Dissataion - Ranga Perera (CB002688)

151
MSc. in TECHNOLOGY MANAGEMENT DISSERTATION ON DEVELOPMENT OF A CONSUMER TECHNOLOGY ADOPTION MODEL FOR MOBILE DATA SERVICES WITH UTILITARIAN AND HEDONIC VALUE PROPOSITIONS BY Ranga Perera CB002688 7 TH SEPTEMBER 2009 1

Transcript of Dissataion - Ranga Perera (CB002688)

Page 1: Dissataion - Ranga Perera (CB002688)

MSc. in TECHNOLOGY MANAGEMENT

DISSERTATION

ON

DEVELOPMENT OF A CONSUMER TECHNOLOGY ADOPTION MODEL

FOR MOBILE DATA SERVICES WITH UTILITARIAN AND HEDONIC

VALUE PROPOSITIONS

BY

Ranga Perera

CB002688

7TH SEPTEMBER 2009

1

Page 2: Dissataion - Ranga Perera (CB002688)

DISSATATION

Student Name: Ranga Naresh Perera

Student Number: CB002688

Email Address: [email protected]

Award Name: MSc in Technology Management

Site Name: APIIT Sri Lanka

Title of Project: DEVELOPMENT OF A CONSUMER TECHNOLOGY

ADOPTION MODEL FOR MOBILE DATA SERVICES WITH UTILITARIAN

AND HEDONIC VALUE PROPOSITIONS

Supervisor: Professor Kennedy Gunawardena

External supervisor: Associate Professor Ms. Geetha Kanaparan

2

Page 3: Dissataion - Ranga Perera (CB002688)

Abstract

This research investigates the adoption of two mobile data services with utilitarian and

hedonic value propositions in Sri Lanka. The proposed Technology adoption model has

been built based on empirical academic research into consumer motives of cognition,

hedonics, social influences and studies into consumer behavior attitude and intension.

The model attempts to explain 40%-54% of the Sri Lankan consumer behavior in the

context of selected mobile data services. There are a number of important findings from

this research, including identification of key determinants of technology adoption in

mobile data service, the importance of variables such as perceived usefulness, perceived

ease of use and comparative advantage in the adoption decisions. The research further

explores the relationship of hedonic motives and their influence of attitude towards

adoption and adoption intension.

3

Page 4: Dissataion - Ranga Perera (CB002688)

Acknowledgement

The writing of this thesis has been one of the most significant academic challenges I have had to

face. Without the patience, support and guidance of the following people I would have not been

able to complete this journey.

Professor Kennedy Gunawardena who undertook to supervise my Dissertation with short notice

in June. Had it not been for the wisdom, knowledge and commitment of Professor Kennedy, I

doubt that I would have been able to present this dissertation. In June when Professor took the

supervision of my research, I was lost and confused. His knowledge and experience guided me,

inspired me and motivated me. I hope this research justifies the support and confidence you

extended to me.

Professor Gordon C. Bruner II from the Southern Illinois University, USA who was kind enough

to provide me research papers and advise on the Consumer Acceptance of Technology model,

which I used as the foundation of this dissertation. Professor Herbjørn Nysveen from Norwegian

School of Economics and Administration for research papers on Mobile Data Services.

The volunteers and provincial coordinators who helped in mammoth task of distributing and

collecting the questionnaires nationally. Thanks to your friendship and interest I was able to

undertake one of the first national surveys on mobile data services adoption and usage in Sri

Lanka.

To Eranga, Sepali, Nayomi, Bashini, Harshini, Chethani, Priyanwada, Janaki, Sadani, Harsha and

Eureka who entered the 450+ questionnaires painstakingly by working day and night, I am

humbled at your friendship and dedication. All the analysis in this research was possible due to

this effort.

To my friends Eranjan and Leshani who extended their valuable support and time to ensure that

this dissertation was a success.

And Finally to my Mom and brother

4

Page 5: Dissataion - Ranga Perera (CB002688)

Table of Contents

1. Introduction ............................................................................................................... 11

1.2 Problem overview ............................................................................................. 13

1.2.1 Problem statement ..................................................................................... 13

1.2.2 Aims .......................................................................................................... 13

1.2.3 Objectives ................................................................................................. 13

1.3 Justification for selection of Mobile Data Services for research .................. 14

1.4 Significance of the study ................................................................................... 17

1.4.1 Theoretical significance ............................................................................ 17

1.4.2 Significance to other stakeholders ............................................................ 19

1.5 Scope and Limitations....................................................................................... 20

2. Background ............................................................................................................... 21

2.1 Mobile telecommunication industry overview ................................................. 21

2.2 Mobile Technology Evolution .......................................................................... 24

2.3 Mobile Data Services ........................................................................................ 27

3. Literature review ....................................................................................................... 32

3.1 Overview of the selected research area ............................................................. 32

3.2 Review of literature on research subject ........................................................... 33

3.2.1 Motives – Utility vs Hedonics .......................................................................... 33

3.2.2 Technology adoption models and Mobile Data Services adoption .................. 35

3.3 Literature review on selected independent variables ............................................ 41

3.3.1 Independent variable 1 - Perceived usefulness ......................................... 41

3.3.2 Independent variable 2 - Perceived ease of use ........................................ 42

3.3.3 Independent variable 3 - Relative advantage ............................................ 43

3.3.4 Independent variable 4 - Pleasure ............................................................. 44

3.3.5 Independent variable 5 - Arousal .............................................................. 44

3.3.6 Independent variable 6 - Dominance ........................................................ 45

3.3.7 Independent variable 7 - Social Influences ............................................... 46

3.3.8 Attitude and Intention ............................................................................... 47

3.3.9 Short Message Service – Mobile Data Service used to test the cognitive

utilitarian value proposition ...................................................................................... 48

5

Page 6: Dissataion - Ranga Perera (CB002688)

3.3.10 Mobile Ringtone – Mobile Data Service used to test the hedonic value

proposition ................................................................................................................ 48

3.3.11 Utilitarian Motives .................................................................................... 50

3.3.12 Hedonic Motives ....................................................................................... 51

4. Solution ..................................................................................................................... 52

4.1 Solution overview ............................................................................................. 52

Proposed model for mobile services adoption in Sri Lanka (Sri Lanka Consumer

Acceptance of Technology Model – SLCAT) .............................................................. 53

4.2 List of developed hypothesis ............................................................................ 54

5. RESEARCH METHODOLOGY.............................................................................. 60

5.1 Research Philosophy ......................................................................................... 61

5.2 Research Approach ....................................................................................... 61

5.3 Research Strategy.............................................................................................. 62

5.4 Pilot study ......................................................................................................... 63

5.5 Time Horizon .................................................................................................... 63

5.6 Determining the Sample and Sample Size ........................................................ 64

5.7 Questionnaire design – Likert scales used ........................................................ 66

5.8 Treatment of data .............................................................................................. 67

6. Deliverable ................................................................................................................ 68

6.1 Descriptive Analysis ......................................................................................... 68

6.1.2 Respondents by Gender ............................................................................ 68

6.1.3 Respondents by Age ................................................................................. 69

6.1.4 Respondents by Province of residence ..................................................... 70

6.1.5 Respondents by Education level ............................................................... 72

6.1.6 Respondents by Employment status ......................................................... 73

6.1.7 Respondents by monthly income level ..................................................... 74

6.1.8 Mobile Data Services Awareness ............................................................. 76

6.2 Statistical analysis of data ................................................................................. 77

6.2.1 Utilitarian model testing using SMS ......................................................... 78

6.2.2 Hedonic model testing using Mobile Ring tone ....................................... 78

6.3 Hypothesis Testing............................................................................................ 79

6

Page 7: Dissataion - Ranga Perera (CB002688)

6.4 Simple liner model building.............................................................................. 89

6.5 Model building .................................................................................................. 95

6.5.1 Utilitarian Product of SMS ....................................................................... 95

6.5.2 Attitude towards adoption ......................................................................... 95

6.5.3 Intension to adopt ...................................................................................... 98

6.5.4 Hedonic Product of Mobile Ringtone ..................................................... 101

6.6 Data Analysis Summary ................................................................................. 104

6.6.1 Utilitarian product – SMS adoption model testing ................................. 104

6.6.2 Hedonic product – Mobile Ringtone adoption model testing ................. 114

7. Discussion ............................................................................................................... 122

8. Recommendations ................................................................................................... 134

10. Future researchReferences .................................................................................. 144

10. References ........................................................................................................... 145

7

Page 8: Dissataion - Ranga Perera (CB002688)

List of Tables

Table 1: Mobile Telephony systems ................................................................................. 26 Table 2: Mobile Data Services classification .................................................................... 29 Table 3: Summary of Litreture review - Utilitarian motives ............................................ 50 Table 4: Summary of literature review - Hedonic motives ............................................... 51 Table 5: : Literature review summary - Attitude and intension ........................................ 51 Table 6: Literature review summary - MDS with utilitarian and hedonic propositions ... 51 Table 7: Hypothesis for utilitarian motives in SMS ......................................................... 54 Table 8: Hypothesis for hedonic motives in SMS ............................................................ 55 Table 9: Hypothesis of social influences in SMS ............................................................. 56 Table 10: Hypothesis attitude and intension in SMS ........................................................ 56 Table 11: Hypothesis for utilitarian motives - M-Ringtone.............................................. 57 Table 12: Hypothesis for hedonic motives in M-Ringtones ............................................. 58 Table 13: Hypothesis for Social influences - M-Ringtones .............................................. 58 Table 14: Hypothesis Attitude and intesion - M-Ringtone ............................................... 59 Table 15: Questionnaire distribution ................................................................................ 64 Table 16: Respondents by Gender .................................................................................... 68 Table 17 : Respondents by Age ........................................................................................ 69 Table 18: Respondents by Province of residence ............................................................. 71 Table 19: Respondents by Education level ....................................................................... 72 Table 20: Respondents by Employment status ................................................................. 73 Table 21: Respondents by monthly income level ............................................................. 74 Table 22: Colour Display vs Black/White display ........................................................... 75 Table 23: Mobile Data Services Awareness ..................................................................... 76 Table 24: Test values for internal consistency – SMS ...................................................... 78 Table 25: Test values for internal consistency - M-Ringtones ......................................... 78 Table 26: Correlation Matrix for SMS.............................................................................. 79 Table 27: Utilitarian model testing using SMS................................................................. 82 Table 28: List of accepted hypothesis (alternative) – Utilitarian product ........................ 83 Table 29: List of Accepted Null Hypothesis..................................................................... 83 Table 30: Correlation Matrix for hedonic motives ........................................................... 84 Table 31: Hypothesis testing for Hedonic model ............................................................. 87 Table 32: List of accepted hypothesis – Hedonic Product ................................................ 88 Table 33: Simple liner model building – SMS ................................................................. 91 Table 34: Simple liner model building - Mobile Ringtones ............................................. 94 Table 35: Variable ranking based on correlation to Attitude towards adoption ............... 95

8

Page 9: Dissataion - Ranga Perera (CB002688)

List of Figures

Figure 1: World mobile subscribers .................................................................................. 22 Figure 2: Cellular subscriber growth rate in Sri Lanka ..................................................... 23 Figure 3: Evolution of GSM Technologies ....................................................................... 25 Figure 4 ............................................................................................................................. 28 Figure 5: Proposed classification of MDS ........................................................................ 29 Figure 6: Techno-centric MDS classification ................................................................... 30 Figure 7: Four tiered MDS classification.......................................................................... 31 Figure 8: Classification of consumer value ...................................................................... 34 Figure 9: Proposed model for mobile services adoption in Sri Lanka .............................. 53 Figure 10: Research Onion (Saunders et al, 2007a) ......................................................... 60 Figure 11: Respondents by Gender ................................................................................... 69 Figure 12: Respondents by Age ........................................................................................ 70 Figure 13: Respondents by Province of residence ............................................................ 71 Figure 14: Respondents by Education level...................................................................... 72 Figure 15: Respondents by Employment status ................................................................ 73 Figure 16: Respondents by monthly income level ............................................................ 74 Figure 17: Colour Display vs Black/White display 75

9

Page 10: Dissataion - Ranga Perera (CB002688)

Abbreviations

10

Page 11: Dissataion - Ranga Perera (CB002688)

1. Introduction

Mobile technology has taken rapid strides in its diffusion across the global. These

quantum leaps in penetration are not only global phenomenon but one also experienced in

the local context of Sri Lanka. In 1992 Sri Lanka had 2,644 mobile phone subscribers.

Today 17 years later the number stands at 11 million (TRC-SL 2008). While mobile

penetration rates are impressive, with 50%-60% average annual growth rates experienced

in Sri Lanka, the strategic prospects of the mobile telecommunication industry are up for

discussion. What comes after you have sold every one a mobile phone?. Signs are

ominous. Across the globe the average revenue per unit (ARPU) are significantly

depreciating (ABI Research 2009; Mälarstig et al. 2007). These issues are compounded

with increase competitive structures and global market competition. The industry seized

on an emerged opportunity in the early 1990 with a new application called Short

Messaging Service. The mobile phone and its use were viewed in a different light than a

simple communication device, rather the gateway to a plethora mobile data services. The

industry spent the next decade investing in high bandwidth, high capacity and new

mobile data services product lines, awaiting the next killer application (C. Carlsson et al.

2005b). However, today after spending billions of dollars into 3G licenses and

sophisticated new services such as MMS, Mobile Internet, Mobile Banking, the “next

killer application” is yet to emerge. SMS still remains the most popular mobile data

service in all markets including the USA(Nielsen Research 2008) and European markets

such as Finland (C. Carlsson et al. 2005b) and Norway(Nysveen et al. 2005b). While

academics and industry in developed countries have focused on studying mobile data

services with new vigor, in developing Countries like Sri Lanka, industry and regulators

seem to be unaware of these global trends and threats.

The aim of this research is analyze the key variables involved in understanding and

predicting consumer behavior of technology adoption. Through this analsys, it is

expected that a behavioral model can be produced which can be identify scientifically the

relationships between the drivers of consumer attitude to adopt and intension to adopt

mobile data services. While there are models researched and developed in countries like

11

Page 12: Dissataion - Ranga Perera (CB002688)

Finland (C. Carlsson et al. 2006), Norway (Pedersen et al. 2002), Korea (B. Kim et al.

2009) and USA, development of an indigenous technology adoption model is essential in

the context of Sri Lanka because of the different socio-economic cultural paradigms.

Further due to the regional similarities in South East Asia, the inter-portability of this

model may help diffusion of mobile data services in similar regional countries.

To undertake this study we recommend identifying key research into information

technology adoption including empirically tested models such as the Technology

Adoption Model (Davis 1989; Davis et al. 1989) and diffusion of innovation models

(Rogers 2005). Further as this model involves operations within the consumer context, it

is proposed that research into better understanding the variables that influence the attitude

towards adoption and intension to adopt be researched. Further, the recent research done

on developing a unified theory for technology adoption (Kulviwat et al. 2007; Kulviwat

et al. 2008; Nasco et al. 2008) provides an important starting point. Therefore it was

decided that the study would focus on the logical motives and hedonic motives of ‘Fun

and entertainment’. While motives guide the decision, what nature of value propositions

influence these motives. The second focus of the research would be on value propositions

and their interrelation to technology adoption.

Based on this analysis it was decided that the research would study two mobile data

services products. One which has primarily a cognitive utilitarian value proposition and

another that has primarily a hedonic value proposition. This research would then enable a

better understanding of the behavior of the model in these different context. The balance

of this document will relate to the building of the proposed model based on empirical

research and testing of the model in the context the Sri Lankan consumer through a

market survey. It is expected that this research path would enable the achievement of this

ultimate objective.

12

Page 13: Dissataion - Ranga Perera (CB002688)

Problem overview

Problem statement The mobile data services adoption in Sri Lanka remains at a very low rate in comparison

to the penetration of mobile phone technology which is estimated to be at 55% (TRC-SL

2008). Research indicates that the future revenues of mobile telecommunication industry

will depend on the provision of mobile data services rather than on voice calls (Kunin et

al. 2005; C. Carlsson et al. 2005b). The dramatic drops in average revenue per user on

voice calls across the globe are an indication of future trends (ABI Research 2009).

Further in most matured telecommunication markets, where mobile penetration has

exceeded 80% reach of the general population, the industry was compelled to look for

more viable sources of revenue other than voice and new subscriber connection fees

(Mälarstig et al. 2007). While the strategic response of the mobile industry was to invest

in expensive 3G technology, the global adoption rates of mobile services that use this

platform remains very low.

Aims To proposition an analytical model that identifies the key attitudinal influences involved

in the adoption of selected Mobile Data Services in the Sri Lankan market context. This

model could be used by the Telecommunication industry and Mobile Data Services

application vendors to identify key consumer relationship variables that influence the

adoption and diffusion of their products and services.

Objectives

– To analyze the nature and behavior of existing relationships between cognitive

utilitarian motives, hedonic motives, social influences and their impact on the

consumers attitude and intension to adopt key Mobile Data Services in Sri Lanka

13

Page 14: Dissataion - Ranga Perera (CB002688)

– To develop a statistical model that analyses the influence of cognitive utilitarian

motives, hedonic motives and social influences to predict the adoption attitude

and intension to adopt the selected Mobile Data Service of Short Message Service

(SMS) which has a dominant utilitarian value proposition in Sri Lanka

– To develop a statistical model that analyses the influence of cognitive utilitarian

motives, hedonic motives and social influences to predict the adoption attitude

and intension to adopt the selected Mobile Data Service of Mobile Ringtone

which has a dominant hedonic value proposition in Sri Lanka

– To analyze the determinant factors that influence the adoption of Mobile Data

Services based on the developed statistical analysis models for utilitarian and

hedonic products

Justification for selection of Mobile Data Services for research

The mobile telecommunication industry has invested billions of dollars in improve the

network bandwidth capacity, mobile phone capacity and overall infrastructure to support

the expanded usage of mobile devices beyond being simple communication devices (.

The anticipation and excitement was that the introduction of 3G would provide mobile

telephone subscribers access to a vast array of mobile data services. However actual

adoption of Mobile Data Services across global markets remain consistently low. On

commenting on this low rate of adoption of Mobile Data Services, states “Our results are

consistent with previous research. Mobile services still have much less users than

envisioned and their usefulness is being questioned by consumers”. While the low

adoption rates of mobile data services are symptoms of consumer perceptions, (Gilbert &

Kendall 2003) outline the need to change the behavioral patterns of consumers to ensure

viable adoption and usage. They state that “MDS are a current example of technology

enabled discontinuous innovation, similar from the economic and behavioral perspectives

to the Internet. Such innovations will succeed only if adopted by a critical mass”. The

researchers highlight the critical need for creating new value and new behavior patterns

14

Page 15: Dissataion - Ranga Perera (CB002688)

to ensure sustainable usage of these innovative products. Elaborating on the behavior,

they state that “..such behaviors include acquiring the enabling technology, learning to

use it, applying it to solve problems or adding value in everyday life, and communicating

what one has learned about it to others”. However industry and academia have been only

starting to recognize the need for identify and build viable and empirically tested

consumer adoption models to enhance the overall adoption of the technology.

The lack of research into mobile data services has been a key issue identified by many

researchers. Umino (2004) in a report on OECD countries notes that there is a general

lack of research into the area of Mobile Data Services both by Government and Industry.

The researcher states “Often mobile data is not yet presented separately from aggregate

data. Industry or government sponsored studies focus only on certain markets or

technologies and definitional constraints make it difficult to compare data across studies.

Further research in this area is worth undertaking”. This lack of research and focus may

be stemming from the industries original concern about new connections. Carlsson et al

(2005) commenting on the lack of industry focus on mobile data services comments

“Gartner Inc. in a recent report still focus on the handset market…. It is strange that not

much is reported on the development of mobile services..”. Thus the evolving nature of

the mobile telecommunication industry, which at its inception presented a value

proposition of a simple communication solution, to todays’ mobile data services, which

are value added services may be key reasons for these gaps in research. While the reasons

behind the lack of research may be varied, this lack of understanding of the mobile data

services phenomena has presented the industry with a major challenge. This challenge

has presented itself as an additional risk into the investments towards mobile data

services in general. Carlsson et al. (2006) observers on this increased risk that has

inherited into the Mobile Data Services market as “Year after year the mobile service

market(s) produce(s) new services and applications that due to complexity or lack of

relevance fail to meet the consumers’ expectations”. Therefore the need for research into

understanding the consumer and technology application has emerged as an important of

the overall mix in the product development cycle of mobile data services.

15

Page 16: Dissataion - Ranga Perera (CB002688)

Therefore there is a clear need to understand the consumers adoption and usage

preferences towards mobile data service. A number of researchers have explored the

applicability of different psychological models within the context of mobile data services.

These notable researchers include, but are not limited to, Carlsson et al. (2006) in the

Finnish mobile services market, Pedersen et al. (2008) in the Norwegian market, Bina et

al. (2007) in the Greek market. However, when exploring the existing research, it is clear

that research are presented within predominantly from Western and European countries

which have mobile penetration rates exceeding 80%.

There are only few research that has been done to analyze the consumers of South East

Asia including consumers within Sri Lanka, India, Bangaladesh and other East Asian

emerging economies. Commenting on the lack of research into developing countries Gao

& Rafiq (2009) state that “We lack knowledge about the characteristics of mobile

telecommunications transformation in developing countries, and the social and

technological factors that impact this process”. In their litreture review covering a period

of 5 years between 2003 and 2008 they have found eight published articles on mobile

telecommunication industry in developing countries. However the critical features of this

analysis are that these literatures have been prepared based on secondary research and not

primary research.

The huge value of investments made into building 3G+ networks and the ubiquitous (ref)

nature of mobile phone technology presents both a threat and opportunity for a country

like Sri Lanka. Emerging from a three decade old conflict situation, the mobile

technology has a huge potential of enriching and thrusting the rural agrarian economy of

the island rapidly into the 21st century. Mobile Internet, Video calls and other ranges of

mobile data services through 3G + networks would provide the stimulus and hope to our

country. However, understanding the intentions and barriers of the Sri Lankan consumer

in the adoption of this innovative technology is crucial for the penetration of mobile data

services products and services in Sri Lanka. It is based on these reasons that this research

area was selected.

16

Page 17: Dissataion - Ranga Perera (CB002688)

Significance of the study

Theoretical significance The research conducted under this project seeks to contribute towards a number academic

interest areas.

Significance of the market data for academic study

The research into adoption of mobile data services is considered by many researchers as a

key gap area in the existing knowledge. Further majority of the available research has

been undertaken in developed markets such as in the USA, Europe and developed

economies in Asia such as Japan and Korea. Therefore this research into the use,

adoption and adoption intension of mobile data services in a country such as Sri Lanka

will be beneficial to understanding the attitudes of a population which has unique

demographic and psychographic characteristics. These characteristics include the high

literacy rate of 96% (G15 2008), the rising Gross Domestic Income of over US$ 1200

which has risen by US$ 150 within the last 3 years, the low computer penetration rate of

and internet penetration of 2% (G15 2008). These characteristics combined with the

estimated mobile phone penetration rate of 54%(TRC-SL 2008) makes this research in to

the study of mobile data services an important and long-term significant study.

The study cover 6 of the 9 provinces and can be used for province wise analysis.

There is currently no available data for academic analysis of the handset types and

capabilities used by the Sri Lankan consumer. This information is particularly important

because the capacity of the mobile phones carried by the consumers in Sri Lanka should

influence decisions on the types of Mobile Data Services that can be promoted in the

island. Further this information should provide a valuable decision and strategic options

consideration tool for mobile telecommunications companies, on whether their current

strategy of not getting involved in the handset market is compatible with their network

17

Page 18: Dissataion - Ranga Perera (CB002688)

investment strategy. In order to successfully launch mobile data services that are accepted

by the consumers, the handsets they use have a major influence on the decision making

process. Therefore it is envisioned that this research would initiate a dialog on this issue.

The research focuses on understanding the existing market share of the five

telecommunication services providers of Sri Lanka.

It studies the switching habits of consumers in terms of change of mobile

telecommunications providers, reasons and switching time frame. The cross referencing

of this information with demographic information of users should provide information

vulnerable market segments that are likely to shift to other telecommunications providers.

The research investigates the current usage of native language features in mobile phones

users. This research information would provide information on the popularity and actual

usage of native language features.

The research focuses on consumer awareness of selected mobile data serives, one time

usage and regular usage and the consumers attitude towards the future adoption of the

services.

Significance of the research proposition and hypothesis testing

The research proposition was built on key Information Systems theories of Technology

Adoption Model, Diffusion of Innovation model and PAD model. The significance of

each of these theories towards the adoption intention in the context of the population of

Sri Lanka will be tested through this research.

The applicability of the Consumer Acceptance of Technology model has not been tested

in a wide national study prior to this research. This would be the first occasion the

propositions applicability is tested within a unique market such as Sri Lanka.

18

Page 19: Dissataion - Ranga Perera (CB002688)

Significance to other stakeholders Provide the government and regulators insights into the importance of promoting

mobile data services through policy frameworks based on the key influences

identified through this research

Provide the mobile telecommunication industry a better understanding of the

influences of motives and attitude towards adoption of utilitarian and hedonic

value propositioned mobile data services.

Provide software and related technology developers of mobile data services

applications a model to test their product prototypes prior to expensive releases to

market.

Help in influencing the technology infrastructure investments done by the

government as a part of developing the information communication technology

infrastructure of urban and rural Sri Lanka.

Provide greater insights to brand and marketing managers in investing their

marketing budgets and understanding of societal influences on adoption.

19

Page 20: Dissataion - Ranga Perera (CB002688)

Scope and Limitations

The research is undertaken within the geographical boundries of Sri Lanka and

may be unique in its findings

The consumers surveyed were primarily from urban and rural areas of Western,

North Western, Southern and Central Provinces

Consumers from other provinces including Northern, Eastern, Uva and North

Central province have not contributed

Due to the small sample size cannot provide analysis at provincial level

implications of the model

20

Page 21: Dissataion - Ranga Perera (CB002688)

2. Background

2.1 Mobile telecommunication industry overview The global penetration of mobile phones reached a new height by the end of 2008 when

the International Telecommunication Union (ITU) declared that it estimates the global

telecommunication subscriber base to be 4 billion ITU (2008). This estimated figure is an

increase of over 1 billion mobile subscribers within a period of one year (ITU, 2007). In

late 2007 the global mobile subscriber base was estimated to be around 3 billion

subscribers and was equivalent to 50% of the global population. The year-on-year

average growth of the global mobile telecommunication subscription between the years

2000 to 2008 has been at an average of 24%. While these figures would indicate that the

global penetration of mobile phones are at 61% and that on average every other person

should have a mobile phone, the information needs to be qualified. Its is noted that the

figures represent subscriptions and not actual persons, an individual may have multiple

subscriptions and mobile phone operators methods of counting the prepaid and post paid

consumer may create duplication. Noting this point, it is estimated that over 30 countries,

predominantly in Europe have mobile penetration rates exceeding their country

populations, the highest being Italy at the rate of 151 subscribers for a population of 100

in 2009(ITU, 2009). While these qualifications are valid, Ratan et al. (2007) in their

research of the Bangladesh mobile market note that through ‘Village Phone Program’

each village is provided with a single mobile phone which is shared between a number of

persons.

These impressive mobile phone penetration figures combined with the analysis by ITU

(2009) that Mobile subscriptions accounted for 61% of the total communication

subscriptions, while standard phone line subscriptions were at a low 26% solidifies the

future importance of mobile technology. Further compounding this trend is the increase

in the average usage minutes of mobile phones. The ITU (2009) analysis of average

minute usage suggests that the number of minutes spent by subscribers on mobile phones

are rising while the usage minutes of fixed phones are reducing. Another important

21

Page 22: Dissataion - Ranga Perera (CB002688)

observation in this analysis is that users of fixed phone lines are spending an increased

number of minutes communicating with mobile phone subscribers. Other important

global trends are in the dramatic reduction in prices of mobile calls. The estimates

indicate that there is an average reduction of over 20% in call charges associated with

mobile phones.

Figure 1: World mobile subscribers

Extracted from ITU (2008)

While the global penetration rates of mobile phones are impressive, these figures are

sustained primarily through the four BRIC countries of Brazil, Russia, India and China.

Based on estimates the total subscription rates of these economies have an estimated 1.3

billion subscription. ie. One third of the world mobile phone subscribers. While China

has over 600 million subscribers, the Indian subscriber rate is estimated to be 296 million.

This represents a very low penetration rate of 20% in comparison with BRIC countries

and regional countries such as Sri Lanka which has an estimated penetration of 55% by

2008. However, these figures indicate that the mobile penetration and growth will remain

healthy over the next few years in the region.

22

Page 23: Dissataion - Ranga Perera (CB002688)

Figure 2: Cellular subscriber growth rate in Sri Lanka

Extracted from TRC-SL (2008)

In Sri Lanka the mobile phone penetration rate has been at a dramatic pace and has

mimicked the global trends closely. In terms of the supply side there are five mobile

telecommunication companies with one new entrant Airtel coming to the market in early

2009 (TRC-SL, 2008). Between the year 2007, where the mobile phone subscribers were

estimated at 7.9 million and 2008 where the figure rose to 11 million, the annual

increment year-on-year has been has been 39%. With the country emerging from a three

decade old conflict situation, Sri Lanka would most likely see the mobile penetration

rates reaching over 80% from the existing rate of 55% of the population within the next 3

years. In comparison to these mobile phone penetration figures the fixed access phone

connection has grown by 20% in 2008 to a figure of 3.4 million phones. It is in the year

2001/2002 that the mobile phone connectivity rate surpassed that of mobile phones in Sri

Lanka. An interesting statistic published is the number of pager connections in the island

which stood at over 10,000 in the year 1996 has seen a complete decline by 2005. While

published research is not available, this may be due to popularity of SMS services.

23

Page 24: Dissataion - Ranga Perera (CB002688)

While Sri Lanka macro economic indicators such as Gross Domestic production (average

rate of 5% to 6% REF) and Gross Income (US$) have see improvements over the last few

years, its internet penetration rate remains at very low level of 2%. Further there is no

available information on Mobile Data Services usage and related trends.

2.2 Mobile Technology Evolution In order to thoroughly appreciate the current mobile industry issues, risks and the

implications of Mobile Data Services, it is crucial to understand the underlying

technology, reasons for technological evolution, the technology evolutionary path, factors

that pushed and pulled the evolution, the current point and future evolutionary path.

The second generation of mobile phones also known as 2G started appearing in the early

1990s. Kunin et al. (2005) states that “Most 2G standards are based on circuit-switched

technology, and they have provided the mobile telecommunications industry with an

exponential growth in terms of the numbers of subscribers as well as new types of

services”. Among the most successful technology variants of the 2G included technology

standards such as CDMA (Code Division Multiple Access), TDMA (Time Division

Multiple Access), Global System for Mobile (GSM). The CDMA technology is a digital

wireless technology that has the capability to provide simultaneous access for subscribers

to share radio frequency. The researcher describing some of the distinguishing features of

CDMA “a voice or data call is assigned a unique code that distinguishes it from others

and all of the signals hop and spread over a shared frequency band”. Kunin et al. (2005)

states that as of 2004 CDMA based mobile telecommunication systems were operational

across 63 countries and services an estimated 200 million users. Originally known as the

IS-54 standard, TDMA technical platform has the capability of delivering as much as six

times more information using the same bandwidth than the first generation analog

technology. It is estimated that the TDMA technology which was simultaneously

developed and implemented with CDMA technology serves approximately 113 million

subscribers. The GSM technology is considered the most widely adopted platform in the

2G family. It uses a combination of Frequency Division Multiple Access (FDMA) and

24

Page 25: Dissataion - Ranga Perera (CB002688)

Time Division Multiple Access (TDMA). This technology has the capacity to deliver

over eight calls over a single channel.

These underlying technologies supported the deployment of a range of value added

services other than voice. Carlsson et al. (2005) in their analysis of the evolution of

mobile applications identifies that SMS which was available with GSM platform since

the early 1990s started to become unexpectedly popular by 1995. Mobile based internet

browsing services was enabled by 1999 through the deployment of Wireless Application

Protocol (WAP) over the GSM networks. While WAP was introduced aimed at linking

the internet with mobile devices its performance and willingness by subscribers to adopt

the technology was poor (Teo & Pok, 2003).

Figure 3: Evolution of GSM Technologies

Extracted from Carlsson et al. (2005) Continuous technology upgrades to the 2G platform continued since its introduction.

These technology upgrades, which positioned between the 2G (GSM) standard and 3G

(UMTS), included enhancements to GSM in the form of General Packet Radio Service

(GPRS) are noted as 2.5G (Carlsson et al. (2005). GPRS is considered a pivitol

technology enhancement as it introduced the concept of “always-on” capability (Kunin et

al. (2005), which mean that users only had to pay for actual downloads instead of

connectivity. Further the use of packet based data transfer meant that the cost of

operating the service was much cheaper than circuit switched networks (Carlsson et al,

2005). OECD commenting on 2.5G technology platform states that “Many operators are

deploying services with these technologies instead of waiting for 3G since they are

capable of delivering many of the 3G services with higher speeds than 2G.”

25

Page 26: Dissataion - Ranga Perera (CB002688)

Table 1: Mobile Telephony systems

Extracted from Kunin et al. (2005)

Kunin et al. (2005) commenting on 3G mobile technology states that “is a generic term

for a set of mobile telephony technologies using a set of high-tech infrastructure

networks, handsets, base stations, switches and other equipment to allow high-speed

Internet access, broadband audio-visual services, and voice and data communications”.

While the 3G technology has a wide bandwidth between 128Kbps to 2 Mbps, the

technology has demonstrated much faster speeds. Among the key distinguishing features

of 3G technology is the wider bandwidth that enables the usages of rich mobile data

services applications such as video calls, mobile internet, high quality audio and visual

services delivery to consumers.

Beyond the 3G technology lies the 4G IP based technology, with an estimated speed 10

time more than that of 3G, with the capability of handling “volatile traffic patterns such

as multiple transmissions of multimedia messages from camera phones” (OECD 2005).

26

Page 27: Dissataion - Ranga Perera (CB002688)

2.3 Mobile Data Services

The global Revenue derived from Mobile Data Services have exceeded US$ 200 billion

in the year 2008. This rise in income is an increase of over US$ 43 billion from the

previous year, an estimated increase of over 22% (Cellular-news, 2008). These revenue

figures represent approximately 20% of the total revenue earned by telecommunications

providers. The Filipino telecom provider Smart Communications recorded 50% of their

total earnings from Mobile Data Services. These revenue trends indicate the important

role that Mobile Data Services will play in the future telecom market. While Short

Messaging Services were the initial driver of growth, the industry has been searching for

new “killer applications” which leverage the network capacities setup through the

institution of 3G technology (C. Carlsson et al. 2005). It is therefore anticipated that the

Mobile Data Services would be the driver of growth in the telecommunication industries

where mobile penetration has achieved saturation level.

Bina et al. (2007) in defining Mobile Data Services states that “encompass all non-voice

value-adding services accessible through mobile networks that are designated to augment

end-user experience with mobility and enrich mobile business models for operators,

service providers and other industry constituents”. While this is a general all

encompassing definition, researchers have sought to better define and understand Mobile

Data Services through consistent study. Kunin et al. (2005) in their early study of Mobile

Data Services sought to categorize them into communications, transactional and content

based services. While this classification attempts to identify the Mobile Data Services

from a technology perspective, it lacks the detailed classifications and categorizations

necessary for detailed study of the products and services. Further the classification by the

researcher is based on technology criteria and not a consumer centric perspective. This

classification lacks the depth of analysis and attempts to basket all mobile data services

into one group. However, for the development and positioning of Mobile Data Services it

is crucial that better understanding and analysis of the portfolio be undertaken.

27

Page 28: Dissataion - Ranga Perera (CB002688)

Figure 4

Extracted from Kunin et al. (2005)

Carlsson et al. (2005) in their analysis of Mobile Data Services in the Finnish market

have rarely attempted to define Mobile Data Services. Rather their focus has been on the

adoption of the technology and therefore the Mobile Services they have used have been

categorized in a more practical classification. Namely, the MDS have been classified into

Communication, Entertainment, Reservation and purchases, and Information. Into these

four major classifications of the services, they have incorporated a total of six-teen (16)

services. However, the problems associated with the definition of mobile data services

could be highlighted through such classification. Under Communication product ranges

the researcher has included SMS services which are primarily interpersonal in nature.

However, Bina et al. (2007) in their definition of MDS specifically state that “all services

afforded through a mobile network except for voice communication and interpersonal

SMS exchanges”. While the researchers have not commented further on this exception, it

is clear from the analysis that they view MDS in the context of business value creation.

However, not withstanding this interpretation by the researcher, SMS is considered to be

the most popular MDS and the foundation of todays’ recognition and pursuit of killer

applicationsCarlsson et al. (2005a). Further commenting on the popularity of the MDS

products in the United States, Nielsen Research (2008) identified that 53% of the US

consumers were using SMS services as oppose to the next most popular MDS which was

MMS has a subscriber base of 26%. In research done in the Finnish mobile market where

penetration rates have exceeded 80%, over 92% of mobile users regularly use SMS

28

Page 29: Dissataion - Ranga Perera (CB002688)

Carlsson et al. (2006) Therefore the exclusion done by Bina et al. (2007) points to the

need to study the context and spatiality of MDS.

Table 2: Mobile Data Services classification Extracted from Gilbert & Han (2004)

A more comprehensive analysis matrix of MDS was presented by Pedersen et al. (2002)

in their analysis of the Norwegian MDS consumer. The matrix attempted to classify MDS

based on the perceived motives and technology characteristics of MDS. The technology

characteristics used by the researchers are communication and transaction. These

dimensions of MDS are cross matched with purpose of usage, where the researchers

introduce the motives of entertainment and utility. This classification is considered by

many researchers as one of the most important cross combinations used in the analysis of

MDS (Nysveen et al., 2005).

Figure 5: Proposed classification of MDS

Extracted from Pedersen et al. (2002)

29

Page 30: Dissataion - Ranga Perera (CB002688)

In contrast to the Communication Vs Transaction and Utilitarian Vs Entertainment

classifications of MDS of Pedersen et al. (2002), Verkasalo (2006) seeks to classify

services based on a technology based classification. He uses the continuums of

Communication Vs Content and Interactive vs Background traffic dimensionality.

However, this classification is also primarily a technology centric analysis of these

portfolios of MDS. Beyond this classification of MDS Verkasalo (2006) presents a more

detailed classification of MDS operating on symbion operating systems. Here the main

categories for the classification of MDS include Browsing, Config, Games, Infotainment,

Messaging, Multimedia, Personal Information Management, Productivity, Unknown and

Utility. While these classification relate to applications available in mobile phones, the

products included as part of the analysis relate to MDS. The researchers definition of

MDS as “mobile services which are based on the IP architecture” confirms the

concentration on technology in stead of consumer perspective or service delivery

perspectives.

Figure 6: Techno-centric MDS classification

Extracted from Verkasalo (2006)

30

Page 31: Dissataion - Ranga Perera (CB002688)

However one of the most comprehensive analysis and classifications of MDS was

presented by Heinonen and Pura (2006). Their complex analysis of MDS attempts to

classify MDS based on type of “consumption, the context, the social setting and

relationship”. Unlike the classifications of MDS by Verkasalo (2006) which was

primarily technology centric, the researchers attempt view MDS from a consumer service

context. In their criticism of the existing literature on MDS classification, they point-out

that no significant effort has been undertaken to study the classifications of MDS, rather

the existing literature have been produced as a part of a specific aspect of study of the

MDS in terms of intension to use, segmentations, sociability etc.

Figure 7: Four tiered MDS classification

Extracted from Heinonen and Pura (2006)

31

Page 32: Dissataion - Ranga Perera (CB002688)

3. Literature review

3.1 Overview of the selected research area Significant research and wide body of knowledge has developed over the past years on

the cogitation and ruminative research of Mobile Data Services adoption across the

globe, in relation to identified markets and on specific mobile data services context.

While the nature of research have been multifarious including industry researchers,

behavioral and social scientists contributing their perspectives, the key thrust area of the

research has remained focused on understanding the adoption of these range of

innovative mobile product and services by the consumer of mobile telephony. Primarily

two significant schools of thought have emerged as the benchmarks for these studies,

namely diffusion research (Rogers 2005) and adoption research (Davis 1989). However

when commenting on research paradigms, it should be noted that compelling research

have been also been undertaken on other promising research directions including (Bina et

al. 2007) on the Triandis (1980) model, the application of Uses and gratifications

research and domestication research by (Pedersen et al. 2002), fit-viability model

proposed by Tjan (2001) which combine the theory of technology and task fit within an

organization, Self-efficacy Theory (Bandura, 2001) were considered during the initial

phased of the LT review.

Excogitating the propositions of the above research, the Consumer Acceptance of

Technology (Kulviwat et al. 2007) distinguishes itself by attempting to build the model

by balancing the logical utilitarian elements of the adoption research (Davis 1989; Rogers

2005) with theories of emotion and affect (Mehrabian & Russell, 1974), to present a

unified theory on technology adoption. The application of this unified theory presents a

potentially powerful prediction and consumer explanation model. This chapter of the

literature review focuses on exploring the critical aspects of the conceptual model

propositioned by this dissertation through a through analysis of the key constructs. It is

hoped that this process would further validate the suitability of exploring the adoption of

32

Page 33: Dissataion - Ranga Perera (CB002688)

Mobile Data Services based on the Consumer Acceptance of Technology theory and

indigenous industry specific variables.

3.2 Review of literature on research subject

3.2.1 Motives – Utility vs Hedonics The motives of utility and hedonics formulate a significant composition of the

proposition hypnotized by this research into MDS. This section of the Literature Review

attempts to provide an analysis and definition to these terms.

Understanding the ‘perceived value’ or the benefits customers intend to derive by

acquiring a product or service has been one of the most researched areas in marketing

theory. The decision by Marketing Sciences Institute (2006) to earmark the definition of

‘value’ as a priority research area highlights the continuing and evolving importance of

the subject. Fernández & Bonillo (2007) in their review of research on the subject

observe that, ‘perceived value’ is a result of “interaction” between the customer and the

selected the product or service. Therefore understanding the motives that drive and

influence this interaction is essential in the context of any exchange between a customer

and the provisioning of products or services. On motives and the nature of ‘perceived

value’, the researchers indicate that it may be “...preferential, perceptual, and cognitive-

affective”. It should therefore be appreciated that utilitarian and hedonic motives are only

two key motives that are part of a large portfolio of possible motives that underline the

consumers buying decision. Fernández & Bonillo (2007) identify a large body of research

into ‘perceived value’ while categorizing them into uni-dimensional and multi-

dimensional approaches. While they differentiate between the uni-dimensional and multi-

dimensional approaches because the former propositions a single overall measure to

‘perceived value’, while the latter accepts that multiple components may be used to

define value. However a more pertinent observation between these two classifications is

the evolution of importance placed on utilitarian motives in the more classical uni-

dimensional research and the emerging emphasis of hedonics in multi-dimensional

33

Page 34: Dissataion - Ranga Perera (CB002688)

research. It is indeed surprising to note this same evolution of emphasis on utilitarian

motives to hedonics in information systems theory. The once dominant theories such as

Technology Acceptance Model (Davis 1989; Davis et al. 1989) which proposition the

importance of utilitarian motives have now started to incorporate and accept hedonics as

boundary conditions (Heijden 2004; Venkatesh 2000) . Indeed as Ayyagari (2006) notes

on the problems raised due to key Information Systems research such as TAM not

incorporating hedonics, “this might undermine the cumulative results of TAM studies

over the past decade”. Therefore Information Systems researchers such as Kulviwat et al.

(2007) and MDS researchers (C. Carlsson et al. 2005a; C. Carlsson et al. 2005b; C.

Carlsson et al. 2006; Bina et al. 2007; Heinonen & Pura 2006; Nysveen et al. 2005b)

have continued to incorporate hedonics to improve the prediction capabilities of their

research constructs.

Figure 8: Classification of consumer value

Extracted from Fernández & Bonillo (2007)

When considering the different motives that influence consumer decisions, the research

undertaken by Sheth et al. (1991) on consumer value, which has been classified by

34

Page 35: Dissataion - Ranga Perera (CB002688)

Fernández & Bonillo (2007) as multi-dimensional, identifies five key ‘values’ that

influence the choice of consumers. Namely, functional, conditional, social, emotional and

epistemic values. It is however important to note that the researcher defines functional

value by stating “ ..functional, utilitarian and physical performance”, this statement

underpins the utilitarian motive selected as a part of the MDS research. Further in

defining emotional value, the researcher states “..arouse feeling or affective state”. It

should be noted that hedonic motives are also known as affect and are part of this

research into MDS. It is therefore necessary to appreciate that the two motives of utility

and hedonics considered as part of this research have significant and empirical theoretical

bases.

Fernández & Bonillo (2007) in defining utilitarian value based on Babin et al. (1994)

research as “instrumental, task-related, rational, functional, cognitive, and a means to an

end” and hedonic value as “reflecting the entertainment and emotional worth of shopping

non-instrumental, experiential, and affective”. hedonic value derived from the usage of a

product or service could be identified with fun or entertainment motive. Bina et al. (2007)

in defining affect “the feelings of joy, elation, or pleasure, or depression, disgust,

displeasure, or hate associated by an individual with a particular act”.

3.2.2 Technology adoption models and Mobile Data Services adoption

The Technology Adoption Model (Davis 1989) is one of the most widely used models in

explaining user adoption behavior in relation to innovative technologies especially within

the context of mandatory settings (Pedersen et al. 2008). Technology Acceptance Model

(TAM) proposed by Davis (1989) conjectured that an individuals cognitive behavioral

intent to adopt a given technology is influenced by two main constructs of perceptions,

namely, perceived usefulness and perceived ease of use. TAM further postulates the

significance of behavioral intension on the attitude of the individual towards adoption.

Davis (1989) defines perceived usefulness as “the degree to which a person believes that

using a particular system would enhance his or her job performance” and perceived ease

35

Page 36: Dissataion - Ranga Perera (CB002688)

of use as “the degree to which a person believes that using a particular system would be

free of effort”. The significance of these definitions buttress on the individuals

perceptions and not if the system in consideration is actually useful or easy to use.

Although the original postulation of TAM has been used to research and explain users

intention to use in organizational or mandatory context, Davis et al. (1989) describe the

universal adoptability of the TAM variable in computer and information systems by

users. However, the emphasis of cognitive process and its application within mandatory

settings has meant that researchers (Pedersen et al. 2002) have concluded that TAM is a

utilitarian theory on adoption of technology.

The original construct of TAM is primarily based on Theory of Reasoned Action (TRA)

proposed by Fishbein and Ajzen (1975). The application of TRA is general in comparison

to TAM, and focuses on explaining conscious behavior (Davis et al., 1989). Out of the

four variables identified in the TRA model, namely, Attitude towards behavior

(influenced by Beliefs and Evaluations), Subjective Norms (influenced by Normative

Beliefs and Motivations to comply), Behavioral Intension and actual behavior, TAM

focuses on the variables of Attitude toward use and Behavioral Intension. Taylor & Todd

(1995) in their evaluation of this proposition of Davis et al. (1989) suggest that TAM is a

special case of TRA in its application within technology adoption context. While Davis et

al. (1989) invited research into the investigation of influences of social influences, the

exclusion of this variable in the TAM due to lack of evidence of influence, remained a

point of vigorous discussion by researchers. Researches such as Mathieson (1991)

findings supported the assertions of Davis et al. (1989) on the exclusion of the subjective

norm variable within mandatory setting. However, recent studies by the original authors

and other researchers have significantly changed this proposition. Venkatesh & Davis

(2000) in their extension of the Technology Acceptance Model included the variable of

subjective norm. Further Researches such as Venkatesh & Morris (2000), Lucas & Spitler

(1999) have supported this inclusion of the social norms variable as their individual

research has identified strong influences between this variable and attitude towards

adoption within mandatory settings. It should also be noted that Theory of Planned

Behavior (TPB) is an extension to TRA by Ajzen (1991) and proposes the variable

36

Page 37: Dissataion - Ranga Perera (CB002688)

“behavioral control” to explain instances where the individuals behavior is influenced by

internal and external constraints. This inclusion of behavioral control variable has

significantly improved the predictive power of TBP considering the fact that Behavioral

intension is explain as a weighted factor of intension to use and behavioral control

(Taylor & Todd 1995).

The original TAM theory has been extensively changed and modified to improve the

validity of its predicting capability. Venkatesh & Davis (2000) included subjective norms

as “peer pressure”, that influence the persons beliefs in using the IS. Venkatesh et al.

(2003) proposed Unified Theory of Acceptance and Use of Technology (UTAUT) claims

to explain over seventy percent of variance in intention of usage behavior in both

voluntary and non-voluntary settings. However, the application of the TAM theory within

mandatory and organizational setting has meant that TAM has been categorized as a

rational, cognitive theory (Pedersen et al. 2002). (Kulviwat et al. 2007) in their construct

of Consumer Acceptance of Technology model, have pointed out that in two research

undertaken by Davis et al. (1992) and (Riemenschneider et al. 2002), the construct of

affect has been deliberately excluded, as the researchers believed that the inclusion of

hedonic variable was inappropriate within organization settings. The consistent

exclusions of affect from the primary proposition of TAM and its various subsequent

flavors have meant that researchers seeking to understand consumer behavior, which is in

many regards voluntary, have supplemented the main TAM construct. Pedersen et al.

(2002) in their analysis of E-commerce and Mobile data services adoption have used

domestication research (Haddon, 2001)(as cited by Pedersen et al. (2002) and uses and

gratification research (Leung & Wei 2000), whereas, the Consumer Acceptance of

Technology theory has used TAM with the Pleasure, Arousal and Dominance theory of

Mehrabian & Russell (1974).

Kulviwat et al. (2007) et al contend in their analysis that theories such as diffusion of

innovation (Rogers, 1995) and TAM ((Davis et al.,1989)) in their application within

consumer adoption of innovations have not considered the impact of affect, rather depend

on cognition to fully explain behavior. Heijden (2004) and Venkatesh (2000) have

37

Page 38: Dissataion - Ranga Perera (CB002688)

attempted to incorporate non-utilitarian aspects into TAM, their main problem has been

that they have been built on the cognitive model. Bina et al. (2007) criticize these

developments by pointing out that “they do not differentiate the affective from the cognitive

dimension and further assume that a person is located on an affective and cognitive bipolar

evaluative dimension”. Kulviwat et al. (2008) et al highlight the implications in identifying

the moderating influence of the nature of task the individual engaged in, whether it be

hedonic or utilitarian on the acceptance of technology. An individuals cognitive process will

be influenced by either utilitarian motives or hedonic based on the intension and experience

they may have derived prior to adopting the technology. Thus, the intension of individuals

may be equally influenced by hedonic and utilitarian motives. Therefore, in voluntary

settings the exclusion of either motive may not provide a strong construct of evaluating

consumer acceptance of technology. Kulviwat et al. (2008) in defining the utilitarian task

identifies that the task orientation primarily problem solving. This cognitive process

therefore influenced by logical, reason based approach. The need for including affect in

predicting the behavior of consumers was proposed by a number of theories such as the

Triandis (1980) and propositioned by Bina et al. (2007) in relation to Mobile Data

Services. In defining affect “the feelings of joy, elation, or pleasure, or depression,

disgust, displeasure, or hate associated by an individual with a particular act”. Bina et

al. (2007) uses the triandis theory to propose an alternative approach to analyzing the

adoption of mobile data services. Further leading researchers on mobile data services

such as Carlsson et al. (2005) use hedonic factors such as enjoyment to identify consumer

motives, while using TAM as the main construct of the research. Pedersen et al. (2002)

look to the hedonic variables of uses and gratification research to partially explain the

adoption of MDS.

Researchers on the adoption of mobile data services have been using a variety if variables

to construct the influence of hedonic variables on MDS. These variety of constructs to

monitor hedonics have not been limited to MDS but researchers in variety of fields such

as Electronic commerce, telecommunications etc. have been focusing on this regard.

Carlsson et al. (2005) uses two hedonic variables of “Enjoyment” and “new possibilities”

as the basis of evaluating potential user preferences for adoption of mobile data services.

38

Page 39: Dissataion - Ranga Perera (CB002688)

Bina et al. (2007) in incorporating the hedonic variable assessment criteria identify “fun”,

“enjoyment”, “killing time” as the potential candidate emotions towards adoption. In

contrast to these simple approaches, Pedersen et al. (2002) incorporate the uses and

gratification research to correlate the hedonic variable with adoption. While gratification

research is capable of identifying a wide range of gratifications such that was identified

by Leung & Wei (2000) including “fun-seeking”, “entertainment”, fashion and status”,

both these research point to the fact that the emotion continuum of humans are wide and

need to be captured within model that can present it within a parsimonious and

manageable content. Kulviwat et al. (2007) propositions the Consumer Acceptance of

Technology by combining the three dimensions of the Pleasure, Arousal and Domination

model (PAD) by Mehrabian & Russell (1974) to fill the vacuum in the monitoring

construct for affect. The methodology proposed by Kulviwat et al. (2007) to analyze the

affect is through the environmental psychology theory of pleasure, arousal, and

dominance (PAD) by Mehrabian and Russell’s (1974). These researchers contend that the

emotional response signaled by an individual the physical environment and social

environment can be measured within the dimensions of pleasure, arousal, and dominance.

The emotional response of the individual is mapped as a point within the three

dimensions of the PAD variables. The main basis of the Consumer Acceptance of

Technology (CAT) theory is the premise that “Consumers may adopt high-technology

products not only to obtain useful benefits but also to enjoy the experience of using

them”(Kulviwat et al. 2007). Thus unlike the TAM and its related TRB and TPB, CAT

the prediction of consumer adoption of an innovation, especially in the context of

consumer items, the role of affect should be taken into consideration.

The incorporation of relative advantage as a variable that influences the cognitive

utilitarian decision is another distinctive features of the CAT model. The theory focuses

on improving the cognitive conceptualization of belief by introducing the variable –

relative advantage. Kulviwat et al. (2007) in describing relative advantage as “relative

advantage means that the innovation is believed by the adopter to be superior in some

way to what it is intended to supersede”. This is an interesting integration of the diffusion

research with that of the TAM.

39

Page 40: Dissataion - Ranga Perera (CB002688)

Researchers on innovation and hi-technology adoption have acknowledge the causal

relationship that exists between the recognition of society and impact on attitude to adopt.

Rogers (1995) identified social systems variable including, social system norms,

tolerance of deviancy, communication integration, as one of the key groups of variables

that influence the knowledge variable/dimension of consumer. Venkatesh & Davis (2000)

identified the variable of subjective norms in their extension to the TAM model.

Kulviwat et al. (2008) in their theory recognizes the role of social influences on adoption

behavior. In their research on “private” and “public” consumption and the influence on

attitude, they observe “It seems, therefore, that adoption decisions regarding

technological innovations are more susceptible to social influence when consumption of

the product is visible to others”. This observation has major implications on the

communication strategy of firms towards their products and services.

40

Page 41: Dissataion - Ranga Perera (CB002688)

3.3 Literature review on selected independent variables

3.3.1 Independent variable 1 - Perceived usefulness

Davis (1989) in defining perceived usefulness states that it is the degree to which using

an information system is thought to improve the activities they are performing. In the

context of TAM, perceived usefulness is considered to be the most powerful predictor of

behavioral intent (Taylor & Todd 1995). In its original application within organizational

context, this variable represented the individual belief that its adoption and usage would

result in an increased performance of the job (Davis et al. 1989). In fact Davis (1989)

suggests that the variable of perceived usefulness is more important than that of

perceived ease of use, this contention was supported by Hu et al. (1999). However, as

MDS represents the consumer context, the validity of the variable may be debatable.

Bruner II & Kumar (2005) in their research on the applicability of TAM in consumer

context found that usefulness could be considered an important variable even in

consumer context. However, Pedersen et al. (2002) have identified that the prediction

capability of the usefulness variable is more strong based on the task context.

Specifically, that the usefulness is more important in relation to utilitarian MDS such as

text messaging and payment than entertainment services which are more hedonic

dependant. These research findings of stronger prediction capability of utilitarian motives

in relation to perceived usefulness rather than hedonics were confirmed by Nysveen et al.

(2005b) and Nysveen et al. (2005a). Kulviwat et al. (2007) supports this finding on the

nature of influence of perceived usefulness in the context of products used for utilitarian

motives rather than for hedonic purposes.

Within regional settings research done by Kim et al. (2007) into the Singapore Mobile

Internet usages market, Kim et al. (2009) in to the Korean wireless pay-per-view market

and Hong et al. (2006) into the Mobile internet market of Hong Kong have empirically

accepted the role played by perceived usefulness.

41

Page 42: Dissataion - Ranga Perera (CB002688)

3.3.2 Independent variable 2 - Perceived ease of use

Jenson (2006) in his critique of the MDS industry points to “default thinking” in

designing and implementing products and services. In his example of the MMS, Jenson

points to the failure of MDS industry to comprehend the value proposition and

complexity of using MMS, and instead proposing it as a natural extension to SMS. This

suggestion of industry pushing forth technology innovations and line extensions without

considering there practical usability and specifically ease of use is propositioned by him

for the failure of many MDS. Carlsson et al. (2006) supports this proposition in the

Finnish MDS market by point to mismatches of expectations between industry experts

and consumers. The survey identifies that while the industry has been introducing new

and more complex applications for MDS, consumers in general have been slow in their

adoption and continued usage of MDS. Davis (1989) defines perceived ease of use as

“the degree to which a person believes that using a particular system would be free of

effort”. This variable points to expectation of effort involved in using product or service.

Kulviwat et al. (2007) while accepting the importance of perceived ease of use as a

determinant in influencing attitude, considers the influence as indirect. They note that

rather than directly influencing the attitude of the user, it has a direct impact on the

perceived usefulness rather than intension directly. This conclusion is empirically tested

by Bruner II & Kumar (2005) who note that perceived ease of use indirectly influences

both the usefulness and fun variables. However the direct influence capability and

indirect influence capability of ease of use in the context of four mobile data services is

recognized by (Nysveen et al. 2005b) and mobile chat services (Nysveen et al. 2005a).

Here too the ease of use was noted to directly influence both attitude toward use and

usefulness.

Significantly Kim et al. (2009) rejects the influence of ease of use in the context of

inexperienced Mobile internet users and experienced users in Korea. However, the

influence of this variable is identified as important in the context of continuing usage

42

Page 43: Dissataion - Ranga Perera (CB002688)

intension. This finding of strong influence of the ease of use variable in post adoption

was confirmed by empirical research done by Hong et al. (2006) in Hong Kong.

3.3.3 Independent variable 3 - Relative advantage

Rogers (2005) included relative advantage as a part of the product variables that

influence the diffusion of innovation. In analyzing the characteristics of innovation to

include relative advantage, compatibility, complexity, triability and observability, Rogers

(2005) note that an innovation being “better” than its existing alternatives is essential.

The decision by Kulviwat et al. (2007) to incorporate relative advantage as a variable in

their research model is interesting because of very few new research literature on the

empirical testing of this variable. Plouffe et al. (2001) in testing the Perceived

Components of Innovation model which Moore & Benbasat (1991) proposed, states that

relative advantage is the most important predictor of adoption intension. The focus of the

formers’ research is comparing the prediction capability of Technology Adoption Model

with Perceived Components of Innovation model. They note that there is similarity

between the variable of perceived usefulness and relative advantage variables. It is stated

in their analysis that dependence on perceived usefulness alone may be misleading as this

variable has a number of sub-classifications – including relative advantage. Kulviwat et

al. (2007) acknowledges the lack of empirical research into the influence of relative

advantage to adoption intension in the context of information systems research, especially

in mandatory settings where uses lack the options of comparing information systems.

However, this decision by the researchers to incorporate the relative advantage variable

was important, as this variable emerged as the most important influencer of intension,

less influential than perceived usefulness and more influential than ease of use.

43

Page 44: Dissataion - Ranga Perera (CB002688)

3.3.4 Independent variable 4 - Pleasure

While there is a wide body of research that acknowledges hedonic motives (Bina et al.

2007; C. Carlsson et al. 2005; Childers et al. 2001; Heijden 2004; Heinonen & Pura 2006;

Hong et al. 2006; Kim et al. 2009) they do not attempt to proceed beyond motives of fun,

entertainment. The research proposition of Kulviwat et al. (2007) is unique in that they

attempt to develop a more deeper analytical model towards hedonic motives by

incorporating Mehrabian & Russell (1974) empirically tested Pleasure-Arousal-

Dominance scales. Lee et al. (2003) describes the pleasure emotion as “the extent to

which a person feels good”. They note of a number of research which indicate that the

emotion of pleasure, in combination with arousal and dominance, have been identified as

a stimulus in increasing purchasing behavior of customers. The research conducted by

Lee et al. (2003) confirmed the validity of pleasure in the context of online shopping. Wu

et al. (2008) in their research into the influence of pleasure and arousal in the context of

online shopping note the validity of these measures in predicting consumer buying

behavior. Wulf et al. (2006) have developed a comprehensive website evaluation model

using pleasure as the key boundary condition between the evaluation of websites and

their success.

3.3.5 Independent variable 5 - Arousal

Arousal formulates the second bipolar variable in assessing hedonic motives as proposed

by Mehrabian & Russell (1974). This bipolar nature is represented within the continuum

of feeling of being aroused to that of un-aroused. Kulviwat et al. (2007) notes that the

state of arousal is a result of a reaction of an individual to presented stimuli, influenced

primarily by the emotion of excitement. Wu et al. (2008) have identified and incorporated

arousal as an essential element in combination with pleasure to influence use buying

behavior in the online shopping and website designing context. These findings were

confirmed in an earlier research into stimulating consumer buying behavior in internet

shopping malls undertaken by Lee et al. (2003). While these research identify the

44

Page 45: Dissataion - Ranga Perera (CB002688)

variable of arousal and its influence and interplay in the consumers buying decision, a

more unique approach to appreciate arousal was proposed by Wirtz et al. (2000). They

introduce the concept of target level arousal as a moderating variable in the satisfaction of

consumers. They proposition that the satisfaction felt by the consumer is based on their

expectation of a given situation or environment. For example, the expectation of the

consumer is selecting a restaurant is for a low arousal experience vs. deciding to go to a

disco is has an embedded high arousal experience. Therefore the level of satisfaction felt

by the consumer is based on the expectation vs actual experience. They empirically

validate this proposition using dinning experience in the Singapore market.

3.3.6 Independent variable 6 - Dominance

The variable of dominance was posited by Russell & Mehrabian (1974) as the third axis

of the PAD dimensional analysis of affect. This bipolar continuum extends from

emotional state of Dominance in which the individual feels greater control over the

innovation to Submissiveness. During the emotional state of Submissiveness the range of

emotion experienced by the individual include those of anger, fear, frustration, confusion

(Russell & Mehrabian, 1977). Kulviwat et al. (2007) when incorporating dominance as

part of the Consumer Acceptance of Technology model noted that there has been

significant debate among researchers on the validity of this variable. This issue of validity

was once again raised when dominance was rejected based on it weak influence on

attitude towards adoption. However, the researchers who propositioned the Consumer

Acceptance of Technology model further investigated the dominance dimension (Nasco

et al. (2008). The researchers note the empirical findings of Yani-de-Soriano & Foxall

(2006) on the role of dominance in the context of consumer setting and their forceful

argument on the validity of this variable. Based on their research into the role of

dominance, they note that in many instances the direct effect of the variable is masked.

45

Page 46: Dissataion - Ranga Perera (CB002688)

3.3.7 Independent variable 7 - Social Influences Glotz et al. (2005) in their international review and research into Mobile phones and their

social and cultural usage note that “enabler of social interactions, hierarchies and

communication”. Bina et al. (2007) incorporate the social influences as part of the

research into the Greek market. In defining the social factor “social factors try to capture

the congruency between social norms and individual beliefs and how the human part of

an individual’s environment affects one in performing a specific behavior”. Venkatesh &

Davis (2000) have incorporated ‘subjective norms’ as an extension to the Technology

Adoption Model, in recognition of influence from the cultural and norms. The lead

researchers of Technology Adoption Model also made further research on the moderating

effects of public and private consumption and adoption of technology. The importance of

social influences have also been highlighted in Rogers (2005) Diffusion of innovation

theory. Here social systems variable is a key variable that influences the inception stage,

identified as Knowledge stage, of the adoption process. Among the sub-variables that

have been identified by the researcher are social systems norms, tolerance of deviancy,

communication integration. López-Nicolás et al. (2008) have also validated the social

influences variable in the Dutch consumer market setting. However, they make an

important observation that communication media has a positive influence effect on social

norms. The identification of this relationship is considered important because MDS

vendors can influence the attitude of society enabling greater adoption. Wei (2008) notes

that the general perception in the USA market for mobile phones is primarily as a

communication device. This general perception has created a social barrier towards the

adoption of MDS. Therefore the researcher suggests the usage of media to change this

perception of society.

46

Page 47: Dissataion - Ranga Perera (CB002688)

3.3.8 Attitude and Intention

The relationship between attitude and intension information systems research was

established as a result of the work done by Davis (1989); Davis et al. (1989) in the

technology adoption model based primarily on the findings of theory of reasoned action

Fishbein & Ajzen (1975). Kim et al. (2009) defines attitude in terms of Information

Systems as “a psychological tendency expressed by evaluating a particular entity in terms

of the degree of positiveness about IS”. Kulviwat et al. (2007) in analyzing the attitude

toward the act in the context of Technology Adoption Model, identifies the cognitive

dimension of the variable by stating that this “refers to the evaluative judgment of

adopting a piece of technology”. Therefore attitude in the context of its role in

influencing the intension of consumers could be viewed as decision or ‘judgment’.

However, Cohen & Areni (1991)(as cited by Kulviwat et al. (2007)) also point to the fact

that like all human emotions, instead of being completely cognitive, hedonics also play a

role in this judgment. This interplay between cognition and hedonics has also been the

basis of the subsequent proposition of Consumer Acceptance of Technology model by

Kulviwat et al. (2007). However, there has been significant debate among researchers of

information systems research on the validity of the influence of attitude to behavioral

intension. Due to the weak empirical evidence to support the influence of attitude on

intension Venkatesh & Davis (2000) decided to remove this motive from their extended

version of Technology Adoption Model. Researchers such as Adams et al. (1992) proved

that the Technology Adoption Model was robust even without the inclusion of attitude.

While it would have been an easier proposition to drop attitude towards adoption from

the research model, it was noted that majority of the research which decided to remove

attitude from the model mix were done in Information System mandatory settings. This

research focuses on consumer behavior and removing an important predictor could

potentially reduce the explanatory power of the model. It was further noted that through

the work of Bruner II & Kumar (2005) pointed to the mediating role between attitude and

intension. Further researchers such as Ayyagari (2006) have started to identify the

47

Page 48: Dissataion - Ranga Perera (CB002688)

presence of hedonics even within the Technology Adoption Model. Therefore both

variable were included as part of this research study.

3.3.9 Short Message Service – Mobile Data Service used to test the cognitive utilitarian value proposition

This research intend to study the behavior and influence of utilitarian and hedonic

motives, social influences on selected mobile data services with utilitarian and hedonic

value propositions. The mobile data service selected to represent utilitarian value

proposition is the popular Short Message Service. This service is known as the “killer

application” which was responsible for the identification of the mobile data services

industry and subsequent investments into 3G technology (C. Carlsson et al. 2005b; Kunin

et al. 2005).Pedersen et al. (2002) in their research into mobile data services classified

SMS as having predominantly utilitarian value propositions. This was primarily due to

the recognition that SMS was used to achieve a specific purpose and the intended value

derived from using the product is task oriented. However, Pedersen et al. (2002) notes

that there is “..potential for entertainment in addition to utility. This classification of SMS

within the value propositions of utility has been confirmed by researchers such as (C.

Carlsson et al. 2005a; C. Carlsson et al. 2005b; Nysveen et al. 2005b). Heinonen & Pura

(2006) in their complex analysis of mobile data services based on consumption, context,

social setting and relationship agree on the classification done by Pedersen et al. (2002)

and Nysveen et al. (2005). It was therefore decided to use SMS as the basis for testing the

behavioural model on utilitarian mobile data services.

3.3.10 Mobile Ringtone – Mobile Data Service used to test the hedonic value proposition

The original selection for hedonic value proposition mobile data services was Mobile

Gaming. However, due to the very low adoption rates among the pilot survey respondent,

it was decided to find a mobile data service which had similar characteristics of value.

Based on the responses from the pilot study it was noted that mobile ringtones were wide

used by the respondents (45%) and therefore used as the basis of testing the research

48

Page 49: Dissataion - Ranga Perera (CB002688)

model. Carlsson et al. (2005) in their classification of mobile data services identified that

mobile ringtones are of the entertainment category. Further using the classifications

matrix used by Pedersen et al. (2002) to categorized data services, mobile ringtones can

be considered to fall within the category of “Entertainment/ Transaction” the transaction

classification is relevant to mobile ringtone as they are used based on fee. Further the

hedonic motives of mobile ringtones were accepted by Heinonen & Pura (2006).

However in Verkasalo (2006)’s techno-centric mobile data service classification mobile

ringtone were not recognized. However, it was decided to use this product as the basis of

assessing hedonic value proposition because users adopt this product primarily due to its

entertainment value.

49

Page 50: Dissataion - Ranga Perera (CB002688)

Summary of Literature Review

3.3.11 Utilitarian Motives

Variable

Literature

Finding

Perceived Usefulness

Kulviwat et al. (2007)

Taylor & Todd (1995)

Davis et al. (1989)

Davis (1989)

Bruner II & Kumar (2005)

Pedersen et al. (2002)

Nysveen et al. (2005b)

Nysveen et al. (2005a)

Kim et al. (2007)

Kim et al. (2009)

Hong et al. (2006)

Positive Relationship

Perceived Ease of use

Davis (1989)

Carlsson et al. (2006)

Kulviwat et al. (2007)

Bruner II & Kumar (2005)

Nysveen et al. (2005b)

Nysveen et al. (2005a)

Kim et al. (2009)

Hong et al. (2006)

Positive Relationship

Relative Advantage

Rogers (2005)

Kulviwat et al. (2007)

Plouffe et al. (2001)

Moore & Benbasat (1991)

Positive Relationship

Table 3: Summary of Litreture review - Utilitarian motives

50

Page 51: Dissataion - Ranga Perera (CB002688)

3.3.12 Hedonic Motives

Variable

Literature

Finding

Pleasure

Kulviwat et al. (2007)

Lee et al. (2003)

Mehrabian & Russell (1974)

Wu et al. (2008)

Wulf et al. (2006)

Positive Relationship

Arousal

Kulviwat et al. (2007)

Wu et al. (2008)

Lee et al. (2003)

Wirtz et al. (2000)

Positive Relationship

Dominance

Nasco et al. (2008)

Yani-de-Soriano & Foxall (2006)

Positive Relationship

Table 4: Summary of literature review - Hedonic motives

Attitude and Intension Variable

Literature

Finding

Attitude and Intension Kulviwat et al. (2007)

Bruner II & Kumar (2005)

Positive relationship

Table 5: : Literature review summary - Attitude and intension Mobile data services with Utilitarian and Hedonic value propositions Mobile Data Service

Literature

Finding

Short Message Service Cognitive utilitarian

value proposition

Mobile Ringtone Hedonic value

proposition

Table 6: Literature review summary - MDS with utilitarian and hedonic propositions

51

Page 52: Dissataion - Ranga Perera (CB002688)

4. Solution

4.1 Solution overview The research focuses on two significant issues associated with mobile data service

technology. First is the analysis of existing behavioral relationships between cognitive

utilitarian motives, hedonic motives and social normative influences and attitude towards

adoption and adoption intension. The key variables identified to represent utilitarian

cognitive motives were perceived usefulness, perceived ease of use and comparative

advantage. These motivational influences were selected primarily based on the core

propositions of the Consumer Acceptance of Technology (Kulviwat et al. 2007) which

was based on the Technology Adoption Model (Davis 1989; Davis et al. 1989) and

Diffusion of innovation (Rogers 2005). The variables of Pleasure, Arousal and

Dominance were selected based on the Consumer Acceptance of Technology model. The

second element of the research is the study of the behavior of the selected variables in the

context of Short Message Services (SMS) which has an established utilitarian value

proposition(Pedersen et al. 2002; Nysveen et al. 2005b; C. Carlsson et al. 2006; Heinonen

& Pura 2006) and Mobile Ringtones (C. Carlsson et al. 2005b; C. Carlsson et al. 2005a;

Pedersen et al. 2002; Nysveen et al. 2005a) which has an established hedonic utilitarian

value proposition.

Based on these variables the following conceptual module will be used as the basis of the

two studies.

52

Page 53: Dissataion - Ranga Perera (CB002688)

Proposed model for mobile services adoption in Sri Lanka (Sri Lanka Consumer Acceptance of Technology Model – SLCAT)

Utilitarian motives Social Influences

Perceived Usefulness

Ease of Use

Attitude towards adoption

Adoption Intension

Comparative advantage

Hedonic motives

Pleasure

Arousal

Dominance

Figure 9: Proposed model for mobile services adoption in Sri Lanka

53

Page 54: Dissataion - Ranga Perera (CB002688)

4.2 List of developed hypothesis Model for analyzing Utilitarian value proposition – Using SMS 4.2.1 Utilitarian motives in the adoption of Utilitarian Services (SMS)

No Variable tested Hypothesis 1. Perceived usefulness in

the context of

Utilitarian Services

adoption and attitude

towards adoption

Hypothesis 1

H01: There is weak influence of perceived usefulness in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

Ha1: There is strong influence of perceived usefulness in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

2. Perceived Ease of use

in the context of

Utilitarian Services

adoption and attitude

towards adoption

Hypothesis 2

H02: There is weak influence of perceived ease of use in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

Ha2: There is strong influence of perceived ease of use in the context of attitude

towards adoption of utilitarian Mobile Data Services in Sri Lanka

3. Relative advantage in

the context of

Utilitarian Services

adoption and attitude

towards adoption

Hypothesis 3

H03: There is weak influence of relative advantage in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

Ha3: There is strong influence of relative advantage in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka Table 7: Hypothesis for utilitarian motives in SMS

54

Page 55: Dissataion - Ranga Perera (CB002688)

4.2.2 Hedonic motives in the adoption of Utilitarian Services (SMS)

No Variable tested Hypothesis 4. Pleasure in the context

of Utilitarian Services

adoption and attitude

towards adoption

Hypothesis 4

H04: There is weak influence of pleasure in the context of attitude towards adoption of

utilitarian Mobile Data Services in Sri Lanka

Ha4: There is strong influence of pleasure in the context of attitude towards adoption of

utilitarian Mobile Data Services in Sri Lanka

5. Arousal in the context

of Utilitarian Services

adoption and attitude

towards adoption

Hypothesis 5

H05: There is weak influence of arousal in the context of attitude towards adoption of

utilitarian Mobile Data Services in Sri Lanka

Ha5: There is strong influence of arousal in the context of attitude towards adoption of

utilitarian Mobile Data Services in Sri Lanka

6. Dominance in the

context of Utilitarian

Services adoption and

attitude towards

adoption

Hypothesis 6

H06: There is weak dominance of arousal in the context of attitude towards adoption of

adopt utilitarian Mobile Data Services in Sri Lanka

Ha6: There is strong influence of arousal in the context of attitude towards adoption of

utilitarian Mobile Data Services in Sri Lanka Table 8: Hypothesis for hedonic motives in SMS

55

Page 56: Dissataion - Ranga Perera (CB002688)

4.2.3 Role of Social Influences in the context of Utilitarian value proposition

No Variable tested Hypothesis 7. Social influence in the

context of Utilitarian

Services adoption and

attitude towards

adoption

Hypothesis 7

H07: There is weak influence of social influences in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

Ha7: There is strong influence of social influences in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka Table 9: Hypothesis of social influences in SMS

4.2.4 Relationship between Attitude towards adoption and adoption intension in the

context of Utilitarian value proposition

No Variable tested Hypothesis

8. Relationship between

attitude towards

adoption of Utilitarian

service and Adoption

Intension

Hypothesis 8

H08: There is weak influence of attitude towards adoption and adoption intension in

the context utilitarian Mobile Data Services in Sri Lanka

Ha8: There is strong influence of attitude towards adoption and adoption intension in

the context utilitarian Mobile Data Services in Sri Lanka Table 10: Hypothesis attitude and intension in SMS

56

Page 57: Dissataion - Ranga Perera (CB002688)

4.2.5 Model for analyzing Hedonic value proposition – Using Mobile Ringtone

No Variable tested Hypothesis 9. Perceived usefulness in

the context of Hedonic

Services adoption and

attitude towards

adoption

Hypothesis 9

H09: There is weak influence of perceived usefulness in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka

Ha9: There is strong influence of perceived usefulness in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka

10. Perceived Ease of use

in the context of

Hedonic Services

adoption and attitude

towards adoption

Hypothesis 10

H010: There is weak influence of perceived ease of use in the context of attitude

towards adoption of hedonic Mobile Data Services in Sri Lanka

Ha10: There is strong influence of perceived ease of use in the context of attitude

towards adoption of hedonic Mobile Data Services in Sri Lanka

11. Relative advantage in

the context of Hedonic

Services adoption and

attitude towards

adoption

Hypothesis 11

H011: There is weak influence of relative advantage in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka

Ha11: There is strong influence of relative advantage in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka Table 11: Hypothesis for utilitarian motives - M-Ringtone 4.2.6 Hedonic motives in the adoption of Hedonic Services (Mobile Ringtone)

No Variable tested Hypothesis 12. Pleasure in the context

of Hedonic Services

adoption and attitude

towards adoption

Hypothesis 12

H012: There is weak influence of pleasure in the context of attitude towards adoption of

adopt hedonic Mobile Data Services in Sri Lanka

57

Page 58: Dissataion - Ranga Perera (CB002688)

No Variable tested Hypothesis Ha12: There is strong influence of pleasure in the context of attitude towards adoption

of hedonic Mobile Data Services in Sri Lanka

13. Arousal in the context

of Hedonic Services

adoption and attitude

towards adoption

Hypothesis 13

H013: There is weak influence of arousal in the context of attitude towards adoption of

hedonic Mobile Data Services in Sri Lanka

Ha13: There is strong influence of arousal in the context of attitude towards adoption of

hedonic Mobile Data Services in Sri Lanka

14. Dominance in the

context of Hedonic

Services adoption and

attitude towards

adoption

Hypothesis 14

H014: There is weak influence of dominance in the context of attitude towards adoption

of adopt hedonic Mobile Data Services in Sri Lanka

Ha14: There is strong influence of arousal in the context of attitude towards adoption of

hedonic Mobile Data Services in Sri Lanka Table 12: Hypothesis for hedonic motives in M-Ringtones 4.2.7 Role of Social Influences in the context of Hedonic value proposition

No Variable tested Hypothesis 15. Social influence in the

context of Hedonic

Services adoption and

attitude towards

adoption

Hypothesis 15

H015: There is weak influence of social influence in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka

Ha15: There is strong influence of social influence in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka Table 13: Hypothesis for Social influences - M-Ringtones

58

Page 59: Dissataion - Ranga Perera (CB002688)

4.2.8 Relationship between Attitude towards adoption and adoption intension in the

context of hedonic value proposition

No Variable tested Hypothesis

16. Relationship between

attitude towards

adoption of hedonic

service and Adoption

Intension

Hypothesis 16

H016: There is weak influence of attitude towards adoption and adoption intension in

the context hedonic Mobile Data Services in Sri Lanka

Ha16: There is strong influence of attitude towards adoption and adoption intension in

the context hedonic Mobile Data Services in Sri Lanka Table 14: Hypothesis Attitude and intesion - M-Ringtone

59

Page 60: Dissataion - Ranga Perera (CB002688)

5. RESEARCH METHODOLOGY

Research in what ever domain looks at enriching the sea of knowledge expanding its

horizon. However Uma Sekaran (2006, p.5) provides a complete definition to it as “an

organized, systematic, data-based, critical, objective, scientific inquiry or investigation in

to a specific problem undertaken with the purpose of finding answers or solutions to it”.

Saunders et al (2007b, p.5) too provides a definition to research that says it’s something

that people undertake in order to find out something in a systematic way, thereby

increasing their knowledge.

Both these two authors identify research to be carried out in a systematic of a methodical

manner. This makes it clear that the research need to be carried out in a very structured

and planned manner where you have a clear understanding of what is required to do and

how to do it. A methodology is required that identify methods, practices and procedures

that helps to carryout the research. The following chapter identify the research

methodology used to carryout this research.

Figure 10: Research Onion (Saunders et al, 2007a)

60

Page 61: Dissataion - Ranga Perera (CB002688)

The following research onion identifies all facets that need to be identified in a research.

The discussion on the methodology would be carried out in accordance to the research

onion looking at each of its layers and discussing how and what was selected for each.

5.1 Research Philosophy

According to the research onion presented by Saunders et al (2007b, 102) the research

philosophies include; Positivism, Realism, Interpretivism, Objectivism, Subjectivism,

Pragmatism, Functionalist, Interpretive, Radical humanist and Radical structuralist. The

research philosophy used for this research can be identified as positivism. The main

reason for the selection of the following philosophy is due to the fact that the research

involves and looks at social aspects in the society and come up with a framework that can

be generalized to a subset of the society.

5.2 Research Approach

The research approach looks at either deductive research, “where you develop a theory

and hypothesis and design a research strategy to test the hypothesis, or inductive

research, in which you would collect data and develop theory as a result of the data

analysis” (Saunders et al 2007b, p.117). The research approach used in this research is

the deductive method of reasoning. This was due to the fact that the research looks at

explaining casual relationships between variables identified it the framework which used

to develop the hypothesis which would be tested with qualitative data.

Two sets of hypothesis were developed to for the two prediction models developed to test

value propositions in cognitive utilitarian mobile data services and hedonic mobile data

services. These hypothesis were developed based on the identified variables of perceived

usefulness, perceived ease of use, comparative advantage, pleasure, arousal, dominance,

social influences, attitude towards adoption and adoption intension.

61

Page 62: Dissataion - Ranga Perera (CB002688)

5.3 Research Strategy In terms of the research strategy several strategies are available such as experimental,

case study, action based, ethnography etc. However due to the nature of this research and

its domain the survey method was selected as the strategy for the research in terms of

collecting the data for the testing of hypothesis deduced from the relationships. The main

reason for selecting this strategy was due to the fact that “survey strategy is usually

associated with the deductive approach” (Saunders et al 2007b, p.138). Another fact for

selecting this was due to the nature of collecting data from a large audience and moreover

due to the fact that it is widely used in social and behavioral science research. Another

reason for choosing survey for this research is due to the fact that it can identify the

“relationships between the data and the unknown in universe” (Kothari, 2002) and also

due to the fact that it is more concerned with formulating hypothesis and testing the

association between the relationships (Kothari, 2002).

Saunders et al (2007b, p. 138) identify the following characteristics of surveys as a

research strategy;

Provides the flexibility to collect large quantities of data from a substantial

population

Considered to be as one of the most economical ways of collecting data from a

large audience

Flexibility of using quantitative analysis using descriptive and inferential statistics

Can be used to propose possible reasons for particular relationships between

variables and to produce models

Provides more control over the research process

These characteristics identified above were some of the reasons for the selection of

survey as the research strategy.

62

Page 63: Dissataion - Ranga Perera (CB002688)

5.4 Pilot study To reduce inconsistency of the research findings and to reduce the errors, a pilot study

was carried out using 40 students and 10 co workers. The main objective of having the

pilot study was to check the reliability of the research questionnaire. Saunders et al

(2007b, p.149) identify reliability as the “extent to which the data collection technique or

analysis procedure will yield consistent findings”. Reliability threats can be identified as;

participant error, participant’s biasness, observer error and observer’s biasness (Saunders

et al 2007b, p.149). Hence, the pilot study was able to identify the reliability as well as

the validity of the survey. The pilot study was conducted between the 15th of June and

20th of June, 2009. In terms of the data collection technique a mono method was used, i.e.

the use of questionnaire which was distributed in all three languages English, Sinhala and

Tamil so that the result would not be bias for one level of people.

Based on the responses of the pilot study a number of features in the questionnaire were

changed. Changes undertaken based on response;

It was identified that the general usage of mobile gaming among the selected sample

population including group of between 18-21 was very low. Only

5.5 Time Horizon

The time horizon for a research identify if the research would follow a snapshot or a diary

perspective. Snapshot is identified as cross sectional studies and a diary perspective is

identified as a longitudinal study (Saunders et al 2007b, p.148). Cross sectional study

refer to as “the study of a particular phenomenon at a particular time” and longitudinal

study that looks at change and development over a period of time (Saunders et al 2007b,

p.148). Out of the two it was decided to use cross sectional study that looks at a snap

shot of a particular effect due to the limited time span allocated for the research as well as

63

Page 64: Dissataion - Ranga Perera (CB002688)

due to the nature of the study. The research was conducted between 15th July, 2009 and

31st July, 2009.

5.6 Determining the Sample and Sample Size

The research study looks at the total population of Sri Lanka which according to the

Census and Statistics Department stands at 20,010,000. Out of which the effective

population (between the ages of 14-64) has been identified as 13,863,950. Saunders et al

(2007b, p.212) further states that a minimum sample of 384 is needed with a margin of

5% error or in other words with a confidence level of 95% for a population above

10,000,000. Hence, looking at a estimated response rate between 40 – 45%,

approximately 1100 questionnaires were distributed among the different provinces of Sri

Lanka. The below distribution was done according probabilistic sampling method and

cluster sampling. The distribution was done according to the following table.

Province No Distributed Number of trained provincial coordinators

Actual responses received by the deadline and used for the research

Actual responses received after the deadline and not used for the research

Western 350 7 230 50 Southern 100 3 25 15 Central 250 7 160 25 North Western 100 3 40 15 North Central 100 3 25 Sabaragamuwa 100 3 20 Uva 100 3 20 1100 29 455 170

Table 15: Questionnaire distribution

In the provinces a total of 29 coordinators were appointed and trained to assist the

questionnaire respondents to coordinates and present clarifications. The questionnaire

was distributed among the volunteers who attended a training workshop conducted by the

researcher. Each volunteer was given five questionnaires to fill with their family and

specific instructions not to distribute the responses among office colleagues. Further it

64

Page 65: Dissataion - Ranga Perera (CB002688)

was instructed to the volunteers that any person between the age range of 18-65 could

participate in filling the responses, irrespective of their ownership of mobile phones. The

questionnaire was designed to enable even participant who did not have mobiles to

respond. The objective of this was to understand the demographics of persons who do not

use mobile phones.

Questionnaire distribution and response information

Distributed questionnaires

(1100)

Collected Responses

(600)

Note received (500)

Accepted for data entry

(430)

Rejected prior to data entry (delayed)

(170)

Accepted for data analysis

(409)

Rejected due errors and incomplete entry

(21)

65

Page 66: Dissataion - Ranga Perera (CB002688)

5.7 Questionnaire design – Likert scales used

The utilitarian motives of the model, perceived usefulness, perceived ease of use,

comparative advantage and social influences variables were measured using five point

Likert scale of “Strongly agree”, “Agree”, “No comment”, “Disagree” and “Strongly

disagree”. The questions were designed based on similar research done by Bina et al.

(2007); López-Nicolás & Molina-Castilio (2008); Kulviwat et al. (2007). A separate set

of questions were developed for SMS and Mobile Ringtone. Each set of these questions

were intended to address the logical motives of the perceived usefulness, perceived ease

of use and comparative advantage variables. The attitude towards adoption was

developed using a five point likert scale including pleasant/unpleasant, bad/good

etc(Kulviwat et al. 2007) bipolar emotions with no comment scale. While the Intension to

use was measured using three point likert scale.

Unlike the utilitarian motives of perceived usefulness, perceived ease of use and

comparative advantage variables, the construct of questionnaire to monitor Pleasure,

Arousal and Dominance was difficult. This was primarily due to the emotions involved

and their bipolar nature. In reviewing variables identified by Kulviwat et al. (2007) to

measure PAD, it was noted that only two states were identified. For example,

Pleased/Annoyed, Satisfied/Unsatisfied etc. This was deemed unsatisfactory and five

emotional points were developed. For example, Very Pleased, Pleased, No comment,

Annoyed and Very Annoyed. However significant effort and language translation

expertise was involved when developing the questionnaire in Singhala. This was due to

some words such as “In Control” not having the same interpreted meaning in the context

of SMS usage. Due to pilot testing of the questionnaire many users complained on the

difficulties in understanding the meaning of the emotions in Singhala. Therefore, a set of

examples were written above each four group of questions that would help the user in

better understanding the question. The example would indicate two statements, for

example, “1. When Sarath uses SMS in general he feels the emotion of Happiness. 2

When Sunil uses SMS in general he does not feel any emotion”. It was noted by many

66

Page 67: Dissataion - Ranga Perera (CB002688)

questionnaire respondents that the states of emotion differ based on the circumstance. It

was based on this input the word “in general” was included.

5.8 Treatment of data

On receiving responses, they were analyzed for completeness and errors. Of the 430

responses 21 were rejected due to incomplete or errors in filling the forms. The

questionnaires were distributed among a group of twelve data entry volunteers. A

Microsoft Access Database was used to enter the data. The group of twelve volunteers

worked in two member teams to ensure quick data entry and quality assurance. One

person reads the responses, while the other enter data. The reader observes the data entry

screen for any typographic errors during the process. Further the data entry operator and

reader changes on a round robin basis to overcome stress in doing the repetitive task. At

the end of the process a printout of the responses are taken and checked randomly to

identify any data entry anomalies. The data entry was completed within two weeks. Data

from the six databases were exported to a main database and readied for analysis after

random checking for quality assurance.

67

Page 68: Dissataion - Ranga Perera (CB002688)

6. Deliverable

6.1 Descriptive Analysis The following section relates to the descriptive statistics generated based on the collected

responses. It was decided to generate descriptive statistics on the gender, age, province of

residence, general education level, current employment status, monthly income level,

type of display used and mobile data services awareness. These information provide an

understanding of the participants and their demographic information which are crucial for

the overall understanding of the proposed research model.

6.1.2 Respondents by Gender Gender

Frequency Percent Valid Percent

Cumulative

Percent

Female 225 55.0 55.0 55.0

Male 184 45.0 45.0 100.0

Total 409 100.0 100.0 Table 16: Respondents by Gender

68

Page 69: Dissataion - Ranga Perera (CB002688)

Figure 11: Respondents by Gender

The survey sample size was noted at 409 where 225 of the respondents were female and

the male respondents constituted 184. The distribution of gender is noted at 55% for

females and 45% for males. The Sri Lankan population distribution was noted at 50.4 %

for females and 49.6 % for Males. Therefore the sample may consist of higher than

normal rates of responses from females than males.

6.1.3 Respondents by Age Age

Frequency Percent Valid Percent

Cumulative

Percent

No comment 12 2.9 2.9 2.9

18-25 180 44.0 44.0 46.9

26-35 163 39.9 39.9 86.8

36-45 45 11.0 11.0 97.8

46-55 7 1.7 1.7 99.5

56-65 2 .5 .5 100.0

Total 409 100.0 100.0 Table 17 : Respondents by Age

69

Page 70: Dissataion - Ranga Perera (CB002688)

Figure 12: Respondents by Age

In 2004 Sri Lankas’ population distribution by age was as follows; between 0-14 years

25%, between 15-64 years 68.3% and above 65 years 6.5%. The age distribution of the

respondents were primarily between the ages of 18-25 representing 44% of the

respondents, ages of 26-35 representing 39% of the respondents. There were 9

respondents between the ages of 46-65 and 12 respondent who did not comment on their

age. This age distribution was considered a valid sample group considering the diffusion

of new technology and their acceptance by younger age groups.

6.1.4 Respondents by Province of residence

Province of Residence

Frequency Percent Valid Percent

Cumulative

Percent

Western Province 207 50.6 50.6 53.5

North Western Province 30 7.3 7.3 60.9

Central Province 142 34.7 34.7 95.6

Southern Province 18 4.4 4.4 100.0

0ther 12 2.9 2.9 2.9

70

Page 71: Dissataion - Ranga Perera (CB002688)

Province of Residence

Frequency Percent Valid Percent

Cumulative

Percent

Western Province 207 50.6 50.6 53.5

North Western Province 30 7.3 7.3 60.9

Central Province 142 34.7 34.7 95.6

Southern Province 18 4.4 4.4 100.0

Total 409 100.0 100.0 Table 18: Respondents by Province of residence

Figure 13: Respondents by Province of residence

The distribution of the questionnaires to provinces other than Western province was

aimed at getting a better understanding of the general population of Sri Lanka instead of

the opinions of Western province residents. However, 50% of the respondents were from

Western province, while 34% from Central, 7% from North Western, 4% from Southern

and 2% of respondents were from other provinces.

71

Page 72: Dissataion - Ranga Perera (CB002688)

6.1.5 Respondents by Education level

Education Level

Frequency Percent Valid Percent

Cumulative

Percent

GCE Ordinary Level 66 16.1 16.1 16.4

GCE Advance Level 173 42.3 42.3 58.7

Diploma Level 37 9.0 9.0 67.7

Degree and Above

qualification 132 32.3 32.3 100.0

Not responded 1 .2 .2 .2

Total 409 100.0 100.0 Table 19: Respondents by Education level

Figure 14: Respondents by Education level

Majority of the respondents had only completed advance level qualifications, while 32%

had completed University degree and above qualifications.

72

Page 73: Dissataion - Ranga Perera (CB002688)

6.1.6 Respondents by Employment status

Current Employment Status

Frequency Percent Valid Percent

Cumulative

Percent

Not responded 1 .2 .2 .2

Employed 277 67.7 67.7 68.0

Unemployed 131 32.0 32.0 100.0

Total 409 100.0 100.0 Table 20: Respondents by Employment status

Figure 15: Respondents by Employment status

67% of the respondents were employed while the balance 32% was unemployed. This

statistic is in line with the gender distribution information where 44% of the respondents

were between the ages of 18-25. Therefore their employability level would be lower than

more matured aged groups.

73

Page 74: Dissataion - Ranga Perera (CB002688)

6.1.7 Respondents by monthly income level

Monthly income Level

Frequency Percent Valid Percent

Cumulative

Percent

Less than Rs.5000 62 15.2 15.2 19.8

More than Rs5000 and Less

than Rs.10,000 73 17.8 17.8 37.7

More than Rs. 10,000 and

Less than Rs.30,000 211 51.6 51.6 89.2

More than Rs.30,000 and

Less than Rs. 50,000 35 8.6 8.6 97.8

More than Rs.50000 9 2.2 2.2 100.0

Not responded 19 4.6 4.6 4.6

Total 409 100.0 100.0 Table 21: Respondents by monthly income level

Figure 16: Respondents by monthly income level

Majority of the respondents were with the income range of Rs.10,000 to 30,000 monthly.

This income category represented 51% of the respondent. 17.8% of the respondents were

between the income range of Rs.5,000-Rs.10,000. A significantly small portion of the

respondents had an income level above Rs.50,000.

74

Page 75: Dissataion - Ranga Perera (CB002688)

Colour Display vs Black/White display

Frequency Percent Valid Percent

Cumulative

Percent

Valid

Black and White Display 283 69.2 69.2 73.6

Color Display 108 26.4 26.4 100.0

Not responded 18 4.4 4.4 4.4

Total 409 100.0 100.0 Table 22: Colour Display vs Black/White display

Figure 17: Colour Display vs Black/White display

This information was obtained to assess if the preference of the respondents were for

basic phones or more sophisticated ones. The more sophisticated the nature of the phone,

the customer has a better potential to use mobile data services.

75

Page 76: Dissataion - Ranga Perera (CB002688)

6.1.8 Mobile Data Services Awareness

No Mobile Data Service

Frequency Percentage

Aware Not aware

No comment Aware

Not aware

No comment

1 SMS 395 14 96.6 3.4 2 MMS 315 54 40 77 13.2 9.8 3 Local language messaging 353 18 38 86.3 4.4 9.3 4 Mobile E-mail 345 24 40 84.4 5.9 9.8 5 Ringtones 365 4 40 89.2 1 9.8

6 Download Icons, Logos and wallpapers 336 29 44 82.2 7.1 10.8

7 Mobile Games 361 10 38 88.3 2.4 9.3 8 Listen to music/radio 343 12 54 83.9 2.9 13.2 9 Ask/ Send credit 312 47 50 76.3 11.5 12.2

10 Book movie tickets through mobile 226 135 48 55.3 33 11.7 11 Mobile Banking 316 48 45 77.3 11.7 11 12 Mobile internet 335 29 45 81.9 7.1 11

Table 23: Mobile Data Services Awareness This information indicates that majority of the respondents are aware of the availability

of mobile data services. This high awareness level may be attributed to the high literacy

rate in Sri Lanka estimated to be 96%. Out of the selected mobile data services surveyed

knowledge of SMS was the highest at 96.6%. While awareness of Book movie tickets

through mobile (55%), mobile baking (77%) and MMS (77) were low in comparison.

These mobile data services are considered technical and require more complex

knowledge. Jenson (2006) commenting on the complexity and stages involved in using

MMS identifies that unlike SMS, using this technology is difficult and there is a high

degree of complexity.

76

Page 77: Dissataion - Ranga Perera (CB002688)

6.2 Statistical analysis of data

This section presents a summary of the model testing results for responses on utilitarian

product of SMS and Hedonic product of Ringtone. Data analysis followed the two-step

structural equation modeling approach recommended by Anderson & Gerbing (1988).

Further the recommendations of Sekaran (2006) in analyzing data were used as the basis

of analysis. The software packaged used for analysis was SPSS version 16.

The first step in the model testing and building approach involved the data being tested

for consistency and stability between the variable and responses using Cronbach’s alpha

(Sekaran, 2006). This best fit testing enabled the refining of the model and scales that did

not have high reliability with the individual variables were eliminated. The Cronbach’s

alpha measure of above 0.7 were considered acceptable according to Nunnally &

Bernstein (1994) (cited by Kulviwat et al. (2007)) was maintained in two cycles of this

best-fit model building was undertaken to ensure the quality of the test prior to hypothesis

testing. The descriptive statistics of frequency distribution table and measures of central

tendency and dispersion of maximum, minimum, means and standard deviations,

frequency distribution tables and Cronbach’s alpha values were calculated and tabulated

as indicated in Annex 2. The next stage of the analysis involved testing of the hypothesis.

This testing was done using the Pearson Correlation calculations. These were suited

based in the scaled nature of the data involved (314p, Sekaran,2006).Hypothesis were

accepted or rejected based on 95% confidence levels (p value less than 0.05).

Prior to building the prediction models for the Utilitarian and Hedonic products, it was

decided to study the individual relationship and predictability through simple linear

regression models. The models were tested based on confidence level of 95% and

ANOVA tests (F value) were conducted. In addition to these tests the residuals were

analyzed through normal probability plots (based on residuals) and scatter diagrams to

analyze if the normality assumptions were violated by the models.

77

Page 78: Dissataion - Ranga Perera (CB002688)

Summary table on internal consistency and validity testing

6.2.1 Utilitarian model testing using SMS

No Variable Cronbach’s alpha values

1. Perceived Usefulness 0.919 2. Perceived Ease of use 0.960 3. Comparative advantage 0.848 4. Social Influences 0.836 5. Pleasure 0.668 6. Arousal 0.71 7. Dominance 0.859 8. Attitude Towards Adoption 0.534 9. Adoption Intension 0.273

Table 24: Test values for internal consistency – SMS Detailed statistics are available in Annex 2.

6.2.2 Hedonic model testing using Mobile Ring tone

No Variable Cronbach’s alpha values

1. Perceived Usefulness 0.989 2. Perceived Ease of use 0.988 3. Comparative advantage 0.934 4. Social Influences 0.984 5. Pleasure 0.954 6. Arousal 0.934 7. Dominance 0.923 8. Attitude Towards Adoption 0.808 9. Adoption Intension 0.342

Table 25: Test values for internal consistency - M-Ringtones Detailed statistics are available in Annex 2.

78

Page 79: Dissataion - Ranga Perera (CB002688)

6.3 Hypothesis Testing The hypothesis testing was done using Person correlation and hypothesis were accepted

or rejected based on 95% confidence level. The correlation matrix provides an overview

of the nature of correlations between the tested variables. It was decided to add the

statistical correlation matrix generated through SPSS in Annex 03 for further information.

The below correlation matrix does not contain information on significance levels. This

information is provided in the table on Utilitarian model testing using SMS and hedonic

model testing using Mobile Ringtone.

Correlation Matrix for Utilitarian motives in attitude towards adoption (SMS)

Construct PU EOU RA SI PL AR DO ATA AI

Perceived

Usefulness (PU) 1

Perceived Ease

of Use (EOU) 0.81 1

Relative

Advantage (RA) 0.82 0.83 1

Social

Influences(SI) 0.40 0.45 0.33 1

Pleasure(PL) 0.41 0.37 0.33 0.27 1

Arousal(AR) 0.35 0.30 0.28 0.21 0.68 1

Dominance(DO) 0.25 0.13 0.17 0.15 0.30 0.34 1

Attitude toward

adoption(ATA) 0.79 0.74 0.71 0.41 0.52 0.53 0.30 1

Adoption

Intension(AI) 0.45 0.49* 0.41 0.17 0.14 0.21 0.57 0.38 1

Table 26: Correlation Matrix for SMS

79

Page 80: Dissataion - Ranga Perera (CB002688)

Utilitarian model testing using SMS

No Hypothesis Person correlation (two tailed test) result

Tested Significance level

Status of acceptance or rejection of hypothesis at 95% confidence level

1. H01:

There is weak influence of perceived usefulness in

the context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

Ha1:

There is strong influence of perceived usefulness in

the context of attitude towards adoption of

utilitarian Mobile Data Services in Sri Lanka

0.785

p value less than 0.05

Reject null hypothesis ACCEPT ALTERNATIVE HYPOTHESIS

2. H02:

There is weak influence of perceived ease of use in

the context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

Ha2:

There is strong influence of perceived ease of use in

the context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

0.739

p value less than 0.05

Reject null hypothesis ACCEPT ALTERNATIVE HYPOTHESIS

3. H03:

There weak influence of relative advantage in the

context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

Ha3: There is strong influence of relative advantage in the

context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

0.711

p value less than 0.05

Reject null hypothesis ACCEPT ALTERNATIVE HYPOTHESIS

80

Page 81: Dissataion - Ranga Perera (CB002688)

No Hypothesis Person

correlation (two tailed test) result

Tested Significance level

Status of acceptance or rejection of hypothesis at 95% confidence level

4. H04:

There is weak influence of social influences in the

context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

Ha4:

There is strong influence of social influences in the

context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

0.407

p value less than 0.05

ACCEPT NULL HYPOTHESIS Reject alternative hypothesis

5. H05:

There weak influence of pleasure in the context of

attitude towards adoption of utilitarian Mobile Data

Services in Sri Lanka

Ha5:

There is strong influence of pleasure in the context

of attitude towards adoption of utilitarian Mobile

Data Services in Sri Lanka

0.524

p value less than 0.05

Reject null hypothesis ACCEPT ALTERNATIVE HYPOTHESIS

6. H06:

There weak influence of arousal in the context of

attitude towards adoption of utilitarian Mobile Data

Services in Sri Lanka

Ha6:

There is strong influence of arousal in the context

of attitude towards adoption of utilitarian Mobile

Data Services in Sri Lanka

0.527

p value less than 0.05

Reject null hypothesis ACCEPT ALTERNATIVE HYPOTHESIS

81

Page 82: Dissataion - Ranga Perera (CB002688)

No Hypothesis Person

correlation (two tailed test) result

Tested Significance level

Status of acceptance or rejection of hypothesis at 95% confidence level

7. H07:

There weak influence of dominance in the context

of attitude towards adoption of adopt utilitarian

Mobile Data Services in Sri Lanka

Ha7:

There is strong influence of dominance in the

context of attitude towards adoption of utilitarian

Mobile Data Services in Sri Lanka

0.303

p value less than 0.05

ACCEPT NULL HYPOTHESIS Reject alternative hypothesis

8. H08:

There is weak influence of attitude towards adoption

and adoption intension in the context utilitarian

Mobile Data Services in Sri Lanka

Ha8:

There is strong influence of attitude towards

adoption and adoption intension in the context

utilitarian Mobile Data Services in Sri Lanka

0.383

p value less than 0.05

ACCEPT NULL HYPOTHESIS Reject alternative hypothesis

Table 27: Utilitarian model testing using SMS

82

Page 83: Dissataion - Ranga Perera (CB002688)

List of accepted hypothesis (alternative) – Utilitarian product

No Hypothesis

Hypothesis 1

There is strong influence of perceived usefulness in the context of attitude

towards adoption of utilitarian Mobile Data Services in Sri Lanka

Hypothesis 2

There is strong influence of perceived ease of use in the context of attitude

towards adoption of utilitarian Mobile Data Services in Sri Lanka

Hypothesis 3

There is strong influence of relative advantage in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

Hypothesis 5

There is strong influence of pleasure in the context of attitude towards adoption

of utilitarian Mobile Data Services in Sri Lanka

Hypothesis 6

There is strong influence of arousal in the context of attitude towards adoption

of utilitarian Mobile Data Services in Sri Lanka Table 28: List of accepted hypothesis (alternative) – Utilitarian product

List of Accepted Null Hypothesis

No Hypothesis

Hypothesis 4

There is weak influence of social influences in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

Hypothesis 7

There is weak influence of dominance in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka

Hypothesis 8

There is weak influence of attitude towards adoption and adoption intension in

the context utilitarian Mobile Data Services in Sri Lanka Table 29: List of Accepted Null Hypothesis

83

Page 84: Dissataion - Ranga Perera (CB002688)

Hedonic model testing using Mobile Ringtone Two models were tested in the research. This section relates to the test data on the

Hedonic value proposition model testing. The statistical analysis table generated by SPSS

was included in Annex 3 for further reference.

Correlation Matrix for hedonic motives in attitude towards adoption (Mobile Ring tones)

Construct PU EOU RA SI PL AR DO ATA AI

Perceived

Usefulness (PU) 1.00

Perceived Ease

of Use (EOU) 0.96 1.00

Relative

Advantage (RA) 0.94 0.97 1.00

Social

Influences(SI) 0.96 0.98 0.97 1.00

Pleasure(PL) 0.90 0.92 0.94 0.92 1.00 Arousal(AR) 0.78 0.74 0.75 0.79 0.72 1.00 Dominance(DO) 0.51 0.51 0.51 0.51 0.48 0.49 1.00 Attitude toward

adoption(ATA) 0.92 0.93 0.97 0.92 0.94 0.74 0.54 1.00

Adoption

Intension(AI) 0.69 0.71 0.72 0.72 0.68 0.59 0.42 0.67 1.00

Table 30: Correlation Matrix for hedonic motives

84

Page 85: Dissataion - Ranga Perera (CB002688)

Hypothesis testing for Hedonic model

No Hypothesis Person

correlation (two

tailed test)

result

Tested

Significance

level

Status of

acceptance or

rejection of

hypothesis at 95%

confidence level

1. H09:

There is weak influence of perceived usefulness in

the context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

Ha9:

There is strong influence of perceived usefulness in

the context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

0.922

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

2. H010:

There weak influence of perceived ease of use in

the context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

Ha10:

There is strong influence of perceived ease of use

in the context of attitude towards adoption of

hedonic Mobile Data Services in Sri Lanka

0.930

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

3. H011:

There weak influence of relative advantage in the

context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

Ha11:

There is strong influence of relative advantage in

the context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

0.965

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

85

Page 86: Dissataion - Ranga Perera (CB002688)

No Hypothesis Person

correlation (two

tailed test)

result

Tested

Significance

level

Status of

acceptance or

rejection of

hypothesis at 95%

confidence level

4. H012:

There weak influence of social influence in the

context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

Ha12:

There is a strong influence of social influence in the

context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

0.917

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

5. H013:

There is weak influence of pleasure in the context

of attitude towards adoption of adopt hedonic

Mobile Data Services in Sri Lanka

Ha13:

There is strong influence of pleasure in the context

of attitude towards adoption of hedonic Mobile

Data Services in Sri Lanka

0.941

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

6. H014:

There weak influence of arousal in the context of

attitude towards adoption of hedonic Mobile Data

Services in Sri Lanka

Ha14:

There is strong influence of arousal in the context

of attitude towards adoption of hedonic Mobile

Data Services in Sri Lanka

0.742

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

86

Page 87: Dissataion - Ranga Perera (CB002688)

No Hypothesis Person

correlation (two

tailed test)

result

Tested

Significance

level

Status of

acceptance or

rejection of

hypothesis at 95%

confidence level

7. H015:

There weak influence of dominance in the context

of attitude towards adoption of adopt hedonic

Mobile Data Services in Sri Lanka

Ha15:

There is strong influence of dominance in the

context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka

0.538

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

8. H016:

There is weak influence of attitude towards

adoption and adoption intension in the context

hedonic Mobile Data Services in Sri Lanka

Ha16:

There is strong influence of attitude towards

adoption and adoption intension in the context

hedonic Mobile Data Services in Sri Lanka

0.672

p value less

than 0.05

Reject null

hypothesis

ACCEPT

ALTERNATIVE

HYPOTHESIS

Table 31: Hypothesis testing for Hedonic model

87

Page 88: Dissataion - Ranga Perera (CB002688)

List of accepted hypothesis – Hedonic Product

No Hypothesis

Hypothesis 9 There is strong influence of perceived usefulness in the context of attitude

towards adoption of hedonic Mobile Data Services in Sri Lanka

Hypothesis 10 There is strong influence of perceived ease of use in the context of attitude

towards adoption of hedonic Mobile Data Services in Sri Lanka

Hypothesis 11 There is strong influence of perceived ease of use in the context of attitude

towards adoption of hedonic Mobile Data Services in Sri Lanka

Hypothesis 12 There is a strong influence of social influence in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka

Hypothesis 13 There is strong influence of pleasure in the context of attitude towards adoption

of hedonic Mobile Data Services in Sri Lanka

Hypothesis 14 There is strong influence of arousal in the context of attitude towards adoption

of hedonic Mobile Data Services in Sri Lanka

Hypothesis 15 There is strong influence of dominance in the context of attitude towards

adoption of hedonic Mobile Data Services in Sri Lanka

Hypothesis 16 There is strong influence of attitude towards adoption and adoption intension

in the context hedonic Mobile Data Services in Sri Lanka Table 32: List of accepted hypothesis – Hedonic Product

88

Page 89: Dissataion - Ranga Perera (CB002688)

6.4 Simple liner model building Variables in Model tested for Utilitarian value proposition (SMS)

No Variables Model

predictabilit

y percentage

(%) based

on the

Adjusted R

squared

value

ANOVA F

value test and

acceptability

at 95%

confidence

level – Accept

or Reject

status

Simple Liner

regression

equation

Comments on residual

value analysis

Reference in

Annex 6

1. Perceived

usefulness and

attitude

towards

adoption

61.5% Accept y = .092 + .332 X1 Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 1-4

and Figures

1-2

2. Perceived ease

of use and

attitude

towards

adoption

54.5 % Accept y = .133 + .305 X1 Model Acceptable.

45 degree upward sloping

plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 5-8

and Figures

3-4

3. Comparative

advantage and

attitude

towards

adoption

50.4% Accept y = .266 + .343 X1 Model Acceptable.

45 degree upward sloping

plot observed and

Randomly scattered

residuals an even

horizontal band around

Tables 9-12

and Figures

5-6

89

Page 90: Dissataion - Ranga Perera (CB002688)

No Variables Model

predictabilit

y percentage

(%) based

on the

Adjusted R

squared

value

ANOVA F

value test and

acceptability

at 95%

confidence

level – Accept

or Reject

status

Simple Liner

regression

equation

Comments on residual

value analysis

Reference in

Annex 6

residual value of zero

observed

4. Social

influences and

attitude

towards

adoption

16.3% Accept y = .749 + .201 X1 Reject Model.

Sparse and broken 45

degree upward sloping

plot observed. Low

random nature of the

scattered instances shown

in the plot

Tables 13-15

and Figures

7-8

5. Pleasure and

attitude

towards

adoption

27.3% Accept y = .933 + .653 X1 Reject Model.

Sparse and broken 45

degree upward sloping

plot observed. Low

random nature of the

scattered instances shown

in the plot

Tables 16-18

and Figures

9-10

6. Arousal and

attitude

towards

adoption

27.6% Accept y = 1.004 + .611 X1 Reject Model.

Sparse and broken 45

degree upward sloping

plot observed. Low

random nature of the

scattered instances shown

in the plot

Tables 19-21

and Figures

11-12

7. Dominance

and attitude

8.9% Accept y = 1.088 + .282 X1 Reject Model.

Sparse and broken 45

Tables 22-24

and Figures

90

Page 91: Dissataion - Ranga Perera (CB002688)

No Variables Model

predictabilit

y percentage

(%) based

on the

Adjusted R

squared

value

ANOVA F

value test and

acceptability

at 95%

confidence

level – Accept

or Reject

status

Simple Liner

regression

equation

Comments on residual

value analysis

Reference in

Annex 6

towards

adoption

degree upward sloping

plot observed. Low

random nature of the

scattered instances shown

in the plot

13-14

8. Attitude

towards

adoption and

Adoption

Intension

14.4% Accept y = 3.923 + .569 X1 Reject Model.

Sparse and broken 45

degree upward sloping

plot observed. Low

random nature of the

scattered instances shown

in the plot

Tables 25-27

and Figures

15-16

Table 33: Simple liner model building – SMS The simple linear regression models were tested between the identified variables and

attitude towards adoption. Based on these linear regression models and residual analysis

it was identified that the variables of Social influence, pleasure arousal, dominance and

the relationship between attitude towards adoption and adoption intension had weak

model building capabilities. Even thought the models were acceptable within the 95%

confidence level of ANOVA test, their analysis of residuals indicate that these models

will not successfully provide the required prediction capability.

91

Page 92: Dissataion - Ranga Perera (CB002688)

Variables in Model tested for Hedonic value proposition (Mobile Ringtone)

No Variables Model

predictabilit

y percentage

(%) based

on the

Adjusted R

squared

value

ANOVA F

value test and

acceptability

at 95%

confidence

level – Accept

or Reject

status

Simple Liner

regression

equation

Comments on residual

value analysis

Reference in

Annex 6

1. Perceived

usefulness and

attitude

towards

adoption

85%

Accept y = .039+ .516X1 Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 28-30

and Figures

17-18

2. Perceived ease

of use and

attitude

towards

adoption

86.5%

Accept y = .029+ .494X1 Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 31-33

and Figures

19-20

3. Comparative

advantage and

attitude

towards

adoption

93.1%

Accept y = .005+ .564X1 Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 34-36

and Figures

21-22

92

Page 93: Dissataion - Ranga Perera (CB002688)

No Variables Model

predictabilit

y percentage

(%) based

on the

Adjusted R

squared

value

ANOVA F

value test and

acceptability

at 95%

confidence

level – Accept

or Reject

status

Simple Liner

regression

equation

Comments on residual

value analysis

Reference in

Annex 6

4. Social

influences and

attitude

towards

adoption

84%

Accept y = .041+ .489X1 Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 37-39

and Figures

23-24

5. Pleasure and

attitude

towards

adoption

88.5% Accept y = .044+ .684X1

Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 40-42

and Figures

25-26

6. Arousal and

attitude

towards

adoption

54.9% Accept y = .331+ 1.226X1

Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 43-45

and Figures

27-28

93

Page 94: Dissataion - Ranga Perera (CB002688)

No Variables Model

predictabilit

y percentage

(%) based

on the

Adjusted R

squared

value

ANOVA F

value test and

acceptability

at 95%

confidence

level – Accept

or Reject

status

Simple Liner

regression

equation

Comments on residual

value analysis

Reference in

Annex 6

7. Dominance

and attitude

towards

adoption

28.8%

Accept y = .578+ 1.162X1

Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 46-48

and Figures

29-30

8. Attitude

towards

adoption and

Adoption

Intension

45% Accept y = 2.953 + .726 X1 Model Acceptable.

45 degree upward

sloping plot observed and

Randomly scattered

residuals an even

horizontal band around

residual value of zero

observed

Tables 49-50

and Figures

31-32

Table 34: Simple liner model building - Mobile Ringtones The simple linear regression models were tested between the identified variables and

attitude towards adoption. Based on these linear regression models and residual analysis

it was identified that the model build based on all the selected variables will provide a

successful prediction model.

94

Page 95: Dissataion - Ranga Perera (CB002688)

6.5 Model building

6.5.1 Utilitarian Product of SMS

6.5.2 Attitude towards adoption

Based on the statistical analysis below only five variables were identified to have strong

correlation of above 0.5 with the attitude towards adoption. These variable were

Perceived usefulness, perceived ease of use, comparative advantage, pleasure and

arousal. It was decided to analyzed the validity of the prediction model between the

incorporation of all selected variables and only utilitarian motives. The object of the

study was to evaluate if the prediction capability of the model was different between the

two selected options.

Variable ranking based on correlation to Attitude towards adoption

No Measured Variable Correlation (at 95% significance – two tailed test)

R squared value Ranking

1. Perceived Usefulness and attitude towards adoption in the context of an utilitarian product

0.79 61.50% 1

2. Perceived ease of use and attitude towards adoption in the context of an utilitarian product

0.74 54.50% 2

3. relative advantage and attitude towards adoption in the context of an utilitarian product

0.71 50.40% 3

4. Arousal and attitude towards adoption in the context of an utilitarian product

0.53 27.60% 4

5. Pleasure and attitude towards adoption in the context of an utilitarian product

0.52 27.30% 5

Table 35: Variable ranking based on correlation to Attitude towards adoption

95

Page 96: Dissataion - Ranga Perera (CB002688)

Option 1: Testing of the prediction capability of attitude towards adoption with the

incorporation of Perceived Usefulness, Perceived Ease of Use, Comparative

advantage, Pleasure and Arousal.

Option 2: Testing of the prediction capability of attitude towards adoption with only

the incorporation of Perceived Usefulness, Perceived Ease of Use and Comparative

advantage.

Result

Test Model 1 : Option 1

Pearson Correlation

Significance (Two tailed) (N=409)

.849 0.000

Liner Model y = .031+ .169 PU+ .098 EOU+ .044 CA+ .109 PL+ .266 AR

where,

PU – Perceived Usefulness

EOU – Perceived Ease of Use

CA – Comparative Advantage

PL – Pleasure

AR – Arousal

The constant value, variable of comparative advantage and

Pleasure recorded p values greater than 0.05.

Model predictability (Adjusted R squared) 71.7%

ANOVA F value was noted at 207.720 indicating the significance is under

95% confidence. Model Accepted.

Acceptability of model based on Residual

analysis

ACCEPTED

Normal probability plot (Residual) 45 degree upward sloping plot observed

Scatter plot of the standardized residuals vs

standard fitted (Residual)

Randomly scattered in an even horizontal band around residual

value of zero

Reference for detailed test data ANNEX 06, 14-16 pages

96

Page 97: Dissataion - Ranga Perera (CB002688)

Test Model 1 : Option 2

Model predictability (Adjusted R

squared)

64.4%

Liner Model y = .002+ .217 PU+ .107 EOU.035 CA

where,

PU – Perceived Usefulness

EOU – Perceived Ease of Use

CA – Comparative Advantage

The constant value and variable of comparative

advantage recorded p values greater than 0.05.

ANOVA F value was noted at 247.463 indicating the significance

is under 95% confidence. Model Accepted.

Pearson Correlation

Significance (Two tailed) (N=409)

.804

.000

Acceptability of model based on

Residual analysis

ACCEPTED

Normal probability plot (Residual) 45 degree upward sloping plot observed

Scatter plot of the standardized

residuals vs standard fitted

(Residual)

Randomly scattered in an even horizontal band

around residual value of zero

Reference for detailed test data ANNEX 06, 17-19 pages

97

Page 98: Dissataion - Ranga Perera (CB002688)

6.5.3 Intension to adopt

The objective of this analysis using the multiple regression model building statistical

methodology is to identify a prediction model for intension to adopt SMS. Two models

will be developed for comparison purposes. The first model selected for evaluation will

consist of all the variables involved in the original model. The first model consists of the

six variables of perceived usefulness, perceived ease of use, comparative advantage,

pleasure, arousal and attitude towards adoption. The first five variables listed above

helped in developing the strong prediction model for attitude towards adoption, which

had a prediction capability of 71.7%.

Option 1: Testing of the prediction capability of the adoption intension with the

incorporation of perceived usefulness, perceived ease of use, comparative advantage,

social influence, pleasure, arousal, dominance and attitude towards adoption

Option 2: Testing of the prediction capability of the adoption intension with the

incorporation of perceived usefulness, perceived ease of use, comparative advantage,

pleasure, arousal and attitude towards adoption

98

Page 99: Dissataion - Ranga Perera (CB002688)

Test Model 1 : Option 1

Model predictability (Adjusted R

squared)

40.0 %

Multiple regression Model Y = 3.139+ 0.165 EOU + 0.175 DO + 0.316 SO – 0.119 ATA +

0.85 PU – 0.032 AR + 0.158 PL

Where,

EOU – is Perceived Ease of Use

DO – Dominance

SO – Social Influences

ATA – Attitude towards adoption

PU- Perceived usefulness

AR- Arousal

PL- Pleasure

The variable of Perceived Usefulness, Perceived Ease of Use,

Comparative advantage, Pleasure and Arousal all recorded p

values greater than 0.05.

ANOVA F value was noted at 89.323 indicating the significance is under

95% confidence. Accepted

Pearson Correlation

Significance (Two tailed) (N=409)

.631

.000

Acceptability of model based on

Residual analysis

Accepted

DECSION ACCEPT MODEL

Reference for detailed test data ANNEX 06, 2-7 pages

99

Page 100: Dissataion - Ranga Perera (CB002688)

Test Model 1 : Option 2

Model predictability (Adjusted R

squared)

24.2%

Multiple regression Model When building the model it was noted that the variables of

Attitude towards adoption, Comparative advantage, pleasure

and arousal all failed in their t test values recording

significance above 95% confidence. Therefore it was decided

not to proceed with building the model.

ANOVA The F value is 22.680. This is significant under 95 % confidence (p

value is less than 0.05) Pearson Correlation

Significance (Two tailed) (N=409)

.503

.000

Acceptability of model based on

Residual analysis

Rejected

DECSION REJECT MODEL

Reference for detailed test data ANNEX 06, 7-10 pages

Based on the above analysis it was decided to accept model 1 as the prediction model

for Adoption intension for SMS

100

Page 101: Dissataion - Ranga Perera (CB002688)

6.5.4 Hedonic Product of Mobile Ringtone The statistical analysis identified that both utilitarian and hedonic motives were

influencing the adoption intension of the survey respondents. The below analysis is

aimed at identifying key variables and their influence on the prediction capability of the

Hedonic product adoption model. In order to test the models two alternative options

were developed. The first option consists variables from the original model. The second

option consists of variables representing hedonic motives. Through this analysis it would

be possible to identify if hedonic motives have a better prediction capability than

utilitarian and hedonic motives combined.

Option 1: Testing of the prediction capability of the adoption intension with the

incorporation of perceived usefulness, perceived ease of use, comparative advantage,

social influence, pleasure, arousal and attitude towards adoption

Option 2: Testing of the prediction capability of the adoption intension with the

incorporation of only Hedonic motives.

101

Page 102: Dissataion - Ranga Perera (CB002688)

Test Model 2 : Option 1

Model predictability (Adjusted R

squared)

54.2 %

Liner Model y = 2.872 + .014 HePU + .200 HeEOU + .720 HeCA -.273 HeSO

+ .221 HePL + .286 HeAR + .241 HeDO -.841 HeATA

where,

HePU – Perceived Usefulness HeEOU – Perceived Ease of Use

HeCA – Comparative Advantage

HeSO – Social influences

HePL – Pleasure

HeAR - Arousal

HeDO - Dominance

HeATA – Attitude towards adoption

When building the model it was noted that the variables of

Perceived Usefulness, perceived ease of use and social

influences all failed in their t test values recording significance

above 95% confidence.

ANOVA The F value is 61.471. This is significant under 95 % confidence (p

value is less than 0.05) Pearson Correlation

Significance (Two tailed) (N=409)

0.743 0.000

Acceptability of model based on

Residual analysis

ACCEPTED

Normal probability plot (Residual) 45 degree upward sloping plot observed

Scatter plot of the standardized

residuals vs standard fitted (Residual)

Randomly scattered in an even horizontal band around residual

value of zero

Reference for detailed test data ANNEX 06, 8-11 pages

102

Page 103: Dissataion - Ranga Perera (CB002688)

Test Model 2 : Option 2

Model predictability (Adjusted R

squared)

49.1 %

Liner Model y = 2.913 +.345 HePL + .311HeAR + .181HeDO+.095 HeATA

where,

HePL – Pleasure HeAR - Arousal

HeDO - Dominance

HeATA – Attitude towards adoption

When building the model it was noted that the variables of

Dominance and Attitude towards adoption failed in their t test

values recording significance above 95% confidence.

ANOVA The F value is 99.335. This is significant under 95 % confidence (p

value is less than 0.05)

Pearson Correlation

Significance (Two tailed) (N=409)

.743 0.000

Acceptability of model based on

Residual analysis

ACCEPTED

Normal probability plot (Residual) 45 degree upward sloping plot observed

Scatter plot of the standardized

residuals vs standard fitted

(Residual)

Randomly scattered in an even horizontal band around

residual value of zero

Reference for detailed test data ANNEX 06, 12-14 pages

Based on this analysis test model 2: Option 1 was selected as its prediction capability was

above that of its alternative.

103

Page 104: Dissataion - Ranga Perera (CB002688)

6.6 Data Analysis Summary

6.6.1 Utilitarian product – SMS adoption model testing

Perceived Usefulness

The questions to measure perceived usefulness were developed on a five point likert scale

with points between ‘Strongly Agree’, ‘agree’, ‘no comment’, ‘disagree’, ‘strongly

disagree’. Coding of the responses were done from left to right starting with 5 marks for

‘Strongly agree’ and 1 mark for ‘Strongly disagree’. 3 marks were awarded for the

response ‘no comment’. The mean value of the responses was measured at 3.4 with a

standard deviation of 1.28 (annex 2, table 2). This response mean indicates that the

majority of respondents agreed with the variables assigned to measure perceived

usefulness in the context of SMS usage. The internal consistency and reliability of the

variable were tested using Cronbach’s alpha, which was measured at 0.919 (table 24)

indicating a very good fit as noted by Hair et al. (1998)(as cited by Kulviwat et al. (2007).

The two tailed Pearson Correlation test between the variable and Attitude towards

adoption indicated a correlation of 0.785 with p value less than 0.05(annex 3). This

indicated that there is significant correlation between the two variables at 95% confidence

level. Therefore the Null hypothesis was rejected and alternative hypothesis of “There is

strong influence of perceived usefulness in the context of attitude towards adoption of

utilitarian Mobile Data Services in Sri Lanka” was accepted. To further analyse the

nature the relationship between these two variables regression analysis was conducted.

The adjusted R squared value was 0.615 indicating that the predictability of the linear

model at 95% confidence level was 61.5% . Based on the regression analysis an

individual linear prediction model was built with the equation of y = .092 + .332 X1. (y

being the Attitude towards adoption and X being the Perceived usefulness). Two tests

were conducted prior to the acceptance of the simple linear regression model. First,

ANOVA testing of the model was conducted at 95% confidence level. The model was

accepted based on this test. Residual analysis of the model was conducted as the second

104

Page 105: Dissataion - Ranga Perera (CB002688)

test. The Normal probability plot indicated a 45 degree sloping plot while the scatter plot

indicated even horizontal band around residual value of zero. In addition to these

statistical relationships the correlation between Perceived Usefulness and Intension to

adopt was observed at 0.451. This strength of correlation was above that between the

Attitude towards adoption and Intension to adopt.

Perceived Ease of Use

The questions to measure perceived ease of use were developed on a five point likert

scale with points between ‘Strongly Agree’, ‘agree’, ‘no comment’, ‘disagree’, ‘strongly

disagree’. Coding of the responses were done from left to right starting with 5 marks for

‘Strongly agree’ and 1 mark for ‘Strongly disagree’. 3 marks were awarded for the

response ‘no comment’. The mean value of the responses was measured at 3.5 with a

standard deviation of 1.3(annex 2, table 5). This mean value indicates that the majority of

respondents agreed with the variables assigned to measure perceived ease of use in the

context of SMS service usage. The internal consistency and reliability of the variable

were tested using Cronbach’s alpha, which was measured at 0.96 (table 24) indicating a

very good fit. The two tailed Pearson Correlation test between the variable and Attitude

towards adoption indicated a correlation of 0.739 with p value less than 0.05(annex 3).

This indicated that there is significant correlation between the two variables at 95%

confidence level. Therefore the Null hypothesis was rejected and alternative hypothesis

of “There is strong influence of perceived ease of use in the context of attitude towards

adoption of utilitarian Mobile Data Services in Sri Lanka” was accepted. To further

analyse the nature of the relationship between these two variables, regression analysis

was conducted. The adjusted R squared value was 0.545 indicating that the predictability

of the linear model at 95% confidence level was 54.5%. Based on the regression analysis

an individual linear prediction model was built based on the equation of y = .133 + .305

X1. (y being the Attitude towards adoption and X being the Perceived Ease of use). Two

tests were conducted prior to the acceptance of the simple linear regression model. First,

ANOVA testing of the model was conducted at 95% confidence level. The model was

accepted based on this test. Residual analysis of the model was conducted as the second

105

Page 106: Dissataion - Ranga Perera (CB002688)

test. The Normal probability plot indicated a 45 degree sloping plot while the scatter plot

indicated even horizontal band around residual value of zero. In addition to these

statistical relationships the correlation between Ease of use and Intension to adopt was

observed at 0.493 (annex 3). This is the highest noted strength between the identified

independent variables and the Intension to use variable.

Comparative Advantage

The questions to measure comparative advantage were developed on a five point likert

scale with points between ‘Strongly Agree’, ‘agree’, ‘no comment’, ‘disagree’, ‘strongly

disagree’. Coding of the responses were done from left to right starting with 5 marks for

‘Strongly agree’ and 1 mark for ‘Strongly disagree’. 3 marks were awarded for the

response ‘no comment’. The mean value of the responses was measured at 2.9 with a

standard deviation of 1.1(annex 2, table 8). This response indicates that the majority of

respondents did not agree with the proposition. However a detailed analysis of the

frequency charts and histogram indicates that over 60% of the sample respondents agreed

with the variables associated with comparative advantage in using SMS. The internal

consistency and reliability of the variables were tested using Cronbach’s alpha which was

measured at 0.85 (table 24) indicating a very good fit. The two tailed Pearson Correlation

test between the variable and Attitude towards adoption indicated a correlation of 0.711

with p value less than 0.05(annex 3). This indicated that there is significant correlation

between the two variables at 95% confidence level. Therefore the Null hypothesis was

rejected and alternative hypothesis of “There is strong influence of relative advantage in

the context of attitude towards adoption of utilitarian Mobile Data Services in Sri Lanka”

was accepted. To further analyse the nature of the nature of the relationship between

these two variables regression analysis was conducted. The adjusted R squared value was

0.504 indicating that the predictability of the linear model at 95% confidence level was

50.4%. Based on the regression analysis an individual linear prediction model was built

based on the equation of y = .266 + .343 X1. (y being the Attitude towards adoption and

X being the comparative advantage). Two tests were conducted prior to the acceptance of

the simple linear regression model. First, ANOVA testing of the model was conducted at

106

Page 107: Dissataion - Ranga Perera (CB002688)

95% confidence level. The model was accepted based on this test. Residual analysis of

the model was conducted as the second test. The Normal probability plot indicated a 45

degree sloping plot while the scatter plot indicated even horizontal band around residual

value of zero. In addition to these statistical relationships the correlation between

comparative advantage and Intension to adopt was observed at 0.408(annex 3). This is

the third highest strength identified between the independent variables and the Intension

to use variable in utilitarian product adoption model. This strength of correlation was

above that between the Attitude towards adoption and Intension to adopt.

Social Influences

The questions to measure social influences were developed on a five point likert scale

with points between ‘Strongly Agree’, ‘agree’, ‘no comment’, ‘disagree’, ‘strongly

disagree’. Coding of the responses were done from left to right starting with 5 marks for

‘Strongly agree’ and 1 mark for ‘Strongly disagree’. 3 marks were awarded for the

response ‘no comment’. The mean value of the responses was measured at 2.39 with a

standard deviation of 1.1(annex 2, table 11). This response indicates that the majority of

respondents did not agree with the proposition. This notion is confirmed as 65 % of the

respondents disagreed with the influence of social factors as a reason for using SMS. The

internal consistency and reliability of the variables were tested using Cronbach’s alpha

which was measured at 0.836 (table 24) indicating a very good fit. The two tailed Pearson

Correlation test between the variable and Attitude towards adoption indicated a

correlation of 0.407 with p value less than 0.05(annex 3). This indicated that there is

significant correlation between the two variables at 95% confidence level. Therefore the

Null hypothesis was accepted. The accepted hypothesis was “There is a weak influence

of social influences in the context of attitude towards adoption of utilitarian Mobile Data

Services in Sri Lanka”. To further analyse the nature of the nature of the relationship

between these two variables regression analysis was conducted. The adjusted R squared

value was 0.163 indicating that the predictability of the linear model at 95% confidence

level was 16.3%. Based on the regression analysis an individual linear prediction model

was built based on the equation of y = .749 + .201 X1. (y being the Attitude towards

107

Page 108: Dissataion - Ranga Perera (CB002688)

adoption and X being the Social influences). Two tests were conducted prior to the

acceptance of the simple linear regression model. First, ANOVA testing of the model was

conducted at 95% confidence level. The model was accepted based on this test. Residual

analysis of the model was conducted as the second test. Sparse and broken 45 degree

upward sloping plot observed. Low random instances were observed in the scatter plot.

Therefore it is recommended that this liner model be rejected. In addition to these

statistical relationships the correlation between social influences and Intension to adopt

was observed at 0.567. This is the strongest identified correlation other than motives of

Utility and Hedonics, between the independent variables and the Intension to use variable

in utilitarian product adoption model. This strength of correlation was above that between

the Attitude towards adoption and Intension to adopt.

Pleasure

The questions to measure pleasure was developed on a five point likert scale. Two states

of positive pleasure and two states of negative pleasure were identified with the option of

‘no comment’ as the scales. Marks were given as 2 and 1 for both positive and negative

emotional states, where ‘highly pleased’ or ‘highly displeased’ emotions received 2

marks and the ‘no comment’ response was allocated 0 marks. The mean value of the

responses was measured at 0.44 with a standard deviation of 0.44 (annex 2, table 14).

This response mean indicates that the majority of respondents did not feel any emotion

associated with pleasure when using SMS. However a detailed analysis of the frequency

charts and histogram indicates that over 62% of the sample respondents registered a

response associated with the variables pleasure in using SMS, while 152 (37%) of the

respondents indicate that they did not have any emotions identified with pleasure when

using SMS. The internal consistency and reliability of the variables were tested using

Cronbach’s alpha which was measured at 0.66 (table 24). This measure indicates a less

than optimum fit (Hair et al. (1998)(as cited by Kulviwat et al. (2007)). Three additional

rounds of cross matching the responses were conducted to asses if the Cronbach’s alpha

could be improved if certain responses were removed. But this effort was a failure as

none of these cross matching improve the level beyond 0.66. The two tailed Pearson

108

Page 109: Dissataion - Ranga Perera (CB002688)

Correlation test between the variable and Attitude towards adoption indicated a

correlation of 0.524 with p value less than 0.05(annex 3). This indicated that there is

significant correlation between the two variables at 95% confidence level. Therefore the

Null hypothesis was rejected and alternative hypothesis of “There is strong influence of

pleasure in the context of attitude towards adoption of utilitarian Mobile Data Services in

Sri Lanka” was accepted. To further analyse the nature of the relationship between these

two variables, regression analysis was conducted. The adjusted R squared value was

0.273 indicating that the predictability of the linear model at 95% confidence level was

27.3%. Based on the regression analysis an individual linear prediction model was built

based on the equation of y = .933 + .653 X1. (y being the Attitude towards adoption and X

being the pleasure). Two tests were conducted prior to the acceptance of the simple linear

regression model. First, ANOVA testing of the model was conducted at 95% confidence

level. The model was accepted based on this test. Residual analysis of the model was

conducted as the second test. Sparse and broken 45 degree upward sloping plot observed.

Low random instances were observed in the scatter plot. Therefore it is recommended

that this liner model be rejected. In addition to these statistical relationships the

correlation between Pleasure and Intension to adopt was observed at 0.170. This was the

weakest relationship identified between the independent variables and the Intension to

use variable in utilitarian product adoption model.

Arousal

The arousal variable was developed based on a five point likert scale and marks were

allocated based on the same technique used with the Pleasure variable. The mean value of

the responses was measured at 0.36 with a standard deviation of 0.46 (annex 2, table 17).

This response indicates that the majority of respondents did not feel any emotions

associated with arousal when using SMS. In a detailed analysis of the frequency charts

and histogram indicates that only 47% of the sample respondents registered a response

associated with the variable of arousal in using SMS, while 215 (52%) of the respondents

indicate that they did not have any emotions identified with pleasure when using SMS.

The internal consistency and reliability of the variables were tested using Cronbach’s

109

Page 110: Dissataion - Ranga Perera (CB002688)

alpha which was measured at 0.71(table 24). Nunnally & Bernstein (1994) (as cited by

Kulviwat et al. (2007) notes that internal consistency reliability levels between 0.7-0.8

are considered acceptable levels. The two tailed Pearson Correlation test between the

variable and Attitude towards adoption indicated a correlation of 0.527 with p value less

than 0.05(annex 3). This indicated that there is significant correlation between the two

variables at 95% confidence level. Therefore the Null hypothesis was rejected and

alternative hypothesis of “There is strong influence of arousal in the context of attitude

towards adoption of utilitarian Mobile Data Services in Sri Lanka” was accepted. To

further analyse the nature of the relationship between these two variables regression

analysis was conducted.

The adjusted R squared value was 0.276 indicating that the predictability of the linear

model at 95% confidence level was 27.6%. Based on the regression analysis an

individual linear prediction model was built based on the equation of y = 1.004 + .611 X1.

(y being the Attitude towards adoption and X being the arousal). Two tests were

conducted prior to the acceptance of the simple linear regression model. First, ANOVA

testing of the model was conducted at 95% confidence level. The model was accepted

based on this test. Residual analysis of the model was conducted as the second test.

Sparse and broken 45 degree upward sloping plot observed. Low random instances were

observed in the scatter plot. Therefore it is recommended that this liner model be rejected.

In addition to these statistical relationships the correlation between arousal and Intension

to adopt was observed at 0.143. This was the weakest relationship identified between the

hedonic independent variables and the Intension to use variable in utilitarian product

adoption model.

Dominance

The arousal variable was developed based on a five point likert scale and marks were

allocated based on the same technique used with the Pleasure variable. The mean value of

the responses was measured at 0.48 with a standard deviation of 0.58. This response

mean indicates that the majority of respondents did feel any emotions associated with

110

Page 111: Dissataion - Ranga Perera (CB002688)

arousal when using SMS. In a detailed analysis of the frequency charts and histogram

indicates that 51% of the sample respondents registered a response associated with the

variable of dominance in using SMS, while 48% of the respondents indicate that they did

not have any emotions identified with arousal when using SMS.

The internal consistency and reliability of the variables were tested using Cronbach’s

alpha which was measured at 0.859(table 24). Nunnally & Bernstein (1994) (as cited by

Kulviwat et al. (2007)) notes that internal consistency reliability levels above 0.8 are

considered good. The two tailed Pearson Correlation test between the variable and

Attitude towards adoption indicated a correlation of 0.303 with p value less than

0.05(annex 3). This indicated that there is no significant correlation between the two

variables at 95% confidence level. Therefore the Null hypothesis was accepted. The

accepted null hypothesis is “There is weak influence of dominance in the context of

attitude towards adoption of utilitarian Mobile Data Services in Sri Lanka”.

To further analyse the nature of the nature of the relationship between these two variables

regression analysis was conducted. The adjusted R squared value was 0.089 indicating

that the predictability of the linear model at 95% confidence level was 8.9%. Based on

the regression analysis an individual linear prediction model was built based on the

equation of y = 1.004 + .611 X1. (y being the Attitude towards adoption and X being the

dominance). Two tests were conducted prior to the acceptance of the simple linear

regression model. First, ANOVA testing of the model was conducted at 95% confidence

level. The model was accepted based on this test. Residual analysis of the model was

conducted as the second test. Sparse and broken 45 degree upward sloping plot observed.

Low random instances were observed in the scatter plot. Therefore it is recommended

that this liner model be rejected. In addition to these statistical relationships the

correlation between arousal and Intension to adopt was observed at 0.205.

Attitude towards adoption and intension The five point likert scale used in this instance registered responses based on “bad/good”,

“negative/positive”, “favorable/ unfavorable”, “pleasant/unpleasant” scales. Marks were

111

Page 112: Dissataion - Ranga Perera (CB002688)

allocated from 5 to 1, where the positive responses earned 5-4 marks while the negative

responses earned 2-1 marks. 3 marks were awarded for the response ‘no comment’. The

mean value of the responses for attitude towards adoption was measured at 1.22 with a

standard deviation of 0.54. (annex 2, table 24). This indicates that the attitude towards

adoption was low.

For intension to adopt the likert scales used were “unlikely/likely”,

“improbable/probable”, “impossible/possible”. With marks awarded using the same

technique as attitude. The mean value for intension to adopt was measured at 4.62 with a

standard deviation of 0.81 (annex 2, table 25). The internal consistency and reliability of

the variables were tested using Cronbach’s alpha which was measured at 0.534 for

Attitude towards adoption and 0.273 for intension. Both of these tests indicate that the

internal consistency of these variables are poor.

The two tailed Pearson Correlation test between the Attitude towards adoption and

intension to adopt is 0.383 with p value less than 0.05 (annex 3). This indicated that there

is no significant correlation between the two variables at 95% confidence level. Therefore

the Null hypothesis was accepted. The accepted null hypothesis is “There is weak

influence of attitude towards adoption and adoption intension in the context utilitarian

Mobile Data Services in Sri Lanka”. To further analyse the nature of the nature of the

relationship between these two variables regression analysis was conducted. The adjusted

R squared value was 0.144 indicating that the predictability of the linear model at 95%

confidence level was 14.4%. Based on the regression analysis an individual linear

prediction model was built based on the equation of y = 3.923 + .569 X1. (y being the

Intension to adopt and X being the Attitude towards adoption). Two tests were conducted

prior to the acceptance of the simple linear regression model. First, ANOVA testing of

the model was conducted at 95% confidence level. The model was accepted based on this

test. Residual analysis of the model was conducted as the second test. Sparse and broken

45 degree upward sloping plot observed. Low random instances were observed in the

scatter plot. Therefore it is recommended that this liner model be rejected.

112

Page 113: Dissataion - Ranga Perera (CB002688)

113

Page 114: Dissataion - Ranga Perera (CB002688)

6.6.2 Hedonic product – Mobile Ringtone adoption model testing

The general construct of the questions and scales used in testing the variables are similar

to those described in the earlier section. Therefore comments on the likert scaling will not

be included in the following analysis of response data. Detailed statistical information

relating to mean, standard deviation and frequencies are available in Annex 2.

Perceived Usefulness

The mean value of the responses was measured at 1.56 with a standard deviation of 1.8

(annex 2, table 28). This response mean indicates that the majority of respondents

strongly disagreed with the variables assigned to measure perceived usefulness in the

context of Mobile Ringtone usage. The internal consistency and reliability of the

variables were tested using Cronbach’s alpha which measured at 0.989 (table 25). The

two tailed Pearson Correlation test between the variable and Attitude towards adoption

indicated a correlation of 0.922 with p value less than 0.05(annex 4). This indicated that

there is significant correlation between the two variables at 95% confidence level.

Therefore the Null hypothesis was rejected and alternative hypothesis of “There is strong

influence of perceived usefulness in the context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka” was accepted. To further analyse the nature of the

nature of the relationship between these two variables regression analysis was conducted.

The adjusted R squared value was 0.855 indicating that the predictability of the linear

model at 95% confidence level was 85.5%. Based on the regression analysis an

individual linear prediction model was built based on the equation of y = .092 + .332 X1.

(y being the Attitude towards adoption and X being the Perceived usefulness). Two tests

were conducted prior to the acceptance of the simple linear regression model. First,

ANOVA testing of the model was conducted at 95% confidence level. The model was

accepted based on this test. Residual analysis of the model was conducted as the second

test. The Normal probability plot indicated a 45 degree sloping plot while the scatter plot

indicated even horizontal band around residual value of zero. Therefore it is

recommended that this liner model be accepted. In addition to these statistical

114

Page 115: Dissataion - Ranga Perera (CB002688)

relationships the correlation between Perceived Usefulness and Intension to adopt was

observed at 0.69. This strength of correlation was above that between the Attitude

towards adoption and Intension to adopt.

Perceived Ease of Use

The mean value of the responses was measured at 1.6 with a standard deviation of 1.9.

This response indicates that the majority of respondents disagreed with the variables

assigned to measure perceived ease of use in the context of Mobile Ringtone usage. The

internal consistency and reliability of the variables were tested using Cronbach’s alpha

which was measured at 0.988 (table 25) indicating a very good fit. The two tailed Pearson

Correlation test between the variable and Attitude towards adoption indicated a

correlation of 0.93 with p value less than 0.05(annex 4). This indicated that there is

significant correlation between the two variables at 95% confidence level. Therefore the

Null hypothesis was rejected and alternative hypothesis of “There is strong influence of

perceived ease of use in the context of attitude towards adoption of hedonic Mobile Data Services

in Sri Lanka” was accepted. To further analyse the nature of the relationship between these

two variables regression analysis was conducted. The adjusted R squared value was 0.865

indicating that the predictability of the linear model at 95% confidence level was 86.5%.

Based on the regression analysis an individual linear prediction model was built based on

the equation of y = .029+ .494X1 (y being the Attitude towards adoption and X being the

Perceived Ease of use). Two tests were conducted prior to the acceptance of the simple

linear regression model. First, ANOVA testing of the model was conducted at 95%

confidence level. The model was accepted based on this test. Residual analysis of the

model was conducted as the second test. The Normal probability plot indicated a 45

degree sloping plot while the scatter plot indicated even horizontal band around residual

value of zero. Therefore it is recommended that this liner model be accepted. In addition

to these statistical relationships the correlation between Ease of use and Intension to

adopt was observed at 0.714. This is the second highest noted strength between the

identified independent variables of utilitarian motive and the Intension to use variable.

115

Page 116: Dissataion - Ranga Perera (CB002688)

This strength of correlation was above that between the Attitude towards adoption and

Intension to adopt.

Comparative Advantage

The mean value of the responses was measured at 1.47 with a standard deviation of 1.75.

This response mean indicates that the majority of respondents did not agree with the

proposition. The internal consistency and reliability of the variables were tested using

Cronbach’s alpha which was measured at 0.93 (table 25) indicating a very good fit. The

two tailed Pearson Correlation test between the variable and Attitude towards adoption

indicated a correlation of 0.965 with p value less than 0.05(annex 4). This indicated that

there is significant correlation between the two variables at 95% confidence level.

Therefore the Null hypothesis was rejected and alternative hypothesis of “There is strong

influence of relative advantage in the context of attitude towards adoption of hedonic Mobile

Data Services in Sri Lanka” was accepted. To further analyse the nature of the nature of the

relationship between these two variables regression analysis was conducted. The adjusted

R squared value was 0.931 indicating that the predictability of the linear model at 95%

confidence level was 93.1%. Based on the regression analysis an individual linear

prediction model was built based on the equation of y = .005+ .564X1 (y being the

Attitude towards adoption and X being the comparative advantage). Two tests were

conducted prior to the acceptance of the simple linear regression model. First, ANOVA

testing of the model was conducted at 95% confidence level. The model was accepted

based on this test. Residual analysis of the model was conducted as the second test. The

Normal probability plot indicated a 45 degree sloping plot while the scatter plot indicated

even horizontal band around residual value of zero. Therefore it is recommended that this

liner model be accepted. In addition to these statistical relationships the correlation

between comparative advantage and Intension to adopt was observed at 0.721. This

strength of correlation was above that between the Attitude towards adoption and

Intension to adopt.

116

Page 117: Dissataion - Ranga Perera (CB002688)

Social Influences

The mean value of the responses was measured at 1.63 with a standard deviation of 1.9.

This response mean indicates that the majority of respondents did not agree with the

proposition. The internal consistency and reliability of the variables were tested using

Cronbach’s alpha which was measured at 0.984 (table 25) indicating a very good fit. The

two tailed Pearson Correlation test between the variable and Attitude towards adoption

indicated a correlation of 0.725 with p value less than 0.05(annex 4). This indicated that

there is significant correlation between the two variables at 95% confidence level.

Therefore the Null hypothesis was rejected. The accepted alternative hypothesis was

“There is strong influence of social influence in the context of attitude towards adoption of

hedonic Mobile Data Services in Sri Lanka”. To further analyse the nature of the nature of

the relationship between these two variables regression analysis was conducted. The

adjusted R squared value was 0.840 indicating that the predictability of the linear model

at 95% confidence level was 84.%. Based on the regression analysis an individual linear

prediction model was built based on the equation of y = .041+ .489X1. (y being the

Attitude towards adoption and X being the Social influences). Two tests were conducted

prior to the acceptance of the simple linear regression model. First, ANOVA testing of

the model was conducted at 95% confidence level. The model was accepted based on this

test. Residual analysis of the model was conducted as the second test. The Normal

probability plot indicated a 45 degree sloping plot while the scatter plot indicated even

horizontal band around residual value of zero. Therefore it is recommended that this liner

model be accepted. In addition to these statistical relationships the correlation between

social influences and Intension to adopt was observed at 0.725.

This strength of correlation was above that between the Attitude towards adoption and

Intension to adopt.

117

Page 118: Dissataion - Ranga Perera (CB002688)

Pleasure

The mean value of the responses was measured at 1.16 with a standard deviation of 1.41

indicating that most respondents felt a pleasurable emotion associated with using mobile

ringtones. The internal consistency and reliability of the variables were tested using

Cronbach’s alpha which was measured at 0.954 (table 25) for the measures indicating

good fit. The two tailed Pearson Correlation test between the variable and Attitude

towards adoption indicated a correlation of 0.941 with p value less than 0.05(annex 4).

This indicated that there is significant correlation between the two variables at 95%

confidence level. Therefore the Null hypothesis was rejected and alternative hypothesis

of “There is strong influence of pleasure in the context of attitude towards adoption of hedonic

Mobile Data Services in Sri Lanka” was accepted. To further analyse the nature of the

relationship between these two variables regression analysis was conducted. The adjusted

R squared value was 0.885 indicating that the predictability of the linear model at 95%

confidence level was 88.5%. Based on the regression analysis an individual linear

prediction model was built based on the equation of y = .044+ .684X1. (y being the

Attitude towards adoption and X being the pleasure). Two tests were conducted prior to

the acceptance of the simple linear regression model. First, ANOVA testing of the model

was conducted at 95% confidence level. The model was accepted based on this test.

Residual analysis of the model was conducted as the second test. The Normal probability

plot indicated a 45 degree sloping plot while the scatter plot indicated even horizontal

band around residual value of zero. Therefore it is recommended that this liner model be

accepted. In addition to these statistical relationships the correlation between Pleasure

and Intension to adopt was observed at 0.685.

Arousal

The mean value of the responses was measured at 0.411 with a standard deviation of 0.62

indicating that most respondents felt emotions of arousal associated with using mobile

ringtones. The internal consistency and reliability of the variables were tested using

Cronbach’s alpha which was measured at 0.93. This was noted to be a good fit. The two

118

Page 119: Dissataion - Ranga Perera (CB002688)

tailed Pearson Correlation test between the variable and Attitude towards adoption

indicated a correlation of 0.742 with p value less than 0.05(annex 4). This indicated that

there is significant correlation between the two variables at 95% confidence level.

Therefore the Null hypothesis was rejected and alternative hypothesis of “There is strong

influence of arousal in the context of attitude towards adoption of hedonic Mobile Data Services

in Sri Lanka” was accepted. To further analyse the nature of the nature of the relationship

between these two variables regression analysis was conducted. The adjusted R squared

value was 0.549 indicating that the predictability of the linear model at 95% confidence

level was 54.9%. Based on the regression analysis an individual linear prediction model

was built based on the equation of y = .331+ 1.226X1. (y being the Attitude towards

adoption and X being the arousal). Two tests were conducted prior to the acceptance of

the simple linear regression model. First, ANOVA testing of the model was conducted at

95% confidence level. The model was accepted based on this test. Residual analysis of

the model was conducted as the second test. The Normal probability plot indicated a 45

degree sloping plot while the scatter plot indicated even horizontal band around residual

value of zero. Therefore it is recommended that this liner model be accepted. In addition

to these statistical relationships the correlation between arousal and Intension to adopt

was observed at 0.595.

Dominance

The mean value of the responses was measured at 0.22 with a standard deviation of 0.47

indicating that most respondents felt emotions associated with dominance while using

mobile ringtones.. The internal consistency and reliability of the variables were tested

using Cronbach’s alpha which was measured at 0.923. The two tailed Pearson Correlation

test between the variable and Attitude towards adoption indicated a correlation of 0.538

with p value less than 0.05(annex 4). This indicated that there is no significant correlation

between the two variables at 95% confidence level. Therefore the Null hypothesis was

rejected and the accepted alternative hypothesis is “There is strong influence of dominance

in the context of attitude towards adoption of hedonic Mobile Data Services in Sri Lanka”.

119

Page 120: Dissataion - Ranga Perera (CB002688)

To further analyse the nature of the nature of the relationship between these two variables

regression analysis was conducted. The adjusted R squared value was 0.288 indicating

that the predictability of the linear model at 95% confidence level was 28.8%. Based on

the regression analysis an individual linear prediction model was built based on the

equation of y = .578+ 1.162X1. (y being the Attitude towards adoption and X being the

dominance Two tests were conducted prior to the acceptance of the simple linear

regression model. First, ANOVA testing of the model was conducted at 95% confidence

level. The model was accepted based on this test. Residual analysis of the model was

conducted as the second test. The Normal probability plot indicated a 45 degree sloping

plot while the scatter plot indicated even horizontal band around residual value of zero.

Therefore it is recommended that this liner model be accepted. In addition to these

statistical relationships the correlation between arousal and Intension to adopt was

observed at 0.421.

Attitude towards adoption and intension The mean value of the responses for attitude towards adoption was measured at 0.837

with a standard deviation of 1.03. The mean value for intension to adopt was measured at

3.56 with a standard deviation of 1.12. The internal consistency and reliability of the

variables were tested using Cronbach’s alpha which was measured at 0.808 for Attitude

towards adoption and 0.342 for intension. While the goodness of fit of the attitude

towards adoption was within acceptable range acceptability and consistency of intension

to adopt failed. The two tailed Pearson Correlation test between the Attitude towards

adoption and intension to adopt is 0.672 with p value less than 0.05(annex 4). This

indicated that there is significant correlation between the two variables at 95% confidence

level. Therefore the Null hypothesis was rejected. The accepted alternative hypothesis is

“There is strong influence of attitude towards adoption and adoption intension in the

context hedonic Mobile Data Services in Sri Lanka”. To further analyse the nature of the

nature of the relationship between these two variables regression analysis was conducted.

The adjusted R squared value was 0.450 indicating that the predictability of the linear

model at 95% confidence level was 45.0%. Based on the regression analysis an

individual linear prediction model was built based on the equation of y = 3.923 + .569 X1.

120

Page 121: Dissataion - Ranga Perera (CB002688)

(y being the Intension to adopt and X being the Attitude towards adoption). Two tests

were conducted prior to the acceptance of the simple linear regression model. First,

ANOVA testing of the model was conducted at 95% confidence level. The model was

accepted based on this test. Residual analysis of the model was conducted as the second

test. The Normal probability plot indicated a 45 degree sloping plot while the scatter plot

indicated even horizontal band around residual value of zero. Therefore it is

recommended that this liner model be accepted.

121

Page 122: Dissataion - Ranga Perera (CB002688)

7. Discussion Based on the accepted hypothesis, the following tables were produced to identify existing

literature that support these findings.

Utilitarian value proposition and variables accepted

Hypothesis

No

Measured Variable Correlation

(at 95%

significance

– two tailed

test)

R squared

value

Relevant Literature

Hypothesis

1

Perceived Usefulness and attitude

towards adoption in the context of an

utilitarian product

0.79 61.5% Pedersen et al. (2002)

Nysveen et al. (2005)

Kulviwat et al. (2007)

Bruner II & Kumar (2005)

Kim et al. (2009)

Hypothesis

2

Perceived ease of use and attitude

towards adoption in the context of an

utilitarian product

0.74 54.5 % Pedersen et al. (2002)

Nysveen et al. (2005)

Kulviwat et al. (2007)

Bruner II & Kumar (2005)

Kim et al. (2009)

Hypothesis

3

Relative advantage and attitude towards

adoption in the context of an utilitarian

product

0.71 50.4% Kulviwat et al. (2007)

Rogers (2005)

Hypothesis

5

Pleasure and attitude towards adoption

in the context of an utilitarian product

0.52 27.3% Kulviwat et al. (2007)

Wu et al. (2008) Hypothesis

6

Arousal and attitude towards adoption in

the context of an utilitarian product

0.53 27.6% Kulviwat et al. (2007)

Wu et al. (2008)

122

Page 123: Dissataion - Ranga Perera (CB002688)

Hedonic value proposition and variables accepted

No Measured Variable Correlation

(at 95%

significance

– two tailed

test)

R squared

value

Relevant Literature

Hypothesis 9

Perceived Usefulness and

attitude towards adoption in

the context of an hedonic

product

0.92 85% Pedersen et al. (2002)

Nysveen et al. (2005a)

Kulviwat et al. (2007)

Hypothesis 10

Perceived ease of use and

attitude towards adoption in

the context of an hedonic

product

0.93 86.5% Pedersen et al. (2002)

Nysveen et al. (2005a)

Kulviwat et al. (2007)

Hypothesis 11

relative advantage and attitude

towards adoption in the

context of hedonic product

0.97 93.1% Kulviwat et al. (2007)

Hypothesis 12

social influences and attitude

towards adoption in the

context of an hedonic product

0.92 84% Kulviwat et al. (2007)

Kulviwat et al. (2008)

Hypothesis 13

Pleasure and attitude towards

adoption in the context of an

hedonic product

0.94 88.5% Kulviwat et al. (2007)

Hypothesis 14

Arousal and attitude towards

adoption in the context of an

hedonic product

0.74 54.9% Kulviwat et al. (2007)

Wu et al. (2008)

Hypothesis 15

Dominance and attitude

towards adoption in the

context of an hedonic product

0.54 28.8% Nasco et al. (2008)

123

Page 124: Dissataion - Ranga Perera (CB002688)

Discussion

This research into the development of models to analyze the relationship between the

value propositions of mobile data services and motives towards adoption in Sri Lanka has

national and international significance. Within the national context, there are no

published research or statistics on Mobile Data Services adoption rates, important and

emerging mobile data services trends, important mobile data services infrastructure or

demographic trends on mobile data services popularity. Commenting on this point

Carlsson et al. 2005) notes that the mobile telecommunication industry is still focused on

selling handset instead of on mobile data services diffusion. The focus on mobile data

services in Sri Lanka also remains at an embryonic stage. This can be observed by the

publications of the national telecommunications regulator of Sri Lanka only focusing on

mobile handset penetration rates (TRC-SL 2008). The area of mobile data services truly

remains a blind spot within the regulatory and industry context in the island. This

weakness in industry focus is indeed concerning considering that the future survival of

the telecommunication industry will depend on mobile data services as indicated by the

steadily declining average revenue per user on voice charges across the global

telecommunication industry (ABI Research 2009).

Within a global context it is noted by Gao & Rafiq (2009) that there is general lack of

primary research on the mobile telecommunication industry in developing countries as

oppose that of developed countries. While the underlying technology remains the same,

the cultural and social influences need to be better understood within developing

countries. This is very important from mobile data services perspective because these

services go beyond the homogeneous nature of voice and propel the notions of

convenience and personalization (Clarke & Flaherty 2003)(as cited by (Heinonen & Pura

2006). Therefore this research attempt to use a soundly tested and accepted technology

adoption model and observe its behavior within the Sri Lankan user context. It is hoped

that this research would lay the foundation towards building of a localized adoption

model for the country.

124

Page 125: Dissataion - Ranga Perera (CB002688)

The research findings indicate that the Sri Lanka Consumer Acceptance of Technology

model (SL-CAT) presents a valid basis for analysis and prediction of technology

adoption intension. The prediction capability of SL-CAT was derived based on multiple

regression modeling. Here the prediction capability of the SL-CAT for SMS was

recorded at 40% and for mobile ringtones at 54%.This prediction capability of SL-CAT is

better than Technology Adoption Model prediction rates of 17%-33% (Davis 1989; Davis

et al. 1989; Chau & Hu 2001)(as cited by (Kulviwat et al. 2007). Further in comparison

to adoption model presented by Kulviwat et al. (2007) where the prediction capability

was listed at 53%, SL-CAT results for hedonic product of Mobile ringtones was on par.

However, in comparison to the models presented by Pedersen et al. (2002) and Nysveen

et al. (2005a) these results of SL-CAT are relatively weak. In the two comparable studies

for SMS (62%) and mobile gaming (67%) the recorded capabilities of Pedersen et al.

(2002) models are better than SL-CAT. Nysveen et al. (2005a) models noted an average

prediction capability of 72% for SMS and Mobile Games.

When analyzing the reasons for deviation between the SL-CAT and those of Pedersen et

al. (2002) and Nysveen et al. (2005), the method in which hedonic motives are analyzed

are a key factor. The SL-CAT uses the complex dimensions of Mehrabian & Russell

(1974) model of pleasure, arousal and dominance motives to record hedonic motives. In

comparison the models of Pedersen et al. (2002) and Nysveen et al. (2005) use a single

variable of “perceived enjoyment” to capture hedonic motives. Therefore, instead of

understanding the dimensionalities of the hedonic motives, these theoretical models can

be observed as “rounding-off” all the hedonic motives into one basket. However, if

researchers and mobile telecommunication industry are to better understand the

dimensionality of hedonics it is crucial that researchers go beyond the all encompassing

basket of “fun and enjoyment”. While accepting that the pleasure, arousal and dominance

variable may not be comprehensive, they do present an equitable starting point. Therefore

it could be argued that SL-CAT has the ability to better understand the motive of

hedonics in comparison to the existing research models.

125

Page 126: Dissataion - Ranga Perera (CB002688)

The significant deviation between the prediction results for SMS (predictability 40%) and

Ringtones (predictability 54%) was considered as a point of concern. One of the key

reasons for the low predictability rate of the SMS model was the low internal consistency

and reliability of the variables as tested using Cronbach’s alpha. The variables of pleasure

(0.66), arousal (0.71), attitude towards adoption (0.534) and adoption intension (0.273)

recorded the lowest internal consistency rates. The lowest rates of internal consistence

acceptable was above 0.7 as noted by Hair et al. (1998)(as cited by Kulviwat et al.

(2007). The unacceptably low rates of Cronbach’s alpha for attitude towards adoption

and adoption intension had directly resulted in diminishing the integrity of the SL-CAT

model. While efforts were undertaken to optimize the internal consistency rate, these

were unsuccessful in significantly inducing improvement. One possible reasons for these

low alpha scores were considered to be related to the translation of the questionnaire from

English to Sinhala and Tamil. However in comparison the internal consistency rates in

the SMS model, the variables of attitude towards adoption and intension to adopt motives

were recorded alpha rates of at 0.808 and 0.342 in the mobile ringtone. Unlike in the SL-

CAT SMS model, all other independent variables recorded healthy rates above 0.8 in the

SL-CAT mobile ringtone study. Therefore these identified anomalies in the design of the

questionnaire and its relation to the testing variable needs to be improved.

When presenting the Consumer Acceptance of Technology model, Kulviwat et al. (2007)

notes that all variables had internal consistency rates between 0.76 to 0.93. The internal

consistency rate noted by Pedersen et al. (2002) and Nysveen et al. (2005a) were above

0.75. Based on this analysis it is important to note that due to the poor design of the SL-

CAT questionnaire the overall validity of the model has got effected adversely. This is

especially acute in the context of the variables of attitude towards adoption and adoption

intension. Therefore an improved and refined questionnaire design may provide better

insights into the prediction capabilities of the SL-CAT model.

Noting the failure of the SMS model in significantly predicting adoption intension, it was

decided to further analyze the relationship between the identified independent variables

and attitude towards adoption. The objective of this process was to present a scale-down

126

Page 127: Dissataion - Ranga Perera (CB002688)

model of prediction for discussion. Based on the analysis, two models were built and the

accepted final model incorporated the five variables of Perceived Usefulness, Ease of use,

comparative advantage, pleasure and arousal. The multiple linear regression model

presented a prediction capability of 71.7%. Therefore in considering these results, it

could be stated with reliability that the low internal consistency rates and low correlation

between the design constructs of attitude towards adoption and adoption intension

resulted in the SMS model having an overall predictability of 40%.

Model for SMS and individual variables

The proposed adoption model for SMS product identified five key variables have a direct

effect on influencing the adoption of this mobile data service. The strongest influencer of

attitude is the variable of perceived usefulness. Revalidating the theoretical propositions

of (Davis 1989; Davis et al. 1989) the perceived usefulness emerged as the strongest

variable with a direct relationship to attitude towards adoption. While Kulviwat et al.

(2007) observed with surprise the strong relationship between usefulness and adoption

intension, this same nature of relationship was identified in this research where the

correlation between usefulness and intension was 0.451. Usefulness also showed strong

relationship between the ease of use (0.815) and comparative advantage (0.821)

variables. This indicates that more appreciation the customer has about the usefulness of

this product, greater will be their motivation to use and consume the service.

Comparative advantage is a variable with limited research beyond those conducted by

Rogers (2005). Until the variable was adopted into the Consumer Acceptance of

Technology model, there was no available research into the nature of relationship

between comparative advantage and perceived usefulness. While Kulviwat et al. (2007)

research confirmed the strong correlation between these two variables, this SL-CAT

research in Sri Lanka also re-discovers this strong relationship. However the research

conducted by Kulviwat et al. (2007) did not identify any significant interrelationship

between relative advantage and perceived ease of use. However the SL-CAT research

discovers that there is indeed a strong relationship between comparative advantage and

127

Page 128: Dissataion - Ranga Perera (CB002688)

ease of use (0.832). This would mean that improving the ease of use of the product

improves its comparative advantage in the perception of the customer. This was noted by

Jenson (2006), although without market research that, the consumers attraction towards

SMS instead of MMS due to its simple design and few steps involved in operating the

service. This relationship needs to be further explored in the context of other mobile data

services, especially during the product designing stages and product localizing stages.

This research indicates that the adoption of mobile services, with cognitive utilitarian

motives such as SMS could be influenced by focusing on this relationship.

While the relationship between perceived ease of use and attitude towards adoption was

not established by research conducted by Kulviwat et al. (2007), the SL-CAT research

model firmly establishes this relationship. Extending beyond the observed correlation

between ease of use and attitude towards adoption, the research indicates that ease of use

may also influence intension, at a moderate level (correlation of 0.49 was observed). It

should also be noted that from the group of utilitarian motives that were tested in this

research, perceived ease of use had the strongest relationship with adoption intension.

These relationships are new discoveries in the available literature and may be unique in

the Sri Lankan consumer environment. It was also noted with surprise that ease of use

had a moderate relationship (0.45) with social influences. This is indeed unique

relationship which could point to the general perceptions or societal attitude towards the

ease of using a given mobile data service positively or negatively influencing the final

adoption decision. This research finding was also observed by Lu et al. (2005) in a recent

research into social influences and adoption of technology. Further this research

discovery also has a very positive potential towards the diffusion of mobile data services

in Sri Lanka., indicating that if the societies perceptions could be changed through

sustained education and information such as by advertising of the ease of use of mobile

data services, this may have a positive impact on adoption.

The role of societal influence has been a key discussion issue in information systems

research. While the initial Consumer acceptance of technology model (Kulviwat et al.

2007) did not incorporate societal influences subsequent research by (Kulviwat et al.

128

Page 129: Dissataion - Ranga Perera (CB002688)

(2008) identified the importance of this variable in adoption. This research into adoption

of SMS shows, surprisingly that social influences have a strong influence on adoption

intension (0567) than on attitude toward adoption. This finding is different from those of

Pedersen et al. (2002) where the social influences were considered very weak in

influencing adoption intension. The researcher points to the fact that in Scandinavian

markets where mobile penetration rates have exceeded 80%, mobile data service products

such as SMS are considered mature technologies and is used by a majority of the

population. In comparison to this setting, in Sri Lanka mobile data services may be

considered a relatively new technology that is rapidly gaining ground. Therefore the

influence of society on the adoption intension is significant. Indeed in this research,

societal influences pointed to a stronger relationship to adoption intension than the

attitude towards adoption. This finding suggests that an individual may adopt the

technology primarily due to societal influences rather than a specific utilitarian or

hedonic motive. Here again, this research is important for future strategy building and

marketing budget allocations of mobile telecommunications providers. The greater effort

that is invested to influence peers and opinion leaders to adopt the technology, would

result in greater overall adoption.

In selecting the Consumer Acceptance of technology model, one of the main interests of

the research was to identify the behavior of motives of cognition and hedonics on

products categorized as having predominantly utilitarian and hedonic value propositions.

Therefore, it was not expected to have significant readings of the pleasure, arousal and

dominance motives in SMS product. The research finding also confirmed these

assumptions to a greater extent due to only pleasure (0.524) and arousal (0.527) having

significant correlation with attitude towards adoption. However, unlike the research

findings of Kulviwat et al. (2007) arousal indicated to have the same level of correlation

as pleasure with attitude towards adoption. In further analyzing this relationship it was

noted that there is a significantly strong correlation between pleasure and arousal of

0.675. this relationship was well above the levels identified by Kulviwat et al. (2007) of

0.54. It is indeed interesting as to why the researchers have not commented on this

relationship. While the measures to detect pleasure and arousal are closely aligned, this

129

Page 130: Dissataion - Ranga Perera (CB002688)

research finding requires further study and analysis to identify how increasing pleasure

would also result in an increase in arousal. Based on the presented research data it could

be noted that these two variable behave similarly, even in their influence of adoption

intension (pleasure – 0.17, Arousal .143). It is also noteworthy that the motives of

pleasure and arousal were present in users of a predominantly utilitarian product.

Therefore it could be concluded that in order to successfully launch a product with

utilitarian dimensionalities, the presence of hedonic motivators are also important.

It was not surprising to observe that the dominance motive was not present in the strong

influencers of attitude or intension. This variable had failed to be present in the original

Consumer Adoption of Technology model presented by Kulviwat et al. (2007). However

the importance or lack of it should not be discounted in the overall analysis. While

dominance is in essence relating to being in control or being under the control of another,

this research finds that dominance motive was not present among the surveyed

respondents. It is however noteworthy of Nasco et al. (2008) research finding of

dominance being a hidden covert motive than an overt motive and being task specific.

Therefore, it is suggested that further research into understanding the role of dominance

be undertaken.

130

Page 131: Dissataion - Ranga Perera (CB002688)

Model for Ringtones and individual variables

The selection of mobile ringtones to test the hedonic motive was not the original plan of

the research. It was originally planned to use mobile games, as this was the product of

choice to study hedonics in global markets. However, to the surprise of the researcher

only ten of the forty respondents of the pilot study group who were aged between 18-21

played mobile games regularly. The balance ten respondents of aged between 22-35,

while acknowledging their awareness of mobile games did not use mobile games

frequently. Therefore it was necessary to find a viable alternative which displayed the

characteristics of a product used with hedonic intension and which was being used by a

majority of the test population. The selection of the mobile ringtone was undertaken

based on these criteria. It was also noted by the researcher (subjectively) that the only

mobile data service aggressively being position to the Sri Lankan consumers are mobile

ringtones. While this observation needs to be proven or disproved based on market

research, mobile ringtones were selected for the research as the next best alternative to

mobile games.

The proposed model for predicting Mobile Ringtone adoption was considered more

successful in its overall capability than that of the SMS model. Unlike the SMS model

this model uses the utilitarian variables, Hedonic and social influences. The strength of

the correlation between perceived usefulness, attitude towards adoption and adoption

intension was noted at 0.922 and 0.690 respectively. These relationships well above those

identified by Kulviwat et al. (2007). Further unlike in the context of SMS the usefulness

variable indicates to have strong correlations with utilitarian and hedonic variables alike.

This is indeed an important relational discovery. Further usefulness is not the variable

with the strongest relationship to attitude towards adoption. This is an interesting

development because all research findings of Pedersen et al. (2002), Nysveen et al.

(2005a) and Kim et al. (2009) found that perceived usefulness as the most important

indicator of relationship between attitude and adoption intension even in the context of

products with predominantly hedonic values propositions. Therefore this finding may be

unique in the context of Sri Lanka.

131

Page 132: Dissataion - Ranga Perera (CB002688)

The variable of comparative advantage also recorded similar relationships with

usefulness (0.941), ease of use (0.965), social influences (0.974), pleasure (0.935),

arousal (0.754) and dominance (0.513). In Kulviwat et al. (2007) research there was no

significant relationship between the utilitarian motives and hedonics in general. However

in research done by Pedersen et al. (2002) there is a relatively weak correlation identified

between the move of perceived enjoyment and usefulness of 0.37, comparably, these

finding point to a stronger relationship in the Sri Lankan market context. The ease of use

variable also recorded strong relationships between the utilitarian and hedonic motives on

par with those of comparative advantage. These finding would indicate that while

consumer appreciates the hedonic motives of fun associated with Mobile ringtones, they

also have selected the product based on logical cognitive reasons.

Based on these observations of the behavior of utilitarian motives on predominantly

hedonically motivated products, can an explanation to minimal usage of mobile games be

propositioned?. These findings do suggest that in order for even hedonically valued

products to propagate into mass circulation, there needs to be a logical reasoning. Further

it was noted that the utilitarian motives of usefulness, ease of use and comparative

advantage were more strongly present than hedonic motives even in mobile ringtone

users. The utilitarian motives had strong correlations with the attitude towards adoption

and adoption intension than the relationship of hedonic motives with these same

dependant variables.

Unlike in the context of SMS, the correlation between social influences and adoption

intension was relatively weak (0.421). The relation between attitude towards adoption

and social influences were marginally stronger (0.538). While this would indicate that

society has a relatively moderate to weak influence on the adoption of the technology,

this conclusion requires further study. It should be noted that the overall promotion,

branding and market positioning of mobile data services in Sri Lanka remains very low.

Therefore the society and its opinion leader’s perception towards the adoption of this

132

Page 133: Dissataion - Ranga Perera (CB002688)

technology may be weak to indifferent. Thus, this weakness could be leveraged by the

mobile telecommunication industry to their advantage to induce diffusion.

The variables of pleasure and arousal in the context of Mobile ringtones also behaved

with the same level of correlation between them, as in SMS (0.722). However, unlike in

the context of SMS, their correlations to adoption intension was much stronger

(pleasure:0.725; Arousal:0.685). Therefore the observation that the variables of pleasure

and arousal have the same behavior patterns, as in the case of SMS is reconfirmed. These

observations required further analysis and decision making during future research. Do we

maintain the same model of Pleasure and Arousal? or in order to improve the Consumer

Acceptance of Technology, do we substitute one variable with a better predictor variable?

In order for the general acceptance of the Consumer Acceptance of Technology model, it

needs to be able to provide greater and more robust prediction capability. Therefore it is

suggested that the models proposed by Pedersen et al. (2002) and Nysveen et al. (2005)

be evaluated with the performance of CAT and the model improved accordingly.

Once again the dominance variable behaved differently than that of pleasure and arousal.

The correlation of dominance to attitude towards adoption (0.742) and intension (0.595)

was significantly higher than in the context of SMS. It has also maintained significantly

strong relations with utilitarian and hedonic motives. While this study has identified the

behavior of dominance in the context of utilitarian and hedonic products, it is indeed

difficult to explain the implications of this motive on the overall research model. The

researcher agrees with the notion of Nasco et al. (2008), that this variable has a hidden

nature to it. Rather than discounting its importance, it is suggested that future research

focus on better understanding dominance.

133

Page 134: Dissataion - Ranga Perera (CB002688)

8. Recommendations

8.1 Key determinants of Utilitarian value proposition based mobile data services

adoption variables

Utilitarian value proposition and variables accepted

Hypothesis

No

Measured Variable Correlation

(at 95%

significance

– two tailed

test)

R squared

value

Relevant Literature

Hypothesis

1

Perceived Usefulness and attitude

towards adoption in the context of an

utilitarian product

0.79 61.5% Pedersen et al. (2002)

Nysveen et al. (2005)

Kulviwat et al. (2007)

Bruner II & Kumar (2005)

Kim et al. (2009)

Hypothesis

2

Perceived ease of use and attitude

towards adoption in the context of an

utilitarian product

0.74 54.5 % Pedersen et al. (2002)

Nysveen et al. (2005)

Kulviwat et al. (2007)

Bruner II & Kumar (2005)

Kim et al. (2009)

Hypothesis

3

Relative advantage and attitude towards

adoption in the context of an utilitarian

product

0.71 50.4% Kulviwat et al. (2007)

Rogers (2005)

Hypothesis

5

Pleasure and attitude towards adoption

in the context of an utilitarian product

0.52 27.3% Kulviwat et al. (2007)

Wu et al. (2008) Hypothesis

6

Arousal and attitude towards adoption in

the context of an utilitarian product

0.53 27.6% Kulviwat et al. (2007)

Wu et al. (2008)

134

Page 135: Dissataion - Ranga Perera (CB002688)

8.2 Key determinants of hedonic value proposition based mobile data services

adoption variables

Hedonic value proposition and variables accepted

No Measured Variable Correlation

(at 95%

significance

– two tailed

test)

R squared

value

Relevant Literature

Hypothesis 9

Perceived Usefulness and

attitude towards adoption in

the context of an hedonic

product

0.92 85% Pedersen et al. (2002)

Nysveen et al. (2005a)

Kulviwat et al. (2007)

Hypothesis 10

Perceived ease of use and

attitude towards adoption in

the context of an hedonic

product

0.93 86.5% Pedersen et al. (2002)

Nysveen et al. (2005a)

Kulviwat et al. (2007)

Hypothesis 11

relative advantage and attitude

towards adoption in the

context of hedonic product

0.97 93.1% Kulviwat et al. (2007)

Hypothesis 12

social influences and attitude

towards adoption in the

context of an hedonic product

0.92 84% Kulviwat et al. (2007)

Kulviwat et al. (2008)

Hypothesis 13

Pleasure and attitude towards

adoption in the context of an

hedonic product

0.94 88.5% Kulviwat et al. (2007)

Hypothesis 14

Arousal and attitude towards

adoption in the context of an

hedonic product

0.74 54.9% Kulviwat et al. (2007)

Wu et al. (2008)

Hypothesis 15

Dominance and attitude

towards adoption in the

context of an hedonic product

0.54 28.8% Nasco et al. (2008)

135

Page 136: Dissataion - Ranga Perera (CB002688)

8.1 SL-CAT Relational Model 1 – Relationship between Utilitarian, Hedonic and

Social influences on Utilitarian value propositioned Mobile Data Services

Product - SMS

Based on the model building exercise to identify an optimum solution between the

identified variables of perceived usefulness, ease of use, comparative advantage,

pleasure, arousal, dominance and social influences, the SL-CAT relational model for

SMS adoption is recommended. The diagram below notes the identifies correlations.

8.2 SL-CAT Relational Model 2 – Relationship between Utilitarian, Hedonic and

Social influences on Hedonic value propositioned Mobile Data Services Product

– Mobile Ringtones

Based on the model building exercise to identify an optimum solution between the

identified variables of perceived usefulness, ease of use, comparative advantage,

pleasure, arousal, dominance and social influences, the following model is recommended.

136

Page 137: Dissataion - Ranga Perera (CB002688)

0.785

0.711

0.524

0.527

0.303

0.383

0.407 0.451

0.408

0.567

0.567

0.815

0.832 0.821

SL-CAT model 1-correlation diagram of Utilitarian Product of SMS adoption model

137

Page 138: Dissataion - Ranga Perera (CB002688)

0.421

0.538 0.922

0.714 0.930

0.965 0.721

0.917

0.941

0.742 0.595

0.685

0.725

0.627

SL-CAT Model 2 - Correlation diagram of Hedonic Product of Mobile Ringtone adoption model

0.941

0.722

0.491

138

Page 139: Dissataion - Ranga Perera (CB002688)

Proposed staircase model for adoption of Mobile Data Services

139

Page 140: Dissataion - Ranga Perera (CB002688)

8.3 Prediction Model 1 – Utilitarian Product of SMS adoption intension

Proposed Model Two models were developed in relation to the attitude towards adoption and adoption

intension. This was due to the failure of the adoption intension model to develop strong

prediction capability. The attitude towards adoption model was prepared as a secondary,

scaled-down version of the original model. The adoption intension model had a

prediction capability of 40% while incorporating the independent variables of perceived

usefulness, perceived ease of use, comparative advantage, social influences, pleasure and

arousal. While the overall significance of the model in terms of ANOVA analysis was

acceptable at 95% confidence level, individual variable of of Perceived Usefulness,

Perceived Ease of Use, Comparative advantage, Pleasure and Arousal all recorded p

values greater than 0.05. The multiple liner regression equation for the model as noted

below. The second scale down model on attitude towards adoption presented a prediction

capability of 71.4% . This model incorporated the variables of Perceived Usefulness,

Perceived Ease of Use, Comparative Advantage, Pleasure and Arousal. These selected

variables had a correlation of above 0.5 between them and attitude towards adoption.

while the overall significance of the model in terms of ANOVA analysis was acceptable

PREDICTION CAPABILITY = 40%

Y = 3.139+ 0.165 EOU + 0.175 DO + 0.316 SO – 0.119 ATA + 0.85 PU – 0.032 AR + 0.158 PL

Where,

EOU – is Perceived Ease of Use

DO – Dominance

SO – Social Influences

ATA – Attitude towards adoption

PU- Perceived usefulness

AR- Arousal

PL- Pleasure

140

Page 141: Dissataion - Ranga Perera (CB002688)

at 95% confidence level, individual variable of comparative advantage and pleasure

recorded individual p> 0.05.

These models can be used to predict the adoption of mobile data services.

8.4 Prediction Model 2 – Utilitarian Product of SMS attitude towards

adoption

PREDICTION CAPABILITY = 71.7%

y = .031+ .169 PU+ .098 EOU+ .044 CA+ .109 PL+ .266 AR

where,

PU – Perceived Usefulness

EOU – Perceived Ease of Use

CA – Comparative Advantage

PL – Pleasure

AR – Arousal

141

Page 142: Dissataion - Ranga Perera (CB002688)

8.5 Prediction Model 2 – Prediction model for SMS based on the selected variables

of

PREDICTION CAPABILITY = 54%

y = 2.872 + .014 HePU + .200 HeEOU + .720 HeCA -.273 HeSO + .221 HePL + .286

HeAR + .241 HeDO -.841 HeATA

where,

HePU – Perceived Usefulness HeEOU – Perceived Ease of Use

HeCA – Comparative Advantage

HeSO – Social influences

HePL – Pleasure

HeAR - Arousal

HeDO - Dominance

HeATA – Attitude towards adoption

142

Page 143: Dissataion - Ranga Perera (CB002688)

8.6 Proposed staircase model

The staircase model presented above represents the interaction of significant variables

involved in the adoption process. For adoption to take place each of the minimum

required variables identified through the model building process must be present.

Therefore as the customer overcomes and interacts with each of the variables, their

potential to increase using mobile data services will also increase.

143

Page 144: Dissataion - Ranga Perera (CB002688)

9 Future research Refine the SL-CAT model by identifying key weakness in its construct

Research into the correlation of demographic variables of age, income, gender and the

potential to adopt new mobile data services.

144

Page 145: Dissataion - Ranga Perera (CB002688)

References ABI Research, 2009. Economic conditions reducing mobile end-user spending up to

15%. IntoMobile. Available at: http://www.intomobile.com/2009/05/15/abi-research-economic-conditions-reducing-mobile-end-user-spending-up-to-15.html [Accessed May 25, 2009].

Adams, D., Nelson, R. & Todd, P., 1992. Perceived usefulness, ease of use,

and usage of information technology: A replication. MIS Quarterly, 16(2), 227–247.

Anderson, J.C. & Gerbing, D.W., 1988. Structural Equation Modeling in Practice: A

Review and Recommended Two-Step Approach. Psychological Bulletin, 103(3), 411-423.

Ayyagari, R., 2006. Examination of hedonism in TAM research. In Southern Association

for Information Systems Conferance. Florida, USA. Babin, B., Darden, W. & Griffin, M., 1994. Work and/or Fun: Measuring

Hedonic and Utilitarian Shopping. Journal of Consumer Research, 20(4), 644–656.

Bina, M., Karaiskos, D. & Giaglis, G., 2007. Motives and Barriers Affecting the Use of

Mobile Data Services. In Toronto, Ontario, Canada. Bruner II, G.C. & Kumar, A., 2005. Applying T.A.M. to consumer usage of

handheld Internet devices. Journal of Business Research, 58, 553–558. Carlsson, C., Carlsson, J. & Walden, P., 2006. Nice Mobile Services do not Fly.

Observations of Mobile Services and the Finnish Consumers. In eValues. Slovenia.

Carlsson, C. et al., 2005a. Adoption of Mobile Services across Different Technologies. In

eIntegration in Action. Bled, Slovenia. Carlsson, C. et al., 2005b. Asynchronous Adoption Patterns of Mobile Services. In

Hawaii, USA . Chau, P. & Hu, P., 2001. Information technology acceptance by individual professionals:

A model comparison approach. Decision Sciences, 32, 699–719. Childers, T.L. et al., 2001. Hedonic and utilitarian motivations for online retail shopping

behavior. Journal of Retailing, 77(4), 511-535.

145

Page 146: Dissataion - Ranga Perera (CB002688)

Clarke, I. & Flaherty, T., 2003. Mobile Portals: The Development of M-Commerce Gateways. In Mobile Commerce: Technology, Theory, and Applications. Idea Group Publishing, pp. 185-201.

Cohen, J. & Areni, C., 1991. Affect and consumer behavior. In Handbook of consumer

behavior. Englewood Cliffs, NJ: Prentice-Hall. Davis, F., 1989. Perceived usefulness, perceived ease of use, and user acceptance of

information technology. MIS Quarterly, September, 318-340. Davis, F., Bagozzi, R. & Warshaw, P., 1989. User acceptance of computer technology: A

comparison of two theoretical models. Management Science, 35, 982-1003. Fernández, R.S. & Bonillo, M.Á.I., 2007. The concept of perceived value: a systematic

review of the research. Marketing Theory, 7(4), 427-451. Fishbein, M. & Ajzen, I., 1975. Belief, attitude, intention and behavior: An introduction

to theory and research, Boston: Addison-Wesley. G15, 2008. THE ICT REVOLUTION AND G-15 COUNTRIES

Suggestions for bridging the Digital Divide, G-15 COUNTRIES. Available at: www.g15.org/workingpaper4.pdf [Accessed June 7, 2009].

Gao, P. & Rafiq, A., 2009. The transformationofthemobiletelecommunicationsindustryin

Pakistan:Adevelopingcountryperspective. TelecommunicationsPolicy, 33, 309–323.

Gilbert, L. & Han, H., 2004. Understanding mobile data services adoption: Demography,

attitudes or needs ? Technological Forecasting and Social Changes, 72, 327-337. Gilbert, L. & Kendall, J.D., 2003. A Marketing Model for Mobile Wireless Services. In

Hawaii, USA . Glotz, P., Bertschi, S. & Locke, C., 2005. Thumb Culture: The Meaning of Mobile

Phones for Society, Bielefeld: transcript Verlag. Hair, J.F. et al., 1998. Multivariate

Data Analysis with Readings 5th ed., Englewood Cliffs, NJ: Prentice Hall. Heijden, V., 2004. User acceptance of hedonic information systems. MIS Quarterly,

28(4), 695-704. Heinonen, K. & Pura, M., 2006. Classifying Mobile Services. In Helsinki.

146

Page 147: Dissataion - Ranga Perera (CB002688)

Hong, S., Thong, J.Y. & Tam, J.Y., 2006. Understanding continued information technology usage behavior: A comparison of three models in the context of mobile internet. Decision Support Systems, 42(3), 1819–1834.

Hu, P. et al., 1999. Examining the technology

acceptance model using physician acceptance of telemedicine technology. Journal of Management Information Systems, 16, 91–113.

ITU, 2009. Executive Summary. OECD Communications Outlook 2009, 10. Available at:

http://www.oecd.org/dataoecd/24/32/43472431.pdf [Accessed August 1, 2009]. ITU, 2007. ITU expects global mobile penetration rate to reach 50 percent in early 2008.

ITU, Press Release. Available at: http://www.itu.int/ITU-D/ict/newslog/ITU+Expects+Global+Mobile+Penetration+Rate+To+Reach+50+Percent+In+Early+2008.aspx [Accessed May 6, 2009].

ITU, 2008. Worldwide mobile cellular subscribers to reach 4 billion mark late 2008. ITU,

Press Release. Available at: http://www.itu.int/newsroom/press_releases/2008/29.html [Accessed August 1, 2009].

Jenson, S., 2006. Default Thinking: Why Mobile Services are setup to fail, "The Inside

Text: Social, Cultural and Design Perspectives on SMS". Available at: http://www.jensondesign.com/DefaultThinking.pdf [Accessed June 7, 2009].

Kim, B., Choi, M. & Han, I., 2009. User behaviors toward mobile data services: The role

of perceived fee and prior experience. Expert Systems with Applications, 36(4), 8528-8536.

Kim, H., Chan, H.C. & Gupta, S., 2007. Value-based Adoption of Mobile Internet:

An empirical investigation. Decision Support Systems, 43(1), 111-126. Kulviwat, S., Bruner II, G.C. & Al-Shuridah, O., 2008. The role of social influence on

adoption of high tech innovations: The moderating effect of public/private consumption. Journal of Business Research, 62(7), 706-712.

Kulviwat, S. et al., 2007. Toward a Unified Theory

of Consumer Acceptance Technology. Psychology & Marketing, 24(12), 1059–1084.

Kunin, C., Vickery, G. & Wunsch-Vincent, S., 2005. DIGITAL BROADBAND

CONTENT: MOBILE CONTENT NEW CONTENT FOR NEW PLATFORMS, France: Organisation for Economic Co-operation and Development.

147

Page 148: Dissataion - Ranga Perera (CB002688)

Lee, M., Suh, K. & Whang, J., 2003. The impact of situation awareness information on consumer attitudes in the Internet shopping mall. Electronic Commerce Research and Applications, 2(3), 254–265.

López-Nicolás, C. & Molina-Castilio, F., 2008. An assessment of advanced mobile

services acceptance: Contributions from TAM and diffusion theory models. Information and Management, 45, 359-364.

Lu, J., Yao, J. & Yu, C., 2005. Personal innovativeness, social influences and adoption of

wireless internet services via mobile technology. Journal of Strategic Information Systems, 14, 245-268.

Mälarstig, Y., Holmström, O. & Davidsson, P., 2007. The Swedish Telecommunications

Market 2007, Stockholm, Sweden: National Post and Telecom Agency. Marketing Sciences Institute, 2006. Research Priorities (2006–2008) A guide to MSI

research programs and procedures, Marketing Sciences Institute. Available at: http://www.msi.org/pdf/MSI_RP06-08.pdf [Accessed June 1, 2006].

Mehrabian, A. & Russell, J.A., 1974. An Approach to Environmental Psychology.

In MIT Press. Moore, G.C. & Benbasat, I., 1991. Development of an instrument to measure the

perceptions of adopting an information technology innovation. Information Systems Research, 2, 192–222.

Nasco, S.A. et al., 2008. The CAT model: Extensions and moderators of dominance in

technology acceptance. Psychology and Marketing, 25(10), 987-1005. Nielsen Research, 2008. Realizing Potential: Overcoming Barriers to the U.S. Mobile

Advertising Market, USA: The Nielsen Company. Nunnally, J.C. & Bernstein, I., 1994. Psychometric theory 3rd ed., New York: McGraw-

Hill. Nysveen, H., Pedersen, P.E. & Thorbjørnsen, H., 2005a. Explaining intension to use

mobile chat services: Moderating effects of gender. Journal of Consumer Marketing, 25(5), 247-256.

Nysveen, H., Pedersen, P.E. & Thorbjørnsen, H., 2005b. Intentions to Use Mobile

Services: Antecedents and Cross-Service Comparisons. Journal of the Academy of Marketing Science, 33(3), 330-346.

148

Page 149: Dissataion - Ranga Perera (CB002688)

Pedersen, P.E., Methlie, L. & Nysveen, H., 2008. An exploratory study of business model design and costumer value in heterogeneous network services, Bergen, Norway: Foundation for Research in Economics and Business Administration.

Pedersen, P.E., Nysveen, H. & Thorbjørnsen, H., 2002. The adoption of mobile services:

A cross service study., Norway: Foundation for Research in Economics and Business Administration.

Plouffe, C.R., Hulland, J.S. & Vandenbosch, M., 2001. Research report: Richness

versus parsimony in modeling technology adoption decisions: Understanding merchant adoption of a smart card-based payment system. Information Systems Research, 12, 208–222.

Ratan, S.R.A., Alam, M.S. & Sohel-Uz-Zaman, A.S.M., 2007. Innovation Strategy: A

Realistic Example in Bangladesh Mobile Telecommunication Service Marketing. The Cost and Management, 35(1), 49-58.

Rogers, E.M., 2005. Diffusion of Innovations, New York: Free Press. Sekaran, U., 2006. Research menthods for business - A skills building approach 4th ed.,

Wileys Publication. Sheth, J., Newman, B. & Gross, B., 1991. Why We Buy What We Buy: A

Theory of Consumption Values. Journal of Business Research, 22(2), 159-170. Taylor, S. & Todd, P., 1995. Understanding information technology usage: A test of

competing models. Information Systems Research, 6, 144-176. TRC-SL, 2008. Statistical Overview - 2008 December, Sri Lanka: Telecommunication

Regulatory Commission of Sri Lanka. Available at: http://www.trc.gov.lk/images/pdf/stodec1.pdf [Accessed January 8, 2009].

Umino, A., 2004. DEVELOPMENT OF THIRD-GENERATION MOBILE SERVICES IN

THE OECD, Organisation for Economic Co-operation and Development. Venkatesh, V., 2000. Determinants of Perceived Ease of Use: Integrating Perceived

Behavioral Control, Computer Anxiety and Enjoyment into the Technology Acceptance Model. Information Systems Research, 11, 342-365.

Venkatesh, V. & Davis, F., 2000. A Theoretical Extension of the Technology Acceptance

Model: Four Longitudinal Field Studies. Management Science, 46(2), 186-204. Verkasalo, H., 2006. Mobile Data Service Evolution - Empirical Observations on Packet

Data Service Adoption. In Denmark.

149

Page 150: Dissataion - Ranga Perera (CB002688)

Wirtz, J., Mattila, A.S. & Tan, R.L.P., 2000. The Moderating Role of Target-Arousal on the Impact of Affect on Satisfaction—An Examination in the Context of Service Experiences. Journal of Retailing, 76(3), 347-365.

Wu, C., Cheng, F. & Yen, D.C., 2008. The atmospheric factors of online storefront

environment design: An empirical experiment in Taiwan. Information & Management, 45(7), 493–498.

Wulf, K.D. et al., 2006. The role of pleasure in web site success. Information &

Management, 43, 434-446. Yani-de-Soriano, M. & Foxall, R., 2006. The emotional power of place: The fall and rise

of dominance in retail research . Journal of Retailing and Consumer Services, 13(6), 403-416.

Bandura, A., 2001. Social cognitive theory of mass communication. Media Psychology, 3, 265-298.

Heijden, V., 2004. User acceptance of hedonic information systems. MIS Quarterly,

28(4), 695-704. Leung, L. & Wei, 2000. More than just talk on the move: Uses and gratifications of the

cellular phone. J&MC Quarterly, 77, 308-320. Lucas, H. & Spitler, V., 1999. Technology use and performance: A field study of broker

workstations. Decision Sciences, 30, 291-311. Mathieson, K., 1991. Predicting user intentions: Comparing the technology acceptance

model with the theory of planned behavior. Information Systems Research, 2(3), 173-191.

Riemenschneider, C.K., Hardgrave, B.C. & Davis, F., 2002. Explaining software

developer acceptance of methodologies: A comparison of five theoretical models. IEEE Transactions on Software Engineering, 28, 1135–1145.

Taylor, S. & Todd, P., 1995. Understanding information technology usage: A test of

competing models. Information Systems Research, 6, 144-176. Tjan, A., 2001. Finally, a Way to Put Your Internet Portfolio in Order. Harvard Business

Review, February, 76-85. Venkatesh, V. & Morris, M.G., 2000. Why Don’t Men Ever Stop to Ask For Directions?

Gender, Social Influence, and Their Role in Technology Acceptance and Usage Behavior. MIS Quarterly, 24, 115-139.

Wei, R., 2008. Motivations for using the mobile phone for mass communications and

entertainment. Telematics and Informatics, 25, 36-46.

150

Page 151: Dissataion - Ranga Perera (CB002688)

Sekaran. U, 2006, Research Methods For Business: A Skill Building Approach, 4th ed,New Delhi:John Wiley & Sons Kothari. C.R., 2002. Research Methodology: methods & Techniques. 2nd ed. New Delhi: New Age International Saunders, M., Lewis, P. & Thornhill, A., 2007a,"Research Onion", [Online], Available from:http://wps.pearsoned.co.uk/ema_uk_he_saunders_resmethbus_4/65/16695/4274089.cw/content/index.html, [Accessed 4th July 2009] Saunders, M., Lewis, P. & Thornhill, A., 2007b, Research Methods for Business Students, 4th ed, England:Pearson Education Limited.

151