in Sri Lanka, Vidyodaya Journal of Humanities and Social Science, 2, 49-
82.

45
Gupta, P. (2015). Comparative Study of Online and Offline Shopping: A Case Study
of Rourkela in Odisha, Department of Humanities and Social Sciences National
Institute of Technology Rourkela

Henseler, J., Ringle, C. M., & Sarstedt, M. (2012). Using partial least squares path
modeling in international advertising research: Basic concepts and recent issues,
Handbook of research in international advertising, 252-276

Hair J., F., Hult, G., T., M., Ringle, C., M., & Sarstedt, M. (2013). A Primer on
Partial Least Squares Structural Equation Modeling (PLS-SEM), SAGE
Publications, Inc.

Javadi M., H., M., Dolatabadi, H., R., Nourbakhsh , M., Poursaeedi, A., &
Asadollahi, A., R. (2014). An Analysis of Factors Affecting on Online Shopping
Behavior of Consumers, International Journal of Marketing Studies, 4(5), 81.

Kapurubandara, M., & Lawson, R. (2008). Availability of E-commerce Support for

SMEs in Developing Countries, The International Journal on Advances in ICT for
Emerging Regions, 1(1), 3 – 11.

Kabango, M. C., & Asa, A. R. (2015). Factors influencing e-commerce development:

Implications for the developing countries, International Journal of Innovation and
Economic Development, 1, 59-66.

Kanyaru M., P., & Kyalo K. J. (2015). Factors Affecting the Online Transactions in
the Developing Countries: A Case of E-commerce Businesses in Nairobi County,
Kenya, Journal of Educational Policy and Entrepreneurial Research (JEPER), 2, 1-7

Lin H. F., & Lee G. G. (2005). Impact of organizational learning and knowledge
management factors on e-business adoption, Management Decision, 43, 171-188.

Lu J. (2014), Research shows men are more likely to shop on mobile than women,
Retrieved January 25, 2019, from https://www.washingtonpost.com/news/
technology/wp/2014/08/06/research-shows-men-are-more-likely-to-shop-on-mobile-
than-women/?noredirect=on

46
Makhitha K., M. (2014). Factors Influencing Generations Y Students‟ Attitude
towards Online Shopping, Mediterranean Journal of Social Sciences, 5(21),39.

Nanson, M. (2018). Gen X, Millennial and Gen Z Shopping Behaviors, Compared,

Retrieved June 08, 2019, from https://www.volusion.com/blog/gen-x-millennial-and-
gen-z-shopping-behaviors-compared/

Parveen, H., & Showkat, N. (2017). Non-Probability and Probability Sampling,

Retrieved June 08, 2019, from http://optimierung.mathematik.unikl.de/mamaeusch
/veroeffentlichungen/ver_texte/sampling_en.pdf

Schneider, G. P., & Perry J. T. (2000). Electronic Commerce Cambridge, MA:

Course Technology, 388

Senarathna, R. P., & Wickramasuriya, I. R. (2011). Organizational Factors Affecting

E-commerce Adoption in Small and Medium-sized Enterprises, Tropical
Agricultural Research, 22(2), 204 – 210.

Suriyapperuma, H. P., Ab Yajid, M. S., Khatibi, A., & Premarathne, S. P. (2015).

The Impact of Internet Adoption on SME performance in Sri Lanka: Development of
a Conceptual Framework, International Journal of Arts and Commerce, 4(1), 46- 68.

Telecommunications Regulatory Commission of Sri Lanka. (2019). The Internet

Statistics - December 2018. Retrieved June 08, 2019, from Telecommunications
Regulatory Commission of Sri Lanka website: http://www.trc.gov.lk /images/pdf/
statis_q42018.pdf

United Nations Department of Economic and Social Affairs (UN/DESA), United

Nations Conference on Trade and Development (UNCTAD), United Nations
regional commissions, & United Nations World Tourism Organization (UNWTO).
(2018). Country classification. World Economic Situation and Prospects, 140-147.

Whitely, D. (1998). E-Commerce: Strategy, Technologies and Applications, McGraw

Hill

47
World Bank Group, The World Bank‟s Classification of Countries by Income,
Retrieved January 25, 2019, from http://documents.worldbank.org/curated/en/
408581467988942234/pdf/WPS7528.pdf

48
 APPENDIX A: PRE- SURVEY QUESTIONNAIRE

1. Could you please explain the type of business your organization is engaged in?
(Focus areas, types of products, customer base, etc.)
2. Could you please explain the role that e-commerce plays in your business? (How
critical e-commerce is for success)
3. Could you please explain your e-commerce strategy? (i.e., your approach to e-
commerce)
a. What type of items your business focuses on?
b. What items are easier to sell online?
c. Is the business also supported by shop or an experience center?
4. What do you think about customer bases?
a. Are they e-commerce savvy/ready?
b. Are there any problems due to customer knowledge, attitude and
practices?
 Knowledge – Lack of e-commerce knowledge, competencies
in using modern tools, return policies, lack of awareness of
available legal framework for e-commerce in Sri Lanka, etc.
 Attitude -No confident in online transaction, Resistance to
change, thinking that it causes extra cost
 Practice - Lack of e-commerce experience
c. As per your experience, are Sri Lankans ready for online only or
mostly online type of businesses?
5. Do you think they are mature?
a. Are they concerned about convenience, options, and time saving that
e-commerce provide or just looking for bargains?
b. Who would be your ideal customer?
6. What are the challenges you face in promoting and growing your online
business?
a. Any issues in handling delivery and returns?
b. What about stock management and product options?

49
 c. Are you satisfied with the available online payment options? If not
suggestions.
d. Is your online business highly affected by the customers‟ buying
capacity?
e. Do you see the absence of a strong legal framework as a hindrance?
7. How do you propose to overcome those challenges?
a. What steps you have taken to minimize the customer complaints?
b. How about government involvement? (facilitating strong ICT infrastructure)
8. What should the government, ICTA, or industry bodies (CA Sri Lanka), banks,
or even ISPs/Telcos could do to help your business?
9. What do you think about the impact of factors such as:
Factors How Relevant (1-3)
Cause extra cost
No confident on delivery service
Poor return policies
No confident on online transaction
Resistance to change
Lack of e-commerce experience
Lack of legislations and rules for e-commerce in Sri Lanka
Lack of awareness of available legal framework for e-commerce
in Sri Lanka
Could not check the quality of items manually
Accountability of service providers
Current customer habits of people in Sri Lanka does not suit
online transactions
Lack of online payment options in sales desk
Low capacity in e-money expense
Poor ICT infrastructure

10. Anything else you wish to share?

11. Professional Title
a. MD, CEO, CIO, CTO
b. IT Manager
c. Mobile operator

50
 APPENDIX B: SURVEY QUESTIONNAIRE

Survey on Identifying the factors affecting the Adoption of E-

commerce in Sri Lanka: A Customer Perspective
Main purpose of this study is to identify your experience as a consumer of e-
commerce services in Sri Lanka.
1. I have (You may select multiple options)
 One or more Social Media Accounts
 One or more Email Accounts
 My own computing device (e.g., smartphone, Laptop, Tablet, or
Desktop computer)
 My own Internet connection (e.g., Wi-Fi connection, dial-up
connection, ADSL connection, 3G/4G dongle)
 Online Bank Account / PayPal Account
 Mobile Banking Account (e.g., Easy Cash or M-Cash)
2. How do you get to know about online purchasing? (You are able to select
multiple answers)
 SMS by retailers
 Social media promotions by retailers
 Social media posts that others share
 Email flyers by retailer
 TV and radio advertisements
 Credit card promotions
 Other; Specify………………………………………….
3. How many online purchasing, transactions, and service websites do you
know and used? (e.g., – Sampath vishwa, WOW, Pick me)
Number I'm
1 2 3 4 5 6 7 8 9 10 10+
aware of

Number I
1 2 3 4 5 6 7 8 9 10 10+
have used

51
4. Which types of goods/ services do you typically purchase/ consume online?
(You may select multiple answers)
 Clothing and accessories
 Electronic equipment
 Sport items
 Stationary
 Bank and other financial services
 Tourist services (hotels, transport, car hire)
 Health-care services
 Other; Specify………………………………………….
5. How often do you execute online purchasing/transactions?
 Once or several times a week
 Once or several times a month
 Once or several times a year
 Rarely
 Never
6. What is the estimated value of all your online purchases over the last 12
months (in rupees)?
 0 – 5,000
 5,001 – 15,000
 15,001 – 50,000
 50,001 – 100,000
 Above 100,000
7. How do you pay for your online purchases? (You are able to select multiple
answers)
 Using Credit or Debit card
 Prepaid card
 Online Account (PayPal, Paytm)
 Cash on delivery
 Mobile phone payment (e.g., – Dialog easy cash, Mobitel Mcash)
 Crypto currency (e.g., bitcoin)

52
 8. Compared to your all expenses related to purchases, what percentage is
carried out as online purchases over the last 12 months (please provide best
guess)?
 0 - 2%
 3 - 5%
 6 - 10%
 11- 25%
 26 -50 %
 51% or above

From following question onward, please rate your perception on a scale of 1-10. “1”
indicates the lowest important/relevant whereas “10” indicates the highest
important/relevant.
e.g., –
1 2 3 4 5 6 7 8 9 10

9. Following factors are important to me in deciding what to purchase online:

1 2 3 4 5 6 7 8 9 10
1. Price

2. Promotions(e.g., – 10% 1 2 3 4 5 6 7 8 9 10
off)
3. Time taken for 1 2 3 4 5 6 7 8 9 10
checkout
4. Home delivery and 10
multiple pick up 1 2 3 4 5 6 7 8 9
options
1 2 3 4 5 6 7 8 9 10
5. Low delivery Fees

1 2 3 4 5 6 7 8 9 10
6. Delivery time

1 2 3 4 5 6 7 8 9 10
7. Brand and quality

1 2 3 4 5 6 7 8 9 10
8. Peer reviews

1 2 3 4 5 6 7 8 9 10
9. Customer reviews

1 2 3 4 5 6 7 8 9 10
10. Availability of options

10. I have faced difficulties in online purchases/transactions due to:

53
 ‘’Record the relevance 1 – least relevance, 10 – most relevance’’
1. Lack of choices and services 1 2 3 4 5 6 7 8 9 10
2. Lack of knowledge in using
1 2 3 4 5 6 7 8 9 10
technology
3. Payment issues (e.g., –pay
with a different currency, lack 1 2 3 4 5 6 7 8 9 10
of payment options)
4. Delivery issues 1 2 3 4 5 6 7 8 9 10
5. Lack of confidence on the
1 2 3 4 5 6 7 8 9 10
product quality
6. Internet connectivity issues
1 2 3 4 5 6 7 8 9 10
(e.g., – speed, break downs)
7. Cost associated with online
transactions (e.g., – Internet 1 2 3 4 5 6 7 8 9 10
charges)
8. Customer support 1 2 3 4 5 6 7 8 9 10
9. Safety and return policies 1 2 3 4 5 6 7 8 9 10
10. Security and privacy issues 1 2 3 4 5 6 7 8 9 10
11. Personal Financial Difficulties 1 2 3 4 5 6 7 8 9 10

11. To what extent you agree or disagree with following statements about Sri
Lankan websites/applications related to online purchases/transactions?

Strongly Strongly
Disagree Neutral Agree
disagree agree

1. I am well aware of Sri Lankan

customer protection laws with
regard to online transaction /
purchases
2. I have necessary knowledge
and skills to perform online
transactions/purchases
3. I trust online transaction
methods than traditional
methods
4. I'm satisfied with the
government involvement in
promoting online business
5. I believe websites/applications
are user friendly
6. I'm satisfied with the number,
variations, and quality of
payment methods offered to
me

54
 7. I'm satisfied with the quality of
products/services offered to me
8. I believe available payment
methods are secure
9. I'm satisfied with the
confidentiality and privacy
level maintained by the online
service providers
10. I believe websites/applications
provide good customer support

Demographic Questions
1. Gender
 Male
 Female
2. Age Group (in years)
 16 – 25
 26 – 35
 36 – 45
 46 – 55
 Above
3. What is the highest level of formal education you have completed? (If you’re
currently enrolled in school/university, please indicate the highest degree you
have received.)
 G.C.E. O/L
 G.C.E. A/L
 Certificate level
 Diploma level
 Bachelor‟s degree
 Master‟s degree
 Professional degree (e.g., MD)
 Doctorate (e.g., PhD)
 Other
o Please specify

55
 …………………………………………………………
4. What is your current employment status?
 Employed full time (35 or more hours per week)
 Employed part time (up to 35-hours per week)
 Self-employed
 Unemployed
 Housewife
 Student
 Retired
5. If you are employed, which employment category your occupation belongs
to?
 Health Sector
 Engineering Sector
 Information Technology Sector (including IT engineers)
 Legal Sector
 Finance Sector
 Educational Sector
 Administrative Sector
 Other
o Please specify
…………………………………………………………
6. What is your monthly income level (In rupees)?
 Below 25,000
 25,000 – 49,999
 50,000 – 74,999
 75,000 – 99,999
 100,000 – 149,999
 Over 150,000
7. Anything you would like to share about your experiences related to e-
commerce in Sri Lanka and how it should improve?

56
 APPENDIX C: HISTOGRAMS FOR LIKERT SCALE
QUESTIONS

SURVEY QUESTIONNAIRE; Question 11 - To what extent you agree or disagree

with following statements about Sri Lankan websites/applications related to online
purchases/transactions?
Statement 01: “I am well aware of Sri Lankan customer protection laws with
regard to online transaction / purchases”

Strongly Disagree 78
Disagree 99
Neutral 108
Agree 96
Strongly Agree 21

0 20 40 60 80 100 120
No of respondents

Statement 02: “I have necessary knowledge and skills to perform online

transactions/purchases”

Strongly Disagree 18
Disagree 18
Neutral 51
Agree 180
Strongly Agree 135
0 20 40 60 80 100 120 140 160 180 200
No of respondents

Statement 03: “I trust online transaction methods than traditional methods”

Strongly Disagree 7
Disagree 58
Neutral 159
Agree 105
Strongly Agree 50

0 20 40 60 80 100 120 140 160 180

No of respondents

57
Statement 04: “I'm satisfied with the government involvement in promoting online
business”

Strongly Disagree 137

Disagree 130
Neutral 75
Agree 40
Strongly Agree 20

0 20 40 60 80 100 120 140 160

No of respondents

Statement 05: “I believe websites/applications are user friendly”

Strongly Disagree 36
Disagree 60
Neutral 126
Agree 159
Strongly Agree 21

0 20 40 60 80 100 120 140 160 180

No of respondents

Statement 06: “I'm satisfied with the number, variations, and quality of payment
methods offered to me”

Strongly Disagree 48
Disagree 81
Neutral 111
Agree 135
Strongly Agree 27

0 20 40 60 80 100 120 140 160

No of respondents

Statement 07: “I'm satisfied with the quality of products/services offered to me”

Strongly Disagree 42
Disagree 87
Neutral 129
Agree 123
Strongly Agree 21

0 20 40 60 80 100 120 140

No of respondents

58
Statement 08: “I believe available payment methods are secure”

Strongly Disagree 45
Disagree 69
Neutral 129
Agree 144
Strongly Agree 15

0 20 40 60 80 100 120 140 160

No of respondents

Statement 09: “I'm satisfied with the confidentiality and privacy level maintained
by the online service providers”

Strongly Disagree 44
Disagree 72
Neutral 138
Agree 130
Strongly Agree 18
0 20 40 60 80 100 120 140 160
No of respondents

Statement 10: “I believe websites/applications provide good customer support”

Strongly Disagree 36
Disagree 84
Neutral 138
Agree 114
Strongly Agree 30

0 20 40 60 80 100 120 140 160

No of respondents

59
 APPENDIX D: FACTOR ANALYSIS BASED ON
DEMOGRAPHGIC CHARACTERISTICS

1. Gender

Path Coefficients Path Coefficients

(Female) (Male)
AF -> LSA 0.379 0.031
B -> LSA 0.060 0.199
DI -> LSA 0.018 -0.014
FR -> LSA 0.092 0.420
KAF -> LSA -0.389 -0.258

Path Coefficients- t-Value (Female P Value (Female

diff (Female - Male) vs Male) vs Male)
AF -> LSA 0.348 2.538 0.012
B -> LSA 0.139 0.715 0.475
DI -> LSA 0.032 0.380 0.704
FR -> LSA 0.328 1.730 0.084
KAF -> LSA 0.131 1.619 0.106

2. Educational Level
Path
Path Coefficients
Coefficients
(Masters)
(Bachelors)
AF -> LSA 0.213 0.161
B -> LSA 0.115 0.611
DI -> LSA 0.103 0.004
FR -> LSA 0.298 -0.250
KAF -> LSA -0.362 -0.331

60
 t-Value p-Value
Path Coefficients-diff (
(Bachelors vs (Bachelors vs
| Bachelors - Masters |)
Masters) Masters)

AF -> LSA 0.051 0.244 0.807

B -> LSA 0.495 1.882 0.055
DI -> LSA 0.099 0.899 0.359
FR -> LSA 0.548 2.296 0.021
KAF -> LSA 0.031 0.280 0.779

4. Employment Sector
Path Path Path
Coefficients Coefficients Coefficients
(Engineering ) (Health) ( IT)
AF -> LSA 0.484 0.125 0.043
B -> LSA 0.167 -0.150 0.091
DI -> LSA -0.345 0.008 0.597
FR -> LSA 0.588 0.689 -0.039
KAF -> LSA -0.443 -0.396 -0.325

Path Path Path t- t- t-Value p-Value p-Value p-Value

Coefficie Coefficie Coefficie Value Value (Health (Enginee (Enginee (Health
nts-diff nts-diff nts-diff (Engine (Engin vs IT) ring vs ring vs vs IT)
(Enginee (Enginee (Health - ering vs eering Health) IT)
ring - ring - IT) Health) vs IT)
Health) IT)

AF 0.358 0.440 0.082 1.652 2.068 0.429 0.100 0.040 0.669

-> LSA
B -> 0.352 0.942 0.589 1.142 3.177 1.898 0.255 0.020 0.059
LSA
DI -> 0.317 0.076 0.241 1.956 0.446 1.405 0.052 0.656 0.161
LSA
FR 0.102 0.627 0.729 0.339 2.128 2.809 0.735 0.035 0.005
-> LSA
KAF - 0.047 0.118 0.071 0.406 0.833 0.504 0.685 0.406 0.615
> LSA

61