CULTURE INFLUENCE ON HUMAN COMPUTER
INTERACTION
Cultural Factors Toward User’s Preference on Groupware Application Design
Rein Suadamara, Stefan Werner and Axel Hunger
Institute of Computer Engineering, University of Duisburg-Essen, Oststr. 99, Duisburg, Germany
Keywords: Culture, Cultural Dimension, Computer Supported Cooperative Work, Groupware, Human Computer
Interaction.
Abstract: This paper reports about on-going research on how cultural dimensions affect user’s preference in
intercultural collaboration using computer supported cooperative work (CSCW) tools. It proposes how
selected cultural dimensions should be applied when designing a synchronous groupware application aimed
for multicultural users. Using four cultural dimensions, which are Collectivist-Individualist, Power
Distance Index, Uncertainty Avoidance, and Low- and High Context communication from Hofstede,
Gudykunst, Triandis, and Edward T.Hall, this research will try to analyse how culture influences the way
users prefer to interact using a groupware as a remote collaboration tools.
1 INTRODUCTION
During the last decades, there has been a tremendous
change in our ways of communicating. Nowadays
technology plays an important role in our daily lives.
Technology has replaced most of the communication
practices, where people often use email as a
communication medium instead of traditional mail
and where teleconference meeting replaces
traditional face-to-face meeting. The current
business trend integrates technological issues and
human factors; the internet enables us to have a
lunch meeting with colleagues from New York and
dinner conference with staff in Frankfurt in the same
day. Communication has move towards a “without
limits” trends (the world without limits).
With the expanding growth and development of
the information and communication technology,
cultures have been dramatically affected, especially
as a result of the increasing accessibility and speed
of communication platform. Many multinational
companies rely on multicultural teams call for an
effective functioning of geographically dispersed
work-related activities. Developing a groupware that
supports multicultural teams is one strategy to
improve intercultural communication hindrance in
the global business environment. Groupware aims to
save time and cost in coordinating group work,
however, developers are often more focussed on the
technology and produce highly sophisticated
software systems that may not be necessary at all in
real-life multicultural groups. The design of
groupware has to be centred on users and should
behave as a “co-work” environment that provides
flexibility, adaptability, and intelligent system
behaviour.
Earlier research had shown an increase of
scholarly interest on cultural factors affecting
Human Computer Interaction (HCI). Research
exploring issues related to cross-cultural and user-
interface design had quickly spread (Marcus 2000).
However, there is not much research done on how
culture influences user’s preference and behaviour
on groupware application. Although basic principles
of groupware tools for supporting social mechanism
for intercultural remote collaboration such as
awareness, coordination and conversation
(Vilhjalmsson 2003) has been conducted, no special
intention has been paid to investigating how cultural
dimensions influences user preferences on using
groupware as a collaboration tools. In addition to
technical problems, cultural problems also play an
important role among the problem with the tools,
especially when interaction between users from
different nationalities occurs. This paper shows the
influence of culture toward user’s preferences on
working collaborately using groupware as a
communication solution. It will try to analyse the
186
Suadamara R., Werner S. and Hunger A. (2010).
CULTURE INFLUENCE ON HUMAN COMPUTER INTERACTION - Cultural Factors Toward User’s Preference on Groupware Application Design.
In Proceedings of the 12th International Conference on Enterprise Information Systems - Human-Computer Interaction, pages 186-191
DOI: 10.5220/0002972701860191
Copyright
c
 SciTePress

effect of selected cultural dimensions on user’s
preferences on how tools and features that should be
include in a groupware application.
2 RESEARCH OBJECTIVE
At University Duisburg-Essen, Germany, a
synchronous groupware named PASSENGER has
been developed at the Institute of Computer
Engineering throughout the last years. This client-
server based groupware application enables student
teams to communicate and cooperate via internet,
even if the members are located at distributed sites
(Werner 2003). The system has not applied any
social aspect which allow multicultural user to
collaborate conveniently using the available tools
with less suitability for working environment. It did
not provide flexibility for user to change the settings
based on user’s preferences. To counter these
limitations, a new system approach is under
development, namely PASSENGER 2.
PASSENGER 2 relies on user-defined specification
to present optimal environment and configuration
during its operation. In order to do so, we will need
to know the user’s requirement on how to make
PASSENGER 2 become a tailor-able and flexible
synchronous groupware design that can adapt to
user’s preference settings to optimally to support day
to day cooperative in working environment.
Therefore, in this paper, user’s requirement set on
how human factors should be taking under
consideration when designing the tools and features
for the groupware design will be defined.
Integrating theories from cultural studies and
Computer Supporter Cooperative Work (CSCW)
may lead to an adaptable groupware application
designed accordingly with user’s preference. This
paper will try to explain how several cultural
dimensions may influence human as the user
interacting with the computer as the supporting tool
for collaborative working from the HCI point of
view. This is an on-going research; in this paper
only four dimensions are being discussed, which are:
Collectivism vs. Individualism, Uncertainty
Avoidance Index, Power Distance Index, and Low-
and High-context from Hofstede (Hofstede 1997),
Gudykunst(Gudykunst 1991), Triandis (Triandis et
al. 1985), and Edward T. Hall (Hall 1976).
3 THEORETICAL FOUNDATION
3.1 Culture
There are many definitions of culture in the
literature, but there is no agreement on a specific
definition of culture (Hoft 1996). Most of the
definitions refer culture as influencing the way in
which communication takes place. Thus, for the
purpose of this paper, culture is defined in terms of
cultural aspects that affects the communication style
which influence human (user) to human interaction
through the use of computer as its tool.
3.2 Cultural Dimension
3.2.1 Cultural Metamodels
The Iceberg Model - shown in Fig. 1, identifies three
layers of culture: (1) Surface: visible and easy to
research; (2) Unspoken rules: obscured and need
context of situation to understand the rules; (3)
Unconscious rules: out of conscious awareness.
Figure 1: The Iceberg Model (Hoft 1996).
The Pyramid Model - shown in Fig. 2 is introduced
by Geert Hofstede (Hofstede 1980): (1) Personality:
specific to an individual and is both learned and
inherited; (2) Culture: it is learned and not inherited;
(3) Human Nature: it is universal, inherited, and not
learned.
The Onion Model
- shown in Fig. 3 is introduced by
Fons Trompenaars (Trompenaars 1993): (1) Outer
layer: explicit products and artifacts of culture; (2)
Middle layer: defines norms and values; (3) Core:
implicit and consists of how people adapt to their
environments.
3.2.2 The Focus of the Four Cultural
Dimensions
In this paper, we will review four cultural
dimensions, which are: (1) Collectivism vs.
Individualism: Collectivistic cultures emphasize
goals, needs, and views of the ingroup over those of
the individual (Gudykunst 1991); (2) Uncertainty
CULTURE INFLUENCE ON HUMAN COMPUTER INTERACTION - Cultural Factors Toward User's Preference on
Groupware Application Design
187

Figure 2: The pyramid model (Hofstede 1980).
Figure 3: The Onion Model (Trompenaars 1993).
Avoidance Index: focuses on “the extent to which
people feel threatened by uncertain or unknown
situation” (Hoft 1996); (3) Power Distance Index
: is
described by “the extent to which less powerful
members of institutions and organization accept that
power is distributed unequally” (Hofstede 1997).;
(4) Low vs. High Context Communication
: Edward
T. Hall differentiates cultures on the basis of the
communication that predominates in the culture.
Context refers to the amount and specificity of
information in a given situation.
4 METHODS
In order to define how cultural dimensions
influenced user preferences in the field CSCW, a
survey was conducted. We used qualitative survey
as the method used in this research. Two methods
that we used in conducting this qualitative research
were: (1) In-depth interviews with the key person in
the industry/company; and (2) Distribution of
questionnaires (paper based and web based).
The survey was carried out for two months in
Indonesia in 2009. Indonesia was selected as the
subject respondent since Indonesia is the largest
archipelago and the fourth most populated country in
the world which is rich with culture. The
questionnaires were distributed for two months.
Respondents were from multicultural environments
working in the globalised industries, organizations,
and academic institutions. More than 20 companies
from different fields or companies, such as
Information and Technology, Advertising, and
Telecommunication, as well as Academic
Institutions were visited to conduct the in-depth
interview with the key person. The total respondents
of both paper based and web based questionnaires
were 142 respondents from 150 that were
distributed. The position or levels of the responded
varies from staff until executives/director. The age
of the respondents were dominated with the age of
25-45 years old. The data gained from the
questionnaires was translated into numerical
information in order to produce statistical result. An
appropriate statistical method is applied in
processing the data obtained from the
questionnaires. The method for analysis is
qualitative method using SPSS software version 12.0
to create a descriptive statistic of the data.
5 DISCUSSION
5.1 Individualism vs. Collectivism
Hofstede scored Indonesia as a country belonging to
the collectivistic culture. As shown in Fig. 4,
Indonesia has one of the lowest world rankings for
Individualism with a 14, compared to the greater
Asian rank of 23, and world rank of 43. The score on
this Dimension indicates the Indonesian society is
Collectivist as compared to Individualist (Hofstede
1980).
Figure 4: Hofstede Cultural Dimension for Indonesia
(Hofstede 1980).
Individualistic cultures focus on the “I” identity and
collectivistic cultures focus on the “we” identity.
The emphasis is placed on individuals’ goals in
individualistic cultures, while group goals have
precedence over individuals’ goals in collectivistic
cultures (Gudykunst and Ting-Toomey, 1988). This
is true, as shown in Table 1, most respondent agreed
that members have strong relationship in the group
and everyone takes responsibility for fellow
members of their group as much as 90.1% or 118
respondents and 9.9% or 13 respondents disagreed.
Gudykunst (1988) decribed collectivistic cultures
emphasize group harmony; individualist cultures
CORE
ICEIS 2010 - 12th International Conference on Enterprise Information Systems
188

Table 1: Indonesia as collectivistic culture.
Answer Frequency
V
alid Percent
Disagree 13 9,9
Agree 118 90,1
Total 131 100,0
emphasize the importance of verbal self-assertion
(Gudykunst & Ting-Toomey 1988). This is true,
most respondent answered that when working in
group, conflicts are very much avoided, in order to
keep the group harmony. Words are expressed as
polite as possible. For instance, to say “no” is
avoided and undesirable, since
it can lead to a
confrontation. To behave in such a way that would
result in the group ‘losing face’ is absolutely
displeasing. But, on the other hand, the word ‘yes’
does not necessarily denote approval, but serves as
maintenance of the communication line. In
Hofstede’s analysis on Indonesia, Hoftede
mentioned that “When someone says yes, does he or
she mean yes or no. Indonesia is a gracious culture
that is polite. Wanting to be agreeable and never
wanting to embarrass another”, this is untrue. Most
of the respondent disagree with the statement above
(62,3% or 66 respondents), while as much as 37,7%
or 52 respondents agreed, as shown in Table 2.
Table 2: The word “yes” may not mean yes.
In collectivistic culture, decision making are usually
done collectively using voting or the group’s
majority voice. To avoid conflicts, polling tools will
be useful for this type of user to be able to obtain a
fair result of discussion whenever ambiguity persist.
Triandis (1986) contends that the larger the number
of in-groups, the narrower the influence and the less
depth of influence. This affects the number of
groups that user will allow when a session starts
when working in a groupware. When the user
belongs to a high rating of individualistic culture, the
tendency to limit the group occur, and not allowing a
non-invited users to enter and join a session. Users
that belong in a high rank of collectivistic culture
would have the tendency to invite sessions directly
to an open list (or mailing list) and would allow any
other participants to join the session even if the
session has already been started.
5.2 Uncertainty Avoidance Index (UAI)
The second highest Hofstede ranking for Indonesia
is UAI at 48, compared to the greater Asian average
of 58 and a world average of 64. This reflects a more
moderated influence of this dimension within the
Indonesian society. In a country with low
uncertainty characteristic, societies are open toward
changes or ready to accept changes. This is true, as
shown in Table 3, most respondents are open in
accepting the changes in their society (66,7%),
whereas 42% answered very open, 24,6% answered
open, 15,2% answered average, 8% not open, and
10,1% answered very not open.
Table 3: Open in accepting changes in the society.
Answer Frequency Valid Percent
Very open 58 42,0
Open 34 24,6
Average 21 15,2
Not open 11 8,0
Very not open 14 10,1
Total 138 100,0
In Table 4 below, it shows that most of the
respondent, 93,2% or 124 respondents, answered
that they are willing to follow the changes and the
other 6% answered they prefer to leave it as it is.
One respondent answered that he preferred not to
have any changes, and 6, 3% or 9 respondents refuse
to answer.
Table 4: Willing to follow the changes.
Answer Frequency Valid Percent
Willing to follow the changes 124 93,2
Just leave it as it is 8 6,0
Do not want to have changes 1 0,8
Total 133 100,0
Users belonging to the high UAI culture are tend to
be more rigid type of people and would have the
tendency to work using the default settingUsers
belong to this type of culture resist change more,
therefore would like everything to do in order and
using the default settings, and also insist to have
clear instructions. Therefore manuals or guidelines
should really be available for users to refer to. On
the other hand, users belonging in the low UAI
culture are more of the loosely type of people and
more a risk taker. User of this type would prefer to
be able to configure or set up the configuration
manually based on their taste and preference. This
type of users will have the tendency to ignore rules;
therefore all settings should be made as flexible as it
can be. Since low UAI culture are risk takers, then
user are more of the “experiment” type. Users of this
type will have the tendency to use the most of the
provided tools available in the groupware. Users of
this type also have the tendency to negotiate;
therefore groupware may be useful for as a
negotiating tool.
Agree Frequency Valid Percent
No 86 62,3
Yes 52 37,7
Total 138 100,0
CULTURE INFLUENCE ON HUMAN COMPUTER INTERACTION - Cultural Factors Toward User's Preference on
Groupware Application Design
189

5.3 Power Distance Index (PDI)
In PDI dimension, Indonesia scored 78. It indicates
that in Indonesia, there is a far distance between
those who have high power to those who have the
least power in the society. The average PDI for the
greater Asian countries is 71 (Hofstede 1997). In
cultures with low power distance, bosses and
subordinates work closely together and consult with
each other. In culture with high power distance, the
relationship between the boss and the subordinate is
strictly ruled and dependent on the decision of the
boss. This is true, as shown in Table 5, most of the
respondents answered that usually the boss or leader
(with 34,5%) and the supervisor (with 26,8%) that
dominates the interaction as Indonesia is as a
country with high PDI score.
Table 5: Who dominates the interaction.
Answer Frequency Valid Percent
Boss/leader 49 34,5
Supervisor 38 26,8
Both member & leader 24 16,9
Member only 20 14,2
No answer 11 7,6
Total 142 100,0
Most of the respondents agreed that rank and status
is very important in the Indonesian culture, as shown
in Table 6 below. This opinion reach 61,3% or as
much of 87 respondents agreed, while 31% or 44
respondents did not agreed, and the other
respondents that did not answer is as much as 7,7%.
This affects the usage of the user profile in a
groupware application. Rank and status is important
for this type of user, therefore user would like to be
able to see other user’s profile in other to be more
respected by the other users. This also affects the
communication style in a groupware, as the boss,
usually would want to have more special treatment.
This will affect the video screen window on the
groupware interface layout. The boss’ screen should
be different from its subordinates, for instance the
size of the window would only be accepted if it is
made larger. Individuals in high power distance
cultures being more relaxed and open to accept
changes compared to their low power distance
counterparts. Most respondents are willing to try and
accept new application that enables them
communicate and coordinate better with their
colleagues or with the other companies. Although in
term of web-based application, the bandwidth should
also be increased considering the infrastructure in
Indonesia to avoid delay of the system.
Table 6: Rank and status is important.
Answer Frequency Valid Percent
No 44 31,0
Yes 87 61,3
No answer 11 7,7
Total 142 100,0
5.4 Low vs High Context
Communication
The dimension of context is firstly introduced by
Edward T. Hall (Hall 1976). Members of low
context, individualistic cultures tend to communicate
in a direct fashion, while members of high-context,
collectivistic culture tend to communicate in an
indirect fashion (Gudykunst 1991). In low-context
cultures, open confrontation of ideas and direct,
issue-oriented discussion are valued modes of
human expressiveness. This relates to the usage of
tools in groupware, whereas in low context culture,
user would prefer to be able to use external tools to
express their creative ideas while working.
Users belonging in the high-context culture are
often indirect when stating a purpose. This will
affect the timing of the session in a groupware. The
session should not be given a strict ending time and
flexible. Users belonging to high-context culture
would probably use emoticons more often than
people from low-context culture. High context
culture would also prefer to use video conference to
coordinate within a group and the tendency to make
use of PowerPoint during discussion to highlight
pointers rather than a long documentation reports.
Indonesia has the tendency to belong in the high-
context culture. As shown in Table 7 below, most
respondents answered that they prefer to use Instant
Messaging (IM) as a good way of communicating
with the other colleagues. Most of the respondents
use IM for both business and personal reasons
(Table 8). On the contrary, users belonging in the
low-context culture are the “straight to the point”
type of user. Time is used as effective and efficiently
as it should be. Therefore each session should be
given a limitation of time. Low-context cultures
have the tendency to use notepad, and word
applications for collaboration. Hall’s (1976)
description of low-and high-context cultures can
also be argued that there is greater need for
coordination of activity in high-context cultures than
in low-context cultures. It can also be assumed that
user belonging in the high-context cultures will
make use of the groupware for collaboration
activities rather than user belonging in the low-
context cultures.
ICEIS 2010 - 12th International Conference on Enterprise Information Systems
190

Table 7: Usage of IM.
Answer Frequency Valid Percent
No 24 16,9
Yes 88 62,0
Total 112 78,9
Missing answer 30 21,1
Total 142 100,0
Table 8: Reasons to use IM.
Answer Frequency Valid Percent
Business Purpose 88 62,0
Personal Reasons 71 50,0
6 CONCLUSIONS
Culture awareness is the crucial topic in
international collaboration. Each culture has its own
cultural values and style of communication. To
better communicate with each other, one must
distinguish some enormous difference between the
cultural backgrounds. Most people must adapt to the
technology rather than adapting to their needs (Day
1996). Since there is not enough effort to adapt
software culturally, people from other cultures tend
to adapt to the technology (Cagiltay 1999). Below is
the summary on how cultural dimensions affect
user’s preference upon the usage of tools and feature
in a synchronous groupware application: (a) In
collectivistic culture polling or voting tools is useful
as one of a tool for decision making and during
running session will allow late joiners and early
leavers; (b) In Individualistic culture, the limitation
of group size is more preferred and will not allow
late joiners nor early leavers; (c) In High UAI
culture, user tend to work using the default setting,
manuals and guidelines should be available since
user insist to have clear instructions; (d) In Low UAI
culture, the setting should be made flexible since
user prefer to be able to set up the configuration
manually based on their own taste and preferences
and will use groupware as a negotiating tool; (e) In
High PDI culture, the availability of user’s profile
information and differentiation of the video screen
size window between boss and the subordinates; (f)
In Low PDI culture, all the settings should be made
flexible and not strictly defined prior to its use; (g)
In High Context culture, the session timing should
be made flexible, make most use of video
conference rather than chatting tools, also use
PowerPoint rather than long documentation reports
to highlight pointers; (h) In Low Context culture, the
session timing should be limited and will make use
of notepad or word applications for collaboration.
As the result of this study, it will be use as a
requirement set for next generation groupware
PASSENGER 2 that is currently under development
at the Institute of Computer Engineering, University
Duisburg-Essen. By implementing the cultural
dimensions analysis to the PASSENGER 2 system,
it will make this new groupware as an intercultural
collaboration supporting tool, as well as a flexible
synchronous groupware that can adapt to user’s
preference setting.
REFERENCES
Cagiltay, K., 1999. Culture and its effects on human
computer interaction. Proceedings of World
Conference on Educational Multimedia, Hypermedia
and Telecommunications.
Day, D., 1996. Cultural based of Interface Acceptance.
Proceedings of the 11th annual European Human-
Computer Interaction Conference, (People and
Computers XI), 35-47.
Gudykunst, W.B., 1991. Bridging Differences: Effective
Intergroup Communication, Newbury Park, CA:
SAGE Publications.
Gudykunst, W.B. & Ting-Toomey, S., 1988. Culture and
Interpesonal Communication, SAGE Publications.
Hall, E.T., 1976. Beyond culture, New York: Doubleday.
Hofstede, G., 1997. Cultures and Organizations Software
of the Mind. Intercultural cooperation and it
importance for survival, UK: McGraw Hill.
Hofstede, G., 1980. Culture's consequences: International
differnces in work-related values, Beverly Hills, CA:
Sage.
Hoft, N., 1996. Developing a Cultural Model. In
International User Interface. New York: John Wiley
and Sons.
Marcus, A., 2000. Cultural Dimensions and Global Web
User Interface Design: What? So What? What Now?
In South African Human-Computer Interaction
Conference. Available at: http://chi-
sa.org.za/abstracts.htm.
Triandis, H. et al., 1985. Allocentric vs. idiocentric
tendencies. Journal of Research in Personality, 19,
395-415.
Trompenaars, F., 1993. Riding the waves of culture:
Understanding cultural diversity in business, London:
Nicholas Brealey.
Vilhjalmsson, H.H., 2003. Avatar Augmented Online
Conversation.
Werner, S., 2003. Synchrone Groupware für die Software-
Engineering-Ausbildung. Germany: Universität
Duisburg-Essen.
CULTURE INFLUENCE ON HUMAN COMPUTER INTERACTION - Cultural Factors Toward User's Preference on
Groupware Application Design
191