DF

Culture and Website

User experience in website redesign based on
cultural differences
Master’s thesis in Interaction Design and Technologies

Zihua Yang

Department of Computer Science and Engineering

C HALMERS U NIVERSITY OF T ECHNOLOGY
U NIVERSITY OF G OTHENBURG
Gothenburg, Sweden 2019
 Master’s thesis 2019

Culture and Website

User experience in website redesign based on cultural differences

Zihua Yang

DF

Department of Computer Science and Engineering

Interaction Design and Technologies
Chalmers University of Technology
University of Gothenburg
Gothenburg, Sweden 2019
Culture and Website
User experience in website redesign based on cultural differences
Zihua Yang

© Zihua Yang, 2019.

Supervisor: Gordana Dodig Crnkovic, Department of Computer Science and Engi-

neering
Supervisor: Peter Bladh, Linn Burman, Kollmorgen Automation AB
Examiner: Staffan Björk, Department of Computer Science and Engineering

Master’s Thesis 2019

Department of Computer Science and Engineering
Interaction Design and Technologies
Chalmers University of Technology and University of Gothenburg
SE-412 96 Gothenburg
Telephone +46 31 772 1000

Cover: Wind visualization constructed in Matlab showing a surface of constant wind

speed along with streamlines of the flow.

Typeset in LATEX, template by David Frisk

Printed by Chalmers Reproservice
Gothenburg, Sweden 2019

iv
Culture and Website
User experience in website redesign based on cultural differences
Zihua Yang
Department of Computer Science and Engineering
Chalmers University of Technology and University of Gothenburg

Abstract
The growth of multinational corporations has lead to the increased importance of
cross-cultural communication across the globe. Therefore, the effect of cultural dif-
ferences on user requirements has become an essential issue. Hofstede’s six dimen-
sions and Hall’s concept of high and low context cultures have been widely discussed
and accepted, in relation to marketing and consumption. However, both theories
were conceived in the 1970s and may not apply to today’s environment. The investi-
gations for this thesis sought to discover and identify how cultural factors influence
user’s preferences for user experience attributes on websites, combining Hofstede’s
and Hall’s theories with Holliday’s grammar of culture and Edwin Hoffman’s Culture
mode theories. Design attributes such as color and layout were determined using
online surveys, usability tests and interviews with middle-class subjects from both
Sweden and China. A redesigned website and a guideline for cultural localization in
both countries are provided. The results confirm and extend prior research, as sig-
nificant differences were found in some of the listed design attributes, but tend to be
similar in some other attributes. Different user experience design is still suggested
for successful communication in different cultural groups. Although this study was
applied to cultural differences between China and Sweden, the methodology and
design procedure can be adapted for other cultures.

Keywords: Cross-cultural design, User experience design, Web design, Usability,

Communication, China, Sweden.

v
Acknowledgements
I would like to thank Gordana Dodig Crnkovic, my supervisor at Chalmers, for pro-
viding strong support, clear guidelines and advice during the project. I am obliged
to Peter Bladh and Linn Burman, my supervisors at Kollmorgen, for selecting such
an interesting topic and making the study possible, also helping me to organize user
tests and giving comments and feedback. Also thanks to all the testers, who spent
time to fill out surveys, perform user tests and provide their opinions honestly. Fi-
nally, I would like to give special thanks to my family and friends, who never stop
encouraging and supporting me, giving me the strength to pursue my dream to
become a UX designer.

Zihua Yang, Gothenburg, June 2019

vii
 Contents

List of Figures xiii

1 Introduction 1
1.1 Industry Background . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Kollmorgen . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 The website . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Research question . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 Deliverable . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.2 Delimitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Stakeholders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.1 Chalmers University of Technology . . . . . . . . . . . . . . . 3
1.3.2 Kollmorgen Automotion AB . . . . . . . . . . . . . . . . . . . 3
1.3.3 Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.4 Researcher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.5 Users . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 Ethical Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4.1 Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4.2 Informed consent . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4.3 Usability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4.4 Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Theory 7
2.1 Culture and website design . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Hofstede’s six cultural dimensions theory . . . . . . . . . . . . . . . . 9
2.2.1 Power Distance (PDI) . . . . . . . . . . . . . . . . . . . . . . 9
2.2.2 Individualism vs. Collectivism (IDV) . . . . . . . . . . . . . . 10
2.2.3 Uncertainty avoidance (UAI) . . . . . . . . . . . . . . . . . . 10
2.2.4 Masculinity vs. Femininity (MAS) . . . . . . . . . . . . . . . 10
2.2.5 Long Term vs. Short Team Orientation (LTO) . . . . . . . . . 10
2.2.6 Indulgence vs. restraint (IND) . . . . . . . . . . . . . . . . . . 10
2.3 Hall and Hall’s low-context and high-context culture model . . . . . . 11
2.3.1 High-context vs. Low-context . . . . . . . . . . . . . . . . . . 11
2.3.2 Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Holliday’s grammar of culture . . . . . . . . . . . . . . . . . . . . . . 11
2.5 Edwin Hoffman’s culture mode . . . . . . . . . . . . . . . . . . . . . 13
2.6 Cultural globalization . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

ix
Contents

2.7 User Experience Elements . . . . . . . . . . . . . . . . . . . . . . . . 14

2.7.1 Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.7.2 Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.7.3 Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.7.4 Aesthetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.7.4.1 Color . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.7.4.2 Font style . . . . . . . . . . . . . . . . . . . . . . . . 16
2.7.4.3 Visual presentation . . . . . . . . . . . . . . . . . . . 16
2.7.5 Multimedia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.7.6 Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3 Methodology 17
3.1 Design Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.1 Diary study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1.2 Questionnaire . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1.3 Prototyping . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1.4 Thinking aloud . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1.5 Usability tests . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1.6 Pilot test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

4 Planning 21
4.1 Methods selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2 Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.3 Time plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

5 Design Process 23
5.1 Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.1.1 An Analysis of McDonald’s websites . . . . . . . . . . . . . . 24
5.1.2 Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.1.2.1 Animation . . . . . . . . . . . . . . . . . . . . . . . . 24
5.1.2.2 Promotion of values . . . . . . . . . . . . . . . . . . 31
5.1.2.3 Individuals separate from or together with the product 31
5.1.2.4 Level of transparency . . . . . . . . . . . . . . . . . 32
5.1.2.5 Linear vs. parallel navigation on the website . . . . . 32
5.1.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
5.2 User research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.2.1 Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.2.1.1 Participants . . . . . . . . . . . . . . . . . . . . . . . 34
5.2.1.2 Preparation of questions . . . . . . . . . . . . . . . . 34
5.2.1.3 Analysis of responses . . . . . . . . . . . . . . . . . . 34
5.2.1.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . 38
5.3 Concept ideation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.3.1 Low-fidelity prototyping . . . . . . . . . . . . . . . . . . . . . 38
5.4 Pilot testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.4.1 Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.4.2 Prepare tasks and questions . . . . . . . . . . . . . . . . . . . 44
5.4.3 Feedback and responses . . . . . . . . . . . . . . . . . . . . . 45

x
 Contents

5.4.4 Result analysis and evaluation . . . . . . . . . . . . . . . . . 46

5.5 Usability testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.5.1 Participants recruitment . . . . . . . . . . . . . . . . . . . . . 51
5.5.2 Test procedure and question preparation . . . . . . . . . . . . 52
5.5.3 Test results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.5.4 Analyzing responses . . . . . . . . . . . . . . . . . . . . . . . 56
5.5.4.1 Color tone . . . . . . . . . . . . . . . . . . . . . . . . 56
5.5.4.2 Text with/without icons . . . . . . . . . . . . . . . . 56
5.5.4.3 Layout design . . . . . . . . . . . . . . . . . . . . . . 56
5.5.4.4 Link opening . . . . . . . . . . . . . . . . . . . . . . 57
5.5.5 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

6 Results 59
6.1 Redesign websites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
6.1.1 Login page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
6.1.2 Home page . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.1.3 Knowledge base page . . . . . . . . . . . . . . . . . . . . . . . 64
6.1.4 Knowledge based detail page . . . . . . . . . . . . . . . . . . . 68
6.1.5 Submit request page . . . . . . . . . . . . . . . . . . . . . . . 70
6.2 Design guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
6.2.1 Design guidelines for Chinese . . . . . . . . . . . . . . . . . . 73
6.2.1.1 Orientation . . . . . . . . . . . . . . . . . . . . . . . 73
6.2.1.2 Main navigation menu . . . . . . . . . . . . . . . . . 74
6.2.1.3 Layout . . . . . . . . . . . . . . . . . . . . . . . . . . 74
6.2.1.4 Color . . . . . . . . . . . . . . . . . . . . . . . . . . 74
6.2.1.5 Links . . . . . . . . . . . . . . . . . . . . . . . . . . 74
6.2.1.6 Text . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.1.7 Other . . . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.2 Design guidelines for Swedish . . . . . . . . . . . . . . . . . . 75
6.2.2.1 Orientation . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.2.2 Main navigation menu . . . . . . . . . . . . . . . . . 75
6.2.2.3 Layout . . . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.2.4 Color . . . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.2.5 Links . . . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.2.6 Text . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

7 Discussion 77
7.1 Result discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
7.2 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
7.2.1 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
7.2.2 Types of Websites . . . . . . . . . . . . . . . . . . . . . . . . . 79
7.2.3 Target user groups . . . . . . . . . . . . . . . . . . . . . . . . 79
7.2.4 UX attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
7.2.5 Sample size . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
7.3 Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

8 Conclusion 81

xi
Contents

Bibliography 83

A Appendix I

B Appendix VII

C Appendix IX

D Appendix XIII

xii
 List of Figures

2.1 IKEA’s website in Sweden and China . . . . . . . . . . . . . . . . . . 8

2.2 Online shopping websites in Sweden and China . . . . . . . . . . . . 9
2.3 Cultural dimension comparison between China and Sweden [10] . . . 11
2.4 Grammar of culture, based on the figure by Hoffman [29] . . . . . . . 12
2.5 Humanity, culture and personality: a conceptual scheme (based on a
figure by Geert Hofstede, 1991) . . . . . . . . . . . . . . . . . . . . . 13

3.1 The UX design process [25] . . . . . . . . . . . . . . . . . . . . . . . 17

5.1 McDonald’s website in Mainland China . . . . . . . . . . . . . . . . . 25

5.2 McDonald’s website in Japan . . . . . . . . . . . . . . . . . . . . . . 26
5.3 McDonald’s website in Hong Kong . . . . . . . . . . . . . . . . . . . 27
5.4 McDonald’s website in Korea . . . . . . . . . . . . . . . . . . . . . . 27
5.5 McDonald’s website in Denmark . . . . . . . . . . . . . . . . . . . . . 28
5.6 McDonald’s website in Norway . . . . . . . . . . . . . . . . . . . . . 29
5.7 McDonald’s website in Sweden . . . . . . . . . . . . . . . . . . . . . . 30
5.8 McDonald’s website in Finland . . . . . . . . . . . . . . . . . . . . . 31
5.9 Summary of cultural differences connect to UX design in McDonald’s
websites in 2019 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5.10 Text with or without icons . . . . . . . . . . . . . . . . . . . . . . . . 35
5.11 Hierarchical Menu or Flat menu . . . . . . . . . . . . . . . . . . . . . 35
5.12 Hue preference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.13 Image with or without individuals . . . . . . . . . . . . . . . . . . . . 36
5.14 Video preference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.15 Text density preference . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.16 Layout preference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.17 Link opening preference . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.18 Home page of prototype Chinese . . . . . . . . . . . . . . . . . . . . . 39
5.19 Menu bar for Chinese website . . . . . . . . . . . . . . . . . . . . . . 39
5.20 Request submission page for Chinese website . . . . . . . . . . . . . . 40
5.21 Interaction between Chinese prototype pages . . . . . . . . . . . . . . 40
5.22 Home page-1 of prototype Swedish . . . . . . . . . . . . . . . . . . . 41
5.23 Home page-2 of prototype Swedish . . . . . . . . . . . . . . . . . . . 42
5.24 Menu bar for prototype Swedish . . . . . . . . . . . . . . . . . . . . . 42
5.25 Request submission page for prototype Swedish . . . . . . . . . . . . 43
5.26 Interaction between Swedish prototype pages . . . . . . . . . . . . . . 43
5.27 Information of pilot testing participants . . . . . . . . . . . . . . . . . 44

xiii
List of Figures

5.28 Questions for the pilot testing . . . . . . . . . . . . . . . . . . . . . . 45

5.29 Answers received from participants . . . . . . . . . . . . . . . . . . . 45
5.30 Dark tone design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.31 Light tone design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.32 Text with icons in light tone design . . . . . . . . . . . . . . . . . . . 48
5.33 Text without icons in light tone design . . . . . . . . . . . . . . . . . 48
5.34 Text with icons in dark tone design . . . . . . . . . . . . . . . . . . . 49
5.35 Text without icons in dark tone design . . . . . . . . . . . . . . . . . 49
5.36 Three to four column layout in dark tone design . . . . . . . . . . . . 50
5.37 One to three column layout in dark tone design . . . . . . . . . . . . 50
5.38 Participants background in usability testing . . . . . . . . . . . . . . 51
5.39 Questions asked to participants in usability testing . . . . . . . . . . 52
5.40 The education background of the participants . . . . . . . . . . . . . 52
5.41 The education background of the participants . . . . . . . . . . . . . 53
5.42 The education background of the participants . . . . . . . . . . . . . 53
5.43 Color tone preference . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.44 Text with/without icons preference . . . . . . . . . . . . . . . . . . . 55
5.45 Layout style preference . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.46 Link opening preference . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.47 Usability testing survey results . . . . . . . . . . . . . . . . . . . . . . 57

6.1 Old login page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

6.2 New login page for Chinese user . . . . . . . . . . . . . . . . . . . . . 60
6.3 New login page for Swedish user . . . . . . . . . . . . . . . . . . . . . 61
6.4 Old home page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
6.5 Home page for Chinese user . . . . . . . . . . . . . . . . . . . . . . . 62
6.6 Home page for Swedish user V1 . . . . . . . . . . . . . . . . . . . . . 63
6.7 Home page for Swedish user V2 . . . . . . . . . . . . . . . . . . . . . 64
6.8 Old knowledge base page . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.9 Knowledge base page for Chinese users . . . . . . . . . . . . . . . . . 66
6.10 Knowledge base page for Swedish users . . . . . . . . . . . . . . . . . 67
6.11 Old knowledge based detail page . . . . . . . . . . . . . . . . . . . . 68
6.12 Knowledge based detail page for Chinese . . . . . . . . . . . . . . . . 69
6.13 Knowledge based detail page for Swedish . . . . . . . . . . . . . . . . 70
6.14 Old submit request page . . . . . . . . . . . . . . . . . . . . . . . . . 71
6.15 Submit request page for Chinese users . . . . . . . . . . . . . . . . . 72
6.16 Submit request page for Swedish users . . . . . . . . . . . . . . . . . 73
6.17 Main navigation menu . . . . . . . . . . . . . . . . . . . . . . . . . . 74
6.18 Submenu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

xiv
List of abbreviations
HCI Human-Computer Interaction

UI User Interface

UX User Experience

PDI Power Distance Index

IDV Individualism versus Collectivism

UAI Uncertainty Avoidance Index

MAS Masculinity versus Femininity

LTO Long-term Orientation versus Short-term Orientation

IND Indulgence versus Restraint

xv
List of Figures

xvi
 1
Introduction

The rapid development of globalization and international trade have brought criti-
cal challenges to human-computer interaction (HCI). Designing websites and online
services to fulfill the needs of people from different cultural backgrounds is one of
the success factors in today’s global economy [1].

The most important principle in the HCI field is “know your user”, in other words,
human-centered design (HCD) is the foundation of commercial products [2]. How-
ever, humans usually live in social groups with different languages, cultures, social
norms and values. So adopting diversity in user experiences (UX) for diverse peo-
ple is essential for the globalized market. Simply providing the local language or
currency for websites as cultural adaptation is not enough. Cultures affect the way
users interact with a product and interpret its features [5], and different cultures
lead users to have different web design preferences, such as layout, color, graph-
ics, links, navigation, multimedia, and text. The failure of eBay in both Chinese
and Japanese markets is a corollary of the grave consequences of underestimating
the cultural differences between Asian and U.S. markets[3]. It also indicates that
Levitt’s theory “one size will fit all” is not the solution for global corporations [4].
Understanding these cultural preferences and translating these design elements for
local conditions is essential for cross-culture UX designers.

Kingsley Davis (1949) argues in detail how culture is not static but is dynamically
changing in society. The invention and popularisation of new technology is one way
to change people’s behavior. The Chinese market has 800 million internet users
which keep increasing every year and is the largest population on the internet in the
world [6]. To understand Chinese users in the context of the rapid growth of the
internet industry can be a big challenge for international companies who have the
desire to open huge market opportunities in China.

1.1 Industry Background

1.1.1 Kollmorgen
Kollmorgen Automation AB is a Swedish company based on Goteborg, Sweden,
which focus on producing motion systems and components for manufacturers. In
response to the rapid growth in Chinese market, the company has a keen desire to

1
1. Introduction

understand the cultural differences between China and Sweden to better expand the
business and make promotion.

1.1.2 The website

https://support.ndcsolutions.com/hc/en-us is a website belong to Kollmor-
gen, focus on customer support and product documentation. The target user groups
are clients and company employee based on different countries all over the world.

1.2 Research question

This thesis will focus on the research question:

How different cultural factors influence user’s preferences for UX attributes between
Chinese and Swedish websites.

To help answer this, a supporting question will be asked:

How to adopt UX design for Chinese and Swedish users?

To give answers to these questions, interviews will be held in China and Sweden to
analyze how different cultures affect users’ surfing behavior. A detailed UX design
guideline will be provided to compare the different preferences of design elements.
Furthermore, young people and middle age people are considered to have different
web performance [7]. So when recruiting the participates, age will be selected and
try to be balanced. The number of the participants will be equally divided (half of
Chinese while another half of Swedish).

1.2.1 Deliverable
A Swedish website (https://support.ndcsolutions.com/hc/en-us) was redesigned
based on the research results. Both Swedish and Chinese versions were implemented
respectively which towards user’s usability and applicability to increase user satis-
faction based on different cultural background. As young people and middle age
people are considered to have different web performance [7]. When recruited the
participates, age was selected and try to be balanced. The number of the partici-
pants was equally divided (half of Chinese while another half of Swedish).

A design guideline was provided for general website design as a clear instruction to
designers and developers on how to achieve principles for both Chinese and Swedish
websites. The guideline priority focuses on UX elements design, dealing with the
layouts, navigation, links, aesthetics part, including color, font style and visual
representation, multimedia and content.

2
 1. Introduction

1.2.2 Delimitation
The content of https://support.ndcsolutions.com/hc/en-us a business website
mainly focus on commerce, corporate information and product introduction. There-
fore, the design guideline is only valid for websites in the same category with the
same or similar functions and features.

The target user group for the website has a certain limitation. For Chinese users,
it can be categorized to be middle class, who are well educated, have a full-time
job, lives in mid-sized or The target user group for the website has a certain lim-
itation. For Chinese users, it can be categorized to be middle class, who are well
educated, have a full-time job, lives in mid-sized or big cities in China. For users
from Sweden, they are well educated, have a full-time job and live in cities. The
participants selected for this study followed the target user group pattern, who hold
an undergraduate degree or above, and live in cities both in China and Sweden.
Therefore, the result and conclusion of this research can only be seen as a sample
for users limited in a certain scope.

1.3 Stakeholders
The stakeholders of this thesis including Chalmers University of Technology, Koll-
morgen Automotion AB, participants of the research study, users, researcher of this
thesis, and any other people who may have a relationship with this thesis.

1.3.1 Chalmers University of Technology

The school supervisor of this thesis is Gordana Dodig Crnkovic, professor in Inter-
action Design at Chalmers University of Technology. The examiner of this thesis is
Staffan Björk, professor in Interaction Design at Chalmers University of Technology.

1.3.2 Kollmorgen Automotion AB

The thesis study corporated with Kollmorgen Automotion AB, a Swedish company,
redesign a website for its Chinese clients to fit their cultural background and website
design preference. Peter Bladh, director of product management is the supervisor
in Kollmorgen.

1.3.3 Participants
Participants from China and Sweden was recruited for research study. They accom-
plished a series of tests and interviews for the redesign work.

1.3.4 Researcher
An interaction design student from Chalmers University of Technology was re-
searcher of this thesis.

3
1. Introduction

1.3.5 Users
The end users of the redesign website are the colleagues and clients of Kollmorgen
Automotion AB.

1.4 Ethical Issues

It is not unusual that designers sometimes have to meet the ethical conflicts in a
project. According to the book “Human values, ethics, and design” [26], there are
twelve human values related to different ethical issues that should be considered
during a design process. Designers should mutually understand the influence they
might exert, which would result in a long-term impact on society. For this thesis,
redesign the websites for both Chinese and Swedish, the most important ethics
aspects are privacy, informed consent, usability and bias.

1.4.1 Privacy
The study is about how cultural differences influence user’s behavior and preferences
for UX design. In this case, participants are persuaded to provide a list of personal
information for detecting user behaviors and preferences. How to protect users’
privacy, how to reduce the risks from divulging this information to other people or
organizations become one of the most severe problems in our design process. The
famous Facebook data scandal [27] is an example how people’s life and activities
were affected by data breach. Target users might be easily influenced by advertise-
ment and be persuaded purchasing the particular commodity or attending particular
events. As designers, we have the responsibilities to inform users of the risks they
may meet in connection to the gathering individual data.

1.4.2 Informed consent

Back to the Facebook data scandal [27] mentioned above, the third-party app gath-
ered and used personal data not only from users but also from their friends without
their knowledge or any informed consent. This scandal was an invasion of users’
privacy and also lead to other issues such as lack of trust and autonomy [26]. In
this study, the informed consent was signed based on the principle of voluntariness
before they engage in the whole process for this study. Participants have the rights
to know what information they would provide, and fully understand the reason and
the purpose of data. In addition, participants had the right to modify or withdraw
the consents once they feel unsafe or uncomfortable. They also had the right to
delete their personal data if they consider it is necessary. All these rules mean that
user’s personal data are controlled by law and can only be used for a certain purpose.

1.4.3 Usability
Undoubtedly, usability is the key point of the design of interaction of the user with
the website. It should be user-friendly and easy to achieve tasks. According to

4
 1. Introduction

“Human values, ethics, and design” [26], UX design should consider user diversity
more than aesthetics. Since the website can be visited by a range of different people
with different background, how to create a product that can satisfy most of the
users is one of the most important tasks for this project. For instance, the website
should consider users’ social environment, culture, age, job, disability, education
background to guide all kinds of users to understand how to visit it and meet their
needs. Usability aims to narrow the gap in user’s background and make sure they
enjoy the website.

1.4.4 Bias
Bias usually hides in details and is easy to be ignored. People with biased thoughts
tend to refuse to open their mind and take other people’s opinions, which can influ-
ence user’s behaviors and beliefs and cause a long-term effect. Designer should take
the responsibility to avoid harms and contribute to improving human situations.
Since this thesis aimed to design a website based on cultural differences, a cus-
tomized UX design was published based on the cultural preferences. On one hand,
it is our goal to design for local users and meet their needs, on the other hand, this
local customized cultural design might lead to the increase of social cultural tensions
or increasing misunderstanding. For instance, if the social distance with respect to
ethicsin the culture is high, should the design condone this phenomenon and enlarge
the social inequality? Designers should pay attention and always try to improve
communication and equity as basic principles in modern democratic society..

5
1. Introduction

6
 2
Theory

In this chapter, relevant theoretical concepts and modes regarding how culture in-
fluences UX design will be brought up. Most of the thesis mentioned here is about
cultures and how cultures impact human behaviors. Two classical culture theories,
including Hofstede’s six cultural dimensions theory and Hall and Hall’s low-context
and high-context culture model is introduced first, which were widely applied in
the researcher similar area. Followed up with Holliday’s grammar of culture and
Edwin Hoffman’s culture mode, which giving suggestions on new aspects to con-
sider the relationship between individuals and cultures. Followed are the concept
of culture globalization and the introduction of user experience elements, for better
understanding this study.

2.1 Culture and website design

There is a variety of understanding of culture. A broad definition of culture is that
it is the general behaviors, beliefs, values and other common characteristics to the
people live a specific group or society [8]. Culture influences the way people think
and behave as well as the way they understand and interact with a product. For
website design, culture also play an important role to impact people’s behaviour and
preferences, as people live in different regions are influenced by the local values and
other characteristics.

Figure 2.1 shows IKEA’s website in Sweden and China. We can see that IKEA sells
similar products in China and Sweden, and it is easy to figure out the two websites
are from the same company at the first glance. However, these two pages applied
a series of differences UX design, including layout, navigation, images, font style,
color, etc. It can be assumed that a successful international company such as IKEA
would do detailed user research before they launch the websites. Does IKEA apply
different user interface (UI) based on cultural preferences? How many of the UX
differences on the web pages are connected to cultural differences?

Figure 2.2 shows two online shopping websites: Adlibris and Taobao, designed by
Swedish and Chinese companies respectively, which show a totally different design
style. These two online shopping websites have a similar business strategies that
they sell a large range of products cover different categories. Users can search and
find similar items with different prices, compare them and purchase what they like.
It can be seen from the pictures that the UX design styles are totally different. It

7
2. Theory

is one of the evidence that cultures may play an important role in UX design.

Figure 2.1: IKEA’s website in Sweden and China

8
 2. Theory

Figure 2.2: Online shopping websites in Sweden and China

2.2 Hofstede’s six cultural dimensions theory

Hofsted [9] defined culture as “the collective programming of the mind which distin-
guishes the members of one group or category of people from another.” Hofstede and
Minkov’s cultural dimensions theory [9] is widely applied as the most comprehensive
framework of national culture for business purpose. It is also used in cross-cultural
web design for analyzing how cultures influence web design. He organized and de-
termined cultures along six dimensions:

2.2.1 Power Distance (PDI)

People’s attitude to the degree of hierarchical structure in the social and how they
manage the inequalities. In high power distance context, people tend to accept a

9
2. Theory

large degree of wealth or status disparity without any reason, while in low power
distance society, people require equality and demand justification for power inequal-
ities.

2.2.2 Individualism vs. Collectivism (IDV)

People belong to individualism society are considered to take care of themselves and
their immediate family members. In opposite, people live in collectivism society can
be expected to take care of relatives and other members for extra resources.

2.2.3 Uncertainty avoidance (UAI)

Uncertainty avoidance measures social member’s tolerance for uncertainty and am-
biguity. Societies with high uncertainty avoidance expect members to follow certain
social rules and behaviors while low uncertainty avoidance countries are more relax
and allow people to explore themselves.

2.2.4 Masculinity vs. Femininity (MAS)

This dimension does not represent gender but the way people concern achievement
and financial rewards to success. People are used to a competitive environment in
a high masculinity society. By contrast, high femininity dimension means people
prefer work-life balance lifestyle.

2.2.5 Long Term vs. Short Team Orientation (LTO)

Societies with lower scores in this dimension would rather keep their traditions and
norms and are hesitant about social changes. However, societies with higher scores
in this dimension are more open to modern education and social revolution.

2.2.6 Indulgence vs. restraint (IND)

Societies have high Indulgence scores tend to have a tolerant attitude towards people
enjoy life and have fun. Societies have low indulgence scores are considered as the
restriction that people are regulated by strict social rules.

As the dimension scores are relative to each culture, this model can be used through
comparing with different cultures. We can see the cultural differences between Swe-
den and China in figure 2.3. It can be seen that marked differences display in almost
all the dimensions in these two cultures. Sweden owns low power distance, masculin-
ity, long term orientation scores, and high individualism and indulgence scores, while
the scores show in an opposite way in China. These results match those observed in
UX design in both cultures mentioned above, that culture may have played a vital
role in bringing about how cultures influence UX design.

10
 2. Theory

Figure 2.3: Cultural dimension comparison between China and Sweden [10]

2.3 Hall and Hall’s low-context and high-context

culture model
Hall and Hall’s cultural model [11] analyze cultures by two key factors: context and
time, and dividing cultures into two categories: high-context (HL) and low-context
(LC). He considered that culture is all about communication, and all the acts, events
and material things behind the communication have certain meanings [11].

2.3.1 High-context vs. Low-context

In HC societies, people prefer a long-term relationship and may use a more implicitly
and ambiguity communication style. By contrast, communication in LC societies
is more direct, precise, open and rely more on the literal meaning of the words.
According to this culture model, China is a typical HC society while Sweden belongs
to LC society.

2.3.2 Time
It connects to how people used the time to organize tasks. It is divided into two
categories: single task focus, and multi-tasking. According to the culture model, the
Chinese prefer to perform multitasking (polychronic) and Swedish tend to conduct
single tasks (monochronic).

2.4 Holliday’s grammar of culture

According to the grammar of culture created by Holliday [29], culture is not a single
structure but composed of three domains: particular social and political structures,
underlying universal cultural processes and particular cultural products. In his cul-
ture theory, Holliday indicates that Culture is influenced by multiple factors, people

11
2. Theory

would generate different culture concepts based on family, politics, literature, media,
etc., even they live in the same society.

As can be seen in figure 2.4, underlying universal cultural processes is in the middle
of the grammar, combined with small culture formation and personal trajectories.
Small culture formation is structured by small groups such as families and work
groups, which is the foundation for personal culture concept. Meanwhile, individu-
als who belong to small culture formation would generate the concepts of their own
culture from personal trajectories such as society and ancestors.

Figure 2.4: Grammar of culture, based on the figure by Hoffman [29]

Particular social and political structures are on the left of the grammar, which is
constructed by two blocks, cultural resources and global position and polities. Cul-
ture resources, including education, languages, government and the overall ideolog-
ical of the living environment impact people’s behave. Global position and politics
including the nation and its history, and how individuals consider the position of
themselves in the world. The overall understanding of social and political structures
impact the way an individual’s thoughts of culture.

On the right of the grammar is the particular cultural products, which is made up
of artifacts and statements about cultures. Art, literature, media, architecture and
cultural practices group together as artifacts. Cultural practices are the things indi-
viduals do every day, for instance, eat, greet and show respect, which are the most
common cultures in a nation. And for statements about cultures, is the way how

12
 2. Theory

individuals represent and display cultures by themselves.

According to this theory, it is not surprised that individuals can generate a variety
of cultures, even they live in the same society.

2.5 Edwin Hoffman’s culture mode

In the book Diversity competence [30], Hoffman describes intercultural communi-
cation should follow the pattern as shown below (Fig 2.5). First of all, individuals
from different background have and share the basic needs, emotions and capabilities
with others, as illustrated in the lower level in the figure. In this level, people’s
thoughts and behaviors are similar no matter where they come from. Then in the
second level is how people deal with and express these basic needs, emotions and
capabilities. In this level, cultures would influence people’s behaviors from different
ethnic, nations or religious background. In the last and highest level is personalities,
which is the ultimate expression of an individual to communicate with others. This
should be a unique way not like any others in the world.

In Hoffman’s opinion, interaction and communication are in encounters between

unique personalities which are also impacted by human nature and cultures.

Figure 2.5: Humanity, culture and personality: a conceptual scheme (based on a

figure by Geert Hofstede, 1991)

13
2. Theory

2.6 Cultural globalization

Cultural globalization refers the phenomenon that ideas, meanings, and values are
transmitted around the world, propelled by the rapid advancement of intent, social
media, entertainment, international travel, immigration, food, etc [31]. The devel-
opment of wireless communications promotes conditions for inexpensive air travel,
as well as internet, telephones and television, breaks the barriers of time and space,
allows people to receive and send news and information in every corner of the world,
which accelerate the spread of globalization [33].

Globalization has a trend toward homogeneity that will eventually lead people to
live in different countries or regions experience essentially the same culture [32]. For
instance, nowadays, it is easy to find Starbucks, Coca-Cola and Adidas in cities all
over the world. These brands apply the same design styles for their visual Identity
including color, logo, stores, service, etc, As well as selling similar products with
similar packages.

Furthermore, globalization of fast food does not only introduce exotic flavor to local
residents but also influence public conduct [35]. According to the book “Golden
Arches East: McDonald’s in East Asia”, Watson describes how fast food changed
local people behaviors and preferences. One example is in some Asian countries,
such as Japan and China, eating food without spoon or chopsticks was considered
as a breach of etiquette. However, with the popularization of McDonald’s, it is
common to see locals to eat hamburgers with hands in public place nowadays.

Moreover, the power of entertainment, such as social media and digital games also
have a huge impact on people’s idea and behaviors. People use the same mobile
apps, play the same games, watch the same series all over the world lead people to
have the tendency to generate the same preferences and tasty of entertainment style.
The popularization of Japanese video game in Sweden is an example that people
first interested in one thing from a foreign country, then they become interested in
other cultures from Japan. Now it is not a surprise that people are familiar with
Bushido and kimono in Sweden.

2.7 User Experience Elements

According to the Nielsen Norman group [7], user experience is about all aspects of
how user interact with a company, including its service and products [7]. User ex-
perience usually connects with HCD, that focus on understanding users’ needs and
requirements. Designers are required to create specific user experiences to improve
the user’s daily life [15]. The ultimate goal of HCD design is to improve efficiency
and usability by observing people’s behaviors and conducting a series of research to
analyze user’s desires and needs.

Researchers have suggested a series of design attributes that can represent how users

14
 2. Theory

interact with a product. These key attributes are listed as layout, navigation, links,
aesthetics, multimedia and content.

2.7.1 Layout
Layout represents the structure of a web page. How the information organized de-
termines how many efforts users spend to understand the content of the page. Re-
search shows that people from monochronic cultures accept hierarchical structures
more than people from polychronic cultures. By contrast, people from polychronic
cultures would rather visit parallel structures [16]. Alexander, Thompson and Mur-
ray’s study[17] shows that Chinese users prefer three-column web pages and display
density.

2.7.2 Navigation
Navigation is a set of content that allows users to visit information in an effective
way. Navigation includes the main menu on the web pages and the fat footer at
the bottom of the pages, which is usually used as quicklinks. Research shows that
users from high uncertainty avoidance countries prefer clear and logical order of
menu items [18]. In addition, users from HC would like to visit websites with many
sidebars, but users from LC are not interested in sidebars at all [20].

2.7.3 Links
Links is how user visit data from one location to another by click or tap [19]. Re-
search found that users from HC prefer open links in new windows, while users from
LC would rather open links in the same browser windows [17]. Moreover, HC users
tend to apply icons or visuals to represent the meaning of links, while LC users would
like to visit links with precise and logical meaning. Gould, Zalcaria and Yusof [28]
consider that websites from collectivism culture show a lot of links which connect
with other organizations to emphasize strong network with other organizations.

2.7.4 Aesthetics
Aesthetics contains color, font style and visual presentation, and these elements
decide how information is displayed on the website.

2.7.4.1 Color

It is well understood that color contains different meaning in different cultures and
can exert on perception in design [21]. For instance, red means lucky, happiness or
government in China, but means strength in Sweden. White usually represent death
and mourning in China, and in Sweden, it implies peace and nature. National flag
color and traditional color are popular in HC cultures [17].

15
2. Theory

2.7.4.2 Font style

HC prefer to use bold and strong fonts for important context [17].

2.7.4.3 Visual presentation

Images and animations are more likely to be applied in HC websites to convey
information [22]. Moreover, Alexander, Thompson and Murray’s research [17] shows
that images of leaders and skyscraper are popular in collectivism societies with
high power distance. Pictures promote collectivistic characteristics, for example,
group activities images are popular in collectivism cultures. In opposite, images
of people’s daily life or nature are popular among individualism societies with low
power distance. Individual images are more often used in individualism societies.
In addition, cartoons and animal pictures are popular in HC culture [20].

2.7.5 Multimedia
Pop-ups are frequently applied in polychronic cultures but not at all popular in
monochronic cultures [17]. Multimedia displayed on the screen with ‘play’ or ‘pause’
buttons are more popular in individualism societies, but not found in collectivistic
societies.

2.7.6 Content
Individualistic users prefer websites that allow users to customized content and
provide a variety of content options. However, collectivist users are interested in
popular and ranking content so that they can follow the masses [18]. Gould, Zal-
caria and Yusof’s research [28] shows that high power distance culture users prefer
clear structure and description of the company’s hierarchy to better understand the
organization. By contrast, low power distance culture users focus on personal needs
rather than the company’s organization.

16
 3
Methodology

In this section, design methods considered suitable for this thesis are listed. A
description of each method and the reason why it was selected will be provided late
on, as well as their advantages and disadvantages.

3.1 Design Methods

Throughout this project, Gaver’s theory [23] has been adopted, which provides re-
flections to address concerns from research through design. Gaver concluded that
research through design has a better chance to produce provisional, contingent, and
aspirational theories, compared with Philosophy of Science theory. He commented
that research through design is the solution to establish theories of design compre-
hensively, which means design research should be based on realized design products.
Several design methods were considered during the preparatory phase: question-
naire, diary studies, think aloud, pilot test and usability tests. The whole design
process followed the UX process [25] to understand the user’s needs and requirements
(Fig 3.1).

Figure 3.1: The UX design process [25]

17
3. Methodology

3.1.1 Diary study

The Diary study is a design method that allows participants to describe their daily
life and feelings in a short period to designers for future research [24]. As the re-
design work focuses on different cultural preferences, and culture is an abstract
concept related to all the aspects of people’s daily life, a diary study is supposed to
contribute to an understanding of the participant’s values and hidden ideas of how
they perceive things around them.

Participants should be given a list of tasks and be required to achieve them in one
week, one month or longer. The tasks are about how participants make decisions,
relationship with things, preference websites, etc. Participants are asked to sub-
mit pictures or brief descriptions in text for these tasks. Traditionally it has been
achieved by pens in journals through technology affords new solutions that the diary
can be created on digital devices and send back by email.

A Diary study allows participants to provide information based on their daily life in
a central period. It gives designers an opportunity to discover the user’s behavior
in natural usage scenarios, to better understand user abilities and user experience.
Secondly, a diary study reveals external factors that influence user experience, which
are impossible to understand in a closed testing environment. Finally, a diary study
usually lasts a longer duration, which encourages participants to think about their
experience deeply as well as find special ways to express their feeling. However, since
participants follow the diary study independently, designers cannot observe their
behavior when they do the tasks, which may lead to misunderstanding. Furthermore,
longer duration also brings issues, as participants may feel pressure to achieve a series
of tasks in a long period, which may bring bias and moods when they accomplish
tasks. This would lead to an inherent bias in the later study.

3.1.2 Questionnaire
Questionnaire is an efficient method to collect large quantities of data [24]. A ques-
tionnaire is a time and cost-effective solution for data collection. With an online
questionnaire, it is easy to collect results from a large audience. Besides, with on-
line analysis tools, it is easy to analyze the results without a statistics or scientific
research background. One more advantage is that, as questionnaires are most of-
ten anonymous, participants usually accomplish them without any pressure, which
could provide them with an environment to write down their true thoughts.

A Questionnaire also has some disadvantages. Participants may misunderstand or be

unable to interpret the questions and that may lead to skewed results. Short answer
questions should be created in order to ensure individuals fully understanding the
meaning of the questions, which would cause another problem, that it is difficult
to gain explicit and precise answers through the questionnaire. Moreover, it is not
uncommon that questions are ignored or unanswered. If all the questions are set
as required, some individuals may stop answering the questionnaire and close the
window.

18
 3. Methodology

3.1.3 Prototyping
Prototyping is the method that transfers the design concept to the realization [24]
based on the initial data collected from participants. Prototypes are divided by the
level of fidelity: low-fidelity prototype, middle-fidelity prototype and high-fidelity
prototype.

Prototyping allows participants to know the feeling and look of the product before
the final release. Also, it allows designers to test usability and find problems for
ideation. Since designers usually build prototypes with paper or software, it is a
time and cost-effective way for early-stage testing. However, since prototyping is
not an actual website that allows testers to explore every corner on the pages, it
may limit the actual user behavior in the real world as participants can only interact
with the pages assigned by designers. The missing features and details may lead to
misunderstanding.

3.1.4 Thinking aloud

Thinking aloud requires testers to articulate their thoughts, activities and feeling
while achieving tasks assigned by researchers [24]. Usually researchers would repeat-
edly remind participants to speak out what they are thinking when they go through
a task. It allows designers not only to see the process of addressing the tasks but
also to observe testers’ behavior and moods to improve the product.

Think aloud is an effective way to figure out usability problems in the early design
stage. As a face to face interview method, think aloud allows designers to con-
duct open-end interviews based on participants’ behaviors for future improvement.
Moreover, think aloud also gives designers the chance to generate quotes which they
may have ignored. However, when applying the think aloud method, it is common
that testers get stuck when they perform tasks and have to think aloud at the same
time or forget to speak out what they are doing during the test. Sometimes testers
spend a huge amount of time expressing their views about the products, which are
not related to the purpose of the testing.

3.1.5 Usability tests

Usability tests are the way to test a product by empirical evidence while observing
participants’ behaviors and emotions [24]. They are supposed to be conducted after
establishing the website prototype in order to receive feedback for future develop-
ment. The reason for performing usability tests is, that most of the design methods
lack actual user behavior since they only collect data that participants provide them-
selves. Usability tests can contribute to real users’ behaviors and feelings while they
visit the website for the later iteration.

Usability tests give designers an opportunity to know if people would enjoy and go
through the product in an easy way. It can be proof that the design is on the right
track before refining the final design. However, though participants can figure out

19
3. Methodology

the interactive problems in the design, it is difficult for participants to explain what
should be changed. Besides, if the participants cannot achieve the task themselves,
designers may give them hints that help them to accomplish the task, which would
lead to biased results.

3.1.6 Pilot test

A pilot test is a small-scale trial before the formal study with the purpose of testing
if the proposed methods and prototyping are viable. Performing a pilot is a solution
to find out shortcomings and potential confusion and give designers the chance to
modify the methods and prototyping before the real test. The data of the pilot test
should not be included in the final result. Participants who attend the pilot test
are not allowed to join in the subsequent tests as they have already had preexisting
knowledge about the project which can distort the result [36].

20
 4
Planning

The study was planned to start from week three in 2019 and be accomplished in
week 21. Supervised by both supervisors at Chalmers and at Kollmorgen every
week, to ensure the study was on the right track and achieve the goal.

4.1 Methods selection

Initially, diary study, prototype and usability tests were selected. The plan was to
recruit a group of Chinese and Swedish participants, who first did the diary study
to generate a design concept and create the prototypes. They Then were supposed
to test and iterate the prototypes for the final redesign websites. Participants were
supposed to attend all three testings for a consistent result.

However, some of the participants could not achieve all three testings due to work
reasons. Moreover, diary study requires participants to accomplish a number of
tasks in a certain period of time, which made some participants felt pressured or
worried about whether they can finish the tasks on time as most of the participants
have full-time jobs.

Therefore, the diary study method was replaced by an online questionnaire for a flexi-
ble and easy procedure, as it could be done in several minutes. https://surveys.enalyzer.com
was used to create a picture-based questionnaire. It was shared to employees work-
ing in Kollomorgen and interaction design master students studying in Chalmers.
The purpose was to gain as much data as possible at the early stage to see a pat-
tern of how cultural background influences individual web design preference. Two
languages: Chinese and English were shown to support the questionnaire for par-
ticipants from different regions. Low-fidelity prototypes were applied based on the
result of the online questionnaire. A pilot testing with a small scope of users was
added, followed up with an iterative improvement to make sure the formal usability
test was fully prepared. High-fidelity prototypes were implemented for the usability
test, which was retained as the method does not take too much time, so employees
finished them without pressure. They were tested and discussed by participants and
participants’ response received based on aesthetics and usability. For the usability
test, https://www.surveymonkey.com was applied for feedback and comments re-
ceived. Google sheets was used for calculating statistics.

21
4. Planning

4.2 Tools
Several tools and services were applied in the design process. Adobe XD is a tool for
creating prototyping from low-fidelity to high-fidelity for web and mobile apps. In
this study it was mainly used for website wireframing and interactive prototypes for
the usability test and iteration. Compared with photoshop, XD is more straightfor-
ward and effective for low-fidelity prototyping, as it provides a simple-click interac-
tion prototyping solution, which is useful for website design testing. Photoshop was
used for the final website design as it is a precise design tool providing pixel-perfect
level design, which works for final web page development.

4.3 Time plan

This study was planned to be divided into five stages: literature study, analyze,
design, evaluation and thesis writing. Meanwhile, design iteration was conducted in
the whole design process.

20-01-2019 - 14-02-2019 Literature study

Several journal research articles in this area were selected for the literature study.
The Hofstede’s cultural dimensions, Hall’s high/low-context culture model, Holli-
day’s grammar of culture and Hoffman’s culture mode were selected to analyze the
cultural differences between China and Sweden.

15-02-2019 - 28-02-2019 Analyze

Planning report was submitted to Chalmers and presented at Kollmorgen. Mc-
donald’s websites in Asia and the Scandinavian were analyzed with the purpose of
finding out cultural differences between Chinese and Swedish in today’s fast-paced
digital world. User research was conducted based on the analysis results.

01-03-2019 - 27-03-2019 Design

Design concepts and low-fidelity prototypes were created based on the data collected
in the analyze stage. Pilot testing was conducted ahead to ensure the usability test
would work well. The prototypes were refined in the whole process.

28-03-2019 - 12-04-2019 Evaluate (Usability tests)

Redesign pages were created based on the usability test result. Design guideline was
also built in this stage.

15-04-2019 - 31-05-2019 Report Writing

Finishing the thesis report and preparing for the final presentation

22
 5
Design Process

In this chapter, the procedure of this study is described. The design process included
several steps: preparation, user research, ideation and low-fidelity prototyping, pi-
lot testing, usability testing, evaluation, final redesign and design guideline setting.
First of all, McDonald’s website design in part Asian and Scandinavian areas were
analyzed for understanding the different design in different regions. Then a user
survey was published online to gather data from Swedish and Chinese people about
their web design preferences. Prototypes were created based on the user data col-
lected from the survey, followed up with a user test and an analysis for iteration and
the final results.

5.1 Preparation
The initial phase of the study was dedicated to literature research in the domain
of intercultural HCI design. The general concept and structure of the study were
created in this phase.

Literatures relevant to intercultural design were read in this phase. As this research
is mainly focused on commercial and business websites, the goal was to find litera-
ture regarding this subject. Interestingly, there are not so many literature aimed at
HCI in different cultures. Moreover, when it related to the field of market and con-
sumption, most literature focus on the internet of things (IOT) or industrial design
in different cultures. Literature for website analysis tends to have a preference for
government or university websites rather than consumer websites.

Würtz [20] provides an idea for HCI design in different cultures by analyzing Mc-
donald’s websites in different regions. The American fast food chains has more than
26,000 restaurants in 120 countries [34], and applies different website design in dif-
ferent areas, which is a good example for cross-cultural HCI design analysis, as it
can be assumed that a successful international company has done user preference
testing for the website and design pages based on customers preferences to improve
HCI on websites in each culture.

As Würtz’s research was 14 years ago, and website develop technology Is rapidly
growing, all the Mcdonald’s website design displayed in the paper has been changed
since then. It was decided that this study would first analysis nowadays Mcdonald’s

23
5. Design Process

websites across Asia and Europe, especially in the Scandinavian area, to observe
whether the website design still match the results on the paper. If the same data
and results would be received, the low-fidelity prototyping would be designed in the
next phase, otherwise, a questionnaire would be launched for the latest user data.

5.1.1 An Analysis of McDonald’s websites

The analysis of McDonald’s websites contains websites from countries belong to ei-
ther HC or LC cultures. Mainland China, Hong Kong, Japanese and Korean websites
were selected as representatives of HC cultures, while Denmark, Sweden, Norway
and Finland contribute representatives of LC The analysis of McDonald’s websites
contains websites from countries belong to either HC or LC cultures. Mainland
China, Hong Kong, Japanese and Korean websites were selected as representatives
of HC cultures, while Denmark, Sweden, Norway and Finland contribute represen-
tatives of LC cultures.

The analysis of the websites mainly forward on user experience elements including
layout, navigation, links, color, font style, visual presentation, multimedia and con-
tent. For content on the webpage, the analysis refers to the form and expression of
the content on the web pages rather than written text. One thing should be noticed
is, this analysis is mainly focused on a comparison between Würtz’s research result
and nowadays websites design, to figure out whether website design follows the same
design patterns in line with expectations. In this case, there is no in-depth analysis
in this stage.

5.1.2 Result
The following description illustrates how cultures impact McDonald’s website design
in 2019.

5.1.2.1 Animation
According to Würtz [20], HL cultures tend to apply animation connected with hu-
man on web pages while LC cultures only use few animation. In 2019, no significant
difference between the two culture groups was evident. As shown on the screen-
shot, no animation displayed on websites in Mainland China (Fig 5.1), Hong Kong
(Fig 5.3) and Korea (Fig 5.4), only the Japanese website (Fig 5.2) shows a gif about
moving products. Neither Chinese or Japanese websites use animation in connection
with pictures of moving people. Actually, it seems that it is not popular to show
people’s photos on Asain’s website anymore. In all the four asian cultures, products
occupy the biggest and the most important places, only small pictures show people’s
activities.

By contrast, follow the same pattern as Würtz’s results, that none of the four web-
sites in Denmark (Fig 5.5), Sweden (Fig 5.7), Norway (Fig 5.6) and Finland (Fig 5.8),
use any of animations on their pages.

24
 5. Design Process

Figure 5.1: McDonald’s website in Mainland China

25
5. Design Process

26 Figure 5.2: McDonald’s website in Japan

 5. Design Process

Figure 5.3: McDonald’s website in Hong Kong

Figure 5.4: McDonald’s website in Korea

27
5. Design Process

Figure 5.5: McDonald’s website in Denmark

28
 5. Design Process

Figure 5.6: McDonald’s website in Norway

29
5. Design Process

Figure 5.7: McDonald’s website in Sweden

30
 5. Design Process

Figure 5.8: McDonald’s website in Finland

5.1.2.2 Promotion of values

Würtz [20] indicated that website in collectivist cultures prefer apply images related
to family and friend time and people in good body shape, while in individualistic
cultures, people tend to use pictures represent values about freedom and private
time. It is somewhat surprising that none of the website display any image related
to collectivist values in HC cultures. Websites in mainland China, Hong Kong and
Korea mainly show food products without people. Only few images contain family
time and teamwork values are illustrated In Japanese website.

The most surprising aspect is that spending time with family and friend values
are represented on websites in Denmark (Fig 5.5), Sweden (Fig 5.7) and norway
(Fig 5.6). It can be seen that these website emphasize the happy moment with
friends, colleagues and family.

5.1.2.3 Individuals separate from or together with the product

As described in the literature, Würtz [20] believes that relationship between people
and products are highly important in HC cultures, that individuals toward and offer
the products are common in HC cultures. Meanwhile in LC cultures, products are
more like to display separated with people. As can be seen from the screenshots,
things changes in 14 years. On one hand, there are only images of products show
in all the four HC cultures’ websites (Fig 5.5, Fig 5.6, Fig 5.7, Fig 5.8). There are
some individual pictures in Japanese website, but they are not related to the primary
products. One the other hand, products in both Swedish and Finish websites dis-
played with individuals. Especially as shown in Swedish website, that pictures about
individuals toward, hold and receive products are in the most significant position

31
5. Design Process

(Fig 5.6). This finding is contrary to Würtz’s studies which have suggested.

5.1.2.4 Level of transparency

Würtz [20] believed that in LC cultures, information display on websites is clearly
and direct, easy to be found and visited, while in HC cultures, information is hide
and require users to spend extra effort to retrieve it. In contrast to Würtz’s find-
ings, however, no evidence of the same situation was detected. It is indeed that in
Scandinavian area, websites apply the same strategy to design menus and display
information. It can be seen from the screenshots that Denmark (Fig 5.5), Sweden
(Fig 5.7), Norway (Fig 5.6) and Finland (Fig 5.8) follow the same design pattern for
their menus and fat footer on the homepage, provide detailed and clear information
that users can find the content they are looking for effectively, which consistent with
that of Würtz [20]’s result. However, all the websites belong to HC cultures show
a totally opposite result that has not previously been described. Websites in China
and Japan apply a similar main menu on the top of the site, with clear categories
that users can easily find the content they want. Hong Kong and Korean websites
use a sidebar on the left of the websites with bold and clean fonts. In both HC and
LC cultures, websites use hierarchical menu and requires mouse-overs to explore
content, but there is no ambiguous content that requires users to search through
exploration in both HC and LC cultures.

5.1.2.5 Linear vs. parallel navigation on the website

Websites design in China, Japan and all the Scandinavian countries is tabular and
functional, also break the rules to that of Würtz [20] who found only HC cultures
prefer linear design. In other Asian websites, the result matches the observed in
Würtz’s study, that they use the layer-upon-layer look that the products are inte-
grate with the background as a whole, menu and other sections are cover above.

Another unanticipated finding was that all the websites, from HC to LC, open links
in the same window, contrary to Würtz’s study [20] that HC websites tend to open
up links in a new window.

Furthermore, fat footer can be found in three of four Scandinavian websites, expect
the one in Finland. But in Asia, only Japanese website apply the fat footer, none
of the other three use it.

5.1.3 Summary
Together these results provide important insights into McDonald’s website design in
2019. The design pattern changed since 2005, and it seems there are no significant
difference between these two cultures. It is difficult to list the differences between
the two cultures as Würtz [20] did in his study.

However, some other new evidence was found. For instance, we can see in LC cul-
tures, designers tend to use more dark color over a large area on the pages (Fig 5.5,

32
 5. Design Process

Fig 5.6, Fig 5.7, Fig 5.8), compare to the websites in Asia, using bright colors in a
whole (Fig 5.1, Fig 5.2, Fig 5.3, Fig 5.4). Moreover, it can be seen that the layout
design and the content is more density in HC cultures, while the design is more sim-
ple with less text in LC cultures. Three column layout design is the most common
use layout in Asian websites. In China, four column layout design is applied, and in
Japan, each column contains four small columns for more information. LC culture
are more used to simple and less columns design. Two column layout design is the
most common style in Scandinavian, although the website in Finland shows a four
columns layout, they are for less important information showed in second line, and
there are only five columns in total in this page (Fig 5.8).

The observations discussed for website in 2019 above are summarized in Figure 5.9.

parameter Tendency in HC cultures Tendency in LC cultures

Japanese website has moving

Animation No animation
products.
Have more people’s face and group of
Biggest pictures are for products
people than HC culture.
without people.
Promotion of values Social value is on the first screen in
Can see lifestyle (high quality and
Swedish website, but not in main menu
sustainable) in Chinese and Japanese
in all the other Scandinavian websites at
websites.
all.
Individuals separate
Chinese and Japanese websites have Can see lifestyle pictures in Swedish,
or together with
products with people Finland, and Norway’s websites.
product

Chinese website uses heavy text in

menu, only icons for main menu.

Japanese website is the only website

All the websites use big pictures and
doesn’t use second level menu, only
very litter text for the menu.
text
Levels of
transparency All the Scandinavian websites except
Korea and Hong Kong use a very few
Finland use special buttons for each
pictures in menu, mainly text.
column.
But for main pictures, use ambiguous
titles without special buttons for each
column in Chinese website.

layer-upon-layer design for websites

Linear versus Parallel tabular and functional
in Hong Kong and Korea
Use more bright color
Black background with few yellow and
Color White background and yellow, few
red. Few white.
black
4 columns - 5 columns in China, Most 2 columns
layout
Others use 3 columns Full-screen design

navigation No fat footer in china Fat footer is popular

links Open in the same windows Open in the same windows

multimedia No popups No popups

text More density text Less density text

Figure 5.9: Summary of cultural differences connect to UX design in McDonald’s

websites in 2019

33
5. Design Process

With this knowledge in mind, an initial solution for the study was generated. The
results observed differ from Würtz’s 2005 estimate of culture impact user’s prefer-
ences. A user survey should be launched to see whether the results match those
observed in McDonald’s websites comparison in 2019. After that, low-fidelity and
high-fidelity prototyping can be designed and evaluated separately.

5.2 User research

5.2.1 Survey
After the pre-preparation phase with the literature study had concluded, An ques-
tionnaire was sent out for the basic concept of user preferences of UX design on
websites. The data was collected from 2019-February-25th to 2019-March-3rd.

5.2.1.1 Participants
As the target user of https://support.ndcsolutions.com/hc/en-us was well-
educated, full-time employer lives in cities in China and Sweden, participants were
selected based on the requirement of the target user. Additionally, as the purpose
of the questionnaire was to generate the general idea of the study, the quantity
of the data should occur as much as possible to ensure precision and accuracy.
The questionnaire was shared to employer work in Kollmorgen, people work in the
industrial automation field and college students via the internet, and they submitted
the answers back on an anonymous basis. The quantity of participant was balanced
at 50% Chinese and 50% Swedish.

5.2.1.2 Preparation of questions

The survey was created on https://www.enalyzer.com, as a picture-based ques-
tionnaire. The goal is to understand the basic user’s preferences in China and
Sweden based on the observation of McDonald’s websites in China and Sweden, in
order to establish whether user’s UX preferences have changed and what is the latest
trends and developments in UX design field. The questions mainly focus on naviga-
tion style, levels of transparency, information density, brightness, products with or
without individuals, layout, multimedia and links. A bilingual survey was offered in
Chinese and English, in order to facilitate effective understanding by people from
China and Sweden. The questions are outlined in Appendix A.

5.2.1.3 Analysis of responses

The total number of responses for this questionnaire was 108, 72 of the subjects
completed and returned the questionnaire, in which 36 are from China and 36 are
from Sweden. For complete raw data from the 78 participants, see Appendix B .

It can be seen from the data in Figure 5.11, Figure 5.13, Figure 5.14, Figure 5.15
and Figure 5.17, participants from China and Sweden have a similar UX preference
on menu, image, video, text density and link opening, which are in accord with the

34
 5. Design Process

latest McDonald’s studies indicating that people from HC and LC cultures tend to
apply a similar UX design on websites.

There are several obvious findings to emerge from the analysis. It can be seen from
the data in Figure 5.10 that when asked whether they prefer to have text with or
without icons on the navigation bar, 90% of the Chinese respondents reported that
they like to see text with icons. Swedish respond to the same tendency but more
of them prefer text without icons. Secondly, Individuals live in Scandinavian prefer
dark color while Asian people like bright color more (Figure 5.12). In addition, on
the question of layout design preferences, half Swedish selected four columns layout
while half people like the simple layout with two to three columns, which require
a future test for confirmation (Figure 5.16). What is surprising is that although
information on Chinese websites is more density than on Swedish ones, Chinese
participants, especially people from 21 to 30 tend to like simplify layout design
rather than the complex one (Figure 5.16).

Figure 5.10: Text with or without icons

Figure 5.11: Hierarchical Menu or Flat menu

35
5. Design Process

Figure 5.12: Hue preference

Figure 5.13: Image with or without individuals

Figure 5.14: Video preference

36
 5. Design Process

Figure 5.15: Text density preference

Figure 5.16: Layout preference

Figure 5.17: Link opening preference

37
5. Design Process

5.2.1.4 Discussion

As mentioned in the literature study, Würtz [20] provides a clear view of difference
UX design for HC cultures and LC cultures in 2003. However, it seems there is no
significant UX design difference between the two cultures in 2019. Is it caused by
culture changed, or are there other reasons lead to the different? A further study
with more focus on usability testing with face to face interview based on prototyping
design, to identify whether UX preferences in HC and LC cultures trend to reveal
the convergence of views.

5.3 Concept ideation

In this phase, low-fidelity prototypes for Chinese and Swedish websites were created
separately based on the survey results above.

5.3.1 Low-fidelity prototyping

Two versions of prototypes were created according to the survey results for us-
ability testing. Two pages and a navigation bar for Chinese and Swedish were
built separately. They are referenced as prototype Chinese and prototype Swedish
in the following chapters, where they are further analyzed. The prototypes were
created with Adobe XD, pages can interact with each other, which means users
can click on a line in one page and jump to another page. All the informa-
tion displayed on the pages for both prototypes are the same, as only the design
was redesigned and tested. The original text was comes from the pages https:
//support.ndcsolutions.com/hc/en-us/.

As can be seen from Figure 5.18, the home page designed for Chinese users applies a
light grey color background and an overall bright color style. For the navigation bar,
text with icons was designed, and for the main content, bright and colorful images
were used for titles for each column. The page is a four layout design with density
content. All the design style followed the Chinese participants’ preferences received
from the survey achieved in the last chapter. The menu bar applied a hierarchical
design with a dark blue color to emphasize with content is selected. A sub menu
will display when the mouse is hover on any of the navigation text in the same dark
blue color. (Figure 5.19).

For the request a submit page, it can be seen from Figure 5.20 that it applied the
same background color, to consistent web design in every page.

These pages can interact with each other. As shown in Figure 5.21, participants can
click on the text “Products” on the menu bar to unfold the hierarchical menu and
browse a sub menu list. Furthermore, it will jump to the request submission page
when the button “submit a request” on the bottom right corner is clicked on the
home page.

38
 5. Design Process

Prototype Swedish also followed the survey results shown above. As half Swedish
participants selected density layout design, while another half prefer simply layout
design, two layouts with the same content and color style were designed for the
future usability test in order to better understand Swedish user’s layout preference.

Figure 5.18: Home page of prototype Chinese

Figure 5.19: Menu bar for Chinese website

39
5. Design Process

Figure 5.20: Request submission page for Chinese website

Figure 5.21: Interaction between Chinese prototype pages

Figure 5.22 and Figure 5.23 show the two versions home page for prototype Swedish.
It can be seen from Figure 5.22 that It is a simplistic design with a maximum of
three columns on the page. And for Figure 5.23, it is a three to four column layout
design. For both pages, the menu was designed with text without any graphic. And
the second level menu was designed with big background image with title text below

40
 5. Design Process

it (Figure 5.24). For each section on the home page, dark color style background
pictures were provided, with bold font format titles and action buttons with bright
color covey above. All the pages apply an overall dark hue color style in order to
follow the survey result for Swedish users (Figure 5.22, Figure 5.23, Figure 5.24,
Figure 5.25).

The interaction between each page follows the same pattern as the Chinese proto-
type, that once the “products” on the menu bar is clicked, the second level menu
will unfold and accessible. Furthermore, click the “submit a request” button on the
bottom right corner can visit the request submission page as well (Figure 5.26).

Figure 5.22: Home page-1 of prototype Swedish

41
5. Design Process

Figure 5.23: Home page-2 of prototype Swedish

Figure 5.24: Menu bar for prototype Swedish

42
 5. Design Process

Figure 5.25: Request submission page for prototype Swedish

Figure 5.26: Interaction between Swedish prototype pages

43
5. Design Process

5.4 Pilot testing

After the prototyping preparation, a pilot usability testing with an interview was
held. The purpose of the pilot testing was to find out problems and the uncertainty
issues about the prototype, questions or the testing process. It made sure the test-
ing process, including the time consuming, the questions and the design itself was
appropriate. With the feedback and comments received from participants, the test-
ing can be modified and improved. After evaluating the prototypes and questions,
the formal usability testing was launched in Kollmorgen office. Eight company col-
leagues were recruited as participants, conducted the testing, attended the following
interview and fill up a feedback survey at the end of the testing.

5.4.1 Participants
Two interaction design master students, who are referred to as student A and stu-
dent B in this chapter were recruited as the pilot testing participants. All the data
collected in polit test was anonymous (Figure 5.27).

National Age group Education background

Student A Sweden 20-25 Interaction design master student

Student B Sweden 26-30 Interaction design master student

Figure 5.27: Information of pilot testing participants

5.4.2 Prepare tasks and questions

Student A was required to accomplish the tasks on prototype Swedish and later
student B performed the same tasks on prototype Chinese. After they finished the
tasks, they were asked several questions based on their behavior during the testing
process. They were also expected to fill out a Likert-type scales [37] survey after
the interview.

The first task was to find “Design Tools” on the home page. Participants were told
that “Design Tools” was on the navigation bar under “Products”. The second task
was to submit a request on the home page. The purpose of the two tasks was to
observe whether the UX design of the prototypes was in a natural, understandable
and logical order and whether people can accomplish the tasks in a minimize and
efficient way.

Below is a summary of the questions in the survey:

44
 5. Design Process

Do you think the web pages I don’t Strongly

Strongly disagree Disagree Agree
look good over all? know agree

Was it difficult to achieve the

given tasks?
Does the website easy to
navigate?
Were you confused while
performing the tasks?
Are there any improvements
that you feel we could make
to the website?

Figure 5.28: Questions for the pilot testing

5.4.3 Feedback and responses

Below is the survey answers received from student A and student B.

Student A - answer basis on Prototype Swedish

Do you think the web pages look good

Agree
over all?

Was it difficult to achieve the given

Agree
tasks?

Does the website easy to navigate? I don’t know

Were you confused while performing

Agree
the tasks?

Are there any improvements that you

No comments
feel we could make to the website?

Student B - answer basis on Prototype Chinese

Do you think the web pages look good

Strongly disagree
over all?
Was it difficult to achieve the given
I don’t know
tasks?

Does the website easy to navigate? Strongly disagree

Were you confused while performing

Agree
the tasks?

The overall design looks ugly, boring and dull. I do not like
Are there any improvements that you
everything about the UI at all. It looks more like a platform rather
feel we could make to the website?
than a website. I do not feel satisfied when I visit this website.

Figure 5.29: Answers received from participants

45
5. Design Process

5.4.4 Result analysis and evaluation

Student A suggested that the task requirements can be more clear to avoid confu-
sion. As student A thought the submit request task followed the first task, that
participant should clicked in some place in the navigation bar to another page to
achieve the task. Student A was confused that spent time searching on navigation
bar but found nothing. In the final test, the question was modified to be more clear
that participants can search in the whole page to accomplish the task.

The most surprising aspect of the testing results was the feedback received from
student B. As the UI aesthetics did not meet his expectations, he left sharp criti-
cisms only relative the interface design. Although he was reminded that aesthetic
was not included in the testing, he argued that the UI design should be modified to
meet the user’s expectations, otherwise user would feel tired of the UI and do not
have the patience to visit it in a long term. Prototype Swedish was displayed to
him after the testing because it was difficult to receive any useful feedback based on
the Chinese version. Student B strongly agreed that prototype Swedish looks much
more comfortable and acceptable, and he also had the motivation to discuss the UX
part and continued the interview. This experience sent a message that aesthetic
matters a lot in website design. Users look forward a website that at least has an
average mass aesthetic appreciation, otherwise they may refuse to visit it even if the
UX part is acceptable.

Based on the feedback and comments received from the two participants, it is criti-
cal to reduce possible prejudice and errors caused by aesthetics related issues, make
sure participants could focus on the UX testing rather than distract by the UI de-
sign. Moreover, it is also important to take account of quantitative results in the
formal usability testing. It was noticed that it was difficult to measure the proto-
type Chinese and Swedish when they applied different pictures and content. Thus,
prototype Chinese and Swedish were modified to balance the consistency in both UI
design and content.

The new prototyping design can be divided into three groups: the overall light or
dark tone comparison group (Figure 5.30, Figure 5.31), navigation text with or with-
out icons comparison group (Figure 5.32, Figure 5.33, Figure 5.34, Figure 5.35) and
layout comparison group (Figure 5.36, Figure 5.37).

Each group contains two prototypes, and participants were supposed to figure out
the one they like more. As can be seen from the prototyping listed below, images and
content applied on every page are exactly the same in all the comparison groups.
In the overall light and dark tone comparison group, same images with different
tones are applied. For layout comparison group, same images and text are used
but the position of sections are changed. In navigation text with or without icons
comparison group, images and text keep the same size, shape, tone and position.
Applying the same pictures and content can reduce the preferences due to the quality
of the pictures and content, make sure participants can focus on the UX design and
receive accurate quantitative testing results.

46
 5. Design Process

Figure 5.30: Dark tone design

Figure 5.31: Light tone design

47
5. Design Process

Figure 5.32: Text with icons in light tone design

Figure 5.33: Text without icons in light tone design

48
 5. Design Process

Figure 5.34: Text with icons in dark tone design

Figure 5.35: Text without icons in dark tone design

49
5. Design Process

Figure 5.36: Three to four column layout in dark tone design

Figure 5.37: One to three column layout in dark tone design

50
 5. Design Process

5.5 Usability testing

Usability testing was conducted on the high-fidelity prototype showed above. Par-
ticipants were required to answer a set of questions and then conduct two tasks on
the prototype. Additionally, they need to fill out a questionnaire at the end of the
testing.

5.5.1 Participants recruitment

A usability testing invite email was sent in Kollmorgen AB to find participants. A
total of 18 people was selected to be part of the test. Nine of them are Chinese based
on Chinese cities and another nine people come from Sweden live in Goteborg. The
test was conducted anonymously. Table 5.38 below presents the summary statistics
for participants.

Nation Age group Education background

China 31-35 Undergraduate
China 25-30 Postgraduate
Sweden 46-50 Postgraduate
Sweden 41-45 Undergraduate
China 36-40 Undergraduate
Sweden 41-45 Undergraduate
India 25-30 Postgraduate
China 31-35 Postgraduate
China 25-30 Undergraduate
Sweden 25-30 Postgraduate
Sweden 31-35 Postgraduate
China 36-40 Undergraduate
China 36-40 Undergraduate
Sweden 50+ Undergraduate
Sweden 36-40 Postgraduate
China 31-35 Undergraduate
China 31-35 Undergraduate
Sweden 31-35 Undergraduate

Figure 5.38: Participants background in usability testing

51
5. Design Process

5.5.2 Test procedure and question preparation

Firstly, participants were told about the purpose and the general process of the test-
ing. They also were informed that the whole testing was anonymous. Participants
were asked which tone for the prototype they prefer and finished the rest of the
testing with the prototype with the same tone style. They were asked to choose the
design they prefer in each comparison group and were expected to answer questions
related to their selection. The purpose of this step was to achieve the testing on the
pages they feel satisfied, in order to get accurate answers.

Below illustrates the questions asked during the interview:

Which design you prefer in the light or dark tone comparison group and why?
Which design you prefer in the text with or without icons comparison group and why?
(If the participants select the dark ton style design) Which design you prefer in the layout comparison
group and why?
How do you open a link on a website?
Do you have any comments about the UX design on the prototype?

Figure 5.39: Questions asked to participants in usability testing

5.5.3 Test results

General background information of the participants are shown in Figure 5.40, Fig-
ure 5.41, and Figure 5.42. In total, 18 responses, 9 from China and 9 from Sweden
finished the testing. 33.33% of the responses age from 31 to 35, next was age from
25 to 30 and 36 to 40, 22.22% separately. All of them earn a college degree or above.
Further data presented in this section, will have been cross-referenced between dif-
ferent data points derived from the raw data of the survey.

Figure 5.40: The education background of the participants

52
 5. Design Process

Figure 5.41: The education background of the participants

Figure 5.42: The education background of the participants

Unstructured interview hold during the test aimed to find out participants’ prefer-
ence on color tone, navigation icons, layout design and tried to explore the reason
behind their decisions. Figure 5.43, Figure 5.44, Figure 5.45, and Figure 5.46 below
display the results of the interview.

It can be seen that the usability testing results show similar trends as the survey
results displayed. In response to color style preference, six of eight Chinese par-

53
5. Design Process

ticipants (75%) tend to prefer bright tone color style, while six of eight Swedish
participants (75%) considered that dark tone color style looks more comfortable
(Figure 5.43).

Figure 5.43: Color tone preference

However, in the text with or without icons testing section, no significant difference
between the two user groups was evident. Both Chinese and Swedish showed am-
biguous results (Figure 5.44). Three participants in China (37.5%) indicated that
text with icons look better, while another three participants (37.5%) did not care
about whether there were icons with text or not. Swedish participants made similar
decisions, that three people prefer to have text with icons and two does not want
text with icons, the rest of the participants did not care whether there are icons or
not.

The next question aimed to figure out respondents preference for layout design style.
Figure 5.45 presents the results obtained from usability testing. Chinese participants
showed an opposite layout design preference as they did on survey testing. A ma-
jority of participants (62.5%) from China indicated that they would prefer to visit
the four column layout design prototyping rather than the two to three column
prototyping. However, Swedish participants represent the same preference as they
showed in the survey. Half of them indicated that four column design is better while
another half said they prefer two to three column design.

The last question required respondents to give information on how would they open
links when they browser the prototyping. The results as shown in Figure 5.46 in-
dicate that Chinese participants follow the same result pattern as they said in the
survey, that they prefer to open links in a new window. However, the majority of
Swedish respondents (87.5%) who responded to this item felt that links should be

54
 5. Design Process

opened in the same windows rather than in the new windows, totally different from
the previous survey results.

During the whole process, guided questions were asked to encourage participants
to give critical comments and constructive feedback on UX elements including lay-
out, navigation, links, aesthetics, content and other web design related information
participants considered are useful for the website.

Figure 5.44: Text with/without icons preference

Figure 5.45: Layout style preference

55
5. Design Process

Figure 5.46: Link opening preference

5.5.4 Analyzing responses

5.5.4.1 Color tone
The color preference for Chinese users is in the opposite way as the survey result,
that most of the respondents tend to select a bright color style rather than dark style.
One Chinese participant commented: ‘Bright color makes me feel comfortable and
warm’. Other participants who prefer a bright color gave a similar answer. However,
when talking about the dark tone color preference, a Swedish interviewee said: ‘Dark
color makes me feel comfortable.’ It seems that people live in different regions have
a different definition of comfortable feelings. Additionally, what is striking about
the result is Chinese participants age among 25-35 have the tendency to prefer dark
tone color.

5.5.4.2 Text with/without icons

In both cultures, for respondents who left negative comments about the icons, they
believed that those icons do not relate to the text, so it is unnecessary to add icons
with the text as they were used for decorative purposes only but lack of actual
meaning. Meanwhile, participants who did not care about the icons develop the
same attitude toward people who dislike icons, that they did not believe icons would
matter the text or not as they did not have actual meaning. In conclusion, users
prefer a function-oriented UX design rather than aesthetics-oriented pages.

5.5.4.3 Layout design

When asked about the reason to select four column layout design, Chinese partici-
pants were unanimous in the view that all the information was in one screen in the

56
 5. Design Process

four column design, which was effective to find the information they want without
scroll the mouse. Swedish participants who selected four column design hold the
same feeling as the majority Chinese testers that it gave an overview of all informa-
tion on one page which was easy to find things they were looking for. As one Swedish
interviewee put it: ‘I personally prefer the two to three column design, but consider
this is a website for quick search information at work, the four column layout design
concerted all the information in one screen that I can find information quickly. ’
Another half participants considered that two to three column design made pages
look simple and emphasized the important part. For these participants, aesthetics
is more important than functions. In addition, there is a clear trend of interests in
simple layout design between Chinese participants from 25-35. By contrast, elder
Swedish above 45 years old tend to accept density layout design rather than younger
users.

5.5.4.4 Link opening

The majority of Chinese participants agreed with the statement that links should
be opened in new windows automatically, which means links should be designed to
be opened in new windows. Interestingly, all the Swedish participants argued that
links should be opened in the same windows automatically, but they would open a
link in a new window by right click the mouse, or use shortcut on the keyboard if it
is necessary.

5.5.5 Evaluation
After the interview, a survey (See Appendix C) was conducted to examine whether
the prototyping conforms to the usage habits of both Swedish and Chinese users.
The table below shows an overview of the survey result. For complete raw data from
the 18 participants, see Appendix D.

Strongly I don’t Strongly

Questions Agree Disagree
agree know disagree
Overall I think the web pages look good 5 13
It was difficult to finish the given tasks 1 10 7
The website appears easy to navigate 6 12
I was confused while performing the
1 8 9
tasks
The information displayed on the
3 10 5
website is valuable

Figure 5.47: Usability testing survey results

It can be seen from the data in Figure 5.47 that all of the participants gave a
positive evaluation of the prototyping. Of the 18 participants who completed the
survey, all of them agreed or strongly agreed that the pages are aesthetic looking.

57
5. Design Process

When asked whether it was difficult to finish the tasks, only one respondent felt
“I do not know”, all the other respondents reported that there was no difficult to
achieve the given tasks. When the participants were asked whether the website
appears easy to navigate, all of them agreed or strongly agreed that it was easy. In
response to whether they were confused when performing the tasks, most of those
surveyed indicated that they were not confused at all. Lastly, Almost one-third
of the participants (28%) said that they did not know whether the information
displayed on the website is valuable. As one interviewee put it: “The tasks focused
on the UX rather than the content on the pages, I barely paid attention to the
information displayed on the pages, so it is hard for me to say if the information is
valuable or not.”

58
 6
Results

This chapter represents the final outcomes of the thesis. First, the redesigned web-
sites for Swedish and Chinese users are displayed separately, with a description of
the redesign features in each page. The redesigned website mainly includes five
parts for each country: 1) Login page, 2) Home page, 3) Knowledge base page, 4)
Knowledge based detail page, 5) Submit request page. Additionally, a summarized
set of UX design guidelines, including orientation, navigation menu, layout, color,
links, text and other features are presented for future development and extensions.

6.1 Redesign websites

All the redesign pages use the same content as on the original pages, some of the
information follow the old position. For instance, the logo on the top left of the
page, the search bar on the top middle of the page and user’s name on the top
right of the page, which are at the same position as on the original pages but in
an aesthetic way. For all the Chinese pages, the slim footer is applied while the fat
footer is applied for Swedish pages. And on the Chinese version website, links are
open in new windows as default. In contrast, Swedish version website opens links
in the same windows.

6.1.1 Login page

Information on the old sign in/login page is fragmented, all the text and buttons
display on the page without emphasis, that users need to search around the page to
find the information they need for the next step (Figure 6.1).

Based on the UX design trends, Chinese users prefer an overall bright and colorful
style with a light color background on web page design. The input box and activity
buttons are listed on the right sidebar, together with other sign-in options (Fig-
ure 6.2). Therefore, all the essential information are concentrated on the sidebar so
users can find what they need in an effective day.

The Swedish version of sign in/login page follows the test principles, that applies
a dark ton on the background and the overall dark color style on the page. The
login/sign in box and the content display on the box apply the same as they displayed
on the Chinese version, with different button color on the consistency purpose. Users
could easily find what they are looking for as all the important information is con-

59
6. Results

centrated in a small space. Instead of display all the content equally, email address
and password input boxes are emphasized in the redesign page, meanwhile, sign in
from the third-party services are represented with service icons, which is easy to be
understood as well as keep the page looks clean and simple (Figure 6.3).

Figure 6.1: Old login page

Figure 6.2: New login page for Chinese user

60
 6. Results

Figure 6.3: New login page for Swedish user

6.1.2 Home page

Generally, the old homepage has an unstructured navigation bar with unclear con-
tent (Figure 6.4). The layout of the page is not well thought through. Users may
get lost or feel confused when they first visit this page, and they may spend a huge
amount of time to be familiar with the information display on the page. The space
on the home page was not fully used. A banner without any text display on the top
of the page which took big space but did not represent any information. “Recent ac-
tivities” column occupy most of the layout which was not the most important text
users are interested in. In addition, the column “Promoted article” was in black
background with orange text, which is not easy to read.

The homepage was redesigned based on the testing and interview results. In both
versions, the most commonly clicked columns, knowledge base, smart warehouse,
request, recent activity, news and download are displayed in the home page accord-
ing to the feedback from participants.

The Chinese page follows the same design pattern as the sign in/login page, that
apply an overall bright/light color style on the page. Hierarchical menu with only
text is designed for the navigation bar, as according to the feedback from the testing,
the meaning of the text on the menu bar is difficult to represent by icons, and it
is not necessary to put icons there only for aesthetic purposes. Only applying text
keeps the menu looks simple as well as easy to find information. For the layout, the
Chinese homepage applies a three column layout design, all the content is concen-
trated in one screen that users can browse everything without scrolling the page by
mouse.

61
6. Results

Figure 6.4: Old home page

Figure 6.5: Home page for Chinese user

62
 6. Results

Figure 6.6: Home page for Swedish user V1

Meanwhile, as the testing result shows that half of the Swedish participants prefer
density layout design, another half prefer simple layout design, the Swedish home-
page has two versions for future tests. Figure 5.30 shows the one to two layout
design version with an overall dark tone color style. Figure 6.7 presents the same
column layout design Swedish version as the Chinese homepage, as a functional ori-
ented design. Both designs apply buttons with conspicuous orange color, that are
visible and easy to click on dark tone background.

63
6. Results

Figure 6.7: Home page for Swedish user V2

6.1.3 Knowledge base page

As shown in Figure 6.8, the old knowledge base page lists all the information in a
two-column layout without UX design and aesthetics consideration. Users may feel
bored to visit this page and spend a long time scrolling up and down for information
they are looking for.

Figure 6.9 shows the redesign knowledge base page for the Chinese market, that
provides a consistent color style and navigation menu as other Chinese web pages
mentioned above. Following the design principles, a four column layout design was
applied on the page for a density layout feeling. The page is short and concentrated
to meet Chinese user’s preferences. Small icons are applied for each column, rep-
resented the function of this column, and also provide a lively atmosphere for the
page. Each column contains a three lines list, with an arrow icon on the left side in
each line. “Read more” buttons in white background with blue border are displayed
in the columns which contains more than three lines.

64
 6. Results

Figure 6.8: Old knowledge base page

65
6. Results

Figure 6.9: Knowledge base page for Chinese users

The Swedish redesign knowledge base page follows the Swedish design pattern, that
applies a dark background color, white text color, with a minimalist layout design
that only has two columns in each row but has a longer length page (Figure 6.10).
Compared with the old design, this page is more organized, clear, and all the columns
are concentrated in the middle of the page for an aesthetics consideration. Same as
the Chinese page, small icons and arrows are applied to each column and line, so
that users do not feel bored when they visit the page.

66
 6. Results

Figure 6.10: Knowledge base page for Swedish users

67
6. Results

6.1.4 Knowledge based detail page

It can be seen from the Figure 6.11 that the old knowledge based detail page has
a sidebar on the left, “follow”, “previous article” and “next article” buttons on the
far right side of the page, which makes the page looks less structured and it may
take users longer time to search on the page for their needs. Moreover, it is not a
UX design with consideration to reduce the time cost when users visit the page, as
they have to move the mouse from one side to another side for clicking.

Figure 6.11: Old knowledge based detail page

68
 6. Results

The redesign pages, by contrast, are more compact so users can find information
in an effective way. The Chinese page follow the design discipline as other Chinese
pages, that has an overall bright looking. A sidebar with light grey background in
a light grey outline is on the right side, which is obvious and well-structured. The
background color of the sidebar would change to white when it is selected, also a
small arrow will display on the left side in the selected line, so users know which
page they are visiting. Small icons “Print”, “PDF” and “Email” which under the
main article title was redesigned. Text followed the icons was removed as these the
function of them are easy to be recognized. “Share” icon is added in the line, instead
of list at the end of the article for a compact structure (Figure 6.12).

The Swedish version of knowledge based detail page follows the consistency design
pattern that is in a dark background color with bright color text. Other UX at-
tributes and the layout are the same as the Chinese page (Figure 6.13).

Figure 6.12: Knowledge based detail page for Chinese

69
6. Results

Figure 6.13: Knowledge based detail page for Swedish

6.1.5 Submit request page

As shown in Figure 6.14, the user-friendly attributes were not considered for the
page design. The main information is aligned left on the old page, left a big blank
space on the right side, which makes the page looks imbalanced and unstructured.
Navigation bar split into two lines in this page, the one underlying was considered
as a sub menu. However, as this feature does not exist in other pages, users may
get confused when they visit this page, as it is difficult for them to realize that the

70
 6. Results

sub menu is hiding in navigation bar. So in the main space, only the submit request
section shown, but other options such as “my request ”, “Request I’m CC’d on”
and “Organization requests” does not display. It may take users a long time to find
these sections when they visit this page if they are looking for the request results or
other information.

The redesign pages for Chinese (Figure 6.15) follows the same design pattern as
the other Chinese pages that is on a bright style. All the information on this page
is concentrated in the middle of the page for a balanced and structured looking.
Instead of listing all the options in the navigation bar, a right sidebar is presented,
follow the same design discipline as the knowledge based detail page, which contains
all the support options that users can select effectively. The attachments field is
extended, compared with the old version, which allows users to drop a document
into the box from their computer directly. A blue button is applied for submit the
request, which is outstanding in a light background page. All the modification is on
a user-friendly purpose.

Figure 6.14: Old submit request page

71
6. Results

Figure 6.15: Submit request page for Chinese users

The Swedish version redesign page follows the same design pattern as the other swei-
dsh version pages, which has a dark background color and white text. The layout
style and content is the same as Chinese version. An orange color button instead
of a blue button is designed for submit the request, which is more conspicuous on a
dark tone background (Figure 6.16).

6.2 Design guidelines

The UX design guideline are the summarized results based on the three test results
performed above. It can be considered as suggestions or advice to design Chinese
and Swedish websites. Further and iteration work are required as cultures are devel-
oping in both societies so people’s preferences on websites are changing over time.
Moreover, this guideline should be used combined with general UX design guidelines
as it was mentioned above that aesthetics is important in UX design.

72
 6. Results

Figure 6.16: Submit request page for Swedish users

6.2.1 Design guidelines for Chinese

In developing UX design for China, the following guidelines should be considered:

6.2.1.1 Orientation
The design should be functionally oriented, which means the priority of the UX
design is user-friendly and can help users to achieve their goals in an effective way.
On this premise, other guidelines described below can be broken to meet user’s
needs.

73
6. Results

6.2.1.2 Main navigation menu

Websites contain hidden menus which require mouse-overs to explore the content.
This may lead by the demand of an increasing amount of information displayed on
the website. The design should be goal-oriented, with clear, structured but few texts
in connection with the content to make sure they are easy to be understood and to
be found effectively (Figure 6.17).

Submenu follows the same design pattern, applies simply text that is easy to be
selected. Icons and images can be used to support the text, but should with the
purpose to support or represent the text rather than redundant (Figure 6.18).

Figure 6.17: Main navigation menu

Figure 6.18: Submenu

6.2.1.3 Layout
Tabular and compact layouts are commonly applied in Chinese websites. Four to
five columns layout design would meet Chinese user’s needs, as they prefer to search
for information among density content.

6.2.1.4 Color
The overall color style is tend to be light, bright and vivid. White and other light
colors should cover the large area while mages and icons tend to apply bright and
vivid color. The limitation of the this guideline will be presented on the discussion
section below.

6.2.1.5 Links
For Chinese website, links should be opened in new windows or tabs automatically,
which in line with user habits and user satisfaction.

74
 6. Results

6.2.1.6 Text
High-density information in a small space is accepted by Chinese user, which is
consistent with the layout design guideline.

6.2.1.7 Other
Young Chinese users, age among 21-30 have the tendency to accept western UX
design style.

6.2.2 Design guidelines for Swedish

For Swedish website design, the following guidelines apply:

6.2.2.1 Orientation
Design is functionally oriented, the same as for UX design in China. On this premise,
other guidelines described below can be broken to meet user’s needs.

6.2.2.2 Main navigation menu

Main navigation menu follow the same guideline as to the Chinese UX design de-
scribed above. The simple text should be used for both navigation bar and submenu
design. Meaningful icons or images can be used as a bonus.

6.2.2.3 Layout
The minimalist layout design is commonly used for Swedish UX design. Layout
less than four columns are accepted by Swedish users. However, the density layout
design is also acceptable if users find it is high efficiency.

6.2.2.4 Color
Swedish users prefer dark tone color style for the whole website, with text and
buttons with bright and vivid colors cover above.

6.2.2.5 Links
The default design of Link opening should in the same window as the majority of
Swedish users rather decide to open a link in the same or new window by mouse or
shortcut keys by themselves.

6.2.2.6 Text
Swedish users feel satisfied with low content density on the web pages.

75
6. Results

76
 7
Discussion

The discussion chapter is divided into a discussion of the findings and a discussion of
the results of this study. In the discussion of the findings, interpretations of findings
and alternative approaches will be discussed. The result discussion will bring up the
limitation of this study.

7.1 Result discussion

Prior studies that have noted the importance of divided UX design into two cate-
gories for HC and LC cultures, as they emphasized that cross-culture UX design is
strongly related with cultural dimensions based on Hofstede’s six cultural dimen-
sions theory and Hall’s HC and LC culture theory. In contrast to earlier findings,
however, no evidence of such a strong tendency was detected in this thesis.

For instance, Würtz [20] believes that website design in HC culture background
applies a large number of animations, images that imply collectivist societies, hier-
archical menu, many sidebars and open links in new windows. However, this cannot
be detected in Mcdonald’s Asian websites now, where are considered as HL culture
regions, and also cannot be found in subsequent usabilities testing and interview.
Websites in both cultures are goal-oriented. Hierarchical menus are commonly used
in both cultures, and images promote individualistic lifestyle can be found in both
cultures.

This result may be explained by the fact that firstly, technology is developing fast
over the past 30 years, and the web design trend is changing over time. For instance,
animations are not popular in website development all over the world, which may be
caused by the fall of Adobe Flash and the rise of HTML5, rather than the cultural
factors.

Secondly, it seems possible that these results are due to the impact of globalization.
Internet makes the bridges of communication and cultural exchange between dif-
ferent societies, as people are receiving the same information all over the world by
internet, they tend to accept different cultures, and generate a similar preference on
UX website design.

Age is an essential factor that cannot be ignored. It can be seen from the tests
and interview results that Chinese young users, age below 35 years old, are more

77
7. Discussion

inclined to accept western UX design style, compared with older users. Meanwhile,
Swedish elder users, age above 40 years old, hold the opposite view with younger
Swedish users. All of these phenomena indicate that as culture is not static but
changing by time, younger users are easier to accept new technologies and future
movement than elder users. For these users, traditional cultures have little influence
on them. By contrast, older users tend to be more stubborn and less receptive to
new cultures and technologies, which lead to a more culture-oriented UX preference.

Finally, the types of websites determine the UX design strategy. Pilot research of
cross-culture UX design is more focused on analyzing government and university
websites [28], which reflect the hierarchy in the societies and the strong influence of
traditional cultures. This study mainly focuses on analysis business, consumption
and product websites, which are less influenced by local cultures but more emphasis
on individualistic, personality and lifestyles. According to Hoffman’s culture mode
[30], these features belong to humanities that all human beings yearn for them. This
may be a possible explanation why cultures play a less important role in UX website
design in this study.

7.2 Limitations
7.2.1 Theory
Although Hofsted’s theory is generally accepted, it has been extensively criticized.
The culture model was first launched in the 1980s, later modified and extended in
2000. The data it applied might be already out-of-date. As we mentioned below,
culture is changing with time, whether this result meets nowadays‘ rapid growth of
the internet industry is questionable. Individualism parameters can increase with
economic growth [11], and China has been sustaining the world’s highest rates of
economic growth for the last 30 years [12]. Does the high speed of economic growth
change Chinese user’s preferences and behaviors over time?

Furthermore, Hofsted’s theory assumes only one culture for each country, which has
its major limitation. People immigrate from one region to another around Scandina-
vian and other countries all over the world. People from multicultural backgrounds
use to speak native languages at home and be influenced by their home country’s
cultures. Hofsted’s theory applied to Swedish culture dimension is questioned by
this thesis. Several user tests were adopted later to see if the theoretical model
match reality.

Hall’s culture model has the same shortage as Hofsted’s theory, that it was applied
30 years ago. Have cultures been changed during this period or people keep the
same usability and behaviors is the question we explored in this thesis.

Unlike six cultural dimensions and low-context and high-context culture model,
which only describe the culture of one society on a national level, and assume peo-
ple who live in the same society have exactly the same culture concept. Both

78
 7. Discussion

Holliday and Hoffman highlight the mobility and complexity of cultures and the
way of intercultural communication individuals would apply. Factors from small to
big, inside-out and outside-in impact people’s culture concept, and the complexity
of culture implies it is difficult to define only one culture for a given society.

7.2.2 Types of Websites

These findings may be somewhat limited by the specific types of websites. As men-
tioned above, this study is about UX design preferences for business, consumption
and product websites in different cultures. As websites for consumption tend to
invoke user’s desire for a better lifestyle on a promotion purpose, the UX style can
be similar all over the world. However, for other types of website, for instance,
government, authorities, universities, etc., the results might be different. It is possi-
ble to hypothesize that power distance, collectivism and individualism, HC and LC
cultures may play important roles in UX preferences for these kinds of websites.

7.2.3 Target user groups

A note of caution is due here since the Chinese participants in this study are well
educated middle class who live in big cities in China, which cannot represent the
Chinese market. China is a huge market with an expanding gap between rich and
poor. China’s Gini Coefficient is 0.465, ranking top most unequal countries [38].
Moreover, household income in Chinese urban areas is 2.72 times that of rural
areas [39], which means well educated Chinese users who live in big cities are more
likely to receive, accept and can afford western cultures, agree that personalities
and individualism are crucial, while users who stay in rural areas may have fewer
opportunities to get in touch with new things, or the environment around them holds
the opinion against accepting new things. They may have different preferences on
UX website design and that segment of users require future research.

7.2.4 UX attributes
Only a few UX attributes were tested in the usability testing and interview. Mul-
timedia, promotion of values, animations were not included in the study because of
the lack of resources on the redesigned website. As the study can only apply image
specified relative to the content, it was difficult to test these attributes with this
website.

7.2.5 Sample size

There were only 16 participants attend the usability testing. With a small sample
size, caution must be applied, as the findings might not represent the real situation.

79
7. Discussion

7.3 Future work

To develop a full picture of the UX preferences in different cultures, additional stud-
ies will be needed. First, other types of website can be tested to contribute to the
results. Users may represent different UX preferences for different types of websites.
It is not surprising that people consider personal consumption website should be
leisure-centered and highlight personality, while government or authority websites
should be formal and represent the nation.

Secondly, the Chinese user group should be extended. Individuals who live in less
developed areas, and with low levels of education may hold a different UX prefer-
ences on website design.

More UX attributes, including multimedia, and promotion of values, animations

should be tested for the whole picture. Moreover, the size of the test sample should
be increased for a more accurate result.

For Swedish users, A/B test for the redesigned layout is necessary for better under-
standing of their preferences for layout design.

80
 8
Conclusion

This study was undertaken to redesign a website with the UX attributes to meet
the needs of Chinese and Swedish users, and evaluate how different cultural factors
influence user’s preferences for UX attributes between Chinese and Swedish websites.

One of the more significant findings to emerge from this study is that culture is not
stagnant but keeping change in society, also the same as the user’s UX preferences
on website design. With the development of web technologies, individuals all over
the world live in an era of multiple cultures, are impacted and constantly change
their UX preferences. Therefore, Hofstede’s and Hall’s theories do not apply to the
UX website trend at this stage. New theories should be included for better under-
standing of user’s UX preference nowadays.

Hoffman (2018) suggests that human nature and personalities should be taken into
account when considering the influence of culture, as they can also be big factors on
UX website design. This study has identified it. Users in both culture background
are function-oriented when they were required to operate tasks in short time. And
for some questions, users’ personalities overcome the impact of cultures in both user
groups.

The results of this study indicate that young middle-class Chinese users tend toward
Swedish users’ UX preferences, as they have a big opportunity to receive and accept
foreign cultures from the internet. However, limitations need to be considered as
China has a big population and a widely distributed social structural levels. Different
UX strategies may need to apply to different target user groups.

81
8. Conclusion

82
 Bibliography

[1] Reinecke, K. and Bernstein, A. (2011). Improving performance, perceived us-

ability, and aesthetics with culturally adaptive user interfaces. ACM Transac-
tions on Computer-Human Interaction, 18(2), pp.1-29.
[2] Norman, D. (2005). Human-centered design considered harmful. interactions,
12(4), p.14.
[3] Keegan, W. J., Green, M. C. (2011). Global marketing: 2011 custom edition
(6th ed.). Upper Saddle River, NJ: Prentice Hall / Pearson.
[4] Levitt, T. (1983), “The globalisation of markets”, Harvard Business Review,
May/June.
[5] Hiller, M. (2003). The role of cultural context in multilingual Website usability.
Electronic Commerce Research and Applications, 2(1), 2–14.
[6] Cia.gov. (2019). The World Factbook — Central Intelligence Agency.
[online] Available at: https://www.cia.gov/library/publications/the-world-
factbook/rankorder/2153rank.htmlin [Accessed 7 Feb. 2019].
[7] Nielsen Norman Group. (2019). Middle-Aged Users’ Declining Web Perfor-
mance. [online] Available at: https://www.nngroup.com/articles/middle-aged-
web-users/ [Accessed 7 Feb. 2019].
[8] Hoft, N. (1999). “Global Issues, Local Concerns.” Technical Communication 46
(2): 145–148.
[9] Hofstede, G., G. J. Hofstede, and Michael. Minkov. (2010). Cultures and Orga-
nizations: Software of the Mind. 3rd ed. New York: McGraw-Hill Education.
[10] Hofstede Insights. (2019). Country Comparison - Hofstede Insights.
[online] Available at: https://www.hofstede-insights.com/country-
comparison/china,sweden/ [Accessed 7 Feb. 2019].
[11] Hall, E. T., Hall, M. R. (1990). Understanding cultural differences. Yarmouth,
ME: Intercultural Press Inc.
[12] Imf.org. (2019). Report for Selected Countries and Subjects.
[13] International Organization for Standardization (2009). Ergonomics of human
system interaction - Part 210: Human-centered design for interactive systems
(formerly known as 13407). ISO F±DIS 9241-210:2009.
[14] Nielsen Norman Group. (2019). The Definition of User Experience (UX).
[online] Available at: https://www.nngroup.com/articles/definition-user-
experience/ [Accessed 8 Feb. 2019].
[15] Giacomin, J. (2014). What Is Human Centred Design?. The Design Journal,
17(4), pp.606-623. Wildemuth, B. M. (2003). Why conduct user studies? The
role of empirical evidence in improving the practice of librarianship. In INFO-
RUM (Vol. 2003, p. 9th).

83
Bibliography

[16] Kralisch, A., M. Eisend, and B. Berendt. 2005. “The Impact of Culture on
Website Navigation Behaviour.” In 11th International Conference on Human-
Computer Interaction, Las Vegas, NV, USA.
[17] Alexander, R., Thompson, N. and Murray, D. (2016). Towards cultural transla-
tion of websites: a large-scale study of Australian, Chinese, and Saudi Arabian
design preferences. Behaviour Information Technology, 36(4), pp.351-363.
[18] Choi, B., Lee, I. and Kim, J. (2006). Culturability in Mobile Data Services:
A Qualitative Study of the Relationship Between Cultural Characteristics and
User-Experience Attributes. International Journal of Human-Computer Inter-
action, 20(3), pp.171-203.
[19] Techterms.com. (2019). Link Definition. [online] Available at:
https://techterms.com/definition/link [Accessed 8 Feb. 2019].
[20] Wurtz, E. (2005). Intercultural Communication on Web sites: A Cross-Cultural
Analysis of Web sites from High-Context Cultures and Low-Context Cultures.
Journal of Computer-Mediated Communication, 11(1), pp.274-299.
[21] De Bortoli, M., Maroto, J. (2001). Colours across cultures: Translating colours
in interactive marketing communications. European Languages and the Imple-
mentation of Communication and Information Technologies.
[22] Chu, S. (2016). Design Factors Affect User Experience for Different Cultural
Populations. Journal of Educational Issues, 2(2), p.307.
[23] Gaver, W. (2012). What should we expect from research through design?. Pro-
ceedings of the 2012 ACM annual conference on Human Factors in Computing
Systems - CHI ’12.
[24] Martin, B. and Hanington, B. (2012). Universal methods of design. Beverly,
MA: Rockport Publishers.
[25] Hartson, H. and Pyla, P. (2012). The UX Book. Amsterdam [etc.]: Elsevier.
[26] Friedman, B. and Kahn, P.H. (2003) Human values, ethics, and design. Hand-
book on Human- Computer Interaction, J. Jacko and A. Sears Eds., Lawrence
Erlbaum Associates Mahwah, NJ, 1177-1201.
[27] En.wikipedia.org. (2018). Facebook–Cambridge Analytica data scandal.
[Online] Available at: https://en.wikipedia.org/wiki/Facebook%E2%80%
93Cambridge_Analytica_data_scandal
[28] Gould, E., Zalcaria, N. and Yusof, S. (2000). Applying culture to Web site de-
sign: a comparison of Malaysian and US Web sites. 18th Annual Conference on
Computer Documentation. ipcc sigdoc 2000. Technology and Teamwork. Pro-
ceedings. IEEE Professional Communication Society International Professional
Communication Conference and ACM Special Interest Group on Documenta-
tion Conferenc.
[29] Holliday, A. (2013). Understanding intercultural communication: Negotiating
a grammar of culture. Routledge.
[30] Hoffman, E., Verdooren, A. (2018). Diversity competence. Bussum: Uitgeverij
Coutinho.
[31] James, P. (2006). Globalism, nationalism, tribalism. London: Sage.
[32] Kraidy, M. Hybridity, or the Cultural Logic of Globalization.
[33] McLuhan, H. (1962). The Gutenberg galaxy. New Amer. Lib.

84
 Bibliography

[34] McDonald’s - Official Global Corporate Website. (2019). Retrieved from

https://corporate.mcdonalds.com/corpmcd.html
[35] Watson, J. (2006). Golden arches east. Stanford, Calif.: Stanford University
Press.
[36] Rogers, Y., Sharp, H. Preece, J. (2011). Interaction Design: beyond human-
computer interaction (3rd ed.). West Sussex: John Wiley Sons.
[37] What is a Likert Scale and How Do You Pronounce Likert?. (2019). Retrieved
from http://core.ecu.edu/psyc/wuenschk/StatHelp/Likert.htm
[38] GINI index (World Bank estimate) | Data. (2019). Retrieved from
https://data.worldbank.org/indicator/SI.POV.GINI?locations=CN
[39] China’s personal income rises 6.3% in 2016 - Business - Chinadaily.com.cn.
(2019). Retrieved from http://www.chinadaily.com.cn/business/2017-
01/20/content_28010029.htm

85
Bibliography

86
 A
Appendix

Cross-cultural user experience research questionnaire

I
A. Appendix

II
A. Appendix

III
A. Appendix

IV
A. Appendix

V
A. Appendix

VI
 B
Appendix

Survey data

Individual Simple 2-3 Video with

Menu style link open in
Nationaliti Flat/hierarchi color with or columns / button /
Age (with icons / same window
es cal menu (dark/bright) product or density 4+ without
without icons) or new window
not design columns button
36-40 Chinese Text with icons Flat menu Bright Without Simple 4+ with Same window
Hierarchical
21-25 Swedish Text with icons Dark Without Simple 4+ with New window
menu
Hierarchical
21-25 Chinese Text with icons Bright Without Simple 4+ with New window
menu
Text without Hierarchical
21-25 Swedish Dark Without Simple 2-3 with New window
icons menu
26-30 Swedish Text with icons Flat menu Dark Without Density 2-3 No video New window
Hierarchical
26-30 Chinese Text with icons Bright Without Simple 2-3 No video New window
menu
Hierarchical
26-30 Chinese Text with icons Dark Without Simple 4+ with New window
menu
Hierarchical
36-40 Chinese Text with icons Dark Without Density 2-3 with Same window
menu
Hierarchical
26-30 Chinese Text with icons Dark Without Simple 2-3 Whatever New window
menu
Hierarchical
50+ Swedish Text with icons Bright With Simple 4+ with New window
menu
Text without
21-25 Swedish Flat menu Dark Without Simple 4+ with New window
icons
Hierarchical
21-25 Swedish Text with icons Dark Without Simple 2-3 with New window
menu
Hierarchical
21-25 Swedish Text with icons Dark Without Simple 2-3 with Same window
menu
Hierarchical
41-50 Swedish Text with icons Dark With Simple 2-3 Whatever New window
menu
Text without
21-25 Swedish Flat menu Dark With Simple 2-3 with New window
icons
Hierarchical
21-25 Swedish Text with icons Bright Without Simple 4+ with Same window
menu
Hierarchical
21-25 Swedish Text with icons Dark Without Density 2-3 with New window
menu
Hierarchical
31-35 Chinese Text with icons Dark Without Simple 4+ with New window
menu
Text without Hierarchical
21-25 Swedish Dark Without Density 2-3 Whatever New window
icons menu
Text without Hierarchical
41-50 Chinese Bright With Simple 4+ with New window
icons menu
Text without Hierarchical
26-30 Swedish Dark Without Simple 4+ with Same window
icons menu
Hierarchical
36-40 Chinese Text with icons Dark Without Simple 4+ with Same window
menu
Hierarchical
21-25 Swedish Text with icons Bright Without Density 2-3 with New window
menu
31-35 Chinese Text with icons Flat menu Dark Without Density 2-3 with New window
21-25 Chinese Text with icons Flat menu Bright Without Simple 2-3 No video New window
Hierarchical
26-30 Chinese Text with icons Bright Without Simple 2-3 with New window
menu
Hierarchical
26-30 Chinese Text with icons Bright With Simple 2-3 with New window
menu
Hierarchical
21-25 Chinese Text with icons Dark Without Simple 2-3 Whatever New window
menu
Hierarchical
31-35 Swedish Text with icons Dark Without Simple 2-3 with Same window
menu
Hierarchical
31-35 Chinese Text with icons Bright Without Simple 2-3 with New window
menu
Hierarchical
21-25 Swedish Text with icons Dark Without Simple 4+ with New window
menu
Hierarchical
26-30 Chinese Text with icons Dark Without Simple 4+ with Whatever
menu
36-40 Chinese Text with icons Flat menu Dark With Simple 2-3 with Same window
Hierarchical
31-35 Chinese Text with icons Bright Without Simple 4+ with New window
menu
Hierarchical
21-25 Chinese Text with icons Dark Without Simple 2-3 with Same window
menu
Text without Hierarchical
31-35 Swedish Dark Without Simple 4+ with New window
icons menu
36-40 Chinese Text with icons Flat menu Dark Without Simple 4+ with Whatever
21-25 Swedish Text with icons Flat menu Dark Without Simple 2-3 Whatever New window

26-30 Chinese
Text without
icons
Flat menu Dark Without Density 2-3 with VII
New window
Text without
26-30 Swedish Flat menu Bright Without Simple 4+ Whatever Same window
icons
Hierarchical
21-25 Chinese Text with icons Bright Without Simple 2-3 No video New window
menu
Hierarchical
21-25 Chinese Text with icons Dark Without Simple 2-3 with New window
menu
31-35 Chinese Text with icons Flat menu Bright Without Density 4+ with Same window
B. Appendix

Individual Simple 2-3 Video with

Menu style link open in
Nationaliti Flat/hierarchi color with or columns / button /
Age (with icons / same window
es cal menu (dark/bright) product or density 4+ without
without icons) or new window
not design columns button
Text without Hierarchical
31-35 Swedish Dark Without Simple 4+ with New window
icons menu
36-40 Chinese Text with icons Flat menu Dark Without Simple 4+ with Whatever
21-25 Swedish Text with icons Flat menu Dark Without Simple 2-3 Whatever New window
Text without
26-30 Chinese Flat menu Dark Without Density 2-3 with New window
icons
Text without
26-30 Swedish Flat menu Bright Without Simple 4+ Whatever Same window
icons
Hierarchical
21-25 Chinese Text with icons Bright Without Simple 2-3 No video New window
menu
Hierarchical
21-25 Chinese Text with icons Dark Without Simple 2-3 with New window
menu
31-35 Chinese Text with icons Flat menu Bright Without Density 4+ with Same window
Hierarchical
21-25 Chinese Text with icons Bright Without Simple 2-3 No video New window
menu
Text without
36-40 Swedish Flat menu Bright Without Simple 4+ Whatever New window
icons
Text without
36-40 Swedish Flat menu Dark Without Simple 2-3 with New window
icons
Hierarchical
21-25 Swedish Text with icons Dark Without Simple 2-3 with New window
menu
Hierarchical
21-25 Chinese Text with icons Dark Without Simple 2-3 with New window
menu
21-25 Swedish Text with icons Flat menu Dark Without Simple 4+ with New window
Hierarchical
41-50 Swedish Text with icons Dark With Simple 4+ with New window
menu
Text without
26-30 Chinese Flat menu Dark Without Simple 4+ with New window
icons
Text without Hierarchical
21-25 Swedish Bright With Simple 2-3 with New window
icons menu
Hierarchical
31-35 Chinese Text with icons Dark With Simple 2-3 with New window
menu
Hierarchical
31-35 Chinese Text with icons Bright Without Simple 4+ with New window
menu
Hierarchical
26-30 Chinese Text with icons Bright With Simple 2-3 with Same window
menu
26-30 Chinese Text with icons Flat menu Dark Without Simple 2-3 with New window
Hierarchical
31-35 Chinese Text with icons Dark Without Simple 4+ with New window
menu
Text without
50+ Swedish Flat menu Bright Without Density 4+ No video Whatever
icons
Text without
21-25 Chinese Flat menu Dark With Simple 4+ with Same window
icons
Hierarchical
41-50 Swedish Text with icons Dark Without Simple 4+ with Whatever
menu
Text without Hierarchical
26-30 Swedish Dark Without Simple 2-3 No video New window
icons menu
21-25 Swedish Text with icons Flat menu Dark Without Simple 4+ with New window
Hierarchical
26-30 Chinese Text with icons Bright Without Simple 2-3 No video Whatever
menu
Text without
21-25 Swedish Flat menu Dark With Simple 4+ with New window
icons
Hierarchical
36-40 Swedish Text with icons Dark Without Simple 2-3 with New window
menu
Hierarchical
41-50 Swedish Text with icons Dark Without Simple 4+ with New window
menu
Hierarchical
31-35 Chinese Text with icons Bright Without Simple 4+ No video Same window
menu
Hierarchical
26-30 Swedish Text with icons Bright Without Simple 4+ with New window
menu
Hierarchical
41-50 Swedish Text with icons Dark Without Simple 2-3 with New window
menu
Hierarchical
26-30 Swedish Text with icons Dark Without Simple 4+ with New window
menu
26-30 Swedish Text with icons Flat menu Dark Without Simple 2-3 with New window
Hierarchical
31-35 Chinese Text with icons Bright Without Simple 2-3 No video New window
menu

VIII
IX
C. Appendix

C
Appendix

X
C. Appendix

XI
XIII
XIV
Nation Age Education color icons Links Layout Overall I It was The The I was
think difficult to website informati confused
the web finish the appears on while
D. Appendix

pages given easy to displaye performing

look tasks. navigate d on the the tasks.
good website
is
valuable.
China 31-35 Undergraduate Bright, the dark i don't care if Open links in prefer 4 columns strongly Disagree Strongly Agree Disagree
version is too there're icons or new windows Agree agree
dark. not
China 25-30 Postgraduated Bright icons make the Open links in prefer 4 columns Agree Strongly Agree Agree Strongly
navigation not that new windows. disagree disagree
boring and the web was
supposed to
open links in new
windows for me.
China 36-40 Undergraduate bright. looks icons are good always onpen prefer 4 columns. Strongly Strongly Strongly Strongly Strongly
comfortable links in new agree disagree agree agree disagree
and warm. windows
China 31-35 Postgraduated Dark. high it doesn't matter open links in new 2-3 columns is better. Agree Disagree Agree Agree I don't know
technology windows easy to focus on the
feeling automaticly important content. 4
columns take too much
time to search for
information
D

China 25-30 Undergraduate Dark, chic and it's easy to find open links in new 2-3 columns is better. Agree Disagree Strongly Strongly Strongly
catagreies with windows clean categraies. big agree agree disagree
icons. icons are automaticly pictures contains more
good information
Appendix

China 36-40 Undergraduate light, feeling don't like icons. open links in new prefer 4 columns. so Agree Disagree Strongly I don't Strongly
comfortable. keep simple is windows by ctrl everything in one page. agree know disagree
the dark one good left click it\s easy to find things
makes me feel whtn they gether
pressure together
China 36-40 Undergraduate light, the color it doesn't matter Open links in prefer 4 columns. so Agree Disagree Agree I don't Strongly
is comfortable. because the icons same windows. everything in one page. know disagree
the dark one is don't make any the 2-3 columns are to
too dusky sense to me. empty and waster space
makes me feel
sad.
China 31-35 Undergraduate bright. make without icons Open links in 2-3 columns is better. Agree I don't Agree Agree Disagree
me feel glad because i don't new windows big pics contain more know
understand the information. and it's not
icons. it's faster to hard to scroll down the
read the text page with mouse
 Nation Age Education color icons Links Layout Overall I It was The The I was
think difficult to website informati confused
the web finish the appears on while
pages given easy to displaye performing
look tasks. navigate d on the the tasks.
good website
is
valuable.
China 31-35 Undergraduate bright is better. without icons. as Open links in 4 columns better. easy Strongly Strongly Agree Agree Disagree
the dark one is it's a simplisic new windows to find everything in one agree disagree
too serious. design screen.

India 25-30 Postgraduated Dark. icons look better Open links in prefer 4 columns. so Agree Disagree Agree Strongly Disagree
comfortable, same windows. everything in one page agree
morden use ctrl + left
click to open in
new windows if
necessiary
Sweden 46-50 Postgraduated Dark.easy to don't like icons. Open links in prefer 4 columns. more strongly Strongly Agree I don't Strongly
read text. keep simple is same windows. information. no scroll Agree disagree know disagree
good right click to down
open in new
windows if
necessiary
Sweden 41-45 Undergraduate Dark. just icons are good Open links in 2-3 columns is better. Agree Disagree Agree agree Disagree
prefer dark. same windows. we can know which one
right click to is more important than
open in new others
windows if
necessiary. it
depends on the
logic whether
open it in same
windows or new
ones.
Sweden 41-45 Undergraduate Dark. just i don't care if Open links in prefer 2-3 columns. Agree Strongly Agree agree Disagree
prefer dark. there're icons or same windows. make the page simple disagree
not use ctrl + left and empsize the
click to open in important part
new windows if
necessiary
Sweden 25-30 Postgraduated light, feeling without icons Open links in prefer 4 columns. so Agree Disagree Agree I don't Strongly
comfortable because the icons same windows. everything in one page know disagree
don't related to use middle
the text mouse click to
open in new

XV
D. Appendix
XVI
Nation Age Education color icons Links Layout Overall I It was The The I was
think difficult to website informati confused
D. Appendix

the web finish the appears on while

pages given easy to displaye performing
look tasks. navigate d on the the tasks.
good website
is
valuable.
Sweden 31-35 Postgraduated dark. it's easy to find Open links in 2-3 columns is better. Agree Disagree Agree Agree Disagree
catagreies with same windows. less desicition make. 4
icons. icons are use ctrl + left columns is to density,
good click to open in makes me don't know
new windows if which one is important
necessiary

Sweden 50+ Undergraduate bright. make it doesn't matter Open links in prefer 4 columns. so Agree Strongly Agree Agree Disagree
me feel glad because the icons same windows. everything in one page. disagree
don't make any
sense to me.

Sweden 36-40 Postgraduated Dark. without icons Open links in 2-3 columns is better. Agree I don't Strongly Agree Strongly
because the icons new windows by show few information in know agree disagree
don't related to right click each screen that I can
the text. but if the know what to read. 4
icons have columns contains too
meaning it would much information in one
be better to have page, that i donät know
icons where to start.
Sweden 31-35 Undergraduate dark one looks with icons. same window. or 4 columns better. give a Agree Strongly Strongly I don't Strongly
serious, visiable. right click for new overview of all disagree agree know disagree
professional window. to help information. the 2-3
and realiable me don't get lost columns is hard to find
as a information
company's
website with
long history