Research on virtual communities: an empirical approach

D. Beckers

University of Amsterdam, Department of Social Science Informatics
Roetersstraat 15 1018 WB Amsterdam, the Netherlands
email beckers@swi.psy.uva.nl

Recent innovations in computers and telecommunication technologies are
changing social interaction between people. However, we have only
vague notions of the precise effects. This leaves space for both
utopian as dystopian views, both seldom founded in empirical data.
Meanwhile, politicians have to make decisions on regulation,
investments and research that could have big implications for the
future of the use of these technologies. But research at community
networks also serve a more direct practical goal: how can a virtual
community try to support their town if they do not know who is using
this system? In this paper, we try to formulate some guidelines for
empirical research on virtual communities based on experience gained
at research on two digital cities in Europe.


1. Thinking about the effects of the use of telecommunication and 
computer technologies

Information and communication technology has invaded all domains of our
society: at work, at home and in public places. The modern culture is
profoundly mediatized (Thompson, 1990). Current innovations in
computers and telecommunication made new kinds of social interaction
and cultural transmission possible across previously impossible
distances. There is little doubt that these rapid advances in modern
telecommunication and computers are changing the way we live our
lives, but the direction of change is still uncertain.

Murdock (1993) discusses some of the social and cultural relations that
are emerging based on older technologies. Telephones for example
helped to support the maintenance of links irrespective of distance
and television 'rituals' such as major sporting occasions support new
social events and gatherings. But the merge of telecommunication and
computers, the computer mediated communication (CMC) might have an
even bigger consequences than the telephone and the television,
because of its unique characteristics. In the first place the ease to
generate and distribute data are unkown to any earlier technique
based on this data  'meta-data' can be generated, for example
searching-mechanisms.

Second, CMC is not limited to only text, but can also transport 
pictures, audio en video. Third, CMC is the first many-to-many medium.
For example the telephone can only be used by two persens at a time 
(one-to-one) and a newspapthe send information from one source to many
(one-to-many). Last, CGC can be used both synchronius (for example
for a telephone call the participants have to use the telephone at
the same time) as asynchronius (for example a letter, that is written
beforehand and is read later). Because of these unique abilities,
computer networks support new sorts of social places in the virtual
reality, where people can meet, work and play without ever seeing
each other in the physical world. For many of them to work and
socialize in cyberspace becomes more and more reality. This makes
cyberspace an interesting subject of research, since: "when people
define a situation as real, than it is real in it's consequences"
(W.I. Thomas).

Much has been written about the social consequences of the use of these
new technologies. However, although the interest on virtual
communities is large, the overall quality and depth of the research
can be questioned. One reason for this is time. It takes time to
build research projects, to ask the right kind of questions and to
adapt research methods to this new field of study (Wellman, 1997;
Parks, 1996). In the mean time, the small amount of empirical
research leaves space for both utopian as dystopian views. First, let
us examine both views.

The keywords for the utopists are liberation, freedom and flexibility.
This view stresses the recognition of the variety, the pluralism and
diversity in the virtual communities, where age, gender and race do
not matter. They also stress the empowering and liberating importance
of the use of the Internet.

"Drawing on simple technological determinism and utopianism, these 
approaches tend to evangelize and hype up the transformative powers of
telematics; there are seen to be virtually inevitable and woven into
some technological logic. (...) Through the humble personal computer
and modem, it is argued, individuals can now shape, participate in
and interact with a vast range of new multimedia spaces rather than
merely being passive consumers of a narrow range of TV, newspapers
and radio as in the past" (Leary, 1994; Negroponte, 1995)

Some authors even go further. Rushkoff (as quoted by Graham and Marvin)
predicts that the diffusion of interactive technologies, such as
email, WWW and virtual reality, will be used to "enhance the
individual ... and to break down the walls between individuals."
 
Opposite to the utopists, the dystopian view stresses the disorienting
and alienating effects of the globalization. According to the 
dystopists, the new media will cause a growing social polarization.
They see the commercialization of the new media as a proof of the
abuse of this technology. Finally, these critiques also stress that
the wealth of information does not necessarily mean that people have
more knowledge.  Postman argues that today's media are obsessed with
providing more information. He calls this the 'elevation of
information to a metaphysical status': information as both a means
and end of human creativity'. "In Technopolis, we are driven to fill
our lives with the quest to 'access' information. For what purpose or
with what limitations, it is not for us to ask' (Postman, 1992).

So who is right? And to whom will politicians listen when they have to
decide about regulations, investments and funding of research, which
could have big implications for the future of the use of these 
technologies. At the moment, we can see different forces, such as large
software companies, telecommunication companies, governmental and 
international organizations and the individual users struggling over 
issues as standards versus open systems, freedom of speech versus 
integrity, local versus global, copyrights etceteras. Without 
comprehensive research tracing the diverse social construction and uses
being made of electronic spaces, set against the broader political 
economy of telematics, the debate seems likely to remain unhelpfully 
polarized between 'visions of heaven and hell' (Harrison, 1995).

2. Empirical research at digital cities and virtual communities

When one is going to do a research on a digital city or a virtual 
community, the researcher has to choose the level of research and the 
suitable technique to collect the data. These techniques can be
borrowed from various disciplines like sociology, psychology and
anthropology. The next table shows the various levels with the
disciplines, techniques that can be 'borrowed' and examples of
researches.

TABLE 1

Level		Discipline	Technique		Examples

Global		Geography, 	mapping, 		Valtersson (1998)
		communication 	organizing 		"Virtual communities"
	   	   science
		information 	spatial data"
	   	   retrieval

City/ 		sociology,	questionnaires		Graham and Marvin
community 	groupware,				"Telecommunications
		communication 	content analysis	and the city"
		   science
		Information 
		   retrieval

Individuals	Psychology	interviews,		Rheingold "The virtual
		Ethnography	participating 		Community" ((on everyday
		Observation				life in the Well)
							Turkle "Live at the screen"

Every technique has its strengths and its weaknesses. For example, a 
questionnaire can give the researcher a lot of information about many 
people, but it is difficult to formulate relevant questions if the 
researcher is new in the field. Interviews yield much information about
a few users, but takes a lot of time for even a few respondents. To
get a good view on a community, different techniques should be
combined, so they can compensate each other weaknesses. In the next
paragraph, a research guide is presented, based on our experience
doing research at two digital cities in Europe.The first digital city
is 'the Digital City of Amsterdam' (DDS). As the biggest digital city
in the Netherlands, they claims to have over 80.000 members. Since
1994 every two years a questionnaire is put on, so we now have quite
a good idea of the characteristics of the population and the way they
use this system. The second one is Parthenay, a very small village in
the south of France with a few thousand inhabitants. Mainly because
of a very enthusiastic major, this town has a very comprehensive
website with about 200 active members.


3. Research at the Digital cities of Amsterdam and Parthenay

The results of the research at DDS show an interesting trend in the 
characteristics of the population. In 1994, the users were mainly
'early adopters'. They were a very homogeneous group, mainly young
males with a high education. Women, elderly, lower educated and
unemployed were strongly underrepresented. Two years later, students
were the biggest group. They were not very interested in information
and politics, but saw the digital city as a way to communicate with
each other. They frequently build their own webpages and used email a
lot. The last research from 1998 shows that the group of students has
shrunk, but the percentage of scholars increased. In 1994, more than
half of the users was resident in Amsterdam. Now, four years later,
it seams that the digital city is used from all over the Netherlands.
The digital city is no longer the digital city of Amsterdam, but
became a real 'virtual city', without connection to a geographical
place.

During the research we encountered a few problems. In the first place 
was it difficult to calculate how representative the results of the 
research were. The DDS had only very little information about their 
users, so it was sometimes difficult to compare the results from the 
questionnaire with information on the whole population.  Also, only 
after the research stopped, we heard of certain subcultures in DDS that
were until then unknown to us.  If we would have known of them 
beforehand, the questionnaire could have contained a question on these
subcultures.

Melis (forthcoming) did an elaborate content analysis of DDS and 
Parthenay. For the Digital city of Amsterdam and Parthenay, it turned 
out that the percentages of the various categories of content where 
quite the same, although these two organizations differ a lot in size, 
amount of users and purpose (see table 1). It would be interesting to 
check this for other virtual communities. 


TABLE 2: Use categories

Categories			% DDS				% Parthenay
Business			13 %				26 %
Education			7 %				11 %
Health				7 %				3 %
Informatics			5 %				3 %
Media				13 %				10 %
Leisure				31 %				28 %
Politics			23 %				20 %


Problems with this analysis were that Parthenay also knew very little
about their users, and they refused to give most of the information
they had because of privacy reasons.

In this paragraph, we present a guide for research in digital cities and 
virtual communities. This is by no means meant as a strict guide, but 
future researchers could use it for their benefit.
- Learn to know the city. Get to know the interface, the features and 
the limitations of the system. Try to find information about the amount 
of users and the size of the system. See how often pages are updated, 
how much traffic there is in discussion lists (high/low, only a few 
persons or many active users), find out who took the initiative (is it a 
grassroots approach of hired the local government a commercial company) 
etcetera. A good relationship with the board of the digital city is very 
important, and it is the only way to get access to the log-files.
- Perform a content analysis. Count how many pages contain information 
on the community, how many on leisure, communication, culture, etcetera. 
Table 3 shows an example of a form for this purpose.
- With the information gathered in the previous steps, now questions for 
a questionnaire can be formulated. When this questionnaire is done in an 
electronic form, the computer can analyze a huge number of responses 
directly, without the need to recode and enter the responses manually. 
If possible, it would be best to repeat this over a period, so changes 
in population and use over time can be identified. To check how 
representative the responses are, the results of the questionnaire 
should be compared with information from the log-files of the digital 
city.
- By now the researcher will have quite a good understanding of the 
virtual community and its users, but some questions may remain. To 
answer these it is best to interview a few key-persons, for example a 
member of the board of the organization or an active user.


Table 3: Example of a score card for content analysis

Square	Site		Cat InfLo InfOth PracLo PracOth CoMail CoChat etc.
Central	Journal	 c	1	1	 0	   0		1	0
Central	Help desk	 c	0	1	 0	   0		0	0
Central	DDS office	 b	0	1	 0	   0		0	0
Central	Info English m	0	1	 0	   1		0	0
Central	Registration c	0	1	 0	   1		0	0
Central	Pub		 l	0	0	 0	   0		0	1=09
Elections	Local 	 p	1	1	 1	   1		1	0
Elections 	National	 p	1	1	 0	   1		0	0
etcetera


Cat -- (see table 2)
NoAc -- access (not found, in construction...)
LiAc -- access (with password)
InfLo -- information (info about Amsterdam)
InfOth -- information (info about elsewhere than Amsterdam)
InfNo -- information
PracLo -- local (practical info about Amsterdam)
PracOth   -- Other (practical info outside Amsterdam)
PracNo -- practical info
Co Mail -- tool e-mail
CoChat -- tool chatroom
CoForm -- tool form
Eng -- language
Othla -- language


4. Conclusion & discussion

Computer mediated communication (CMC) is rearranging our world and will 
change the way we work and our daily live. In this paper we presented 
the utopian and dystopian views on the consequences of the use of this 
technology, but neither are very helpful. To inform design and 
implementation and to answer questions on future regulation knowledge is 
needed about the use, the users and the social effects of these new 
mediasystems. From the results of the research at the Digital City of 
Amsterdam and Parthenay, it seems that until now neither the utopists 
nor the dystopists are right.

The utopists see a world were age, gender, and personal characteristics 
do no matter anymore. However, at the digital city of Amsterdam we found 
a very homogenous population. But it is also not right to say as the 
dystopists do that only an elite has access to this kind of 
technologies. Both the DDS as the community of Parthenay are trying to 
get groups as women, and elders involved by giving courses and 
organizing special events aimed at these groups. It is surprising how 
little these organizations know about their users.

With this paper, we want to encourage further research on digital cities 
and virtual communities. Only with more ideas backed up with empirical 
data we can see what is really happening, without getting lost in 
'visions of heaven and hell'. And only when digital communities know who 
their users are they can build systems that can benefit the community. 
The future is yet undecided, but we are shaping it now. Let us shape it 
in the right direction.

5. References

- Van den Besselaar and Beckers (1998) Demographics and Sociographics of 
the Digital City, in: T. Ishida (ed), Community computing and support 
systems. London: Springer LNCS, 1998. (in press)
- Giddens (1991) Modernity and self-identity, Polity Press, Cambridge, 
UK
- Graham and Marvin (1996) Telecommunication and the city Electronic 
spaces, urban places Routledge, New York, US.
- Melis (In preparation) The social impact of new information and 
communication technology used at a local level: a comperative study of 
two European digital cities.
- Murdock, G (1993) Communication and the constitution of modernity, in: 
Media, Culture and society #15
- Postman (1992) technopolis: The surrender of Culture to Technology, 
Vintage, New York, US
- Schuler (1996) New Communitie Networks Wired for Change 
Addison-Wesley, New York, US.
- Stolterman, Agren and Croon (1998) Virtual communities Why and how are 
they studied, URL http://www.informatik.umu.se%7Emsjon/vircom/stagcr1.html