Conceptualizing and measuring human anxiety on the Internet

(1)

Santosh Kumar Kalwar

CONCEPTUALIZING AND MEASURING HUMAN ANXIETY ON THE INTERNET

Thesis for the degree of Doctor of Science (Technology) to be presented with due permission for public examination and criticism in the Auditorium 1382 at Lappeenranta University of Technology, Lappeenranta, Finland on the 19th of June, 2014 at noon.

Acta Universitatis

Lappeenrantaensis 575

(2)

Supervisors Jari Porras

Professor, Department of Computer Science Lappeenranta University of Technology Finland

Dr. Kari Heikkinen

Associate Professor, Department of Computer Science Lappeenranta University of Technology

Finland

Reviewers Mauri Kaipainen

Professor of Media Technology S¨odert¨orn University

Alfred Nobels all´e 7 S-141 89 Huddinge Sweden

Katherine Blashki

Professor of New Media Technologies Noro↵University College

Postboks 2544, N-4678, Kristiansand S Norway

Opponent Dr. Praminda Caleb-Solly

Senior Lecturer, Department of Computer Science and Creative Technologies

University of the West of England Bristol BS16 1QY

United Kingdom (UK)

ISBN 978-952-265-584-4 ISBN 978-952-265-585-1 (PDF)

ISSN-L 1456-4491 ISSN 1456-4491

Lappeenranta University of Technology Yliopistopaino 2014

(3)

Dedicated to my beloved father,Devendra Kumar Kalwar, who unexpectedly died on 15th Oct, 2012

“I think of you often and make no outward show, But what it means to lose you, no one will ever know.

You wished no one farewell, not even said good-bye, You were gone before I knew it, and only God knows why.

You are not forgotten nor will you ever be, As long as life and memories last, I will remember thee.” Unknown

Also dedicated to a newcomer,Krista Kalwar, who interestingly appeared on 6th Jan, 2013

(4)

(5)

Preface

“Remembering that I’ll be dead soon is the most important tool I’ve ever encountered to help me make the big choices in life. Because almost everything – all external expectations, all pride, all fear of embarrassment or failure – these things just fall away in the face of death, leaving only what is truly important. Remembering that you are going to die is the best way I know to avoid the trap of thinking you have something to lose. You are already naked. There is no reason not to follow your heart.” Steve Jobs ¹, Stanford commencement address (2005).

Struggle

Indeed, time waits for no one and everybody has to die at a point in their life but wait a minute; if you are a graduate student then you are almost dead. By dead, I mean, we are so overloaded with piles of work and we have to sacrifice lots of other things, and you don’t want to know what I have sacrificed to come to this final thesis writing stage. Sometimes, you might wonder if there is any worth in this Darwinian struggle. You can leave me wondering there at Steve’s great quote.

Nokia World

Now let’s go through a quick recap. I remember the day when I first landed in Lappeenranta from the unknown village of Chitwan in Nepal, and today completes the seven years of my stay here. My Bachelor of Engineering studies at Visveswariah Technological University in Bangalore helped me to race through my Master’s degree in Information Technology at Lappeenranta University of Technology (LUT) in merely a year or so, and I graduated, just after Christmas in 2008. I was lucky enough to get an opportunity to do research in the Nokia Research Center (NRC) at Helsinki. I worked as a trainee at the Office of the Chief Technology Officer (OCTO) at Nokia Oyj, Helsinki, Finland. Our team worked on ideas to develop messaging-based services designed for millions of users. The principal research question was “how service concepts and technologies

1http://www.humanity.org/voices/commencements/steve-jobs-stanford-university-speech- 2005

(6)

can scale up without a prohibitive increase in investment?” I am grateful to Anssi Karhinen, ,Otso Alanko, Oskari Koskimies, Mikko Tarkiainen., Tanja Vuoksimaa, Turkka Aijala. These colleagues were not only smart and intelligent but also, open and collaborative. Personally, I am excited about the NRC led by Chief Technology Officer (CTO) Dr. Henry Tirri, and the outstanding research these great researchers are doing. As I’ve worked as a trainee in NRC, I have seen at first hand the process of turning great ideas into products landing in the consumer’s hand. Research centers are about the process not about the final product. Unlike Nepalese ways of doing things, i.e. “talk more, do less,” at NRC I learned a completely new Finnish way of doing things; “talk less, and do more.”

How it all began

After a while, my trainee-ship ended and the company was going through a transformation phase. Despite hard work and above threshold performance, our project was not promoted to a higher level. One good lesson I learned was that there are issues with the technical layer and business layer in many companies.

Technical people usually don’t understand what business people are doing and vice-versa. My manager at Nokia, Mikko Tarkiainen recommended this as a good time to study, if I was committed enough to study. I decided to send e-mails to various professors across Finland. I encountered Professor Pertti Saariluoma from Jyv¨askyl¨a University, who was helpful and keen on giving me a doctoral studies position. But during those recessionary times (in 2008), no funding was available for basic research and I found no motivation to work without funding.

Back to my guru

I decided to contact my Master thesis supervisor and guru, Prof. Jari Porras. He was positive and said he can o↵er “some money to do some studies”. I applied for many jobs while in Helsinki but could get none. After several setbacks and many “unfortunately letters”, I decided to leave the Finnish capital and return to the small town of Lappeenranta. This time, the year was 2009, life was not so bright and the money I received was not enough to put food on my table and pay the room rent. But, I started a completely new journey of doing basic research. The topic I choose was demanding and complicated. It was broader and at higher level. It encapsulated various branches of science ranging from computer science to psychology, from cognitive science to neurology, from algorithmic intelligence to artificial intelligence. I looked deeper into many fields but thought Human-Computer Interaction (HCI) has the answer to the wide range of demanding and challenging questions.

Motivation

How motivating is it when you are the victim of the study you conduct? Not only was I interested in learning more on these subjects, I wanted to explore technical solutions to human su↵ering. Every researcher might pretend he is the next Einstein or Carl Jung but I am not them. Nobody can be like

(7)

anybody. We are di↵erent and let’s keep it that way, because being di↵erent allows us to share, collaborate, learn, create, and innovate. Deep down, I am just a curious child who wants to understand a simple yet complex human phenomenon, human su↵ering. In this case,the su↵ering I am referring to is:

Internet anxiety—pathological feelings, frustration, and anger that a↵ect users online, in our day-to-day interaction with technologies, application, and services.

I believe it is important to understand human psychology, behavior and Internet anxiety to unleash the potential of the human mind and to bridge the gap of cultural di↵erences in IT-related fields, i.e., psychological HCI (Human-Computer Interaction), Educational Psychology, Cognitive Science, and Computer Science.

I believe that we can march forward by understanding contemporary human beings, and/or users, beyond usability in computer science. Our weaknesses in understanding the power of humankind and the mind have resulted in great disasters. The same could be evident for/in the online, connected and networked world. Furthermore, it is important to understand cultural factors which enable users� interaction and engagement in our online over sharing, hyper-connected world. I knew that it would not be an easy task but I had no other choice. Thus, during summer of 2009, or to be more precise, on 8th June 2009, I was officially accepted to pursue my doctoral studies at LUT.

Publish or perish

Given the broad nature of my Master thesis topic, I wanted to publish, and I thought of publishing in Nature or Science but failed. I was happy to fail early because I also learned the value and quality of the work that those reputed journals publish. I learned the value and potential of publishing in the academic world. This thesis is one single bundle of my four years of work, accumulated from a number of publications. While working as a junior researcher (Nuorempi tutkija) at LUT, I was lucky to be accepted as a participant at various conferences, e.g. HCII2009, IHCI2010, e-society2010, IHCI2011, HCII2011, ISCI2011 and HCII2013, with major academic publishers, e.g., Elsevier, IEEE, ACM , Springer, IADIS , and Mary Albert Inc., which are rated well in academia. The premature version of the papers that I submitted to CHI resulted in motivating feedback from the anonymous reviewers:

“This type of research is much needed in HCI and would allow researchers to better study this important construct.”

Similarly, another reviewer said:

“I feel that bringing to the attention of the HCI community concepts such as Internet anxiety is a good thing as it facilitates the fusion between Cyberpsychology (and its mistakes in conception of wide reaching umbrella anxiety concepts to do with technology) and user experience research, which is seriously taking o↵at present.

HCI has a lot to o↵er in the reduction of anxiety both during and about interaction with technology.”

(8)

These anonymous reviewers and many others have been helpful in improving the quality of my work and I thank them from the bottom of my heart. Without their support, this thesis would not have seen the daylight. I was also invited to serve as a reviewer in premier conferences, e.g. CHI’10, ’11,’12, ’13, and INTERACT’11,’13, and in journals such asIEEE Potentials,Cyberpsychology, Behavior and Social Networking, andComputers in Human Behavior (CHB).

Sometimes, I feel cheerful when some of these articles that I read and reviewed get cited (e.g. more than 100 times). By reviewing papers, I learned more from my peers and great researchers alike.

Traveling is fun

I had an opportunity to travel to di↵erent places and mingle with many like- minded people, who I see are working imaginatively on interesting projects to make the world better place. While doing my PhD studies, I visited enchanting venues such as Kathmandu, Kuala Lumpur, Orlando, Porto, Barcelona, Paris and Freiburg. I had an opportunity to discuss my doctoral topic in Germany, where Prof. Katherine Blaski gave excellent advice on conceptualizing and measuring Internet anxieties and conducting the study with larger sample size.

Curious child

As it turns out, a scientific journey is not actually the ripple e↵ect and its impact but an understanding of an individual ripple. Sorry for being more philosophical but it is always the journey which is more important, more than a destination, a process more important than a product, and a passion for new learning. In other words, learning is like throwing a stone into the lake. The ripple created gets bigger and bigger. In the midst of this event, a lonely curious observer stands on the seashore and imagines, “What if. . . ”, and “how”. . . Indeed, time is of essence and someday like Steve, I too will fade away into oblivion. Nevertheless, I do reassure every morning that I have nothing else to lose as I lay naked in pursuit of my life. After four years of studying (like meditation) and meeting great scientific minds, I think time can wait only for learning and my appetite for knowledge will always be unfulfilled.

Legalities

The opinions expressed in this thesis are those of the author and should not be interpreted as representing ComLab, LUT, ECSE, Nokia, or the Finnish Government or any of its agencies.

(9)

Acknowledgments

Guidance

I reaffirm that Newton was right when he said, “If I have seen further it is only by standing on the shoulders of giants.” Moreover, without the help of these giants, I think this thesis would never have reached a satisfactory stage. Therefore, please kindly allow me a few paragraphs to thank these great intellectual minds. First, I would like to thank my adviser, Dr. Kari Heikkinen, for his consistent support, extensive discussions on the thesis, and thoughtful critique on my interdisciplinary research interest in Human-Computer Interaction (HCI). I would also like to thank my principal supervisor, Professor Jari Porras, firstly, for accepting me as a doctoral student in his department; secondly, for his optimism/trust (in my instincts), and thirdly, for providing helpful comments on countless issues.

I would like to thank my thesis reviewers and opponent; Professors Katherine Blashki and Professor Mauri Kaipainen for accepting to review this work and for providing insightful comments leading to notable improvements.

Peer and colleagues support

Heartfelt appreciation goes to Prof. Beverly Woolf for giving permission to reprint a graphic from her paper. I would like to express appreciation to Dr.

Richard Joiner for the IAS instrument that he created during 2005 and that I have used and modified in this thesis. My deep appreciation goes out to Dr.

Manolya Kavakli, Dr. Elvis Mazzoni, Prof. St´ephane Bouchard, Dr. Bruce M.

Gale and John M. Grohol for answering my e-mail query and helping me out with the creation of a self-assessment tool. Tokens of gratitude goes to Päivi Porras for providing valuable feedback especially in statistical methods and analysis. I would like to thank Er. Shiv Mahato, Head of Computer Engineering Department at Khowpa Engineering College in Kathmandu for insightful research-related discussion. I am grateful to Prof. Jen-Her Wu for positive insights on research collaboration issues. I appreciate the helpful comments given by Prof. Heikki Kälviäinen, Prof. Pasi Fränti, and Prof. Jouni Kämäräinen at ECSE Progress Support Meetings (PSM). The work included in this thesis has been carried out in the Communication Software Laboratory (ComLab) at the Department of Computer Science, Lappeenranta University of Technology (LUT) during the years 2009-2013. Therefore, I am indebted to all co-workers of ComLab.

(10)

My Good English Language Friend/Teacher Support

I was not born with English in my pocket. I would like to thank, Peter Jones from our university Language Centre. Peter is a great guy in helping students and friends alike with proper English usage in academic writing. He has helped me immensely in polishing my English, especially in some journal and conference papers. It is always good to turn to Peter for any general advice or to get inspirational advice on any subject matter. He has stayed in this university for more than twenty years and has vast experience in teaching English. He is not only a native Englishman acting as a teacher but also a great friend. If you track down any English mistakes in this thesis, you know who to blame.

Scholarship and grant help

My freedom to investigate interdisciplinary research endeavors was partially supported by an East Finland Graduate School (ECSE) fellowship 2010-2012, for which I am ever grateful, and an incentive grant from Nokia Scholarship, 2011, a LUT Foundation grant (Lauri ja Lahja Hotisen rahasto, 2011), Tekniikan edistämissäätiö (Finnish Foundation for Technology Promotion, 2012) grant, and Väitöskirjan viimeistelyapurahat (LUT Tukisäätiö Dissertation, 2013) grant.

These incentive and dissertation grants have inspired, helped and motivated me to freely investigate and write this dissertation/thesis.

Finally, my family and friends

I must also thank my family and friends, who showed me this “wonderful but, su↵ering world”. It has been said that we choose our parents. One of my greatest gifts was in choosing the two of you, thanks to my mom (¨aiti): Gita Kumari Kalwar and dad (is¨a) to whom this entire thesis is dedicated. My dad,Devendra Kumar Kalwar, recently passed-away, and I specially dedicate this thesis for his eternal love and blessings. Thank you so much for bringing me up in this bright blue planet (or “pale blue dot” as Carl Sagan would call it). Lastly, heartfelt thanks to my wife, Laxmi Kalwar: my biggest fan and true admirer of my ingenuity. Many friends have brightened my life in Lappeenranta and beyond, and I am deeply grateful for their entertainment and poetic melodrama. I would rather ask you to forgive me for not mentioning your specific names in this thesis but if you read this, do remember that I cannot sail a lonesome ship without your continuous support, love and enthusiasm.

Lappeenranta, June 2014 Santosh Kalwar

(11)

List of Figures

1.1 Psychological and HCI approaches (Roto, 2006) . . . 3

1.2 IA phenomenon between HCI, Psychology, HII and the Internet. 6 1.3 Overview of Internet anxiety research . . . 8

1.4 Structure of the thesis . . . 13

2.1 Overview of the studies . . . 16

2.2 Taxonomy of the research approach, mapping with chapters (J¨arvi- nen, 2008) . . . 20

2.3 Research process of the thesis . . . 21

3.1 Overview of human anxiety, computer anxiety, and Internet anxiety 29 3.2 Anxiety analyzed as a word . . . 31

3.3 Illustration of state of anxiety and anxiety disorder . . . 32

3.4 A six-factor model of computer anxiety. Reprinted from (Beckers, 2001), with permission from Elsevier . . . 36

3.5 Illustration of a user having Internet anxiety . . . 38

3.6 Word cloud of all definitions vs. most referenced ones (anxiety, computer anxiety, and Internet anxiety) . . . 44

3.7 Developing Internet anxiety scales . . . 51

3.8 Self-assessment scale and inter-item relationships . . . 54

3.9 MIAS model and inter-item relationships . . . 56

3.10 Conceptual framework of measuring Internet anxiety . . . 57

3.11 Internet anxiety types with example scenarios. . . 58

3.12 User searching for meaning of terms on the ACM digital library. 60 3.13 A user’s Usage Anxiety self-monitor . . . 64

3.14 Example dialog box for pop-up exceptions mitigating Experience Anxiety. . . 65

3.15 MyIAControl App is limiting access to an anxiety-causing website and blocking a specific site. . . 67

3.16 View to user characteristics . . . 70

3.17 Snapshot of a user performing task 1 . . . 71

3.18 A snapshot of user performing task 2 . . . 72

3.19 Modified Yerkes-dodson task performance law (Yerkes, 1908) . . 74 3.20 Anxiety probable associations neurologically (http://www.brainscanr.com/) 75

(12)

3.21 Overview of MyAnxiety tool. . . 78

3.22 MyAnxiety result page . . . 79

3.23 Screen-shots of Intelligentsia tool implemented as a proof of concept 81 3.25 Screenshots of MIAS tool implemented as a proof of concept (Kalwar et al., 2013b) . . . 82

3.24 Screenshots of a self-assessment and MIAS tool implemented as a proof-of-concept . . . 83

3.26 Example use-case ofMyIAControl tool . . . 85

3.27 Screenshot of MyIAControl tool implemented as a proof-of-concept (home view) . . . 86

3.28 Screen-shot of MyIAControl tool implemented as a proof-of- concept (selection view) . . . 87

3.29 Screen-shot of MyIAControl tool implemented as a proof-of- concept (control view) . . . 87

3.30 An anxious user under talk therapy and counseling . . . 91

3.31 PTSD and regions of brain a↵ected by anxiety symptoms . . . . 91

4.1 Illustration of studies, sub-questions (SQ1-8) and number of samples per studies.. . . 96

4.2 Diagram illustrating the basic process of the study (Kalwar, 2012)112 4.3 Statistical analysis of 35 question/items . . . 114

4.4 Emotional pedagogical agents showing range of emotion on left (courtesy of Beverly Woolf; used with permission), process of body gesture extraction utilized on right . . . 117

4.5 Comparison between initial (n=12) vs. later samples (n=140) for the importance of IA types . . . 122

4.6 ITA (d1) histogram . . . 137

4.7 ITDA (d2) histogram. . . 138

4.8 GIFA (d3) histogram . . . 139

4.9 UA (d4) definition . . . 140

4.10 EA (d5) definition . . . 141

4.11 EAA (d6) definition . . . 142

4.12 NSA (d7) definition . . . 143

4.13 Mapping component/factors into self-assessment scale . . . 156

4.14 Mapping component/factors into MIAS (Kalwar et al., 2013b) . 159 4.15 General Internet use . . . 164

4.16 Gender di↵erences and Internet identification . . . 165

4.17 Summary of studies . . . 169

5.1 Common Internet anxiety symptoms discovered . . . 174

5.2 Phases of lesson learned through feedback loop model of feature implementation . . . 178

(13)

List of Tables

2.1 Relationships between the research questions, methods and the

data gathered. . . 25

3.1 Di↵erence between everyday anxiety and anxiety disorder (Source: ADAA) . . . 33

3.2 Examples of Internet anxiety definitions . . . 38

3.3 Summary of studies related to Internet anxiety . . . 41

3.4 Selected definitions of anxiety, computer anxiety, and Internet anxiety . . . 43

3.5 Pre-existing scales used/not used in this thesis . . . 46

3.6 Internet anxiety scale (IAS) . . . 46

3.7 Internet identification scale (IIS) . . . 47

3.8 (Joiner et al., 2007) hypothesis and results. . . 48

3.9 Internet addiction test (IAT) . . . 49

3.10 Comparison of existing Internet anxiety scales . . . 50

3.11 Brief description of user named Johnny and example symptoms . 58 3.12 Comparison between novice and expert task of Johnny . . . 72

4.1 Overview of studies. . . 97

4.2 SQ1 study overview . . . 98

4.3 Study 1 instruments and measures . . . 99

4.4 IADQ score sheet . . . 101

4.5 Comparison result for di↵erent tests . . . 103

4.7 Comparison result between ethnicities and Internet anxiety types 105 4.8 SQ3 study overview . . . 107

4.9 Interview questions . . . 108

4.10 Age groups comparison with fifteen participants. . . 110

4.12 Expert’s feedback and analysis . . . 113

4.13 a test of reliability (Alpha Test - Cronbach’s) . . . 115

4.16 Cultural ethnicity vs. behavioral issues. . . 118

4.15 Age range vs. Observed Symptoms . . . 119

(14)

4.19 Question and measure scale employed (Kalwar et al., 2012c) . . 124

4.20 Descriptive statistics of question and measure scale employed . . 125

4.22 Stages and tasks . . . 128

4.23 Inter-item correlation matrix and descriptive Statistics for Internet identification scale . . . 130

4.24 Total variance explained . . . 131

4.25 Internet identification factor matrix . . . 132

4.26 Inter-item correlation matrix and descriptive statistics for Internet anxiety scale . . . 133

4.27 Inter-item correlation matrix and descriptive statistics for modified Internet anxiety scale (MIAS) . . . 134

4.28 Inter-item correlation matrix and descriptive statistics for self- assessment scale . . . 135

4.29 Inter-item correlation matrix and descriptive statistics for Internet anxiety types . . . 136

4.30 Statistical analysis for Internet anxiety definitions . . . 136

4.31 Internet Terminology Anxiety definition . . . 137

4.32 Internet Time-Delay Anxiety definition. . . 138

4.33 General Internet Failure Anxiety definition . . . 139

4.34 Usage Anxiety definition . . . 140

4.35 Experience Anxiety definition . . . 141

4.36 Environment and Attraction Anxiety definition . . . 142

4.37 Net Search Anxiety definition . . . 143

4.38 Item-total statistics. . . 144

4.39 Inter-item correlation matrix and descriptive statistics for importance of cultural attributes . . . 147

4.40 Total variance explained . . . 153

4.41 Rotated Component Matrix for SAS . . . 154

4.42 Total variance explained for MIAS . . . 158

4.43 Rotated Component Matrix for MIAS . . . 160

4.44 Independent sample t-Test; gender-IA types . . . 161

4.45 Nonparametric test; age range-IA types . . . 162

4.46 Nonparametric test; Internet daily use-IA types . . . 163

4.47 Comparison with (Joiner et al., 2005) . . . 166

4.48 Gender di↵erences in the participants use of the Internet. . . 167

5.1 Common task examples with behavioral symptoms and Internet anxiety types . . . 173

5.2 Comparison of studies with (Shamo, 2001). . . 180

(15)

List of Abbreviations and Terms

AAAS American Association for the Advancement of Science ACM Association for Computing Machinery

ADAA Anxiety Disorders Association of America Anxiety In medicine, it refers to “disease”.

In psychology, it refers to “fear”, and “apprehension”.

In this thesis, it refers to anxiety in general.

AS Anxiety Syndrome

BCI Brain-Computer Interaction

BRUMS Brunel University Mood Scale

CBSN Cyberpsychology, Behavior and Social Networking

CBT Cognitive Behavioral Therapy

CHB Computers in Human Behavior

CTO Chief Technology Officer

DSM Diagnostic and Statistical Manual of Mental Disorders

EA Experience Anxiety

EEA Environment and Attraction Anxiety fMRI functional Magnetic Resonance Imaging GIFA General Internet Failure Anxiety gSAD Generalized Social Anxiety Disorder

HCI Human-Computer Interaction

HII Human-Internet Interaction

Human or user In the scope of this dissertation, human and user are used synonymously. Both refer to people in general (e.g. university students, researcher, IT, non-IT, etc) who are under study and use the Internet.

IA Internet Anxiety

IAD Internet addiction disorder

IADIS International Association for Development of the Informa- tion Society

IAS Internet Anxiety Scale

IAT Internet Addiction Test

(16)

ICT Information and Communications Technology IEEE Institute of Electrical and Electronics Engineers IIS Internet identification scale

IM Instant Messaging

Internet The Internet or the web is used in the common sense through mobile devices and computers.

ITDA Internet Time-Delay Anxiety

LUT Lappeenranta University of Technology MAACL-R Multiple a↵ect adjective check list MIAS Modified Internet Anxiety Scale NIMH National Institute of Mental Health

NRC Nokia Research Center

NSA Net Search Anxiety

OCTO Office of Chief Technology Officer

PET Positron Emission Tomography

PHQ Patient Health Questionnaires

PIU Problematic Internet Use

POMS Profile of Mood State

PTSD Post traumatic stress disorder

PWU Personal web usage

SAD Social Anxiety Disorder

SAS Self-Assessment Scale

SQ1 sub-question 1

SQ2 sub-question 2

SQ3 sub-question 3

SQ4 sub-question 4

SQ5 sub-question 5

SQ6 sub-question 6

SQ7 sub-question 7

SQ8 sub-question 8

Study An approach to establish new or confirm existing knowledge.

Symptoms An observable verbal, non-verbal, physical and/or mental clue or sign of Internet users.

Types A category of Internet anxiety types or classification.

UA Usage Anxiety

UEF University of Eastern Finland

UX User experience

YIAS Yong’s Internet Addiction Scale

(17)

Abstract

Santosh Kumar Kalwar

Conceptualizing and measuring human anxiety on the Internet Lappeenranta, 2014

214 p.

Acta Universitatis Lappeenrantaensis 575 Diss. Lappeenranta University of Technology

ISBN 978-952-265-584-4, ISBN 978-952-265-585-1 (PDF) ISSN-L 1456-4491 ISSN 1456-4491

Since the 1990’s, the Internet has played a central role in our daily lives. The Internet is an integral part of our personal, business, family, research, entertainment, academic and social life. However, there are social implications in using the Internet that are dependent on categories such as gender, age, ethnicity and cultural attributes. This social aspect can play a detrimental role in the expression of human anxiety on the Internet. An anxiety is a complex phenomenon that requires further elaboration. Thus, the aim of this thesis is to investigate human anxiety, or specifically, whetherInternet anxiety can be conceptualized and measured. This thesis utilizes literature, qualitative and quantitative research methodologies, and a triangulation validation approach to conceptualize and measure the Internet anxiety phenomenon.

In particular, the aim is to explore anxiety levels of Internet participants to develop and validate an Internet anxiety scale based on earlier research on Internet anxiety. The results of the dissertation present a two phase study. In Phase I, a smaller set of studies were conducted with a limited sample size. In Phase II, the research topic was investigated using 385 participants. Based on a number of studies or experiments, the state-of-the-art discovered in this thesis is creation, design, and validation of two scales, the Self-Assessment Scale (SAS) and a Modified Internet Anxiety Scale (MIAS) for measuring users’ anxieties on the Internet.

The result of this dissertation is a conceptualization and measurement of various types of Internet anxiety and measurement of a↵ective feelings of users on the Internet. As a proof-of-concept of measuring Internet anxiety, this thesis describes the author’s implementation of three sets of tools: MyAnxiety,introducing Internet anxieties types;Intelligentia,for collecting Internet anxieties types;

andMyIAControl tool, implemented as a browser plug-in, for measuring a↵ective feelings of users on the Internet. Conclusions drawn from the results show that these empirically validated scales and tools might be useful for researchers and practitioners in understanding and measuring the Internet anxiety phenomenon further.

Keywords: Concepts, Measurement, User Studies, Usability Research, Behavior, Internet anxiety, Front-End, Tools

UDC 004.05:004.738.5:159.94:159.938

(18)

(19)

4.5.2.3 Instruments . . . 116 4.5.3 Procedure . . . 117 4.5.4 Results . . . 117 4.5.5 Issues of validity and reliability . . . 118 4.5.6 Implication for the thesis . . . 118 4.6 Study 6 (SQ6) . . . 120 4.6.1 Objective . . . 120 4.6.2 Materials and method . . . 120 4.6.2.1 Participants . . . 120 4.6.2.2 Data gathered . . . 121 4.6.2.3 Instrument . . . 121 4.6.3 Procedure . . . 121 4.6.4 Results . . . 121 4.6.5 Issues of validity and reliability . . . 121 4.6.6 Implication for the thesis . . . 122 4.7 Study 7 (SQ7) . . . 122 4.7.1 Objective . . . 122 4.7.2 Materials and method . . . 122 4.7.2.1 Participants . . . 122 4.7.2.2 Data gathered . . . 123 4.7.2.3 Instrument . . . 123 4.7.3 Procedure . . . 123 4.7.4 Results . . . 123 4.7.5 Issues of validity and reliability . . . 125 4.7.6 Implication for the thesis . . . 125 4.8 Study 8 (SQ8) . . . 126 4.8.1 Objective . . . 126 4.8.2 Materials and method . . . 126 4.8.2.1 Participants . . . 126 4.8.2.2 Data gathered . . . 127 4.8.2.3 Instruments . . . 127 4.8.3 Procedure . . . 127 4.8.4 Results . . . 128 4.8.4.1 Stage 1: Internet Identification Scale (IIS) . . . 129 4.8.4.2 Stage 2: Internet Anxiety Scale (IAS) . . . 132 4.8.4.3 Stage 3: Modified Internet Anxiety Scale (MIAS)132 4.8.4.4 Stage 4: Self-Assessment Scale (SAS) . . . 133 4.8.4.5 Stage 5: Internet Anxiety Types . . . 133 4.8.4.6 Stage 6: Background demographic variables . . 145 4.8.4.7 Stage 7: Importance of cultural attributes . . . 146 4.8.4.8 Stage 8: Quantifying cultural attributes . . . . 147 4.8.5 Validating scales . . . 152 4.8.5.1 Self-assessment scale models validation . . . 152 4.8.5.2 Modified Internet anxiety scale model validation 155 4.8.6 Hypothesis formulation . . . 157 4.8.6.1 Influence of gender. . . 160

(23)

4.8.6.2 Influence of age . . . 162 4.8.6.3 Influence of Internet use. . . 163 4.8.7 Replication of Joiner et al. (2005) . . . 163 4.8.7.1 General Internet use. . . 164 4.8.7.2 Comparison with Joiner et al. (2005) . . . 164 4.8.8 Discussion on cultural attributes . . . 165 4.8.9 Discussion on studies. . . 168 4.8.10 Issues of validity and reliability . . . 168 4.8.11 Implication for the thesis . . . 169

5 Discussion and lessons learned 171

5.1 E↵ect of types of Internet anxieties . . . 171 5.2 E↵ect of cultural backgrounds . . . 175 5.3 E↵ect of measuring Internet anxiety . . . 176 5.4 E↵ect of gender . . . 177 5.5 Lessons learned . . . 177

6 Conclusion and future work 182

6.1 Conclusion . . . 182 6.2 Limitations of the studies . . . 184 6.3 Future work . . . 185

Bibliography 186

Appendix 1: Basic report of advanced statistics 206 Appendix 2: Survey of Internet anxiety/online usage 214

Colophon 214

(24)

(25)

CHAPTER 1 Introduction

“I keep six honest serving-men (They taught me all I knew); Their names are What and Why and When And How and Where and Who.” Rudyard Kipling

1.1 Background

A

nxiety is a mental health construct and brain response that warns of danger and prepares for upcoming challenges. On the one hand, it can act as motivational construct to take actions (e.g., meeting deadlines, thrill, stimulation, and pleasure of games) and on the other hand, it can act as an escapism mechanism (e.g., irrational and obsession with unexpected future outcomes). Many researchers (Eysenck and Payne,2006; Eysenck et al.,2007;Barlow,2000) have argued that anxiety is not merely fear, frustration, depression and emotion. Instead, anxiety is more than fear - threatening situations or moments that might suggest the presence of danger (Barlow,2004). Sometimes anxiety comes before depression and gets complicated and confused. For some people, anxiety might be typical and for others it might be a special case leading to therapy, counselling, and medical assistance. In general, questions such as “what’s going to happen to me?”¹ whispers in the anxious’s mind. Similar to the general anxiety phenomenon, there is a digital phenomenon on the Internet (Kraut et al.,2006;Amichai-Hamburger, 2005b;Joiner et al.,2005,2012) entitled, “Internet anxiety” - a feeling of worries, nervousness, or unease while using or considering the use of the Internet or its application or services. It appears that Internet anxiety might be “evoked by the use of web-enabled technologies” (Thatcher et al.,2007). A general question that the digitally downtrodden person might ask is “what’s going to happen (to me) on the Internet?” It is not completely clear whether technology acts as enabler for a resulting participant’s anxiety. Some studies (Turkle,2011;Rosen, 2012, 1995;Ellis and Blashki,2004) shed light on obsession with technology, especially among young children and teens.²

1http://nymag.com/news/features/xanax-2012-3/

2http://www.adaa.org/living-with-anxiety/children/anxiety-and-depression

(26)

Chapter 1 1.1. Background There is a considerable amount of work hypothesizing that Internet anxiety is part of computer anxiety (Thatcher et al.,2007;Chou,2003), and topics such as instructional techniques to reduce computer/Internet anxiety in classroom settings (Presno,1998), relating with Internet identification and Internet uses (Joiner et al.,2007), impact of pathological levels of Internet use and Internet related anxiety (Brosnan,2012;Thorpe and Brosnan,2007). Academic work on Internet anxiety is, however, still limited (Cowan and Vigentini, 2008;Presno, 1998; Shamo, 2001). Additionally, there is a) no typology or classification of various Internet anxiety or its types, b) no framework to conceptualize Internet anxiety, and c) no e↵ective Internet anxiety measuring scales. As “Internet anxiety” consists of two words, “Internet” and “anxiety” that have roots in multidisciplinary branches of science mainly psychology and Human Computer- Interaction (HCI), it is important to look for the methodological stances of the various interdisciplinary approaches, e.g., psychology and HCI, in dealing with this complex phenomenon.

Roto presented hypothetical approaches of “psychology and HCI ” in her dissertation (Roto,2006) where she outlined the perspectives of how these two fields di↵er in their ways to solve the problem. In the HCI approach, a user reveals much of what is required, much like a requirement engineer. On the other hand, the psychological approach takes user feedback. When we mix these two approaches we obtain both the requirements and the motivational aspects. One should understand the motivational aspects and also the requirement aspects to address users’ real anxiety. The background presented is somewhat analogous to how this thesis approaches the problem and strongly advocates the triangulation research validation technique. The triangulation research validation technique uses literature, and qualitative and quantitative data collection methods to support the findings. The real issue (see Figure 1.1)³ is “user looks frustrated”.

It is solved neither by a purely psychological approach nor by a purely HCI approach. At the outset, HCI obtains various requirements from the user’s mouth (i.e. using a mental model) whereas the psychological approach solves the problem by listening to what the user has to say (i.e talk therapy).

The approach in this thesis should be partly considered in line with Feynman’s scientific method⁴:

Guess - Compute consequence (what this implies) - Compare (studies, experiments, nature)

An educated “guess” or “intuitions” (Saariluoma,2011) is to combine types or classifications of traditional definitions into distinct types of Internet anxieties and explore further the most significant types of Internet anxieties and what they really imply. As Feynman’s articulation, here experiments are about conceptualizing and measuring the Internet anxiety phenomenon using various data collection methods, observations, interviews, surveys and analyses. However, it

3The figure is used with the due permission from Virpi Roto.

4A lecture by physicist Feynman, Richard (1994). The Character of Physical Law. Modern Library. URL: http://research.microsoft.com/apps/tools/tuva/

h 2 i

(27)

Chapter 1 1.2. Research gap

Figure 1.1– Psychological and HCI approaches (Roto,2006)

is also worth mentioning that Feyerabend was critical about physicist Feynman’s approach to science and stated that “they are uncivilized savages, they lack in philosophical depth” (Lakatos et al.,1999). It appears that there is a philosophical debate among scientists about what is considered science in the computing field (Denning,2005) and the scientific method. Additionally, (Medawar,2013), in a rebellious view, states that:

“The fact that scientist do not consciously practice a formal methodology is very poor evidence that no such methodology exists. It could be said - has been said - that there is a distinctive methodology of science which scientist practice unwittingly, like the chap in Moli`ere who found that all his life, unknowingly, he had been speaking prose.”

To understand the Internet anxiety phenomenon several methods or approaches are used to collect data, as this thesis is a descriptive data-based study.

1.2 Research gap

Thirteen years ago a doctoral dissertation was written by (Shamo,2001) entitled,

“University Students and the Internet: Information Seeking Study” where the author studied university students and the Internet related to informational study. His study was based on a scholarly journal paper (Presno,1998), “Tak- ing the bytes out of the Internet anxiety: instructional techniques that reduce computer/Internet anxiety in the classroom” in which the author had outlined and highlighted various types of Internet anxiety (IA). As pointed out in the doctoral dissertation and journal paper, there were limitations and a need for future work (Shamo,2001)(Presno,1998). These limitations are listed below:

1. (Presno, 1998) utilized four types of Internet anxiety and stated that

“further studies in the new area of Internet anxiety” are needed.

2. “Although many of the strategies that are meant to reduce computer anxiety may be applicable to Internet anxiety, there is still a point where

(28)

Chapter 1 1.3. Purpose and scope the two types of anxiety diverge. The one-of-kind virtual world of the Internet is a place without boundaries and a space without a core that will cause users to experience di↵erent kinds of emotions, some of which will be more negative than positive. For this reason, it will be important in the future to address a more specific form of computer anxiety - Internet anxiety.” (Presno,1998), pg. 151

3. (Presno,1998) cautioned on the use of qualitative research methodologies in particular. The study was general in nature using qualitative research methodology with 21 and 17 student participants in an Internet class.

4. No symptoms are mentioned for university students with regards to the Internet anxiety phenomenon (Shamo,2001).

5. The investigation (Shamo,2001) lacked rigorous iterative validation with no new scale proposed or developed.

The limitations pointed out by (Presno,1998) and (Shamo,2001) led to recog- nition of the paucity of research and, thus, this thesis. Although the topic and the research is extremely demanding and challenging, the author of this thesis started to look more deeply at Internet anxiety phenomenon– how people interact with the Internet, and this thesis reports a number of smaller studies (study001-study008). In other words, this thesis is result of curiosity to find out ways in which one couldconceptualize andmeasure the Internet anxiety phenomenon by conducting a number of smaller sets of studies and rigorously validating these studies with scientific research methodologies.

1.3 Purpose and scope

The Internet or the web (terms have been used synonymously)⁵ evolved during the 1990’s and it was taking o↵during 2000’s. The web presently hosts a wide range of application, services, and platforms. Most people in developed and industrialized nations cannot even imagine life without the Internet. The Internet is omnipresent and has touched the lives of billions of humans⁶. Currently, the Internet has grown to not only touch humans in a physical sense but also in an emotional sense. Life without the Internet is unimaginable. There are many purposes for which an individual can use the Internet, e.g., work, fun, social networking, self-therapy, creativity etc., and there are research areas that need to be explored more, e.g., addiction, anxiety, and human su↵ering. The Internet uses that are particularly relevant for the present study are web/Internet browsing, social networking sites, and relationships with other internet users. Therefore, the thesis does not merely describe a conceptualization of the Internet but also enquires about the Internet from a less explored area (anxiety on the Internet),

5In this thesis, the Internet or inside the TCP/IP protocol stack is not examined.

6human or user is synonymously used throughout the thesis.

h 4 i

(29)

Chapter 1 1.3. Purpose and scope and much of the present focus in this area lies on understanding people, theory, and concepts. Basic questions⁷ related to this field are:

• What is the Internet anxiety phenomenon?

• How is students’ Internet anxiety type conceptualized with respect to university participants?

• What types of Internet anxiety can be identified?

• How is the Internet anxiety phenomenon and its types measured?

All these questions are important to understand the nature of Internet anxiety and also to establish ways in which this yet to be fully known construct could be further conceptualized and measured. Conceptualization here is based on G¨ardenfors’ conceptual spaces theory (G¨ardenfors,2000,2004), which defines the term as a collection of one or more dimensions that are independent entities and are correlated in various ways. On the other hand, the measurement is based on quantifiable analysis of various factors. The aims of this thesis were defined as follows:

• This doctoral thesis addresses the conceptualization and measurement of Internet anxiety through a number of studies.

• The conceptualization is based on two phase studies (Phase I and Phase II) that further describe the conceptual framework and empirically validate the newly created scales for the measurement of Internet anxiety.

• The measurement is based on empirical studies of various scales; e.g., IIS (Internet Identification Scale) and IAS (Internet Anxiety Scale) developed

by (Joiner et al., 2007).

A multidisciplinary research approach andsoftware engineering approach is used to tackle the research problem. The combination of these approaches was used because: 1) anxiety is a very complex phenomenon to study (Eysenck and Payne,2006), 2) anxiety stems from various fields (Chou,2003), and 3) “thinking outside of the box”⁸is required in anxiety research. Therefore, the thesis is based on on-going research in this area, and it is especially focused on psychological HCI (Card et al.,1983). Furthermore, the conceptualization part might not be directly visible to the user. However, the measurement part via proof-of-concept tools that are suitable for research purposes (e.g. MyAnxiety,Intelligentsia, and MyIAControl) and newly designed and created scales (e.g. self-assessment scale, modified Internet anxiety scale) is directly visible to the user, thus the approach can be described as user-centric. As pointed out by (Olina and Sullivan,2004), assessment is vital in learning, measuring self-achievements, and evaluating individual programs. It has also been said that self-assessment provides a boost

7These general questions are addressed in the subsequent chapters. However, more specific research questions are devised and described in Chapter 2.

8http://www.fastcompany.com/53187/outside-box-inside-story

(30)

Chapter 1 1.4. Context: HCI and Internet anxiety for learning (Glaser,2001). Thus, a self-assessment tool for measuring Internet anxiety is created and tested. Finally, a conceptual framework is devised to test the identified Internet anxieties types from the literature.

1.4 Context: HCI and Internet anxiety

Human Computer-Interaction (HCI) is a multidisciplinary branch of computer science utilizing a wide variety of methods, approaches and techniques, and sits in the intersection of various related branches of science, e.g., computer science, psychology, and cognitive science.

“Human-computer interaction is a discipline concerned with the design, evaluation and implementation of interactive computing systems for human use and with the study of major phenomena surrounding them.” (Hewett and Verplank,2002)

As quoted, HCI is mainly concerned with humanistic models and experiences that surround systems and products. Knowledge of anxiety on the Internet enables understanding of pitfalls in human-centered design, evaluation and implementation. Therefore, it is important to understand the IA (Internet anxiety) phenomenon, which lies in the midst of HCI, Human-Internet Interaction (HII), Internet and psychology, as shown in Figure 1.2.

Figure 1.2– IA phenomenon between HCI, Psychology, HII and the Internet According to (Saariluoma and Oulasvirta,2010), the main interests of famous scientists such as Alan Turing, Douglas Engelbart, and Allen Newell was intelli-

h 6 i

(31)

Chapter 1 1.4. Context: HCI and Internet anxiety gence. They were also very much interested in psychological concepts. This is significant with early HCI research, which was pillared on psychological concepts (Welford,1968). Since the 1970’s, human cognition and psychological studies for computer use have been a subject of academic study (Sackman,1970). As stated by (Saariluoma and Oulasvirta,2010),Moran(1981) was first to introduce cognition and behavioral studies in HCI and few researchers have elaborated on top of this idea.

According to (Saariluoma and Oulasvirta,2010), “user psychology” (Moran, 1981) is a broader discipline that focuses on how interactive systems and technology “should be designed” for the benefit of e↵ective interaction between users and the machine. The paper states,

“We believe that good research should not be like a black box that reports results from measurements, but rather entail an in-built interest for systematic development of understanding.” (Saariluoma and Oulasvirta,2010)

After reading several foundational papers (Saariluoma,1997,2005;Saariluoma and Oulasvirta,2010) on user psychology and how psychological research (Nor- man,1992) should entail a new conceptualization and measurement for e↵ective understanding of yet unknown constructs, this thesis not only “reports results from measurement” but also tries to generate interest for “systematic development” in-line with the recommendation of (Saariluoma and Oulasvirta,2010).

In this thesis,Human Internet-Interaction, is defined as a new branch that deals with how people interact on the Internet. The conventional wisdom is challenged by replacing “computer” with the “Internet”. The benefit of challenging such a notion is huge. Bargh(2002) broadly defines Human-Internet Interaction as an a↵ect of the Internet on social lives. For example, studies (Rosen,2012;Turkle, 2011;Kraut et al.,2006;Amichai-Hamburger,2003;Amichai-Hamburger and Hayat, 2011) show that people spend a great amount of time on the Internet (as we take the “computers” for granted nowadays), and many researchers and practitioners⁹ have formulated the hypothesis that: people who are distracting themselves with everyday popular technologies might find an adverse a↵ect on their creativity. However, “creativity” is not studied in this particular case. The author of this thesis is interested in the magnitude of the adverse e↵ects such popular technologies may have or have and what are the social implication of using the Internet.

In view of the multidisciplinary nature of the work, a context of the research is presented for conceptualizing and measuring Internet anxieties. The research is mainly conceptual-analytical research with emphasis on building tools to mitigate Internet anxiety symptoms. In the conceptual-analytical element, the study focused on understanding concepts related to HCI , psychology, emotions and feelings. Accordingly, “emotion” and “feeling” are defined such that the term emotion will “designate the collection of responses, many of which are publicly observable” while the term feeling is “reserved for the private, mental experience

9http://worldsciencefestival.com/full programs

(32)

Chapter 1 1.4. Context: HCI and Internet anxiety of an emotion (Damasio,2000),” in particular, an exploration of the present day

“Internet anxiety (IA)” phenomenon. In recent years, the psychological aspect of HCI has been widely studied by Don Norman, who is currently regarded as the pioneer of psychological HCI. He states,

“When simple things need pictures, labels, or instructions, the design has failed. ”(Norman,1992)

From the quoted sentence, it appears as if the design of a system or product should be “simple but not simpler”¹⁰. For example, there is a wide range of sites that “suck”¹¹on the Internet. There is also a wide range of systems and products and not all of these products or systems are well designed as described by Don Norman in his famous book, “The Psychology of Everyday Things” (Norman, 1992). Therefore, to reduce the burden of anxieties of the user(s), one has to consider “conceptualizing” or “measuring” the anxieties arising from various products, applications, services, and designs on the Internet.

A brief overview is provided in Figure 1.3, illustrating how Internet anxiety is studied through classification of the types of IA and through analysis of the various elements which seek to further elaborate on the IA phenomenon by mapping with symptoms (e.g. frustration, anger, addiction). The measures employed are utilized via the conceptual framework and various tools. Overall, the goal is to study suitable techniques in related fields (psychology, cognitive science, culture, behavior, and computer science) by introducing methods and measuring techniques for human anxiety on the Internet.

Figure 1.3– Overview of Internet anxiety research

10http://c2.com/cgi/wiki?EinsteinPrinciple

11http://www.webpagesthatsuck.com/

h 8 i

(33)

Chapter 1 1.5. Terminology

1.5 Terminology

In the course of writing this PhD thesis, some basic terminology has been (mis)used. This terminologies is explained briefly:

• Concept: a general notion or an abstract for formulation and understanding

• Conceptualization: clarification of a concept or various concepts

• Variable: a statistical term used to measure something

• Factor: a statistical term that acts as an independent variable

• Measurement: a numerical representation for the analysis of categories of variables or factors

• Construct: to build or create a model or relationships that meets specific requirements

• Proof-of-concept: concept defined as realization of various types of Internet anxiety

• Proof-of-concept tools: tools designed to test the feasibility of Internet anxiety or its types

1.6 Contributions

In the course of investigating Internet anxiety, this dissertation makes the following contributions.

• Concepts of Internet anxiety (IA) types are introduced and illustrated with example cases.

• A logical conceptual framework for assessing human anxiety on the Internet is developed.

• A new approach for measuring Internet anxiety using a self-assessment scale (SAS) and the Modified Internet Anxiety Scale (MIAS) is introduced and empirically validated.

• Formal development of the concepts of a cultural model and five cultural attributes for understanding human behavior on the Internet is undertaken.

• Proof of concept tools (i.e. MyAnxiety, Intelligentsia, MyIAControl) are designed, implemented and experimented.

(34)

Chapter 1 1.7. List of publications on which the thesis is partly based

1.7 List of publications on which the thesis is partly based

This thesis is written in the form of a monograph which is also partly based on work presented in several scholarly journal/conference papers. Publication II, VI, VII and VIII are scholarly journal/magazine articles. Publication I and V are lightly peer-reviewed (abstract only) conference papers, whereas Publication III, and IV are rigorously double blind peer-reviewed conference papers. A short position paper (Kalwar, 2009) and doctoral consortium paper (Kalwar,2010a) have also been published. A brief overview of these publications is presented in subsequent sections.

In all the publications (P I - VIII), the main author was responsible for planning the study, collecting the data, analyzing the results and writing the papers. The role of co-authors was supervision of the research and writing process.

1.7.1 Publication I

Kalwar, Santosh and Kari Heikkinen. 2009. Study of Human Anxiety on the Internet. InProceedings of the 13th International Conference on Human- Computer Interaction. Part I: New Trends, Julie A. Jacko (Ed.). Springer-Verlag, Berlin, Heidelberg, 69-76.

Overview

This publication was based on existing literature study to understand human behavior on the Internet and types of Internet anxieties. This publication studied four Internet anxiety types conceptualized by (Presno,1998) and as part of the results ascertained three new types of Internet anxiety. In this study, these three new Internet anxiety types are not validated but only conceptualized because of the limited sample size.

1.7.2 Publication II

Kalwar, Santosh, Comparison of Human Anxiety Based on Di↵erent Cultural Backgrounds,Cyberpsychology, Behavior and Social Networking, 2010, vol. 13, nro. 4, p. 443-446, ISSN 2152-2715, 2152-2723

Overview

In order to facilitate understanding of users’ behavior on the Internet, cultural factors were considered, for instance, in the form of di↵erent target groups. A comparison of human anxiety based on two cultural backgrounds was undertaken.

The results showed that participants were very unlikely to change their behavior for browsing contents on the Internet. The results also showed that participants

h 10 i

(35)

Chapter 1 1.7. List of publications on which the thesis is partly based experienced anxiety while performing the given tasks. Based on participant feedback, seven formal definitions of Internet anxiety types were formulated.¹²

1.7.3 Publication III

Kalwar, Santosh & Heikkinen, Kari & Porras, Jari, An evaluation of human anxiety on the Internet, InProceedings of e-Society 2010, 18 - 21 March 2010, Porto, Portugal, IADIS, 2010, p. 272-286, ISBN 978-972-8939-07-6.

Overview

An evaluation of human anxiety on the Internet was carried out in this publication.

The objective was to determine if psychological human-computer interaction analysis is suitable to measure human anxiety on the Internet. A research question and the following hypothesis were presented: RQ1: what are the types of anxieties shown by humans on the Internet? H1: The behavior of humans on the Internet shows seven di↵erent types of anxieties. After identifying seven di↵erent types of anxieties from the Master thesis work, this work evaluated them with ten samples. Furthermore, seven scenarios were sketched and mapped for seven di↵erent types of anxieties and utilized for validating the seven di↵erent types of anxieties.

1.7.4 Publication IV

Kalwar, Santosh & Heikkinen, Kari & Porras, Jari. 2011, Measuring User Reaction to Reduce Internet Anxiety, InProceedings of IEEE Symposium on Computers & Informatics, Kuala Lumpur, Malaysia from 20 - 22 March 2011, p.738-742, IEEE Catalog Number: CFP1163N-CDR, ISBN:978-1-61284-690-3 Overview

The user’s reaction plays a vital role in determining feelings, emotions and behavior. In this work, the objective was to determine users’ reactions, and measure their feelings by introducing the FeelCalc module, a browser plug-in prototype to measure a user’s a↵ective feelings that was designed and tested.

Based on observations and the prototype, a preliminary measurement of users’

feelings was recorded.

1.7.5 Publication V

Kalwar, Santosh & Heikkinen, Kari & Porras, Jari. 2011, Finding a relationship between Internet Anxiety and Human Behavior, In Proceedings of the 14th International Conference on Human-Computer Interaction. Julie. A. Jacko

12This thesis does not study culture in-depth but only utilizes cultural definitions for several types of Internet anxiety. The cultural aspects presented in this thesis should be considered as a “side track” for the reader.

(36)

Chapter 1 1.7. List of publications on which the thesis is partly based (Ed.), Springer-Verlag, in 9-14 July 2011, Hilton Orlando Bonnet Creek, Orlando,

Florida, USA, 2011, p. 359-367, ISBN: 978-3-642-21602-2 Overview

The main objective was to establish the relationship between human behavior and Internet anxiety. Thus, an assessment of human behavior and anxiety patterns among di↵erent age groups was undertaken. The findings showed that a priori understanding and outcome of the relationship between Internet anxiety and human behavior is a determinant of types of Internet anxieties and users’

actions/reactions at the Internet interface.

1.7.6 Publication VI

Kalwar, S.; , “A Self-Assessment Tool for Measuring Internet Anxiety,”Potentials, IEEE , vol.31, no.6, pp.38-44, Dec. 2012

Overview

The main objective was to create a self-assessment tool for measuring Internet anxiety. This was carried out with two sets of users (N=7 and N=35). The reliability and validity of the self-assessment tool is discussed. The major findings was that through the development and validation of a self-assessment tool, Internet anxiety types measurement is feasible.

1.7.7 Publication VII

Kalwar, Santosh & Heikkinen, Kari & Porras, Jari. 2012, Conceptual Framework for Assessing Human Anxiety on the Internet ,Procedia - Social and Behavioral Science, Volume 46, 2012, Pages 4907-4917, ISSN: 1877-0428, Elsevier

Overview

A conceptual framework for assessing human anxiety on the Internet is introduced in this publication. This framework was constructed based on the existing literature on human anxiety and Internet anxiety. “MyAnxiety,” acted as a proof- of-concept tool for the conceptual framework. The elements of the conceptual framework were illustrated with the help of this tool.

1.7.8 Publication VIII

Kalwar, Santosh & Heikkinen, Kari & Porras, Jari. 2012, Intelligentsia: A front-end tool for collecting Internet Anxiety Data,AWERProcedia Information Technology and Computer Science, Volume 2, no. 0, 2012, pp. 475-483, ISSN:

2147-5105

h 12 i

(37)

Chapter 1 1.8. Thesis Overview Overview

A front-end tool“Intelligentsia” for collecting Internet anxiety data is introduced in this article. It is capable of collecting types of Internet anxiety data from user actions. These data were useful in analyzing various anxiety types that users might experience during a given task (s). The findings revealed major behavioral observations, e.g., excitement, enthusiasm, impatient etc., and symptoms, e.g., lack of focus, helplessness and anxiety that were reported and analyzed with seven Internet anxiety types utilizing the proof-of-concept tool.

1.8 Thesis Overview

This thesis provides study of the anxiety of Internet users

,

and a classification, conceptualization, and capability to measure levels of Internet anxiety.¹³ The chapter sequence and the overall structure of the thesis is shown in Figure 1.4.

Figure 1.4– Structure of the thesis

A focused literature review on Internet anxiety is presented in this thesis.¹⁴ Anxiety is a state of helplessness, struggle for information, and a feeling of worry, nervousness, or unease while considering the use of the Internet. The reasons for doing a focused literature review are two-fold: Firstly, there is a vast amount of literature on anxiety research, and this thesis does not broadly study

13This is a thesis statement/claim in line with the rest of the document.

14http://www.stanford.edu/class/cee320/CEE320A/POD.pdf

(38)

Chapter 1 1.8. Thesis Overview

“anxiety” but focuses on “Internet anxiety”. It is important to focus specifically on the Internet anxiety phenomenon when considering the background, relevance, foundational depth, and current trends. Secondly, a focused literature review is usually the norm in the domain of sociological Human-Computer Interaction field (Dunleavy,2003). Following the literature review, the chapter sequence continues with the conceptual framework, discussions, lessons learned and conclusions.

Chapter 1 introduces human anxiety literature that is related to Internet usage. A thesis statement with an overview of the thesis, contribution, and publications are given in this chapter. Chapter 2 deals with research questions and methodologies. The principal research question is divided into sub-questions and these sub-questions are further explained. Chapter 3 outlines literature studies focused on understanding Internet anxiety. Current literature on understanding human anxiety, computer anxiety and Internet anxiety are presented in this chapter. This chapter also introduces a conceptual framework for human anxiety on the Internet. Chapter 4 describes studies carried out and empirical findings to conceptualize and measure human anxiety on the Internet. Chapter 5 discusses the e↵ects of various elements of the conceptual framework and highlights lessons learned. This chapter also summarizes the major findings. The final chapter, Chapter 6 concludes the study by paving the way towards future work.

h 14 i

(39)

CHAPTER 2 Research questions and methodologies

“The methods of science aren’t foolproof, but they are indefinitely perfectible.

Just as important: there is a tradition of criticism that enforces improvement whenever and wherever flaws are discovered.” — Daniel Dennett

T

here have been valuable studies of Internet anxiety using qualitative research methodologies (Presno,1998;Shamo,2001), all of which present evidence of types of Internet anxiety. However, none of these studies provide conceptualization of human behavior on the Internet by building technical enablers (i.e. MIAS, self-assessment tool, and a front-end user interface) for measurement of Internet anxieties. The aim of this section is to formulate a number of research questions that have emerged from this research gap.

2.1 Research questions

The main research question in this thesis is:

How to conceptualize and measure human anxiety on the Internet?

The main research question was divided into sub-questions (SQ1-SQ8) to simplify and clarify the research process (see Figure 2.1). In these sub-questions, the use of hypotheses was considered, however this approach was rejected because there are no preexisting ideas about the nature or the relationship of Internet anxiety, measures and human behavior derived from current data. Hence, these research sub-question do not allow for specific hypotheses.

2.1.1 Sub-question 1

How does Internet anxiety type link with preexisting scales of anxiety and addiction?

Previous studies of (Joiner et al.,2012;Presno,1998;Thatcher et al.,2007) have not: a) conceptualized and measured human behavior on the Internet; b) studied “anxiety” as a web-based phenomenon; c) mitigated Internet anxiety types; d) built a framework for assessing human anxiety and; e) quantified cultural

Conceptualizing and measuring human anxiety on the Internet

Santosh Kumar Kalwar

Acta Universitatis

Lappeenrantaensis 575

Preface

Acknowledgments

List of Figures

List of Tables

List of Abbreviations and Terms

Abstract

Contents

CHAPTER 1

Introduction

1.1 Background

A

1.2 Research gap

1.3 Purpose and scope

1.4 Context: HCI and Internet anxiety

1.5 Terminology

1.6 Contributions

1.7 List of publications on which the thesis is partly based

1.7.1 Publication I

1.7.2 Publication II

1.7.3 Publication III

1.7.4 Publication IV

1.7.5 Publication V

1.7.6 Publication VI

1.7.7 Publication VII

1.7.8 Publication VIII

1.8 Thesis Overview

,

CHAPTER 2

Research questions and methodologies

T

2.1 Research questions

2.1.1 Sub-question 1