Southern Cross University ePublications@SCU eses 2009 Factors influencing the post-adoption consequences of online securities trading in Singapore’s retail investors Anthony Yeong Southern Cross University ePublications@SCU is an electronic repository administered by Southern Cross University Library. Its goal is to capture and preserve the intellectual output of Southern Cross University authors and researchers, and to increase visibility and impact through open access to researchers around the world. For further information please contact [email protected]. Publication details Yeong, A 2009, 'Factors influencing the post-adoption consequences of online securities trading in Singapore’s retail investors', DBA thesis, Southern Cross University, Lismore, NSW. Copyright A Yeong 2009
273
Embed
Factors influencing the post-adoption consequences of online
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Southern Cross UniversityePublications@SCU
Theses
2009
Factors influencing the post-adoptionconsequences of online securities trading inSingapore’s retail investorsAnthony YeongSouthern Cross University
ePublications@SCU is an electronic repository administered by Southern Cross University Library. Its goal is to capture and preserve the intellectualoutput of Southern Cross University authors and researchers, and to increase visibility and impact through open access to researchers around theworld. For further information please contact [email protected].
Publication detailsYeong, A 2009, 'Factors influencing the post-adoption consequences of online securities trading in Singapore’s retail investors', DBAthesis, Southern Cross University, Lismore, NSW.Copyright A Yeong 2009
Submitted to The Graduate College of Management, Southern Cross University Australia
In partial fulfilment of the requirement for the degree of DOCTOR OF BUSINESS ADMINISTRATION
August 2009
- i -
I certify that the substance of the research thesis has not been submitted for any
degree and is not currently being submitted to any other degree.
I also certify that to the best of my knowledge any help received in preparing this
thesis, and all sources used have been acknowledged in this thesis.
Signed:
(Anthony Yeong)
Date: August 2009
STATEMENT OF ORIGINAL AUTHORSHIP
- ii -
The completion of this Doctor of Business Administration dissertation would not
have been possible without the support, advice and encouragement from a number of
people.
First and foremost, I would like to express my deepest appreciation to my supervisor,
Dr Chris McDowell, the Graduate College of Management, Southern Cross
University. His prompt advice and assistance has encouraged me to carry through to
the completion of this research study. He has always shown a keen interest and
patience in my research study.
Secondly, my heartfelt thanks also go to Dr Margo Poole who provided me with
excellent guidance and advice in SPSS and data analysis.
Thirdly, I am thankful to Adjunct Professor Dr. C.S. Teo who has introduced me to
this doctoral program. He has been a great mentor for the past many years and has
provided much advice and support especially during the early stage of this research
study. In addition, I am especially grateful to Sue White and Susan Riordan for their
administrative support. Their genuine support and patience were greatly appreciated.
Last, but not least, I would like to offer a special thanks to my beloved wife,
Jacqueline and daughter, Amanda. They have been supportive and understanding
during my long journey in completion of this program.
Anthony Yeong
ACKNOWLEDGEMENTS
- iii -
This research aims to establish the important dimension of pre-adoption factors’
influence on the consequences, or post-adoption usage behaviour, of online securities
trading by the retail investors in Singapore.
While several studies have been conducted by researchers on the factors that lead
people to adopt new innovations, few have actually explored the consequences of the
innovations or post-adoption usage behaviour. Therefore, this research fills the gap in
the diffusion and adoption studies. The researcher aimed to address the following
three research issues in this study:
RI1) What pre-adoption variables in the Diffusion of Innovations model affect the
post-adoption usage behaviour of retail investors using online securities trading?
RI2) Does Perceived usefulness affect the post-adoption usage behaviour of the
retail investors using online securities trading?
RI3) Will Consumer loyalty affect the post-adoption usage behaviour of retail
investors using online securities trading?
The parent discipline of this research is Consumer Behaviour which has been
elaborated on in the literature review chapter. The immediate disciplines are
Consequences of innovations from the Diffusion model; Perceived usefulness from
the Technology Acceptance Model and Consumer loyalty. They have been discussed
in the literature review chapter as well. Seven propositions have been derived from
ABSTRACT
- iv -
the literature review and have subsequently formed the hypotheses of the research
study.
A positive paradigm has been selected for this research, and the data analysis
undertaken used a quantitative method and an online survey questionnaire to gather
research data from the retail investors using online securities trading in Singapore.
There were 232 data elements collected from the online survey. The data were then
further analysed by: using factor analysis by the Varimax rotation extraction method,
conducting reliability testing using Cronbach Alpha testing, and testing of the
theoretical model and hypotheses using multiple linear regression analysis.
The findings concluded that not all of the variables in the pre-adoption factors of
Roger’s Diffusion model are influencing the post-adoption usage behaviour of the
retail investors trading stock online. The pre-adoption variables tested to have
significant influence on the post-adoption usage behaviour are: Compatibility,
Complexity, Trailability and Observability. The Optional decision variable was found
to have an influence on the post-adoption usage behaviour but not the variables
Authority decision and Collective decision.
Nature of social system and Change agent’s promotion efforts are shown to have
significant influence on post-adoption usage behaviour of the retail investors.
Perceived usefulness and Consumer loyalty variables have also been tested and
concluded to have an influence on post-adoption usage behaviour of the retail
investors trading stock online.
Finally, contributions to the knowledge, research limitations and areas for further
research, especially in the Consequences of Innovations discipline, were discussed.
- v -
From the findings of this research study, a framework has been set for future
researchers to investigate further the pre-adoption factors’ influence on the post-
adoption usage behaviour or consequences of innovations in other products and
services. In addition, this study provides some useful findings and implications for
the retail users, researchers, practitioners and brokerage firms in the area of online
securities trading usage.
- vi -
Page Statement of Original Authorship i Acknowledgements ii Abstract iii Table of Contents vi List of Figures xii List of Tables xiv List of Charts xvi
3.2 Research Paradigms 3.2.1 Positivism 3.2.2 Critical Theory 3.2.3 Constructivism 3.2.4 Realism 3.2.5 Justification of the selected research paradigm
3.3 Research Methods 3.3.1 Qualitative Research 3.3.2 Quantitative Research 3.3.3 Justification of the selected research method 3.3.4 Limitation of the selected research method
Source: developed for this research
Chapter Three: Methodology
- 107 -
3.2 Research Paradigms
As suggested by Perry (2002), it is important to understand the body of knowledge of
methodology and to choose the appropriate research paradigm which is most suitable
for the research undertaking (Perry 2002). A research paradigm is a systemic
conceptual framework which guides researchers in how to conduct appropriate
research (Guba & Lincoln 1994). There are four research paradigms: positivism,
critical theory, constructivism and realism (Perry, Riege & Brown 1999) (Guba &
Lincoln 1994). These research paradigms are reviewed in the following sections.
Guba and Lincoln (1994), define a paradigm as a set of basic beliefs that deal with
ultimates or first principles. It is a basic belief system based on ontological,
epistemological, and methodological assumptions (Guba & Lincoln 1994). Ontology
can be viewed as the reality that researchers investigate, epistemology is the
relationship between that reality and the researcher, and methodology is the
technique used by the researcher to investigate the reality (Healy & Perry 2000).
Krauss (2005) defines ontology as the philosophy of reality, epistemology addresses
how we come to know that reality, while methodology identifies the methods of
attaining the knowledge of the reality (Krauss 2005). A paradigm can also be viewed
as an overall conceptual framework within which a researcher may work (Perry,
Riege & Brown 1999). Easterby-Smith et al. have classified paradigms as positivist
paradigms and phenomenological paradigms. The basic belief of a positivist
paradigm is that the world is external and objective and the observer is independent.
The phenomenological paradigm is to believe that the world is socially constructed
Chapter Three: Methodology
- 108 -
and subjective, and that the observer is part of what is observed (Easterby-Smith,
Thorpe & Lowe 1991).
The following table based on Perry, adapted from Guba and Lincoln (1994)
summarises the four major research paradigms.
Table 3-1 Basic systems of alternative enquiry paradigms
Paradigm Item POSTIVISM CRITICAL
THEORY CONSTRUCTIVISM REALISM
ONTOLOGY naïve realism: reality is real and apprehensible
historical realism: ‘virtual’ reality shaped by social, economic, ethnic, political, cultural, and gender values, crystallised over time
critical relativism: multiple local and specific ‘constructed’ realities
critical realism: reality is ‘real’ but only imperfectly and probabilistically apprehensible and so triangulation from many sources is required to try to know it
EPISTEMOLOGY objectivist: findings true
subjectivist: value mediated findings
Subjectivist: created findings
modified objectivist: findings probably true
METHODOLOGY experiments / surveys: verification of hypotheses: chiefly quantitative methods
dialogic/dialectical: researcher is a ‘transformative intellectual’ who changes the social world within which participants live
hermeneutical/ dialectical: researcher is a ‘passionate participant’ within the world being investigated
case studies / convergent interviewing: triangulation, interpretation of research issues by qualitative and quantitative methods such as structural equation modelling
Note: Essentially, ontology is ‘reality’, epistemology is the relationship between the reality and the researcher and methodology is the technique used by the researcher to discover the reality.
Source: (Perry, Riege & Brown 1999) based on (Guba & Lincoln 1994)
Chapter Three: Methodology
- 109 -
3.2.1 Positivism
According to Easterby-Smith et al. (1991), the positivist view is that the social world
exists externally and that the facts of this social world can be discovered by a set of
scientific methods. The researcher using this world view is usually using a
quantitative method where the technique can provide wider coverage of the range of
situations, and is fast and economical (Easterby-Smith, Thorpe & Lowe 1991). The
investigator and the investigated ‘object’ are assumed to be independent entities. The
researcher is able to study the object without influencing it or being influenced by it
(Guba & Lincoln 1994). The objectives of the research based on a positivism
paradigm include the measurement and analysis of causal relationships between
variables that are consistent across time and context (Perry, Riege & Brown 1999).
The ontology perception of positivism is that the researcher discovers a single
apprehensible reality concerning a research problem based on independent
observation, and the resulting knowledge is considered to be trustworthy (Guba &
Lincoln 1994).
The epistemology of the positivist paradigm indicates that the inquiry takes place as
if through a one-way mirror according to Guba and Lincoln (1994). The researcher
prevents any influence on the outcomes by his own values and biases. The findings
should be repeatable and true in nature. The investigator and the investigated ‘object’
are independent from each other (Guba & Lincoln 1994).
The methodology commonly used in the positivism paradigm includes hypotheses in
the form of propositions which are subjected to an empirical test to verify them.
Chapter Three: Methodology
- 110 -
There are controlled conditions to prevent research outcomes being improperly
influenced (Guba & Lincoln 1994).
3.2.2 Critical Theory
The second paradigm, critical theory, is where reality is held to be based on historical
structures, and the researcher aims at criticising and transforming social, political,
cultural, economic, ethnic and gender values. Marxists, feminists and action research
all fall under this category of research paradigm (Perry, Riege & Brown 1999).
The ontology of critical theory is considered as historical realism, where the reality
was shaped over time (Guba & Lincoln 1994).
The epistemology of critical theory assumes that the investigator and the investigated
object are interactively linked, and that the values of the investigator influence the
inquiry (Guba & Lincoln 1994). Perry et al. state that this paradigm is not appropriate
for marketing research unless the researcher aims to be part of the investigation
(Perry, Riege & Brown 1999).
The methodology of critical theory requires the researcher to have dialogue with the
subjects of the inquiry, the dialogue must be dialectical in nature, and truth should be
reasoned out from the dialogue (Guba & Lincoln 1994).
Chapter Three: Methodology
- 111 -
3.2.3 Constructivism
The third paradigm, constructivism, assumes truth is a construction which refers to a
particular belief system held in a particular context. Meaning has more value than the
measurement, and the perception is the most important reality. It enquires about the
ideologies and values underlying the research findings (Perry, Riege & Brown 1999).
The ontology of constructivism argues that humans construct knowledge and
meaning from their experiences (Guba & Lincoln 1994). The realities in the
constructivism paradigm appear as multiple realities which are socially and
experientially based, and are the intangible mental constructions of individual
persons (Perry, Riege & Brown 1999). Perry et al. argue that the constructivist
approach is rarely appropriate for business research because the approach excludes
concerns about the economic and technological dimensions of business (Perry, Riege
& Brown 1999).
Guba and Lincoln (1994) describe the methodology of constructivism as
hermeneutical and dialectical. It is suggested that the variables of the research can be
elicited and refined through interaction between and among the investigator and the
respondents. The researcher is a ‘passionate participant’ within the world being
investigated (Guba & Lincoln 1994).
Chapter Three: Methodology
- 112 -
3.2.4 Realism
The last research paradigm to be discussed is the realism paradigm. It assumes that
reality exists but is imperfect because of the flaw in human intellectual mechanisms
and the fundamentally intractable nature of phenomena (Guba & Lincoln 1994). It
has the elements of both positivism and constructivism (Perry, Riege & Brown 1999).
The realism paradigm is also known as critical realism or post-positivism (Guba &
Lincoln 1994).
The ontology of the realism paradigm assumes that reality is ‘real’ but only imperfect
and probabilistically apprehensible and so triangulation from many sources is
required to try to understand the reality (Guba & Lincoln 1994; Perry, Riege &
Brown 1999).
The epistemology of the realism paradigm is considered as modified objectivist and
assumes that the findings of the researcher are probably true but always subject to
falsification. The findings of the reality have to fit the pre-existing knowledge (Guba
& Lincoln 1994).
The methodology of the realism paradigm includes case studies and interviewing.
The interpretation of the findings can be by quantitative methods and / or qualitative
methods (Perry, Riege & Brown 1999).
Chapter Three: Methodology
- 113 -
3.2.5 Justification of the selected research paradigm
The above section has reviewed four different types of research paradigms. A
positivism paradigm has been selected for this research for the following reasons:
a. From the ontology perspective, the post-adoption usage behaviour of online
securities trading is an independent reality. A positivist approach is suited to the
measurement and analysis of causal relationships between the pre-adoption variables
and post-adoption variables.
b. The pre-adoption and post-adoption variables related to online securities trading
are quantifiable where personal perceptions are important; they can be quantified by
asking participants to ‘score’ perceptions on a scale.
c. From an epistemological perspective, the researcher is independent of the subjects
in this research. It is free of personal bias or values. The researcher has no influence
on the data collected from the retail investors. So researcher bias or interaction with
subjects will have no influence on the research findings.
d. The researcher does not have any relationship with the survey participants and is
also not part of the object of this particular research.
e. This research is about observable phenomena in a market. It is appropriate
therefore, from a methodology perspective, that the hypotheses of this research are
stated in propositional form and subjected to empirical tests to verify them.
In summary, the research findings are in a quantifiable form to verify the causal
relationships between the variables. The researcher is independent and not
Chapter Three: Methodology
- 114 -
influencing or influenced by this research. Thus, a positivism paradigm has been
selected.
3.3 Research Methods
There are two main types of research methods: quantitative research that can be
associated with a positivist paradigm, and qualitative research that is associated with
a realism paradigm. The strength of quantitative research methods is that they can
provide wide coverage of the range of situations and a survey is fast and economical.
The strength of qualitative research is the ability to look at change processes over
time, to understand people’s meanings, to adjust to new issues and ideas as they
emerge, and the findings can contribute to a new theory (Easterby-Smith, Thorpe &
Lowe 1991).
The following table illustrates the characteristics of quantitative paradigm and
qualitative paradigm.
Table 3-2 Characteristics of quantitative and qualitative paradigms
Qualitative Paradigm Quantitative Paradigm 1. Qualitative method preferred. 1. Quantitative methods preferred. 2. Concerned with understanding human behaviour from the actor’s frame of reference.
2. Seeks the facts or causes of social phenomena without advocating subjective interpretation.
25 UBS Securities Pte. Ltd No 26 UOB Kay Hian Private Limited Yes 27 Westcomb Securities Pte Ltd
No
The proportion of stock trading volume online in Singapore is relatively low as
compared to other Asian countries like Korea and Japan. In one of the Singapore
Exchange conferences, the Head of Retail Investing of SGX has highlighted that only
10 percent of total stock turnover on the SGX is done via online trading (Liew 2006).
Based on the research figures provided by SGX, it is estimated that there were
170,000 active online traders in Singapore for the year of 2006 (Liew 2006).
Source: (SGX 2008)
Chapter Three: Methodology
- 122 -
3.4.2 Accessible Population
As mentioned in the earlier section, the targeted survey objects are the adopters of
Online Securities Trading from the Online Brokerage firms in Singapore. There are
six major Online Securities Trading service providers in Singapore as listed below
(SGX 2006):
Table 3-5 Securities trading members with online trading service
S/No Securities Trading Members with online
trading service
Website
1 POEMS by Phillip Securities www.poems.com.sg
2 Fraser Securities by AmFraser Securities
Pte Ltd
www.amfraser.com.sg
3 Lim & Tan Securities www.limtan.com.sg
4 DBS Vickers Securities Online (S) Pte
Ltd
www.dbsvonline.com.sg
5 DMG Online by DMG & Partners
Securities
www.dmg.com.sg
6 iOCBC by OCBC Securities Pte Ltd www.iocbc.com
(SGX 2006)
Chapter Three: Methodology
- 123 -
The six major brokerage firms account for about 50 percent of the online securities
trading users - around 85,000 subscribers. If permission was granted from all six of
the major brokerage firms to access these subscribers, this would be the estimated
total of users accessible through the brokers who provide an online securities trading
service.
Using this accessible population, a sample could be obtained for further analysis to
test the hypotheses in this research.
The details of selecting the sample and sample size will be discussed in the next
section.
3.5 Sampling
3.5.1 Sampling Design
According to Zikmund (2000), there are 4 common sampling designs used by
researchers (Zikmund 2000):
Simple Random Sampling
A simple random sample requires that every unit in the population has a known and
equal chance of being selected. An example is the drawing of participants’ names
from a hat to get the lucky gift in a party.
Chapter Three: Methodology
- 124 -
Systematic Random Sampling
Systematic random sampling requires the units in the population to be ordered in
some way or another. For example, names that are ordered alphabetically, or
customers who walk into an entrance or follow one another (Zikmund 2000).
Cluster Sampling
A cluster sample is a sample where the units making up the population and sample
are divided into clusters. They are also known as first-stage units and primary
sampling units. Each cluster has more basic units called second-stage units or
secondary sampling units. The cluster is usually divided by geographic area and used
by census surveys (Zikmund 2000).
The geographic region in this research is restricted to Singapore and the sample is
drawn from the retail investors and no further classification is necessary for the
survey.
Stratified Sampling
For stratified sampling, the researcher divides the population into groups and
randomly selects subsamples from each group (Zikmund 2000).
Stratified sampling allows the large population to be divided into several subgroups
or strata to draw a random type sample from each stratum. The online retail investors
can be divided into six stratums represented by the major brokerage firms in
Singapore.
Chapter Three: Methodology
- 125 -
However, the most prominent brokerage firms are Phillip Securities and Fraser
Securities, and they could represent the typical retail investors who trade online in
Singapore.
The online securities trading service by Phillip Securities is called Phillip Online
Electronic Securities Mart System (POEMS). The POEMS customers are the
adopters to be surveyed in this research. POEMS has been selected as it is the earliest
online securities trading tool adopted in Singapore, as well as having the largest
number of subscribers compared with other brokerage firms (Smart Investor
Magazine 2000). POEMS has also been awarded as the Broker of the Year 2000 by
Finance Magazine and Channel News Asia in Singapore (Tan, A. 2000). POEMS
has been awarded the Hitwise Singapore Online Performance Award for 2005 and
2006 as the No.1 ranking website most visited by Singapore Internet users (POEMS
2008).
In addition to POEMS, FraserDirect Online Securities Trading service by Fraser
Securities has been selected, as it has one of the most vibrant chat rooms providing
cyber-advisors for the online investors. (Financial Planner Magazine, July 1999)
FraserDirect is also the most advertised brokerage firm in Singapore. Fraser
Securities is the nation’s oldest brokerage firm and has been established since 1873
(AMInvestment 2007). Both the users of POEMS and FraserDirect Online have been
selected as the survey sample.
Chapter Three: Methodology
- 126 -
3.5.2 Sampling Size
The researcher is not able to get the number of retail investors subscribed to online
securities trading offered by Phillip Securities and Fraser Securities, as both
companies have declined to disclose the numbers. It is estimated that there were
170,000 active online traders in Singapore for the year of 2006 (Liew 2006). As the
samples are gathered from only the retail investors using POEMS and FraserDirect,
the available population will be lower than 85,000.
3.5.3 Validity and Reliability
Sampling Validity
Sampling validity refers to the idea that the sample must allow the research process
to measure what the researcher intend to measure (Zikmund 2000). Pelham and
Blanton (2003) state that validity refers to the relative accuracy or correctness of a
research statement (Pelham & Blanton 2003). This will only be achieved if the
sample is taken from a population which is valid in relation to the research statement.
It is established by the degree to which the measure confirms a network of related
hypotheses generated from the theoretical model. The sampling targets are selected
from the adopter group and so are closely related to the Diffusion model and the
hypotheses derived from it. The hypotheses are about adopter behaviour, and the
post-adoption usage of online securities trading.
Chapter Three: Methodology
- 127 -
Construct Validity
Zikmund (2000) states that construct validity is about the ability of a measure from
the research to confirm the network of variables related to the hypotheses generated
from the theoretical model. To achieve construct validity, the researcher must
establish convergent and discriminant validity. A theoretical model has convergent
validity when it is highly correlated with different measures of similar constructs
(Zikmund 2000). For example, there are repeated questions to be asked which relate
to a single variable in a questionnaire.
The research uses Cronbach’s Alpha coefficient to test scale reliability for the
variables constructed. Shorter-Judson uses Cronbach’s Alpha for reliability testing in
a similar research (Shorter-Judson 2000).
Sampling Reliability
Zikmund (2000) defines reliability as the ability to provide consistent results in
repeated uses of the measuring instrument (Zikmund 2000). It is about the degree of
measures that are free from error. There are two underlining dimensions of reliability:
repeatability and internal consistency.
3.6 Questionnaire Design
3.6.1 Questionnaire Objective
In Chapter Two, the literature on Consumer behaviour, Diffusion of Innovation
Model, Technology Acceptance Model and online securities trading in Singapore
were reviewed. The focus of the research is to investigate the correlation between
Chapter Three: Methodology
- 128 -
pre-adoption factors which influence adoption of online securities trading, and the
consequences, or behaviour, of adopting the online securities trading services by the
retail investors in Singapore. This is to address the research issues identified in
Chapter One:
RI1) What pre-adoption variables in the Diffusion of Innovations model affect the
post-adoption usage behaviour of retail investors using online securities trading?
RI2) Does Perceived usefulness affect the post-adoption usage behaviour of retail
investors using online securities trading?
RI3) Will Consumer loyalty affect the post-adoption usage behaviour of retail
investors using online securities trading?
The literature is built on Roger’s Diffusion of Innovations Model (Rogers 2003) and
Perceived usefulness derived from Davis’ Technology Acceptance Model (Davis, F.
D. 1989). In addition, Consumer loyalty factors have also been taken into
consideration for the study (Oliver 1999).
Generally, it is proposed in Chapter Two that there may be a relationship between
factors preceding adoption and patterns of usage which follow adoption. To test this
relationship, a number of pre-adoption factors are envisaged, derived from the
models of Diffusion of Innovations and Technology Acceptance. These were
explained in Chapter Two. In addition to these, a measure of post-adoption behaviour,
or usage, of the technology is envisaged. Thus, from the general proposition that
post-adoption behaviour might be influenced by specific pre-adoption factors, a
Chapter Three: Methodology
- 129 -
number of specific propositions were developed in Chapter Two. Each proposition
relates to a specific pre-adoption factor. So the form of each proposition is that post-
adoption usage of the technology, or post-adoption behaviour, will be influenced by
each of the specific pre-adoption factors.
From these propositions in Chapter Two, seven hypotheses were developed in this
chapter. Zikmund (2000) states a hypothesis is a proposition that can be tested using
empirical evidence. Thus, hypothesis is an empirical statement concerned with the
relationship among the variables to be studied (Zikmund 2000).
Deriving from the propositions, the hypotheses state that there will be a significant
relationship between each pre-adoption factor and post-adoption behaviour. To test
these hypotheses, it was necessary to find a measurable, or quantifiable construct for
each pre-adoption factor and for post-adoption behaviour.
In order to specify the factor, a questionnaire was devised which asked a large
number of specific questions about what influenced the decision to adopt the
technology of online trading of securities. A large number of questions were devised
so that quantifiable answers to each question could become a variable, where a
number of variables would form a factor as derived from the models in the literature.
As well, for post-adoption behaviour, a number of aspects of behaviour were
incorporated into questions about the extent or degree of post-adoption usage of the
technology. The questionnaire was divided into parts, where each part comprised
questions, as variables, related to the pre-adoption factors outlined in Chapter Two.
Chapter Three: Methodology
- 130 -
There are seven hypotheses established for the research as stated in Figure 3.2. For
the questions 1a, 1b, 1c, 1d, 1e, 1f, 1g , 1h , 1i an 1j, they are addressing the element
related to Perceived attributes of innovation which form hypothesis 1. To answer
questions related to type of innovation-decision associated with hypothesis 2,
questions are found in 2a, 2b, 2c, 2d, 2e and 2f. In regard to Communication channels
related to hypothesis 3, questions are asked in 3a, 3b, 3c and 3d. The Nature of social
system variables are found in hypothesis 4 and the questions are addressed in 4a, 4b,
4c and 4d. For hypothesis 5 that is regarding the Extent of the change agent’s
promotion efforts, questions are found in 5a, 5b and 5c. Hypothesis 1, hypothesis 2,
hypothesis 3, hypothesis 4 and hypothesis 5 are founded on the literature review of
Everett Rogers. Using the Technology Acceptance Model, hypothesis 6 has been
established, and questions 7a, 7b and 7c are to address the correlation between
Perceived usefulness and post-adoption behaviours of the retail investors. In
hypothesis 7, which is to address the question related to Consumer loyalty and post-
adoption behaviours, the questions are found in 8a, 8b, 8c, 9a, 9b and 9c. Table 3.6
summarises the hypotheses in this research and the respective questions indicated in
the questionnaire.
Chapter Three: Methodology
- 131 -
Figure 3-2 Theoretical model developed for this research
Source: developed for this research
Chapter Three: Methodology
- 132 -
Table 3-6 Literature based constructs, hypotheses and survey questions
HYPOTHESIS Variables and Related literature
Questions
H1: A relationship exists between the perceived attribute factors and post-adoption behaviour.
Relative advantage (Rogers 2003)
Q1(a) - Quicker to trade online Q1(b) - Cheaper to trade online
Compatibility (Rogers 2003)
Q1c -Process is not much different from calling the broker Q1d - Trading information is not much different from calling the broker
Complexity (Rogers 2003)
Q1e- Easier to access investment information online Q1f – Easier to trade using Online Securities Trading
Trialability (Rogers 2003)
Q1g -Able to do a trial trade which is not possible via the broker Q1h - Easier to obtain an online demonstration and explanation
Observability (Rogers 2003)
Q1i - More investors signed up for Online Securities Trading system Q1j - More investors started to trade using Online Securities Trading
H2: A relationship exists between the type of innovation-decision and post-adoption behaviour.
Optional (Rogers 2003)
Q2a – Have considered other options like automatic voice trading or WAP trading via phone? Q2b – Have considered online trading as an additional method of trading
Collective (Rogers 2003)
Q2c – Have consulted other investors using online securities trading Q2d – Have consulted my friends using online securities trading
Authority (Rogers 2003)
Q2e – Have been advised by investment experts to sign up online securities trading Q2f – Stock Exchange has liberalised brokerages’ commission rate for online securities trading
H3: A relationship exists between the communication channels and post-adoption behaviour.
Mass media (Rogers 2003)
Q3a - Online advertisements like Internet or email Q3b - Mass media like TV or newspaper advertisements
Interpersonal (Rogers 2003)
Q3c - Broker's explanation and demonstration Q3d - Friends' and other investors' advice
H4: A relationship exists between the nature of the social system and post-adoption behaviour.
Norms (Rogers 2003)
Q4a – Know more friends who used online securities trading Q4b – I feel left out if I do not sign up online securities trading
Degree of network interconnectedness (Rogers 2003)
Q4c – Consult or discuss with friends or other investors when I trade online Q4d – Exchange information with friends or other investors using online securities trading
H5: A relationship exists between the extent of change agent’s promotion efforts and
The change agent’s promotion efforts (Rogers 2003)
Q5a - Constantly received Online Securities Trading information Q5b - Satisfaction with the broker's
Chapter Three: Methodology
- 133 -
post-adoption behaviour. promotional efforts on Online Securities Trading Q5c - Satisfaction with the brokerage firm's promotional efforts on Online Securities Trading
H6: A relationship exists between the Technology Acceptance Model and post-adoption behaviour.
Perceived usefulness (Davis, F. D. 1989)
7a - Trading Online increases my trading profit 7b - The system facilitates diversification of my portfolio 7d - I can react to the stock market quicker
H7: A relationship exists between consumer loyalty and post-adoption behaviour.
8a – Online securities trading will be my major method to trade 8b – I will introduce online securities trading to non-users 8c – I will not consider other new method of trading in near future
As shown in Table 3.6, there are seven groups of independent variables and one
group of dependent variables that is, the post-adoption consequences. In total, there
are 42 variables including both independent variables and dependent variables.
Beside the independent variables constructed from Rogers’ Diffusion of Innovation
Model (2003), Davis’ Acceptance of Technology Model (1998) and the Customer
loyalty, demographic characteristics of the retail investors have been gathered to
illustrate the respondents’ profiles (see Table 3.7).
Table 3-7 Demographic factors and related questions
Demographic Factors Questions Age Part C Q13 Education Part C Q14 Occupation Part C Q15 Income Part C Q16 Marital Status Part C Q17 Gender Part C Q18
3.6.4 Dependent Variable and Components
In this dissertation, the dependent variable is mainly the change of behaviour or the
consequences in the retail stock investors after adopting Online Securities Trading.
The variable has several components that will be described.
The following table, Table 3.8 lists the dependent variable and components versus
the questions to be tested in the questionnaire.
Source: designed for this research
Chapter Three: Methodology
- 137 -
Table 3-8 Post-adoption behaviour and related questions
Dependent Variable Components Questions
Post-Adoption Behaviour Frequency Part B Q11a I trade more frequently now
Volume Part B Q11b I trade in smaller lot sizes now
Trading Type Part B Q11c I buy certain categories of stock (eg. High Tech Stocks)frequently
Location Part C Q11d I trade from more locations like office, Cybercafé and overseas in addition to home
Investment tips exchange Part C Q11e I exchange investment tips with other online investors easily
Investment information Part C Q11f I check investment information frequently
3.7 Mode of Survey
3.7.1 Web Survey / Email
The primary mode of data gathering was to undertake a survey via the Internet. The
questionnaire was firstly designed in a manual form and then converted to HTML
(Hyper Text Mark-up Language) to be hosted on a web server. The web page was
then established with an address that could be accessed via the Internet. For this
research, the web design and hosting for the questionnaire used FormSite, a third
Source: designed for this research
Chapter Three: Methodology
- 138 -
party online survey tool. FormSite is owned by Vronam Systems Inc and has been a
tool to build forms easily on the web since 1998 (Formsite 2008).
The researcher used Formsite to build the questionnaire (appendix C) and publish it
on the Internet. The website of the online questionnaire was :
http://fs3.formsite.com/onlinetrading/index.html
The website of the online questionnaire was sent to Phillip Securities and Fraser
Securities respectively to request them to make it known to their online securities
trading subscribers in Singapore. The researcher also posted the website of the online
questionnaire to the online discussion forum of POEMS provided by Phillip
Securities and that of FraserDirect, provided by Fraser Securities, which were
accessible by their online securities trading subscribers.
Once respondents completed the questionnaire form on the website, the FormSite
server was arranged to send an email to alert the researcher. The result also resided
on the FormSite web server so that it could be downloaded by the researcher.
The Online Survey was preferred in this dissertation as it has been proven successful
in other similar surveys conducted in the areas of adoption related research (Chea &
Luo 2007; Tan, M. & Teo 2000; Vitartas et al. 2007). The samples are drawn from
retail investors using online securities trading and it was convenient for this group of
online users to respond to the survey through an online questionnaire when they saw
the survey website through the online forums.
Chapter Three: Methodology
- 139 -
The following table, Table 3.9 is a comparison between an Internet survey and a mail
survey conducted by the National University of Singapore (Tan, M. & Teo 2000).
Table 3-9 Internet survey versus mail survey
Characteristics Mail Survey Internet Survey
Manpower Insert survey into envelopes, paste stamps or frank envelopes.
Design Web page and write Javascript. For this research, the Web page used FormSite that could automate the web design.
Cost Envelopes, stamps, photocopying of questionnaires.
Rental/maintenance of server space to host Web page. The Online questionnaire was rented on FormSite Web Server.
Sampling frame Restricted to sample that received the questionnaire.
Restricted to people with access to the Internet who chose to respond.
Response rate Can be computed. Percentage of respondents is dependent on follow-up mailings.
Cannot be computed. Response dependent on publicity of the survey as well as follow-up reminders via emails to potential respondents.
Time frame Usually takes about a month for surveys to be returned.
Responses can usually be collected within two weeks.
Quality of data Dependent on whether targeted respondents respond to questionnaire. Systematic bias is reduced with the use of random sample and also by obtaining a high response rate.
Adequate if the targeted respondent is the general Internet user population. Potential for systematic bias if only people with certain characteristics respond.
Generalisability of results Results generalisable to target population if response rate is adequate.
Difficult to determine since there may be systematic bias in terms of who actually responds to the questionnaire.
Suitability Must be able to identify potential respondents. Can reach out to the general public regardless of computer access.
Survey of people with Internet access. Very suitable as only people with Internet access would be part of target population of online traders of securities.
Chapter Three: Methodology
- 140 -
Problems Costly and slow.
Unable to control who responds. Data must be screened for unsuitable respondents.
Pilot Testing
The researcher conducted some pilot testing for the online web survey by sending it
to DBA students as well as MBA students from Southern Cross University in
Singapore. The pilot survey population size was about 100 and 13 responded to the
survey. Based on the feedback, the respondents were able to understand the online
questionnaire easily and were able to complete all the questions without any
difficulty.
3.8 Data Processing Procedures
3.8.1 Descriptive statistics - Cross Tabulation
This questionnaire contained a section to collect the statistics of the demographic
factors of the retail investors.
Zikmund (2000) defines descriptive analysis as the process of transforming raw data
into a format that is easy to understand and interpret (Zikmund 2000).
Zikmund (2000) defines a categorical variable as any variable that has a limited
number of distinct values like gender. A continuous variable, according to Zikmund
(2000), is any variable that has an infinite number of values, like time taken to
complete a test (Zikmund 2000).
Source: (Tan, M. & Teo 2000)
Chapter Three: Methodology
- 141 -
For this study, descriptive statistics about the population are collected with questions
that provide categorical responses.
A cross tabulation method is used to describe the demographic data gathered before
getting into further detailed study of other data.
Cross tabulation can be defined as the tables displaying the number of cases falling
into each combination of the categories of two or more categorical variables. The
table may display the counts, percentages, expected values, and residuals.
Following is an example of cross tabulation (see Table 3.10) comparing the age of
the retail investors and their respective occupations. For example, in the respondents
of between the age ranges of 21 to 30, 29 out of 100 are professional.
Table 3-10 Cross tabulation of age versus occupation
The β coefficients show the relationship between Xi and Y. Each of the coefficients
β1, β2, β3, β4, β5, β6 and β7 is tested for significance using the ‘t’ statistics. An
examination of the t-values indicates the contribution of the independent variables to
the dependent variable (Coakes, Steed & Price 2008). Using the significant t-value of
the β coefficients, it is able to draw a conclusion about the significance of any
attribute. Thus, a test of significance can be used as a test of a hypothesis that a
significant relationship exists between Xi and Y. The statistical analysis software
package SPSS is used here to conduct regression analysis.
SPSS provides the R Square value in multiple regression analysis that indicates how
much of the variance in the dependent variable is explained by the regression model
which made up of the independent variables if the regression equation is to be used
for predictive purpose (Pallant 2007). In the first run of the multiple regressions, if a
certain coefficient of the Xi is not significant, it should be removed and the regression
testing re-run. In the final model, only the coefficients that are statistically significant
are measured to determine the variance between the dependent variable and
Chapter Three: Methodology
- 150 -
independent variables. In this research, the objective is not prediction, but to test the
significance of relationships.
The steps and results of the multiple regression analysis are presented in Chapter
Four.
3.8.5 Data Processing Tools
The data processing tool used is SPSS version 15.0 which is a common statistical
analysis package. The descriptive statistics use cross tabulation to illustrate the
demographic profiles of the respondents. In this research, the factor analysis and
regression analysis of the data gathered is conducted using SPSS.
3.8.6 Ethical Considerations
Ethical considerations are important when dealing with research involving human
beings. The research methodology and questionnaire has been reviewed and
approved by the Southern Cross University’s Human Research Ethics Committee.
The research was conducted in accordance with the guidelines and policies provided
by the committee. The committee’s approval number for this research is ECN-03-20
(see Appendix A).
The respondents to the online survey are not forced into answering the
questionnaires; they are given the free choice to participate in the survey.
At any point of time during the survey, respondents are free to terminate the survey at
their will. Another consideration is the privacy of the respondents; the researcher will
Chapter Three: Methodology
- 151 -
not publish or reuse the respondent’s information anywhere else besides this research
without the consent of the respondents.
Lastly, the researcher will not mislead the respondents in any form but will follow
closely the guidelines and policies provided by the committee.
The researcher will not utilise the results of the research in other sources without
prior consent from the respondents. The researcher is to be honest and keep the
integrity of the survey’s results and not distort or manipulate the results dishonestly.
The researcher will avoid disseminating a conclusion of the research that is
inconsistent with the survey’s results.
Chapter Three: Methodology
- 152 -
3.9 Summary of the chapter
After reviewing the literature in Chapter Two, a unique theoretical model for this
research has been established using Roger’s Diffusion of Innovations Model and the
Perceived usefulness factor derived from Davis’ Technology Acceptance Model,
with the additional independent variable of Consumer loyalty factors.
The model was adapted and used as a basis to test for a proposed relationship
between the attributes of technology, online securities trading, and the degree of post-
adoption usage of the technology.
In this chapter, various research paradigms have been reviewed, and from those a
positive approach was selected. Following a review of alternative methodological
approaches, the research deemed most appropriate was a quantitative approach to test
the model. Data was collected using an online survey of users of the new technology.
The chapter then follows with an outline of the specific way in which questions in the
survey were specified, and how responses were analysed using factor analysis and
regression analysis.
In Chapter Four, the results of data analysis are explained and interpreted.
Chapter Four: Data Analysis
- 153 -
Chapter 4 Data Analysis
4.1 Introduction to Chapter Four
The methods of data collection and data analysis have been discussed in previous
chapters. Proceeding from Chapter Three, this chapter focuses on the analysis of data
with the intention of testing hypotheses formulated from a proposed theoretical
model.
There are seven sections in this chapter as indicated in Figure 4.1. The first section is
the introduction. Section two presents the profile of the sample from which data have
been collected. A cross tabulation technique is used to present the information.
Section three describes the development of constructs in this research and examines
the reliability of the data through Cronbach’s alpha technique.
In section four, factor analysis is used to reduce a large number of variables into
significant factors of online trading behaviour. From these factors, a revised set of
hypotheses is proposed. The factor scores are retained to form construct variables that
are used for multiple linear regression analysis. Regression analysis is used to test the
revised hypotheses. These regression statistics are represented in section five. The
results of hypothesis tests are explained in section six.
The last section summarises the topics covered in this chapter.
Chapter Four: Data Analysis
- 154 -
Figure 4-1 Overview of Chapter Four
4.1 Introduction to Chapter Four
4.2 Data Profile Examination 4.2.1 Data summary 4.2.2 Demographic profiles
4.4 Factor analysis4.4.1 Factor analysis for independent variables4.4.2 Reliability testing of factors (independent variables)4.4.3 Factor analysis for independent variables4.4.4 Reliability testing of factors (dependent variable)
4.5 Multiple regression analysis4.5.1 Multiple regression model4.5.2 Model summary – R Square
4.5.3 ANOVA Table 4.5.4 Model Parameters
4.6 Hypothesis testing 4.6.1 Test of hypothesis 1a 4.6.2 Test of hypothesis 1b 4.6.3 Test of hypothesis 2a 4.6.4 Test of hypothesis 2b
4.6.5 Test of hypothesis three 4.6.6 Test of hypothesis four 4.6.7 Test of hypothesis five 4.6.8 Test of hypothesis six 4.6.9 Test of hypothesis seven 4.6.10 Summary of hypothesis testing
4.7 Summary of the chapter
4.3 Development of Constructs 4.3.1 Reliability Analysis
Source: Developed for this research
Chapter Four: Data Analysis
- 155 -
4.2 Data Profile Examination
4.2.1 Data Summary
There were 246 responses collected via the online survey questionnaires as explained
in Chapter Three. Out of the 246 data elements, 9 records are incomplete and 5
records are inaccurate. The inconsistent records are removed leaving 232 records as
the raw data for analysis.
4.2.2 Demographic Profiles
Data was collected on demographic variables. Presented below is a summary of the
demographic profile of the sample based on age, education, occupation, income,
marital status and gender.
Age
Of the total population (N = 232), 43.1 percent of valid respondents are of the age
range from 21 to 30, while 40.9 percent are aged 31 to 45. Just 12.9 percent of the
sample is 46 to 55 years old, and only 3.0 percent are 56 years old and above. In
summary, around 83 percent of the sample is aged 21 to 45 years. Chart 4.1
illustrates this age distribution.
Chapter Four: Data Analysis
- 156 -
Chart 4-1 Age profiles of respondents
Age
Age
56&up46-5531-4521-30
Freq
uenc
y
120
100
80
60
40
20
0
Source: Survey data
Education
Of the 232 respondents, only 5.2 percent reported having Secondary and below
education. 12.9 percent or 30 respondents had Junior College and equivalent level.
The second largest group of respondents is those with a Diploma and equivalent
which make up 28.0 percent. The largest group of respondents is those with a
Bachelor degree and equivalent which is 38.4 percent of the sample. There are 15.5
percent of respondents with a Master degree and above. Chart 4.2 illustrates the
education levels of the sample respondents. When Chart 4.1 and 4.2 are considered
Chapter Four: Data Analysis
- 157 -
together, it is apparent that the sample is generally below middle age and quite well
educated.
Chart 4-2 Education profiles of respondents
Education
Education
Master degree & abovBachelor degree & eq
Diploma & equivalentJunior College & equ
Secondary & below
Freq
uenc
y
100
80
60
40
20
0
Source: Survey data
Occupation
From the records gathered, 23.7 percent of people who responded are holding
executive and managerial positions. 62 professionals responded which is about 26.7
percent of the total population and it is the largest group of people. People who are
holding administration, sales and services related positions make up 18.1 percent,
while 11.2 percent are self –employed. Only 3.4 percent or 8 cases out of the total
Chapter Four: Data Analysis
- 158 -
population reported an occupation identified as ‘others’. Chart 4.3 illustrates the
occupation categories of the sample respondents.
Chart 4-3 Occupation profiles of respondents
Source: Survey data
Income
There were 11.6 percent of the respondents who reported an income of $20,000 and
below. The next income range, from $21,000 to $35,000, has 19.0 percent of
respondents. 30.6 percent of respondents had an income level of $35,000 to $50,000.
There are 19.0 percent of respondents with an income level range from $50,0001 to
$65,000. The next income range, $65,001 to $80,000 has 10.8 percent of respondents.
There are 4.3 percent of respondents belonging to an income range of $80,001 to
Chapter Four: Data Analysis
- 159 -
$100,000. There are only 4.7 percent of respondents with an income level more than
$100,001. Chart 4.4 illustrates the income distribution of the sample respondents with
a model income range of $30,000 to $50,000; the remainder of the sample is
reasonably even, though quite widely distributed around that.
Chart 4-4 Income profiles of respondents
Income
Income
100,001 & above80,001 - 100,000
65,001 - 80,00050,001 - 65,000
35,000 - 50,00021,000 - 35,000
20,000 & below
Freq
uenc
y
80
60
40
20
0
Source: Survey data
Marital Status
In the marital status factor, the single respondents are the largest group which
is 51.7 percent of the total sample population. Respondents who are married make up
47 percent. Only 1.3 percent respondents indicated ‘others’ in the marital status.
(refer to Chart 4.5)
Chapter Four: Data Analysis
- 160 -
Chart 4-5 Martial status of respondents
marital status
marital status
OthersMarriedSingle
Freq
uenc
y
140
120
100
80
60
40
20
0
Source: Survey data
Gender
From the data of 232 respondents, 72 percent are male while only 28 percent are
female. Chart 4.6 shows that more than twice as many males have responded to the
survey as compared to females.
Chapter Four: Data Analysis
- 161 -
Chart 4-6 Gender profiles of respondents
genderFemaleMale
Cou
nt
200
150
100
50
0
Source: Survey data
Based on the above demographic profiles, most of the respondents belong to the age
group of 21 to 30, with a Bachelor degree and equivalent, working as professionals
with an average income level of $35,000 to $50,000, and are single males.
The average age range of the respondents is between the ages of 31 to 45. The
general education level of the respondents is a Diploma and equivalent. The
respondents are from a broad spectrum of occupational categories.
Chapter Four: Data Analysis
- 162 -
4.3 Development of Constructs
The objective of this study is to test the relationship between the pre-adoption
variables and post-adoption behaviour of retail investors trading stocks online. The
data gathered are to study the research issues addressed in Chapter One:
RI1) What pre-adoption variables in the Diffusion of Innovations model affect the
post-adoption usage behaviour of retail investors using online securities trading?
RI2) Does Perceived usefulness affect the post-adoption usage behaviour of retail
investors using online securities trading?
RI3) Will Consumer loyalty affect the post-adoption usage behaviour of retail
investors using online securities trading?
As the research is based on Roger’s Diffusion Theory and Davis’s Technology
Acceptance Model as well as the inclusion of a Consumer loyalty factor, there are a
large number of variables built into the theoretical model as illustrated in Chapter
Two. The large number of variables, both dependent and independent, was reduced to
meaningful constructs through factor analysis. Cronbach’s alpha tests were used to
test the reliability of the constructs. A small number of construct variables reduces
the large number of dependent variables to a small set of variables that can be used
for further analysis. By ensuring that that the construct variables were reliable, the
researcher was able to have greater confidence in subsequent analysis. The new
construct variables were further analysed using regression analysis to test the
theoretical model.
Chapter Four: Data Analysis
- 163 -
4.3.1 Reliability Analysis
There are several types of reliability testing but the most commonly used is
Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative %Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings
Extraction Method: Principal Component Analysis.
Source: Developed for this research
Chapter Four: Data Analysis
- 175 -
Scree Plot
The Eigenvalues extraction in the previous section indicated that nine factors have
been identified and this can be further verified by using a Scree Plot in which the
Eigenvalues are plot from the largest value to the smallest value. In this Scree Plot,
there seemed to be a few bends, the bends occurred at component three, six and seven.
However, as shown in chart 4.7, the slope seemed to stop changing significantly after
the seven components. The Scree Plot does not appear to be very useful for
interpretation, so it was decided to use nine components for further factor analysis
testing as indicated by choosing factors with an Eigenvalue greater than 1. Some
additional factor analysis tests were done using three factors, six factors and seven
factors extraction, some failed to have coverage in rotation, and some did not provide
results that are meaningful in the context of the analysis. As a result, the nine factors
extraction fell into a pattern that has the most meaningful interpretation. However,
two of the components in the theoretical model have broken into two factors and are
explained in the next section.
Chapter Four: Data Analysis
- 176 -
Chart 4-7 Scree Plot for independent variables
Source: Developed for this research
Rotated Factor Matrix
A common method of rotation is Varimax, which is an orthogonal method of rotation,
in which the loadings for the factors are given by the projections of each plotted point
onto the new axes specified by the rotation. The Varimax criterion essentially drives
squared loadings towards the end of the range 0 to 1, and negative loadings towards -
1, 0 or 1 and away from intermediate values (Hair et al. 1998). In an orthogonally
rotated factor matrix, factors are uncorrelated (Rummel 1967). This makes them
more suitable for use as a basis for creating independent variables for use in multiple
Chapter Four: Data Analysis
- 177 -
regression analysis, as an underlying assumption in regression analysis is that the
independent variables are not correlated with each other. When any independent
variables in a multiple regression are highly correlated a potential problem of
muticollinearity exists (Zikmund 2000). In this research, orthogonally rotation is
selected as an appropriate method. Factor scores are to be used to create construct
variables for regression and it is desirable that these new construct variables should
not be correlated with each other. Regression variable created from factor scores
computed from factor loadings on each orthogonal component resolve the issues of
multicollinearity in multiple regeression (SPSS 2000).
This research will suppress the variables with absolute values less than 0.5 and those
sorted for earlier interpretation (Field 2005). The loadings of the factor analysis are
sorted and the report has suppressed the output with values less than 0.5. The
suppression of the output with values less than 0.5 has facilitated a better
interpretation of the results.
Table 4.5 shows a Varimax rotated factor matrix. The items in the matrix have been
sorted so that items (variables) that load onto a factor are grouped together, listed in
decreasing order of factor loading scores. All factor loading scores less than 0.5 have
been suppressed from this table to enable ease of factor interpretation.
Chapter Four: Data Analysis
- 178 -
Table 4-5 Rotated component matrix for independent variables
In this section, the hypotheses are tested and discussed based on the outputs from the
multiple regression analysis. There are seven null hypotheses in the original
theoretical model derived from the seven research propositions. Two of these were
expanded, and so here nine hypotheses will be tested to be accepted or rejected in the
following sections. Statistical analyses for the multiple regression model to test the
hypotheses were conducted through the use of SPSS version 15.0 for Windows. The
interpretations of the hypotheses are presented below.
4.6.1 Test of Hypothesis 1a
(H1a)0: There is no correlation between FS4-H1a (Trialability and Observability) and
post-adoption behaviour of retail investors using online securities trading.
Table 4.25 shows that for the coefficient of the construct variable FS4-H1a, the t
statistic value is bigger than 1 and the Sig. value is less than 0.05 (p < 0.05), so the
null hypothesis is rejected and the alternative hypothesis is accepted (see Table 4.26).
This means that the independent variables, Trialability and Observability have a
positive and significant impact on the overall post-adoption behaviour of the retail
investors.
4.6.2 Test of Hypothesis 1b
(H1b)0:There is no correlation between FS6-1b (Compatibility and Norm) and post-
adoption behaviour of retail investors using online securities trading.
For the coefficient of construct FS6-1b, the t statistic value is bigger than 1 and the
Sig. value is less than 0.05 (p < 0.05) and so the null hypothesis is rejected and the
Chapter Four: Data Analysis
- 205 -
alternative hypothesis is accepted (see Table 4.26). This means that the independent
variables, Compatibility and Norm have a positive and significant influence on the
dependent variable.
It is concluded that the null hypothesis 1 is rejected and the alternative hypothesis is
accepted. The researcher concludes that the independent construct variable, Perceived
Attributes of Innovation has a positive and significant influence on the post-adoption
behaviour of retail investors using online securities trading. The construct contains
the variables Compatibility and Norm that were included in the original Hypothesis 1
of the theoretical model.
4.6.3 Test of Hypothesis 2a
(H2a)0: There is no correlation between FS8-H2a (Authority decision and Collective
decision) and post-adoption behaviour of retail investors using online securities
trading.
For construct FS8-H2a, the t statistic value is 0.768 which is smaller than 1 and the
Sig. value is 0.443 (p > 0.05) (see Table 4.26). This means that the null hypothesis
for the independent variables, authority decision and collective decision is accepted.
The construct FS8-H2a, that contains the independent variables authority decision
and collective decision, has no significant influence on the dependent variable in the
regression model.
Chapter Four: Data Analysis
- 206 -
4.6.4 Test of Hypothesis 2b
(H2b)0:There is no correlation between FS9-H2b(Optional decision) and post-
adoption behaviours of retail investors using online securities trading.
The finding from the construct FS9-H2b shows that the t value is bigger than 1 and
the Sig. value is 0.001 (p < 0.05) (see Table 4.32). This means that the null
hypothesis for this construct is rejected and the alternative is accepted. The
independent variables, Optional decision1 and Optional decision2 indeed have a
positive and significant impact on the post-adoption behaviour.
In the original theoretical model, the Type of Innovation Decision variable includes
Optional decision, Collective decision and Authority decision. However, in the new
model, the Type of Innovation Decision variable has been divided into two construct
variables. The construct variable FS8-H2a which includes the Authority decision and
Collective decision has accepted the null hypothesis (See Table 4.26). The construct
variable FS9-H2b which includes the variable Optional decision1 and Optional
decision2 has rejected the null hypothesis (See Table 4.26). It can be concluded from
the findings that only the construct variable which consists of the Optional decision
variable has an influence on post-adoption behaviour of the retail investors.
4.6.5 Test of Hypothesis Three
(H3)0: There is no correlation between FS5-H3 (Communication channels) and post-
adoption behaviour of retail investors using online securities trading.
As shown in Table 4.25, the t statistic value for FS5-H3 is 1.710 and the Sig. value is
0.089 (p > 0.05).
Chapter Four: Data Analysis
- 207 -
It is suggested that the predictor FS5-H3 has no significant association to the
dependent variable of the regression model. The null hypothesis is accepted and the
alternative hypothesis is rejected. This means that there is no significant correlation
between the pre-adoption communication channels and the post-adoption behaviour
of retail investors using online securities trading.
4.6.6 Test of Hypothesis Four
(H4)0: There is no correlation between FS2-H4 (Nature of the social system) and
post-adoption behaviour of retail investors using online securities trading.
As shown in Table 4.25, the t value for the coefficient of this construct is 9.557 (t >
1) and the Sig. value is less than 0.05 (p < 0.05) and hence the predictor FS2-H4 has
significant association with the dependent variable of the regression model. The null
hypothesis is rejected and the alternate hypothesis is accepted. It is suggested that the
nature of the social system construct is positively influencing the dependent variable
post-adoption behaviour of retail investors using online securities trading.
4.6.7 Test of Hypothesis Five
(H5)0: There is no correlation between FS7-H5 (Extent of change agent’s promotion
efforts) and post-adoption behaviour of retail investors using online securities trading.
As shown in Table 4.25, the t statistic value for this construct is 3.724 (t > 1) and
the Sig. value is less than 0.05 (p < 0.05) and hence the predictor FS7-H5 has a
positive and significant impact on the dependent variable of the regression model.
The null hypothesis is rejected and the alternative hypothesis is accepted. We can
conclude that the extent of change agent’s promotion efforts has a positive and
Chapter Four: Data Analysis
- 208 -
significance influence on the post-adoption behaviour of retail investors using online
securities trading.
4.6.8 Test of Hypothesis Six
(H6)0: There is no correlation between FS3-H6 (Perceived usefulness and
Complexity) and post-adoption behaviour of retail investors using online securities
trading.
As shown in Table 4.25, the t statistic value for this construct is 7.492 (t > 1) and
the Sig. value is less than 0.05 (p < 0.05) and hence the predictor FS3-H6 has a
positive and significant impact on the dependent variable of the regression model.
The null hypothesis is rejected and the alternate hypothesis is accepted, that is, that
there is a positive correlation between the construct variable FS3-H6, made up of the
variables Perceived usefulness and Complexity or the low level of complexity, and
post-adoption behaviour of retail investors using online securities trading.
4.6.9 Test of Hypothesis Seven
(H7)0: There is no correlation between FS1-H7 (consumer loyalty) and post-adoption
behaviour of retail investors using online securities trading.
As shown in Table 4.25, the value of the t-statistic (7.225) is significant at 0.000 (p <
0.05).
This means that the construct FS1-H7 has a positive and significant impact on the
dependent variable of the regression model. Thus, the null hypothesis is rejected and
the alternative hypothesis is accepted. The researcher concludes that consumer
Chapter Four: Data Analysis
- 209 -
loyalty has a positive and significant influence on the post-adoption behaviour of
retail investors using online securities trading.
4.6.10 Summary of Hypotheses Testing
From the result of testing all the nine hypotheses as shown in Table 4.26; only the
null hypotheses for H2a and H3 were accepted. The research findings suggested that
FS4-H1a (Trialability and Observability), FS6-H1b (Compatibility and Norm), FS9-
H2b (Optional decision), FS2-H4 (Nature of the social system), FS7-H5 (Extent of
change agent’s promotion efforts), FS3-H6 (Perceived usefulness and Complexity)
and FS1-H7 (Consumer loyalty) are the influential factors in the post-adoption
behaviour of retail investors using online securities trading.
The research findings are somewhat different from the theoretical model in Chapter
Two as in H1, the perceived attributes of innovations has been divided into two
Hypotheses, H1a and H1b based on the factor analysis results. However, both H1a
and H1b are rejected, giving support to the original theoretical model.
The Hypothesis 2 in Chapter Two, Type of Innovation-Decision has also been
divided into two Hypotheses, H2a and H2b. H2a consists of the Authority decision
and Collective decision variables while H2b consists of the Optional decision
variables as a result of the factor analysis. H2a is accepted and H2b is rejected,
suggesting that the model in Chapter Two must be adapted to reflect the influence of
optional decisions only.
The hypotheses testing has also accepted (H3)0. This means that Communication
channels have no significant impact on the post-adoption behaviour of the retail
investors using online securities trading.
Chapter Four: Data Analysis
- 210 -
Table 4-26 Summary of the results of hypotheses testing
Null Hypothesis Parameter Result (H1a)0: FS4-H1a (Trialability and Observability) has no association with post-adoption behaviour.
Significant at 0.011 Reject the null hypothesis
(H1b)0: FS6-H1b (Compatibility and Norm) has no association with post-adoption behaviour.
Significant at 0.000 Reject the null hypothesis
(H2a)0: FS8-H2a (Authority decision and Collective decision) has no association with post-adoption behaviour.
Significant at 0.443 Accept the null hypothesis as insufficient significant variables.
(H2b)0: FS9-H2b (Optional decision) has no association with post-adoption behaviour.
Significant at 0.001 Reject the null hypothesis
(H3)0: FS5-H3 (Communication channels) has no association with post-adoption behaviour.
Significant at 0.089 Accept the null hypothesis
(H4)0: FS2-H4 (Nature of the social system) has no association with post-adoption behaviour.
Significant at 0.000 Reject the null hypothesis
(H5)0: FS7-H5 (Extent of change agent’s promotion efforts) has no association with post-adoption behaviour.
Significant at 0.000 Reject the null hypothesis
(H6)0: FS3-H6 (Perceived usefulness and Complexity) has no association with post-adoption behaviour.
Significant at 0.000 Reject the null hypothesis
(H7)0: FS1-H7 (Consumer loyalty) has no association with post-adoption behaviour.
Significant at 0.000 Reject the null hypothesis
Source: Developed for this research
Figure 4.2 below shows the revised theoretical model with the significant
independent variables of Perceived attributes of innovations, Type of innovation-
decision, Nature of the social system, Extent of change agent’s promotion efforts,
Perceived usefulness, Consumer loyalty and the dependent variable, post-adoption
Chapter Four: Data Analysis
- 211 -
behaviour of the retail investors using online securities trading. It is somewhat
different from the conceptual theoretical model described in Chapter Two (see Figure
2.15) as the Communication channels variable has been removed, as well as some
components of other variables. The relative advantage component of the Perceived
attributes of innovations variable has been removed in the revised model. The
Perceived attributes of innovations variable in the original conceptual model is now
divided into two independent variables as a result of the factor analysis. The Type of
innovation-decision in the original model is now left with the Optional decision
variable as shown in Figure 4.2. The Interpersonal component is now merged with
the Nature of the social system independent variable since the component could be
considered as part of the social system. The variable Complexity is referring to a low
level of complexity, or the use of online securities trading being effort-free in this
research context. The Complexity component is now merged with the Perceived
usefulness variable as Complexity is closely related to the Perceived ease of use
component in the Technology Acceptance Model described in Chapter Two. There
are no changes in the Extent of change agent’s promotion efforts variable; Consumer
loyalty variable and post-adoption usage behaviour dependent variable as compared
to the conceptual model developed in Chapter Two shown in Figure 2.17.
Chapter Four: Data Analysis
- 212 -
Figure 4-2 Revised theoretical model
Hypothesis 7
Perceived Attributes of Innovations (H1a)
- Trailability- Observability
Extent of Change Agent’s Promotion Efforts
- Perceived Usefulness
Customer Loyalty
- Loyalty- Confidence
Independent variables Dependent variables
Hypothesis 1a
Hypothesis 2b
Hypothesis 4
Hypothesis 5
Nature of the Social System- Norms- Degree of network interconnectedness
Type of Innovation-Decision- Optional decision
Hypothesis 6
Post-Adoption Usage Behaviour of Online Securities Trading
-
- Type of shares- Location- Volume- Frequency- Investment tips exchange- Investment information frequency
Perceived Attributes of Innovations (H1b)
- Compatibility- Norm
Hypothesis 1b
- Interpersonal
- Complexity
Source: Developed for this research
Chapter Four: Data Analysis
- 213 -
4.7 Summary of the chapter
The chapter started with presenting the data profile for the data collected through the
online survey. The demographic profiles of the sample have been described. The
majority of the respondents are of the age group of 21 to 30, with a Bachelor degree
and equivalent, working as professionals with an income level of $35,000 to $50,000,
and are single males, though a wide cross section of occupations and income brackets
is represented by the sample.
The survey collected data on a large number of variables reflecting factors
influencing the adoption of online security trading. Factor analysis was undertaken
on these many variables to reduce the data to nine common factors. A single factor
representing post-adoption security trading was also derived. Testing was conducted
on these factors to ensure reliability. Factor scores were then used to create nine
independent construct variables and one dependent construct variable.
To test the hypotheses that these nine independent pre-adoption variables had a
significant influence on post adoption behaviour, a regression model was devised.
Standard linear multiple regression was conducted with nine independent variables.
Hypothesis testing was undertaken by testing the statistical significance of the
coefficient associated with the nine independent construct variables. Of the nine null
hypotheses, seven were rejected and two accepted. Thus, the researcher concluded
that seven of the construct variables had a positive and significant influence on post
adoption behaviour of retail security trading.
Chapter Four: Data Analysis
- 214 -
A revised theoretical model, adapted from the one derived in Chapter Two is
presented, incorporating the findings of the statistical analysis.
The interpretation of the results will be discussed in the conclusion, Chapter Five.
The chapter also will summarise this research and its limitations to provide
recommendations for future research.
Chapter Five: Conclusions and implications
- 215 -
Chapter 5 Conclusions and implications
5.1 Introduction to Chapter Five
In this final chapter, the conclusions and implications of the research findings and the
implications derived from the thesis are discussed. The chapter is organised into eight
sections and concludes with a final statement of the thesis. Following the introduction
is a restatement of the research problem and issues. In section three, results of
hypotheses testing are summarised. The theoretical implications of the research
findings are described in section four with a statement of what the findings have
contributed to the body of knowledge. Managerial implications are presented in
section five. Section six acknowledges the limitations of this research. In section
seven, recommendations for future research, based on the current study and findings,
are presented. The last section concludes the findings and summarises this research.
All the sections mentioned are summarised in the Figure 5.1.
Chapter Five: Conclusions and implications
- 216 -
Figure 5-1 Overview of Chapter Five
5.1 Introduction to Chapter Five
5.2 Restatement of the research problem and hypotheses
5.3 Conclusion about the hypotheses and research problem5.3.1 Conclusion for hypothesis one5.3.2 Conclusion for hypothesis two5.3.3 Conclusion for hypothesis three5.3.4 Conclusion for hypothesis four5.3.5 Conclusion for hypothesis five5.3.6 Conclusion for hypothesis six5.3.7 Conclusion for hypothesis seven
5.4 Contributions to the body of knowledge5.4.1 Consequences of Innovations5.4.2 Technology Acceptance Model5.4.3 Consumer loyalty
5.6 Limitations of the research5.5.1 Scope limitation5.5.2 Geographical limitation5.5.3 Online questionnaire
5.7 Recommendations for future research5.6.1 Other products5.6.2 Other geographical regions5.6.3 Other factors affecting post-adoption consequences5.6.4 Other post-adoption consequences
5.8 Research conclusion
5.5 Managerial Implications
Source: Developed for this research
Chapter Five: Conclusions and implications
- 217 -
5.2 Restatement of the research problem and hypotheses
The main objective of this research is to investigate the extent to which pre-adoption
factors influence the consequential behaviour of retail investors using online
securities trading. The following research issues were addressed:
RI1) What pre-adoption variables in the Diffusion of Innovations model affect the
post-adoption usage behaviour of retail investors using online securities trading?
RI2) Does Perceived usefulness affect the post-adoption usage behaviour of the
retail investors using online securities trading?
RI3) Will Consumer loyalty affect the post-adoption usage behaviour of retail
investors using online securities trading?
The following null hypotheses were tested. These are adapted from the literature
review chapter and modified after preliminary data analysis using the methods
described in Chapter Three - Methodology.
Nine hypotheses were tested. They are restated below.
(H1a)0: There is no correlation between the construct variable FS4-H1a (Trialability
and Observability) and post-adoption behaviours of retail investors using online
securities trading.
(H1b)0: There is no correlation between construct variable FS6-H1b (Compatibility
and Norm) and post-adoption behaviours of retail investors using online securities
trading.
Chapter Five: Conclusions and implications
- 218 -
(H2a)0: There is no correlation between construct variable FS8-H2a (Authority
decision and Collective decision) and post-adoption behaviours of retail investors
using online securities trading.
(H2b)0: There is no correlation between construct variable FS9-H2b (Optional
decision) and post-adoption behaviours of retail investors using online securities
trading.
(H3)0: There is no correlation between construct variable FS5-H3 (Communication
channels) and post-adoption behaviours of retail investors using online securities
trading.
(H4)0: There is no correlation between construct variable FS2-H4 (Nature of the
social system) and post-adoption behaviours of retail investors using online securities
trading.
(H5)0: There is no correlation between construct variable FS7-H5 (Extent of change
agent’s promotion efforts) and post-adoption behaviours of retail investors using
online securities trading.
(H6)0: There is no correlation between construct variable FS3-H6 (Perceived
usefulness and Complexity) and post-adoption behaviours of retail investors using
online securities trading.
(H7)0: There is no correlation between construct variable FS1-H7 (Consumer
loyalty) and post-adoption behaviours of retail investors using online securities
trading.
Chapter Five: Conclusions and implications
- 219 -
5.3 Conclusion about the hypotheses and research
problem
The theoretical model of this research was developed and discussed in Chapter Two
(see Table 2.17). From the research propositions formed in Chapter Two, hypotheses
were established and tested using the methodology outlined in Chapter Three. The
research results are detailed in Chapter Four and the conclusions of the hypotheses
test are presented in the following section. A summary of the test results are listed in
Table 5.1
Table 5-1 Summary of results from testing of hypotheses
Null Hypothesis Result Interpretation (H1a)0: FS4-H1a (Trialability and Observability) has no association with post-adoption usage behaviour.
Reject the null hypothesis
Trialability and Observability positively influence post-adoption usage behaviour.
(H1b)0: FS6-H1b (Compatibility and Norm) has no association with post-adoption usage behaviour.
Reject the null hypothesis
Compatibility and Norm positively influence post-adoption usage behaviour.
(H2a)0: FS8-H2a (Authority decision and Collective decision) has no association with post-adoption usage behaviour.
Accept the null hypothesis There is no influence of Authority decision and Collective decision on post-adoption usage behaviour.
(H2b)0: FS9-H2b (Optional decision) has no association with post-adoption usage behaviour.
management and online financial information. Conclusions may be drawn that
similar effects may apply in these online environments, but further research into these
specific areas would be needed to confirm these conclusions.
5.6.2 Geographical limitation
The sampling for this research is limited to retail investors in Singapore only and
does not include other retail investors from other countries. The trading behaviours
and usage of the online securities trading tool could be different for retail investors
from other countries although the research should lead to reasonably general
conclusions that would apply in other countries. The pre-adoption factors influencing
the retail investors from other countries could be different as compared to the retail
investors in Singapore though there would be many similarities. Thus, the pre-
Chapter Five: Conclusions and implications
- 233 -
adoption factors might have a different degree of influence on post-adoption usage
behaviour of retail investors.
5.6.3 Online questionnaire
The use of an online questionnaire limits the pool for sampling to Internet users only.
However, those people who do not have access to the Internet are not the target
sampling for this research. Non-Internet users are not likely to be traders online, so
this limitation of the sample pool is not considered a significant delimitation. The
research is only limited to experienced Internet users who are comfortable in
responding to the online questionnaire, based on the assumption that they are capable
of accessing the Internet since they are adopters of Internet based online securities
trading which is also based on Internet and web browser technologies.
5.7 Recommendations for future research
5.7.1 Other products
The replication of this research can be applied to other online products, for example
online banking or online financial investment. Another possible product for further
research on pre-adoption influence on post-adoption usage is the trading of foreign
exchange currencies online. Future research could also be carried out in other
industries beside financial, for example, online education, or online purchases of
consumer goods, or online purchases of business inputs products. The theoretical
model in this research can be a recommended model to investigate how pre-adoption
factors and Technology acceptance factors influence post-adoption usage behaviour
of other innovations, products or services.
Chapter Five: Conclusions and implications
- 234 -
5.7.2 Other geographical regions
The replication of this research could be applied in other geographical regions. This
research is focused on the retail investors in Singapore and using the online securities
trading tool based in Singapore. The findings might be different if the study was to be
applied in other countries which would add more value to the area of study.
5.7.3 Other factors affecting post-adoption consequences
The research has investigated pre-adoption factors as well as Technology acceptance
factors that affect the post-adoption consequences of online securities trading,
however there may be many other factors that influence trading behaviour. Future
research should build on the findings of this study and should try to identify
additional factors associated with post-adoption consequences of online securities
trading. For example, the latest technology infrastructure and enhanced product
functionality might be of interest to future studies on post-adoption behaviour. The
concerns of security threats in trading stock online might have an influence on post-
adoption behaviour as well.
5.7.4 Other post-adoption consequences
One might also attempt to include other post-adoption behaviour or consequences of
using online securities trading by the retail investors. For example, one might use the
model to study whether there are impacts of pre-adoption factors on financial returns
in using online securities trading. Another possible post-adoption behaviour could be
the trading strategy or patterns of trading by the retail investors trading online.
Chapter Five: Conclusions and implications
- 235 -
5.8 Research conclusion
In this chapter, the final conclusions and implications of this research are presented.
Firstly, the research problem and the related research issues of this research are
restated.
Many researches in the past have been conducted in understanding the factors
affecting the adoption of new innovations. However, few studies have been
conducted in the pre-adoption factors and Technology Acceptance Model factors
influencing the post-adoption usage of an innovation. This study found that there are
positive influences of pre-adoption factors, Technology Acceptance Model factors
and Consumer loyalty on the post-adoption usage of online securities trading by the
retail investors in Singapore. The research also found that external factors like
Authority decision, Collective decision and Communication channels have no
significant influence on post-adoption usage of trading stock online. These imply that
brokerage firms should perhaps focus on internal factors affecting the retail investors,
to increase the probability of higher usage of the online securities trading and thus
generate higher transaction commissions for the company.
Implications were drawn on how this research contributed to the existing body of
knowledge and the field of online securities trading. This was followed by a
discussion of the research limitations, and recommendations for future research were
explored. By understanding which pre-adoption factors influence post-adoption usage
behaviour, online brokerage companies can focus their product development and
marketing communication strategies to increase the use of online securities trading
Chapter Five: Conclusions and implications
- 236 -
and thus expand their businesses. Based on the current theoretical model, future
researchers could investigate how pre-adoption factors influence post-adoption usage
behaviour of other similar products and services.
- 237 -
Bibliography and References 1
Ajzen, I 1985, 'From intentions to actions: A theory of planned behavior.' In J.Kuhi & J. Beckmann (EDs.), Action - control: From cognition to behavior, pp. 11-39.
Ajzen, I & Fishbein, M 1980, Understanding Attitudes and Predicting Social
Behavior., Prentice-Hall, New Jersey. AMInvestment 2007, Fraser Direct Online Securities, <http://www.ambg.com.my/>. Antonides, G & Raaij, WFv 1998, Consumer Behaviour A European Perspective,
First Edition edn, Wiley. Bass, FM 1969, 'A New Product Growth For Model Consumer Durables',
Management Science, vol. 15, no. 5, p. 215. Bhattacherjee, A 2001, 'Understanding Information Systems Continuance: An
Expectation Confirmation Model', MIS Quarterly, vol. 25, no. 3, pp. 351-70. Bryman, A 1984, 'The Debate about Quantitative and Qualitative Research: A
Question of Method or Epistemology?' The British Journal of Sociology, vol. 35, no. 1, pp. 75-92.
CDP 2008, Securities Account for an Individual 2008, <http://www.cdp.com.sg/account/create_individual.html>. Chea, S & Luo, MM 2007, 'Cognition, Emotion, Satisfaction, and Post-Adoption
Behaviors of E-Service Customers', paper presented to 40th Annual Hawaii International Conference, Hawaii.
Cheung, CM, Chan, GW & Limayem, M 2005, 'A Critical Review of Online
Consumer Behaviour: Empirical Research', Journal of Electronic Commerce in Organizations, vol. 3, no. 4, pp. 1-19.
Chua, HH 2008, 'Power failure down broker's trading system', Straits Times, 16 Jan
2008. Coakes, sJ, Steed, L & Price, J 2008, SPSS Version 15.0 for Windows: Analysis
without Anguish, First edition edn, Wiley. Davis, F, Bagozzi, RP & Warshaw, PR 1989, 'User Acceptance of Computer
Technology: A Comparison of Two Theoretical Models', Management Science, vol. 35, no. 8, p. 982.
- 238 -
Davis, FD 1989, 'Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology', MIS Quarterly, vol. 13, no. 3, p. 319.
Dearing, J & Singhal, A 2006, Communication of Innovations: A Journey with Everett
M. Rogers., Sage, London. DeForge, K 2001, Getting Started in Online Broker, 1 edn, Wiley. Deshpande, R 1983, 'Paradigms Lost: On Theory and Method in Research in
Marketing', Journal of Marketing, p. 101. Dillon, A & Morris, MG 1996, 'User Acceptance of Information Technology:
Theories and Models', Annual Review of Information Science and Technology, vol. 31, pp. 3 - 32.
Donio, J, Massari, P & Passiante, G 2006, 'Customer satisfaction and loyalty in a
digital environment: an empirical test', Journal of Consumer Marketing, vol. 23, no. 7, pp. 445-57.
Easterby-Smith, M, Thorpe, R & Lowe, A 1991, The Philosophy of Research Design,
Sage, London. EL-Gayar, O & Moran, M 2007, 'Examing Students' Acceptance of Tablet PC using
TAM', Issues in Information Systems, vol. VIII, no. 1. Field, A 2005, Discovering statistics using SPSS 2nd edition edn, Sage Publications,
London. Fishbein, M & Ajzen, I 1975, Belief, attitude, intention, and behavior: An
introduction to theory and research, Addison-Wesley. Floh, A & Treiblmaier, H 2006, 'WHAT KEEPS THE E-BANKING CUSTOMER
LOYAL?' Journal of Electronic Commerce Research, vol. 7, no. 2. Formsite 2008, Formsite website, viewed 2008 <Error! Hyperlink reference not
valid.. Garber, LL & Dotson, MJ 2002, 'A method for the selection of appropriate business-
to-business integrated marketing communications mixes', Journal of Marketing Communications, vol. 8, pp. 1-17.
Gatignon, H & Robertson, TS 1985, 'A Propositional Inventory for New Diffusion
Research', Journal of Consumer Research, vol. 11, no. 4, p. 849.
- 239 -
---- 1989, 'Technology Diffusion: An Empirical Test of Competitive Effects', Journal of Marketing, vol. 53, pp. 35-49.
Gefen, D, Karahanna, E & Straub, DW 2003, 'TRUST and TAM in online shopping:
An Integrated Model', MIS Quarterly, vol. 27, no. 1, p. 51. Gharavi, H, Love, PED & Cheng, EWL 2004, 'Information and communication
technology in the stockbroking industry: and evolutionary approach to the diffusion of innovation', Industrial Management & Data Systems, vol. 104, no. 9, pp. 756-65.
Glaser, M 2003a, Online Broker Investors: Demographic Information, Investment
Strategy, Portfolio Positions, and Trading Activity, Universitat Mannheim, 2003.
Benefits and Relationship Quality', Journal of Service Research, vol. 4, no. 3, pp. 230-47.
Iyengar, JV 2004, 'A Discussion of current and potential issues relating information
security for Internet communications', Competitiveness Review, vol. 14, no. 1/2, p. 90.
Jacoby, J & Kyner, DB 1973, 'Brand Loyalty Vs. Repeat Purchasing Behavior',
Journal of Marketing Research, vol. 10, pp. 1-9.
- 240 -
Jasperson, JS, Carter, PE & Zmud, RW 2005, 'A Comprehensive Conceptualization Of Post-Adoptive Behaviors Associated With Information Technology Enabled Work Systems', MIS Quarterly, vol. 29, no. 3, p. 525.
Johnson, RB & Onwuegbuzie, AJ 2004, 'Mixed Methods Research: A Research
Paradigm Whose Time Has Come', Educational Researcher, vol. 33, no. 7, p. 14.
Adoption by SMEs in Singapore', paper presented to 34th Hawaii International Conference on System Sciences, Hawaii.
Kong, W 1999, 'Internet Trading: A New Singapore Passtime?' Financial Planner,
July 1999, pp. 10-21. Kotler, P 2000, Marketing Management, International Edition edn, Prentice Hall. Krauss, SE 2005, 'Research Paradigms and Meaning Making: A Primer', The
Qualitative Report, vol. 10, no. 4, pp. 758-70. Kurnia, S & Chien, A-WJ 2003, 'The Acceptance of Online Grocery Shopping', paper
presented to 16th Bled eCommerce Conference, Slovenia, 9-11 June 2003. Lau, A, Yen, J & Chau, PYK 2001, 'Adoption of On-line Trading in the Hong Kong
Financial Market', Journal of Electronic Commerce Research, vol. 2, no. 3, p. 2001.
Lavidge, RJ & Steiner, GA 1961, 'A Model For Predictive Measurements of
Advertising Effectiveness', Journal of Marketing, pp. 59-62. Liew, A 2006, 'SGX StockWhiz Online Trading Challenge: "Seize The Day!"', paper
presented to SGX StockWhiz, Singapore, 6 December 2006. Luo, MM-L, Remus, W & Chea, S 2006, 'Technology Acceptance of Internet-based
Information Services: An Integrated Model of TAM and U&G Theory', paper presented to Twelfth Americas Conference on Information Systems, Mexico, August 4 2006.
MAS 2001, Securities and Futures Act (Chapter 289).
- 241 -
Mason, RD, Lind, DA & Marchal, WG 2000, Statistical Techniques in Business and Economics, Tenth Edition edn, McGraw Hill.
McKechnie, S, Winklhofer, H & Ennew, C 2006, 'Applying the technology
acceptance model to the online retailing of financial services', International Journal of Retail & Distribution Mangement, vol. 34, no. 4/5.
McMullan, R & Gilmore, A 2008, 'Customer loyalty: an empirical study', European
Journal of Marketing, vol. 42, no. 9/10, pp. 1084-904. Naidoo, R & Leonard, A 2007, 'Perceived usefulness, service quality and loyalty
incentives: Effects on electronic service continance', South African Journal of Business Management, vol. 38, no. 3.
Neal, C, Del, P & Haskins, Q 2004, Consumer Behaviour - Implications for
Marketing Strategy 4th Edition edn, McGraw Hill. Oliver, RL 1999, 'Whence Consumer Loyalty?' Journal of Marketing, vol. 63, p. 33. Pallant, J 2007, SPSS Survival Manual, 3rd Edition edn, A&U. Parhasarathy, M & Bhattacherjee, A 1998, 'Understanding Post-Adoption Behavior in
the Context of Online Services', Information Systems Research, vol. 9, no. 4. Park, K 1998, 'INNOVATIVE PRODUCT USAGE BEHAVIOR IN THE POST-
ADOPTION PROCESS', Asia Pacific Advances in Consumer Research vol. 3. Parthasarathy, M & Bhattacherjee, A 1998, 'Understanding Post-Adoption Behavior
in the Context of Online Services ', Information Systems Research, vol. 9, no. 4. Pelham, BW & Blanton, H 2003, Conducting Research in Psychology, Second
Edition edn, Thomson. Perry, C 2002, A Structured Approach To Presenting Theses: Notes For Students And
Their Supervisors. Perry, C, Riege, A & Brown, L 1999, 'Realism's Role Among Scientific Paradigms in
Rigopoulos, G & Askounis, D 2007, 'A TAM Framework to Evaluate Users' Perception towards Online Electronic Payments', Journal of Internet Banking and Commerce, vol. 12, no. 3.
Robertson, TS & Gatignon, H 1986, 'Competitive Effects on Technology Diffusion',
Journal of Marketing, vol. 50, no. 3, p. 1. ---- 1991, 'How Innovators Thwart New Entrants into Their Market', Planning Review,
vol. 19, no. 5, p. 4. Rogers, EM 2003, Diffusion of Innovations, Fifth Edition edn, Free Press. Rummel, RJ 1967, 'Understanding Factor Analysis', The Journal of Conflict
Resolution, vol. 11, no. 4, pp. 444-80. Schiffman, L, Bednall, D, Cass, AO, Paladino, A & Kanuk, L 2005, Consumer
Behaviour, 3rd Edition edn, Pearson. SGX 2006, Annual Report, Singapore. ---- 2008, 2008, <Error! Hyperlink reference not valid.. Shankar, GD 2002, 'Impact of e-Business on stock broking', paper presented to
Australian Computer Society Australia, 26 Jan 2002. Sharma, MK & Bingi, P 2000, 'The Growth Of Web-Based Investement', Information
Systems Management. Sharma, P & Maleyeff, J 2003, 'Internet education: potential problems and solutions',
The International Journal of Educational Management, vol. 17, no. 1, pp. 19-25.
Shorter-Judson, BG 2000, 'Adoption and Diffusion of Innovations in the Airline
Industry: An Investigation of Consumer Preference for Alternative Ticketing Methods (The Case of Electronic Ticketing)', PhD thesis, Golden Gate University.
Smart-Investor 2000, 'e-Trading: Phase Two', Smart Investor, pp. 50-6. Sohn, SY & Ahn, BJ 2003, 'Multigeneration diffusion model for economic
assessment of new technology', Technological Forecasting & Social Change, vol. 70, pp. 251–64.
SPSS 2000, Advanced Statistical Analysis Using SPSS, SPSS Inc. Strong, EK 1925, The Psychology of Selling, McGraw Hill, New York. Talyor, S & Todd, P 1995, 'Decomposition and crossover effects in the theory of
planned behavior: A study of consumer adoption intentions', International Journal Of Research in Marketing, vol. 12, no. 2, p. 137.
Tan, A 2000, 'Phillip's Poems puts the writing on the wall', Business Times, p. 3. Tan, M & Teo, TSH 2000, 'Factors Influencing the Adoption of Internet Banking',
Journal of the Association for Information Systems, vol. 1. Teo, TSH, Tan, M & Peck, SN 2004, 'Adopters and Non-Adopters of Internet Stock
Trading in Singapore', Behaviour and Information Technology, vol. 23, no. 3, pp. 211-23.
Teweles, RJ & Bradley, ES 1998, The Stock Market, Seventh Edition edn, Wiley. TSI 2006a, Online Stock Trading in Asia. ---- 2006b, 'Internet Trading'. UCLA 2009a, What does Cronbach's alpha mean?, UCLA Academic Technology
Services, viewed 24th Jan 2009 2009, <http://www.ats.ucla.edu/stat/spss/faq/alpha.html>.
Information Technology: Toward A Unified View', MIS Quarterly, vol. 27, no. 3, pp. 42-478.
Vitartas, P, Jayne, N, Ellis, A & Rowe, S 2007, 'Student adoption of web base video
conferencing software: A comparison of three student discipline groups', paper presented to ascilite Singapore 2007, Singapore, 2007.
Wells, WD & Presky, D 1996, Consumer Behavior, First Edition edn, Wiley. Werner, M & Murphy, R 2007, 'On-Line Business: Is There Loyalty?' The Business
Review, vol. 9, no. 1, p. 250.
- 244 -
Zhu, K & Kraemer, KL 2005, 'Post-Adoption Variations in Usage and Value of E-Business by Organizations: Cross-Country Evidence from the Retail Industry', Information Systems Research, vol. 16, no. 1, pp. 61-84.
Zhu, K, Dong, S, Xu, SX & Kraemer, KL 2006, 'Innovation diffusion in global
contexts: determinants of post-adoption digital transformation of European companies', European Journal of Information Systems, vol. 15, pp. 601-16.
Zikmund, WG 2000, Business Research Methods, Sixth Edition edn, Dryden.
- 245 -
Appendix A Letter of ethic approval form
- 246 -
- 247 -
Appendix B Letter of Introduction
- 248 -
Appendix C Questionnaire
Online Securities Trading Survey
Please take some time to fill out this survey about online securities trading in
Singapore. Please tick the appropriate answers for the questions below.
PART A: Online Securities Trading Opinions (Please tick the answer according to the scale indicated)
1. To what extent do you agree with the following statements about Online Securities Trading as compared to trading via the broker?
Strongly Disagree
Disagree Slightly Disagree
Neutral Slightly Agree
Agree Strongly Agree
(a) Quicker to trade online (b) Cheaper to trade online (c) Process is not much different from calling the broker
(d) Trading information is not much different from calling the broker
(e) Easier to access investment information online
(f) Easier to trade using Online Securities Trading
(g) Able to do a trial trade which is not possible via the broker
(h) Easier to obtain online demonstration and explanation
(i) More investors signed up for Online Securities Trading system
(j) More investors started to trade using Online Securities Trading
- 249 -
2. To what extent do you agree with the following statements about your selection of Online Securities Trading? Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) Have considered other options like automatic voice trading or WAP trading via phone?
(b) Have considered online trading as an additional method of trading
(c) Have consulted other investors using Online Securities Trading
(d) Have consulted my friends using Online Securities Trading
(e) Have been advised by investment experts to sign up Online Securities Trading
(f) Stock Exchange has liberalized brokerages' commission rate for Online Securities Trading
3. To what extent do you agree that exposure to Online Securities Trading are through the following means: Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) Online advertisements like Internet or email
(b) Mass media like TV or newspaper advertisements
(c) Broker's explanation and demonstration
(d) Friends' and other investors' advice
4. To what extent do you agree with the following statements about your relationship with other Online Securities Trading users? Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) Know more friends who used Online Securities Trading
(b) I feel left out if I do not sign up Online Securities Trading
(c) Consult or discuss with friends or other investors when I trade online
(d) Exchange investment information with friends or other investors using Online Securities Trading
- 250 -
5. To what extent do you agree with the following statements about the promotional efforts of Online Securities Trading? Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) Constantly received Online Securities Trading information
(b) Satisfaction with the broker's promotional efforts on Online Securities Trading
(c) Satisfaction with the brokerage firm's promotional efforts on Online Securities Trading
6. How long did you take to adopt the Online Securities Trading system counting from the day you are aware of it?
less than a week about a month 3 to 6 months about a year
more than a year
7. To what extent do you agree with the usefulness of Online Securities Trading? Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) Trading Online increases my trading profit
(b) The system facilitates diversification of my portfolio
(c) I can react to the stock market quicker
8. To what extent do you agree with the following statements about your loyalty to Online Securities Trading? Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) Online Securities Trading will be my major method to trade
(b) I will introduce Online Securities Trading to non-users
(c) I will not consider other new methods of trading in near future
- 251 -
9. To what extent do you agree with the following statements about your confidence of Online Security Trading? Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) It is a highly secured system (b) It is easily accessible (c) It is very reliable
PART B: Online Securities Trading Usage
10. How often do you trade online in the last 12 months? Daily Weekly Every 2 Weeks Monthly Every 3 months
Every 6 months or more
11. To what extent do you agree with the following statements about using Online Securities Trading as compared to trading via the broker? Strongly
Disagree Disagree Slightly
Disagree Neutral Slightly
Agree Agree Strongly
Agree (a) I trade more frequently now (b) I trade in smaller lot sizes now (c) I buy certain categories of stock (eg. High Tech Stocks)frequently
(d) I trade from more locations like office, Cybercafé and overseas in addition to home
(e) I exchange investment tips with other online investors easily
(f) I check investment information frequently
12. Which of the following information you will usually used to conduct your online trading?
Live prices Historical prices Stock charts Company information