Guidelines for Writing a Policy Brief | 1 Guidelines for Writing a Policy Brief What is a Policy Brief? The Policy Brief is a “short, neutral summary of what is known about a particular issue or problem. Policy briefs are a form of report designed to facilitate policy-making” (Eisele, 2006). The main purpose is to “succinctly evaluate policy options regarding a specific issue, for a specific policy-maker audience” (Eisele, n.d.). Policy-makers need to make practical decisions under time-constraints, so the brief should provide evidence and actionable recommendations (Eisele, n.d.). The issue brief distils or synthesizes a large amount of complex detail, so the reader can easily understand the heart of the issue, its background, the players (“stakeholders”) and any recommendations, or even educated guesses about the future of the issue. It may have tables and graphs; usually, it has a short list of references, so the reader knows something about the sources on which it is based, and where to go for more information. Most of the time, the brief has its own “brief”--a one page “executive summary,” allowing the reader to quickly grasp the essence of the report (Eisele, n.d.). In short, “the purpose of the policy brief is to convince the target audience of the urgency of the current problem and the need to adopt the preferred alternative or course of action outlined and therefore, serve as an impetus for action” (Young & Quinn, n.d.). What are the components of a Policy Brief? (Lifted from Tsai, 2006) Executive summary The executive summary aims to convince the reader further that the brief is worth in-depth investigation. It is especially important for an audience that is short of time to clearly see the relevance and importance of the brief in reading the summary. As such, a 1 to 2 paragraph executive summary commonly includes: 1. A description of the problem addressed; 2. A statement on why the current approach/policy option needs to be changed; 3. Your recommendations for action. Context and importance of the problem The purpose of this element of the brief is to convince the target audience that a current and urgent problem exists which requires them to take action. The context and importance of the problem is both the introductory and first building block of the brief. As such, it usually includes the following: 1. A clear statement of the problem or issue in focus. 2. A short overview of the root causes of the problem 3. A clear statement of the policy implications of the problem that clearly establishes the current importance and policy relevance of the issue. It is worth noting that the length of the problem description may vary considerably from brief to brief depending on the stage on the policy process in focus, e.g. there may be a need to have a much more extensive problem description for policy at the evaluation stage than for one at the option choosing stage. Policy Brief versus Research Paper (T ...

1. Guidelines for Writing a Policy Brief | 1
Guidelines for Writing a Policy Brief
What is a Policy Brief? The Policy Brief is a “short, neutral
summary of what is known about a particular issue or problem.
Policy briefs are a form of report designed to facilitate policy-
making” (Eisele, 2006). The main purpose is to “succinctly
evaluate policy options regarding a specific issue, for a specific
policy-maker audience” (Eisele, n.d.). Policy-makers need to
make practical decisions under time-constraints, so the brief
should provide evidence and actionable recommendations
(Eisele, n.d.). The issue brief distils or synthesizes a large
amount of complex detail, so the reader can easily understand
the heart of the issue, its background, the players
(“stakeholders”) and any recommendations, or even educated
guesses about the future of the issue. It may have tables and
graphs; usually, it has a short list of references, so the reader
knows something about the sources on which it is based, and
where to go for more information. Most of the time, the brief
has its own “brief”--a one page “executive summary,” allowing
the reader to quickly grasp the essence of the report (Eisele,
n.d.). In short, “the purpose of the policy brief is to convince
the target audience of the
urgency of the current problem and the need to adopt the
preferred alternative or
course of action outlined and therefore, serve as an impetus for
action” (Young & Quinn,
n.d.).

2. What are the components of a Policy Brief? (Lifted from Tsai,
2006)
Executive summary The executive summary aims to convince
the reader further that the brief is worth in-depth investigation.
It is especially important for an audience that is short of time to
clearly see the relevance and importance of the brief in reading
the summary. As such, a 1 to 2 paragraph executive summary
commonly includes: 1. A description of the problem addressed;
2. A statement on why the current approach/policy option needs
to be changed; 3. Your recommendations for action.
Context and importance of the problem The purpose of this
element of the brief is to convince the target audience that a
current and urgent problem exists which requires them to take
action. The context and importance of the problem is both the
introductory and first building block of the brief. As such, it
usually includes the following: 1. A clear statement of the
problem or issue in focus. 2. A short overview of the root
causes of the problem 3. A clear statement of the policy
implications of the problem that clearly establishes the current
importance and policy relevance of the issue. It is worth noting
that the length of the problem description may vary
considerably from brief to brief depending on the stage on the
policy process in focus, e.g. there may be a need to have a much
more extensive problem description for policy at the evaluation
stage than for one at the option choosing stage.
Policy Brief versus
Research Paper
(Tsai, 2006)
Some might say that a policy brief is more “professional”
because it is geared towards readers who have a limited amount
of time to make a practical decision, while a research paper is

3. more “academic” because it pays more attention to the scholarly
roots of particular arguments and judges their merit on
intellectual and logical criteria.
Front-loaded!
Policy briefs are front-loaded: the
conclusions are on
the front page! The front page needs an executive summary,
providing a concise (1 or 2 paragraphs) overview of the
brief’s aim and core recommendations (Kopenski, 2010).
Guidelines for Writing a Policy Brief | 2
Critique of policy option(s) The aim of this element is to detail
shortcomings of the current approach or options being
implemented and therefore, illustrate both the need for change
and focus of where change needs to occur. In doing so, the
critique of policy options usually includes the following: 1. A
short overview of the policy option(s) in focus 2. An argument
illustrating why and how the current or proposed approach is
failing. It is important for the sake of credibility to recognize
all opinions in the debate of the issue.
Policy recommendations The aim of the policy
recommendations element is to provide a detailed and
convincing proposal of how the failings of the current policy
approach need to change. As such this is achieved by including:
1. A breakdown of the specific practical steps or measures that
need to be implemented 2. Sometimes also includes a closing
paragraph re-emphasizing the importance of action.
Appendices Although the brief is a short and targeted document,
authors sometimes decide that their argumen t needs further
support and so include an appendix. Appendices should be
included only when absolutely necessary.

4. What a persuasive Policy Brief should be (Lifted from Young
and Quinn, n.d.) As with all good marketing tools, the key to
success is targeting the particular audience for your message.
The most common audience for a policy brief is the decision-
maker but, it is also not unusual to use the document to support
broader advocacy initiatives targeting a wide but knowledgeable
audience (e.g. decision makers, journalists, diplomats,
administrators, researchers). In constructing a policy brief that
can effectively serve its intended purpose, it is common for a
brief to be:
FOCUSED All aspects of the policy brief (from the message to
the layout) need to strategically focused on achieving the
intended goal of convincing the target audience. For example,
the argument provided must build on what they do know about
the problem, provide insight about what they don’t know about
the problem and be presented in language that reflects their
values, i.e. using ideas, evidence and language that will
convince them.
PROFESSIONAL, NOT ACADEMIC The common audience for
a policy brief is not interested in the research/analysis
procedures conducted to produce the evidence, but are very
interested to know the writer’s perspective on the problem and
potential solutions based on the new evidence.
EVIDENCED-BASED The policy brief is a communication tool
produced by policy analysts and therefore all potential
audiences not only expect a rational argument but will only be
convinced by argumentation supported by evidence that the
problem exists and the consequences of adopting particular
alternatives.
LIMITED To provide adequately comprehensive but targeted
argument within a limited space, the focus of the brief needs to
be limited to a particular problem or area of a problem.

5. Do’s and Dont’s
(Kopenski, 2010)
Avoid technical, legalistic, economic or academic jargon!
Presentation needs to be professional as suits a public
document. Use correct grammar and spelling, and appropriate
spacing, font, point, headings and sub-headings. Facilitate
readability through images, catch-phrases, layout choices, and
the provision of data as graphs or charts.
Guidelines for Writing a Policy Brief | 3
SUCCINT The type of audiences targeted commonly do not
have the time or inclination to read an in-depth 20 page
argument on a policy problem. Therefore, it is common that
policy briefs do not exceed 6 – 8 pages in length (i.e. usually
not longer than 3,000 words).
UNDERSTANDABLE This not only refers to using clear and
simple language (i.e. not the jargon and concepts of an
academic discipline) but also to providing a well explained and
easy to follow argument targeting a wide but knowledgeable
audience.
ACCESSIBLE The writer of the policy brief should facilitate
the ease of use of the document by the target audience and
therefore, should subdivide the text using clear descriptive titles
to guide the reader.
PROMOTIONAL The policy brief should catch the eye of the
potential audience in order to create a favourable impression
(e.g. professional, innovative etc) In this way many brief
writers many of the features of the promotional leaflet (use of
colour, use of logos, photographs, slogans, illustrative quotes
etc).
PRACTICAL AND FEASIBLE The policy brief is an action-
oriented tool targeting policy practitioners. As such the brief

6. must provide arguments based on what is actually happening in
practice with a particular policy and propose recommendations
which seem realistic to the target audience
6 Steps for a compelling Policy Brief (Lifted from Young and
Quinn, n.d.) 1. Issue: examine the issue you will be dealing
with. Answer these questions: is the issue general or specific?
How general/specific? 2. Audience: take your primary audience
into serious consideration. Your brief should be tailored to the
needs of your audience. It makes a fundamental difference for
how you must frame your analysis and your recommendation. Is
your audience an individual (i.e. Prime Minister) or an
organization (i.e. the Government as a whole)? 3. Actors:
identify the relevant actors for the issue you are dealing with.
This is an essential step, since you will have to analyze their
interests in order to make sensible and viable policy
recommendations. Identifying the relevant actors is also
essential to produce a good assessment of the context and of the
interests that are plug into the issue. 4. Interests: once you have
identified the relevant actors, it is necessary to analyze their
interests. What are the actors' interests? Which of the relevant
actors have similar interests to your audience? Which ones have
different interests? How different? This step is important both
for the context part of your brief and for the critique of policy
options/policy recommendations. Without a clear identification
of the actors involved in the issue and their interests, your brief
will result vague, and therefore not useful. 5.
Recommendations: your policy recommendations should reflect
the above analysis. Remember that, according to the issue and
the audience, your recommendation(s) might not suggest the
best policy, but instead the most viable one. This should not
limit your recommendation to just compromise policies. If you
want to recommend radical change, you can; remember though
that such radical action has to be implemented in some ways. 6.

7. How-To: the last step is to suggest your audience the way to
'sell' the policy to its public (the public could be other members
of the organizations, voters, other parties, etc.). This last step
helps your audience build support/consensus to implement the
policy you recommended.
Guidelines for Writing a Policy Brief | 4
Online Resources
You can check the following policy briefs online to serve as
your guides in writing
your own policy briefs:
Food and Agriculture Organization of the United Nations
Policy Brief on the 2007-08 Rice Price Crisis:
http://www.fao.org/docrep/013/am172e/am172e00.pdf The
Poverty and Economic Policy Policy Briefs on Food and Fuel
Crises on Cambodia, Ghana and the Philippines:
http://portal.pep-net.org/documents/download/id/16735
http://portal.pep-net.org/documents/download/id/16736
http://portal.pep-net.org/documents/download/id/16737 The
Poverty and Economic Policy Brief on Effects of the GFC in the
Philippines: http://portal.pep-
net.org/documents/download/id/16739 The Royal Children’s
Hospital Policy Brief on Evidence-Based Practice:
http://www.rch.org.au/emplibrary/ccch/Policy_Brief_21_-
_Evidence_based_practice_final_web.pdf
Submission Guidelines: 1. All submissions should be in MS
Word Format (.doc, .docx), 2 pages in length and should be in a
2-column form. 2. Images, tables, charts and graphs present in
the layout of the Policy Brief should be submitted separately,
i.e., all .jpg, .png, .bmp, .gif, .psd, etc., should be submitted
individually. 3. Check policy briefs from the Poverty and
Economic Policy (PEP) Website (www.pep-net.org),

8. particularly those from the Community-Based Monitoring
System (CBMS) to serve as guides in preparing your policy
briefs.
Guidelines for Writing a Policy Brief | 5
References: Eisele, F (n.d.). Preparing a Policy Brief Issue
[PDF Document]. Retrieved from

9. https://www.courses.psu.edu/hpa/hpa301_fre1/IBInstructions_fa
02.PDF Kopenski, Marc (March 2010). Policy Briefs. Retrieved
from the Richmond University Website
http://www.richmond.ac.uk/content/library/subjects/politics/poli
cy-briefs.aspx Tsai (May 2006). Guidelines for Writing a
Policy Brief [PDF Document]. Retrieved from
http://jhunix.hcf.jhu.edu/~ktsai/policybrief.html. Young, E. and
Quinn, L (n.d.). The Policy Brief [PDF Documnet]. Retrieved
from http://www.policy.hu/ipf/fel-pubs/samples/PolicyBrief-
described.pdf Young, E. and Quinn, L (n.d.). The Policy Brief:
Instructions [PDF Document]. Retrieved from
http://sobek.colorado.edu/~salucci/teaching/teaching_portfolio/a
ssets/Policy_Brief_instructions.pdf
Prepared by the Community-Based Monitoring System (CBMS)
Network Coordinating Team
10th Floor Angelo King International Center
Estrada corner Arellano Streets, Malate, Manila
Philippines 1004
Telephone: (02) 5262067
Email: [email protected]; [email protected]
Website: http://www.pep-net.org;
http://www.cbmsphilippines.webs.com/
Small Group Research
42(5) 536 –561
© The Author(s) 2011
Reprints and permission:
sagepub.com/journalsPermissions.nav

10. DOI: 10.1177/1046496410397382
http://sgr.sagepub.com
Creativity in Virtual
Work: Effects
of Demographic
Differences
Luis L. Martins1
and Christina E. Shalley2
Abstract
Organizations are increasingly using virtual teams, in which
individuals
work with their teammates across distance and differences,
using a
variety of information and communication technologies. In this
study,
the authors examined how demographic differences (i.e.,
differences
in race, sex, age, and nationality) between individuals working
virtually
affected their collective creativity. Specifically, the authors
examined how
demographic differences interacted with the nature of
interaction processes
(establishment of rapport, participation equality, and process
conflict) and
difference in technical experience, to affect creativity in short-
term virtual
work interactions. Differences in age interacted with the
processes and
with differences in technical experience to affect creativity.
Differences in

11. nationality had a strong negative direct effect and interacted
with differences
in technical experience to affect creativity. Differences in sex
and race did
not significantly affect creativity. Implications of findings for
managing virtual
teams are discussed.
Keywords
creativity, demographic differences, virtual teams
1The University of Texas at Austin, USA
2Georgia Institute of Technology, USA
Corresponding Author:
Luis L. Martins, University of Texas at Austin, McCombs
School of Business,
1 University Station, B6300, Austin, TX 78712-0210
Email: [email protected]
http://crossmark.crossref.org/dialog/?doi=10.1177%2F10464964
10397382&domain=pdf&date_stamp=2011-02-28
Martins and Shalley 537
With an increase in global competition, companies have been
looking to cre-
ativity and innovation to give them a competitive edge
(Amabile, 1988;
Magadley & Birdi, 2009). Thus, effectively utilizing knowledge
resources
wherever they may reside in the organization has become an
important stra-
tegic priority for organizations (e.g., Dew & Hearn, 2009).

12. Until recently,
much of organizational knowledge was locked within
individuals and units
separated by various boundaries. However, with advances in
information and
communication technologies, organizations are increasingly
using virtual
teams to break down boundaries and connect employees
regardless of their
geographic location and subunit affiliation, so that they can
combine their
knowledge and perspectives to produce creative solutions to
various business
problems (e.g., McDonough, Kahn, & Barczak, 2001;
Townsend, DeMarie, &
Hendrickson, 1998). Indeed, increased creativity and innovation
have been
touted as among the primary benefits of using virtual teams
(e.g., Zakaria,
Amelinckx, & Wilemon, 2004). Thus, for example, consulting
firms such as
Bain and Company and McKinsey and Company use information
technology
(e.g., e-mail, instant messaging, and databases with the contact
information
and areas of expertise of every consultant) and other
communication tools to
enable consultants to reach peers in the company’s globally
distributed work-
force to work collaboratively on client problems as the needs
arise, which
contributes to their organizations’ ability to provide innovative
solutions to
their clients (Hansen, Nohria, & Tierney, 1999).
A natural consequence of the increase in the prominence of

13. global virtual
teams is that individuals are increasingly working virtually with
others who
are demographically different from themselves (Griffith &
Neale, 2001;
Griffith, Sawyer, & Neale, 2003). The extant literature on
creativity has gen-
erally proposed that demographic differences have the potential
to be benefi-
cial for creativity because demographically different individuals
working
together are able to bring to the task different perspectives and
approaches
(e.g., Milliken & Martins, 1996). On the other hand, researchers
have found
that, in the short run, demographic differences make it harder
for team mem-
bers to work together, thus potentially reducing their creative
performance
(e.g., see van Knippenberg & Schippers, 2007, for a recent
review). Whereas
there is a growing literature on the effects of demographic
differences on
creativity in more traditional face-to-face teams (e.g., Hoffman
& Maier,
1961; Nemeth, 1986), the effects have not been examined much
in virtual
teams. This study addresses this gap in the literature by
examining how
demographic differences interact with group process and input
conditions to
affect creativity in short-term virtual work interactions.
538 Small Group Research 42(5)

14. Of the variety of approaches that can be used to study dynamics
in virtual
teams, we focused on dyadic interactions within virtual teams
(i.e., on virtual
collaborations between any two members of a team). Although
there are
groupware systems that enable whole groups to interact at the
same time, much
of the interaction in virtual teams involves two team members at
any one time
collaborating virtually on a component of their team’s task
(e.g., Majchrzak,
Malhotra, Stamps, & Lipnack, 2004). Thus, though meetings of
whole virtual
teams occur episodically, their members’ day-to-day
interactions are typically
dyadic. For example, it is typical for a knowledge worker based
in the United
States to work with a colleague in Spain, one in Ireland, and
one in India, at
different times on different parts of the same task or project.
Indeed, Dew and
Hearn (2009) found that aggregating the creativity of pairs of
collaborators
within teams produced similar results as for the whole team,
leading them to
suggest that “structuring innovation teams into networked,
nominal pairs may
be just as productive as purely nominal group structures” (p.
521). Further-
more, the limited examination of the role played by
demographic differences
in dyadic interactions has been pointed out as a deficiency in
the literature on
diversity in teams in general (e.g., Tsui, Xin, & Egan, 1995) and

15. in virtual
teams in particular (Martins, Gilson, & Maynard, 2004).
Our examination of how the interactions of demographic
differences with
group process and input conditions affect creativity in virtual
collaborations
contributes to the literature in several ways. Whereas there is a
strong interest
among managers in using virtual teams to enhance creativity in
their organi-
zations, very few studies have empirically examined creativity
in a virtual
work context (e.g., Nemiro, 2002; Ocker, 2005), and fewer still
have looked
at how demographic differences affect creativity or innovation
in virtual con-
texts (e.g., Giambatista & Bhappu, 2010; Gibson & Gibbs,
2006). Recently,
researchers have argued that most work in organizations is now
virtual to a
greater or lesser extent, depending on the amount of time the
employees
spend working together on a task, the extent to which they use
technology to
support their interactions, and their geographic and temporal
separation (e.g.,
Griffith & Neale, 2001; Martins et al., 2004). Therefore, these
researchers
have suggested that we need to move beyond comparing virtual
to face-to-
face teams and, instead, empirically examine variation in
behavioral phe-
nomena within virtual teams. The moderated effects examined
in this study
advance understanding of the circumstances affecting the ability

16. of virtual
collaborators to leverage their knowledge resources, which prior
researchers
(e.g., Martins et al., 2004; Ocker, 2005) have suggested is an
important area
in need of further research. Specifically, the examination of
moderated effects
Martins and Shalley 539
enriches theory on the relationships between demographic
differences and
creativity by investigating how various process conditions and
an input fac-
tor determine whether demographic differences benefit or hurt
creativity in
virtual work (van Knippenberg, De Dreu, & Homan, 2004; van
Knippenberg &
Schippers, 2007).
Theory and Hypotheses
It is important to specify the boundary conditions of the model
we develop
and test in this study, as has been recommended in both the
diversity and the
virtual team literatures (Martins et al., 2004; Webber &
Donahue, 2001). Our
unit of analysis was dyadic interactions, which comprise most
of the day-to-
day work dynamics within virtual teams. The demographic
differences we
focused on were differences in race, sex, age, and nationality,
which are among
the major dimensions of demographic difference examined in

17. prior research
(e.g., Tsui & O’Reilly, 1989). The virtual working technology
we focused on
was computer-mediated communication (CMC; specifically,
electronic chat
room), which forms a large component of virtual work,
particularly among
geographically and temporally distributed individuals (Griffith
et al., 2003).
In addition, we focused on short-term dyadic collaborations
aimed at solving
immediate managerial problems. This short-term time
perspective was chosen
for a few reasons. First, virtual interactions for short-term
problem solving are
prevalent in virtual teams (Martins et al., 2004). In addition,
virtual teams
have been found to have a shorter lifecycle than face-to-face
teams, as they
are brought together as needed to work on specific tasks
(Jarvenpaa & Leidner,
1999). Finally, since membership in virtual work groups is often
fluid, dyadic
interactions with any one team member often are of a one-time
or short-term
nature (Zakaria et al., 2004). Finally, we looked at variation
within virtual
work, as has been recommended by researchers (e.g., Griffith et
al., 2003),
rather than comparing virtual to face-to-face work.
Effects of Demographic Differences on Creativity
Creativity is defined as the production of novel, potentially
useful ideas about
work products, practices, services, or procedures (Amabile,
1996; Shalley,

18. 1995). It is a major part of work quality and effectiveness and is
increasingly
valued for a variety of tasks, occupations, and industries
(Amabile, 1988). In
collaborative work, creativity requires the pooling together and
effective
integration of different perspectives, knowledge, skills, and
abilities on a
540 Small Group Research 42(5)
task (Hargadon & Bechky, 2006; Taggar, 2002; Woodman,
Sawyer, &
Griffin, 1993).
Research on diversity suggests two opposing expectations
regarding the
effects of demographic differences on creativity, based on the
contrasting
predictions of what have been termed the information/deci sion-
making per-
spective and the social categorization and similarity/attraction
perspective
(Williams & O’Reilly, 1998). Using the information/decision-
making per-
spective, researchers have proposed that differences in
perspective and expe-
riences underlying demographic differences should result in a
greater range
of information, ideas, and approaches to a problem being
generated and,
in turn, improved creative problem solving (e.g., Nemeth, 1986;
Pelled,
Eisenhardt, & Xin, 1999). Also, they have found that working

19. with different
others may stimulate consideration of nonobvious alternatives
that could
potentially lead to higher creativity (McLeod & Lobel, 1992).
Similarly, the
group brainstorming literature (see Paulus, 2000) has found that
differences
may be beneficial for the generation of more novel ideas. For
example,
McLeod and Lobel found that ethnically diverse groups
produced higher-
quality ideas. Also, culturally heterogeneous groups were found
to generate
more alternatives in the long run (Watson, Kumar, &
Michaelsen, 1993). Thus,
this perspective suggests that demographic differences have the
potential to
contribute to creativity by increasing the number of unique
ideas brought to
bear on a task (Milliken, Bartel, & Kurtzberg, 2003).
On the other hand, using the social categorization and
similarity/attraction
perspective, the literature suggests that demographic differences
can lead to
a variety of process losses leading to negative effects on team
performance.
This expectation is based on a variety of sociocognitive
theories, especially
the argument that individuals are attracted to those who they
perceive to be
demographically similar to themselves (Byrne, 1971) and those
who they
categorize as belonging to the same social category as
themselves (Tajfel,
1981). Social categorization and stereotyping based on

20. demographic charac-
teristics are particularly prevalent when teams are first formed
or in short-
lived teams, since individuals tend to use these cognitive
mechanisms to
make sense of other team members until their stereotypes are
invalidated
through extended positive interactions (e.g., Allport, 1954).
Thus, in the short
term, demographic differences within teams have been found to
result in
greater conflict, communication difficulties, and other negative
processes, as
well as lower cohesion and social integration (e.g., Harrison,
Price, & Bell, 1998;
Pelled et al., 1999), and consequently, lower creative
performance (Ancona &
Caldwell, 1992).
Martins and Shalley 541
Depending on what theoretical and process foundations are
used, demo-
graphic differences can both help and hurt performance (for
reviews, see
Milliken & Martins, 1996; van Knippenberg & Schippers, 2007;
Williams &
O’Reilly, 1998). Commenting on this conundrum, van
Knippenberg and
Schippers noted that these competing predictions and findings
are primarily
caused by focusing on main effects instead of potential
moderators. Consis-
tent with some prior work (e.g., Jackson & Joshi, 2004;

21. Milliken & Martins,
1996; Williams & O’Reilly, 1998), they argue for models that
are more com-
plex and that consider moderator variables in explaining the
effects of diver-
sity. We examine the moderating effects of processes and input
factors that
are likely to determine the extent of cognitive elaboration and
combination of
various perspectives on a collaborative task, which are critical
to translating
the potential benefits of demographic differences into actual
performance
benefits (e.g., van Knippenberg et al., 2004). We argue that in
order for the
potential creativity benefits due to differing perspectives to be
realized, it
is important that interactions among demographically different
individuals
enable the surfacing, pooling together, and integration of their
differing
perspectives (Gilson & Shalley, 2004; Milliken et al., 2003;
Shalley &
Perry-Smith, 2008).
Demographic Differences and Creativity:
Moderation by Processes and Inputs
The literature (e.g., Ocker, 2005) suggests that input and
process factors that
facilitate positive exchanges and the building of a positive
relationship are
important in determining the quality of interactions in virtual
teams. This is
particularly true for problem solving, which benefits from
multiple perspec-
tives but requires collaborators to work through their

22. differences in attitudes
and values to arrive at a consensus on solutions (Straus &
McGrath, 1994).
For instance, Taggar (2002) found that a team’s creativity-
relevant process
moderated the relationship between the average creativity of its
members and
the team’s creative output. Therefore, processes and inputs that
facilitate
information elaboration may be the key to reducing the negative
effects and
accentuating the benefits of demographic differences in virtual
interactions
(van Knippenberg, et al., 2004). For example, a team’s process
skills have
been found to be important to leveraging its members’ creative
resources
(e.g., Stroebe & Diehl, 1994). Also, Payne (1990) found that
communication
patterns in research teams had critical effects on their creative
performance.
The nature of the interaction process in virtual collaborations
also may affect
how members approach a task as well as affect their attention to
the heuristic
542 Small Group Research 42(5)
aspects of the task. The specific process factors we examined as
moderators
of the effects of demographic differences on creativity are the
degree to
which virtual collaborators spend time establishing rapport (i.e.,
bonding

23. through informal conversation) before beginning work on the
task, equality
of participation in their task-related discussions, and degree of
experienced
process conflict. The input factor we examined as a moderator
is the differ-
ence in technical experience between collaborators, which prior
research has
found to be one of the most influential inputs affecting the
quality of virtual
interactions (Sarker & Sahay, 2002). Difference in technical
experience
refers to the difference between virtual collaborators in their
experience with
using the information and communication technologies required
to collaborate
virtually.
Establishment of rapport. Whether individuals spend time
establishing
rapport with each other before working on their task can be
important for the
development of a good working relationship in virtual
collaborations (Coutu,
1998; Saunders, 2000). For example, by spending a few minutes
with intro-
ductions and discussing how they should approach working on
the task,
virtual collaborators can establish a bond of trust that may make
it easier
for them to work together effectively (Jarvenpaa & Leidner,
1999). Such
an establishment of rapport may create a psychologically safe
environment
(Edmondson, 1999) in which demographically different virtual
team members

24. are comfortable raising and discussing their differing
perspectives on a
problem without feeling interpersonally threatened (Griffith &
Neale, 2001).
In keeping with this argument, prior research has found that
virtual teams
whose members spend time at the onset of their work getting to
know each other
experience greater trust among members down the road, which
facilitates the
overall effectiveness of their working together (Jarvenpaa &
Leidner, 1999;
Suchan & Hayzak, 2001). This also should help them to
overcome the interaction
difficulties in working virtually. Therefore, we expect that
demographically
different virtual collaborators who establish rapport to a greater
degree will
be better able to surface, discuss, and integrate differing
perspectives, which
in turn enhances creativity. In contrast, when demographically
different virtual
collaborators do not establish rapport, they may find that their
differences
in perspective lead to difficulties in working together, which in
turn diminishes
creativity.
Hypothesis 1: The relationship between demographic
differences and
creativity will be positive when there is greater establishment of
rap-
port in virtual collaborations and negative when there is less
estab-
lishment of rapport.

25. Martins and Shalley 543
Participation equality. Participation equality reflects the extent
to which
each member of a dyad engaged in virtual collaboration
participates equally
in task interactions. In a diverse group, participation equality
may enable
“cognitive elaboration and information exchange within work
groups, draw-
ing out the different knowledge and skills represented” (Webber
& Donahue,
2001, p. 158). Thus, more equal participation enables better
surfacing and
discussion of different ideas, resulting in greater creativity
(Taggar, 2002).
For example, Kruempel (2000) found that, in order for effective
knowledge
production to occur in a virtual team, the perspectives of all
team members
needed to be raised and debated. Also, Gilson and Shalley
(2004) found that
teams that valued participation by all members were more
creative.
Consideration of the variety of views and ideas represented by
demo-
graphically different collaborators should lead to an expanded
source of
knowledge to use in decision making. Also, the intellectual
stimulation of
considering others’ ideas should encourage exploratory
thinking, which
results in greater creativity. However, effective collaboration

26. and participa-
tion are necessary for virtual teams to successfully integrate
various team
members’ ideas and perspectives (Sarker, Lau, & Sahay, 2001).
Thus, when
demographically different virtual collaborators do achieve
equality of par-
ticipation they may be better able to leverage their diversity of
perspectives
to generate creative solutions. In contrast, when
demographically different
virtual collaborators have unequal inputs they will be less likely
to be work-
ing with a broad range of information and perspectives, which
diminishes
their collective creativity.
Hypothesis 2: The relationship between demographic
differences and
creativity will be positive when there is greater participation
equal-
ity in virtual collaborations and negative when there is less
partici-
pation equality.
Process conflict. Process conflict is defined as “controversies
about aspects
of how task accomplishment will proceed” (Jehn & Mannix,
2001, p. 239;
italics in original). Greater process conflict increases
uncertainty and reduces
the ability of groups working on a task to pool together their
ideas effectively
to come up with collective solutions to problems (Jehn &
Mannix, 2001).
Thus, for demographically different virtual collaborators, a high

27. level of
process conflict between them may worsen the interaction
difficulties caused
by their demographic differences and virtual interaction, which
leads to process
losses (Montoya-Weiss, Massey, & Song, 2001). This argument
is consistent
with Griffith and Neale’s (2001) observation that “more virtual
environments
544 Small Group Research 42(5)
require more attention to procedural matters for success” (p.
401). The greater
the process conflict, the greater the increase in process losses,
which
negatively affects the joint creativity of demographically
different virtual
collaborators. In contrast, based on the diversity literature (e.g.,
Pelled et al.,
1999; Williams & O’Reilly, 1998), when virtual collaborators
do not
experience a great deal of process conflict they may be better
able to reduce
the interaction difficulties caused by their demographic
differences and
virtual interaction, and therefore, enhance effective discussion
and integration
of differing perspec tives to arrive at creative solutions.
Consistent with this
argument, creativity researchers have found that effective
collaboration is a
key determinant of creativity and innovation in teams (Pirola-
Merlo & Mann,

28. 2004). Thus, virtual collaborators who have effective processes
for
integrating their efforts for productive teamwork may be better
able to
overcome the low media richness of virtual work technologies
and to
integrate the differing perspectives on a task resulting from
their
demographic differences, in order to produce creative outcomes.
Hypothesis 3: The relationship between demographic
differences and
creativity will be negative when there is greater process conflict
in
virtual collaborations and positive when there is less process
conflict.
Differences in technical experience. Individuals collaborating
virtually may
be expected to differ in their extent of experience in using the
technologies
needed to interact virtually. Prior research has found that teams
whose
members all have high levels of competence in using virtual
work technologies
perform better than those in which some members are more
proficient in
using the technologies than others (Kayworth & Leidner, 2000;
Sarker &
Sahay, 2002). Similarities in technical experience may thus
create a positive
context for demographically different collaborators to surface
and discuss
their differing perspectives. Differences in technical experience,
on the other
hand, may create communication and interaction problems, as

29. individuals
with stronger technical abilities may feel frustrated or limited in
their virtual
collaborations with others who are not as proficient in using
virtual working
technologies. These difficulties would exacerbate any
interaction diffi-
culties due to demographic differences and the low media
richness of
virtual communication technologies. Thus, differences in
technical experience
between virtual collaborators may create process barriers to
effective virtual
interaction, causing frustration and miscues that reduce
creativity. Similar
levels of technical experience, on the other hand, may provide a
common platform
that establishes the nature of the interaction between virtual
collaborators
Martins and Shalley 545
(i.e., both individuals will face the same difficulties if they both
have low
technical experience or may have the same level of proficiency
if they both
have high technical experience). Therefore, when there are wide
differences
between virtual collaborators in their technical expertise, they
may have
greater difficulty in operating effectively in a virtual context. In
such a
circumstance, it may be expected that they will have difficulties
in establishing

30. interactions that surface and utilize differing perspectives,
which diminishes
their creativity.
Hypothesis 4: The relationship between demographic
differences and
creativity will be negative when there is greater difference in
tech-
nical experience between virtual collaborators and positive
when
there is less difference in technical experience.
Method
Sample, Task, and Procedures
The sample consisted of 94 MBA students in an organizational
behavior
course at a medium-sized urban university in the United States.
As part of
their normal course curriculum, the class worked on a virtual
work project.
The students were asked to volunteer to participate in this
research by filling
out a survey; all did. The sample was demographically diverse:
33% were
women, 45% international (representing 24 countries), and 36%
nonwhite.
The participants were in the age range of 23 to 42 years (M =
27.6 years,
median = 27 years). All participants were proficient in spoken
and written
English; the average TOEFL score for international students
was 637 (out of
a possible 677).
We used a complex heuristic task for which responses were

31. open ended,
did not have correct answers, and required participants to “seek
consensus on
a preferred alternative” (Straus & McGrath, 1994) that has been
used in a num-
ber of prior creativity studies (e.g., Shalley, 1991; Shalley &
Perry-Smith,
2001; Zhou, 1998). Participants were asked to generate
solutions to various
human resource problems (e.g., employee theft, motivating the
sales force)
that typically arise within organizations and that managers need
to be able to
effectively solve. The participants were told that we were
particularly inter-
ested in creative solutions, so they should try to think of unique
ways to solve
the problems that also would work well in the company.
Participants were randomly assigned a partner to collaborate
with, which
yielded 47 dyads engaged in virtual collaboration. Members of
59.57% of the
546 Small Group Research 42(5)
dyads were of different races, 53.19% were of different sexes,
and 63.83%
were of different nationalities. The difference in age between
virtual collabo-
rators ranged from 0 to 16 years (M = 3.89 years). The
participants were only
allowed to communicate with their partner in an electronic chat
room.

32. Participants were given instructions on how to log on to their
assigned chat
room to connect with their partner to work on the task, were
briefly introduced
to their partner face to face (they could see each other, but
could not speak),
and were given 60 min to work on the collaborative task. Before
working on
the task, they were asked to complete a brief survey that
collected informa-
tion on demographics and extent of prior technical experience
with computer-
based interaction (e.g., chat rooms, bulletin boards, and e-mail).
Measures
Demographic differences. All participants were asked to
indicate their race,
sex, age, and nationality. As has been done in previous research
on demo-
graphic differences (Tsui & O’Reilly, 1989), we used
dichotomous mea-
sures for differences in race, sex, and nationality (with 0
indicating no
difference, and 1 indicating a difference in the respective
characteristic) and
computed difference in age as the squared difference between
the ages of
the two collaborators.
Creativity. According to Amabile (1996), a product is creative
if observers
independently agree that it is novel and appropriate. Two
graduate research
assistants independently rated the creativity of all solutions

33. generated on a
7-point scale (1 = not at all creative to 7 = extremely creative).
Interrater reli-
ability was assessed using rWG. The mean rWG(j) for the
creativity ratings was
.96, which is well above the commonly used cutoff of .70. Thus,
the overall
creativity score for each pair of virtual collaborators was
computed as the
average of the two raters’ creativity ratings for the solutions
generated by the
collaborators. Since there may be an association between the
number of solu-
tions provided and the overall level of creativity, we tested the
number of
problems solved as a potential control variable. However, since
it was not
significant as a control, it was excluded from the analyses.
Virtual interaction process factors. We retained a transcript of
the interac-
tions within each chat room as the virtual collaborators worked,
so that we
could code aspects of their interaction process. Two raters
independently
coded the following factors on a 7-point scale: establishment of
rapport,
equality of participation in problem solving, and extent of
process conflict.
The factors were defined for both raters, and their
understanding of the defi-
nitions was ascertained. Establishme nt of rapport was defined as
spending

34. Martins and Shalley 547
time at the start of the virtual collaboration discussing non-task-
related issues
that served to create a positive working relationship (e.g.,
spending time
briefly getting to know each other, reassuring each other about
the upcoming
virtual collaboration). Equality of participation was coded as
the extent to which
each of the virtual collaborators had an equal amount of input
and influence
over the task. Process conflict was defined as the extent of
disagreement
between the virtual collaborators in how the task should be done
(e.g., dis-
agreement was high if one person wanted to read all the memos
before com-
mencing discussion and the other wanted to tackle them one by
one). As was
done with the creativity ratings, interrater agreement for the
three factors was
assessed using rWG. The mean rWG for the process ratings was
.94 for estab-
lishment of rapport, .91 for equality of participation, and .86 for
extent of
process conflict. Thus, for each of the three process factors, the
ratings
assigned by the two raters were averaged to obtain overall
scores.
Virtual interaction input factor. For technical experience,
respondents were
asked to indicate on a 5-point scale (1 = none to 5 = a great
deal) their degree
of experience with using e-mail, chat rooms, bulletin boards,

35. and any other
electronic collaboration technologies (e.g., document sharing).
Responses
were averaged across the four items (alpha = .71). For each set
of collabora-
tors, difference in technical experience was computed as the
variance between
the scores of its members.
Results
Correlations and descriptive statistics are reported in Table 1.
The hypothe-
ses were tested using hierarchical regression analysis
(significant findings
are reported in Table 2). Due to sample size constraints, we ran
separate
regressions for differences in age and nationality and for
differences in race
and sex. Also, we entered each moderator variable separately.
The limita-
tions of the separate tests for the dimensions of difference and
moderator
variables are noted in the discussion section below. The
variables were
entered into the hierarchical regression equation in four steps.
In the first set
of analyses, difference in age and difference in nationality were
entered in the
first step (Table 2, Model 1a-d), the respective moderator
variable was entered
in the second step (Table 2, Models 2a-d), the interaction term
for difference
in age and the respective moderator was entered in the third step
(Table 2,
Models 3a-d), and the interaction term for difference in
nationality and the

36. respective moderator was entered in the fourth step (Table 2,
Models 4a-d).
The second set of analyses repeated the process, but with
differences in race
and sex as the demographic difference variables. A centering
procedure was
548
T
a
b
le
1
.
D
es
cr
ip
ti
ve
s
an
d
C
o
rr

37. el
at
io
ns
V
ar
ia
bl
e
M
ea
n
SD
1
2
3
4
5
6
7
8
1.
D

38. iff
er
en
ce
in
r
ac
e
0
.6
0
0
.5
0
2.
D
iff
er
en
ce
in
s
ex
0

39. .5
3
0
.5
0
-.
25
3.
D
iff
er
en
ce
in
a
ge
(
sq
ua
re
d)
29
.6
3

40. 53
.4
0
-.
07
.0
1
4.
D
iff
er
en
ce
in
n
at
io
na
lit
y
0
.6
4
0

41. .4
9
.4
6
-.
26
-.
26
5.
E
st
ab
lis
hm
en
t
o
f
ra
pp
o
rt
a
4
.0

42. 2
0
.8
1
-.
06
-.
09
-.
11
.0
2
6.
P
ar
ti
ci
pa
ti
o
n
eq
ua
lit
ya

43. 4
.6
9
0
.7
4
-.
17
.2
5
-.
08
-.
17
.0
2
7.
P
ro
ce
ss
c
o
nf

44. lic
ta
3
.4
9
1
.0
6
.0
5
-.
15
.0
6
.1
4
.1
1
-.
65
8.
D
iff
er

45. en
ce
in
t
ec
hn
ic
al
e
xp
er
ie
nc
eb
1
.4
8
1
.3
1
.1
3
-.
20

46. -.
03
.2
6
.1
3
-.
09
.2
9
9.
C
re
at
iv
it
yc
3
.9
8
0
.5
4
-.

47. 09
.0
9
.0
1
-.
38
.3
1
.3
5
-.
20
.1
1
N
o
te
: N
=
4
7.
C
o
rr

48. el
at
io
ns
a
bo
ve
.2
8
ar
e
si
gn
ifi
ca
nt
a
t
th
e
.0
5
le
ve

49. l.
a.
E
st
ab
lis
hm
en
t
o
f
ra
pp
o
rt
, p
ar
ti
ci
pa
ti
o
n
eq
ua

50. lit
y,
a
nd
p
ro
ce
ss
c
o
nf
lic
t
w
er
e
ra
te
d
o
n
a
7-
po
in

51. t
sc
al
e.
b.
T
he
e
xt
en
t
o
f
te
ch
ni
ca
l e
xp
er
ie
nc
e
w

52. as
r
at
ed
o
n
a
5-
po
in
t
sc
al
e.
c.
T
he
le
ve
l o
f
cr
ea
ti
vi

53. ty
w
as
r
at
ed
o
n
a
7-
po
in
t
sc
al
e.
Martins and Shalley 549
Table 2. Results of Hierarchical Regression Analyses
Model 1a 2a 3a 4a
Step 1: Independent variables
Difference in age -.11 -.07 .03 .03
Difference in nationality -.42*** -.41*** -.40*** -.40***

54. Step 2: Moderator
Establishment of rapport — .30** .39*** .39***
Steps 3 & 4: Interaction terms
Establishment of rapport ×
Difference in age
— — .32** .31**
Establishment of
Rapport × Difference in
Nationality
— — — -.03
R2 .17 .26 .34 .34
Adj. R2 .13 .20 .28 .26
F 4.26** 4.79*** 5.28*** 4.13***
R2 change — .09** .09** .00
Model 1b 2b 3b 4b
Step 1: Independent variables
Difference in age -.11 -.07 -.02 -.01
Difference in nationality -.42*** -.36** -.34** -.35**
Step 2: Moderator
Participation equality — .27** .20 .17
Steps 3 & 4: Interaction terms
Participation Equality ×
Difference in Age
— — .29** .32**
Participation Equality ×
Difference in Nationality
— — — .14

55. R2 .17 .24 .31 .33
Adj. R2 .13 .18 .24 .24
F 4.26** 4.34*** 4.58*** 3.87***
R2 change — .07** .07** .02
Model 1c 2c 3c 4c
Step 1: Independent variables
Difference in age -.11 -.09 -.18 -.18
Difference in nationality -.42*** -.40*** -.34** -.35**
Step 2: Moderator
Process conflict — -.15 -.08 -.08
Steps 3 & 4: Interaction terms
Process Conflict ×
Difference in Age
— — -.38*** -.38***
Process Conflict ×
Difference in Nationality
— — -.01
R2 .17 .19 .32 .32
Adj. R2 .13 .13 .25 .23
F 4.26** 3.21** 4.74*** 3.70***
R2 change — .02 .13*** .00
(continued)
550 Small Group Research 42(5)
followed in the computation of interaction terms (Aiken, West,

56. & Reno,
1991). All significant moderated effects found were examined
further, using
recommended procedures (Aiken et al., 1991).
Hypothesis 1 stated that there would be a positive relationship
between
demographic differences and creativity when there was high
establishment of
rapport and a negative relationship when there was low
establishment of rap-
port. This hypothesis was supported (p < .05) for differences in
age (Table 2,
Model 3a; F = 5.28, p < .01) but not for differences in
nationality, race, or
sex. Difference in age was positively related to creativity when
there was
more establishment of rapport but was negatively related to
creativity when
there was less rapport established.
Hypothesis 2 stated that there would be a positive relationship
between
demographic differences and creativity when there was more
equal participation
by both virtual collaborators and a negative relationship when
there was less
equal participation. This hypothesis was supported (p < .05) for
differences
in age (Table 2, Model 3b; F = 4.58, p < .01) but not for
differences in nation-
ality, race, or sex. Difference in age was positively related to
creativity when
there was relatively equal participation in the virtual w ork
interaction but was
negatively related when there was less equal participation.

57. Table 2. (continued)
Model 1d 2d 3d 4d
Step 1: Independent variables
Difference in age -.11 -.12 -.11 -.13
Difference in nationality -.42*** -.49*** -.47*** -.55****
Step 2: Moderator
Difference in technical
experience
— .25* .17 .22
Steps 3 & 4: Interaction terms
Difference in Technical
Experience × Difference
in Age
— — –.28** -.35***
Difference in Technical
Experience × Difference
in Nationality
— — — -.33**
R2 .17 .22 .29 .39
Adj. R2 .13 .17 .23 .32
F 4.26** 3.98** 4.26*** 5.14****
R2 change — .06* .07** .10**
Note: N = 47.
*p < .10. **p < .05. ***p < .01. ****p < .001.

58. Martins and Shalley 551
Hypothesis 3 stated that there would be a negative relationship
between
demographic differences and creativity when there was high
process conflict
and a positive relationship when there was low process conflict.
This hypothe-
sis was supported (p < .01) for differences in age (Table 2,
Model 3c; F = 4.74,
p < .01), but not for differences in nationality, race, or sex.
Difference in age
was negatively related to creativity when there was high process
conflict in
the virtual interaction but was positively related to creativity
when there was
low process conflict.
Hypothesis 4 stated that there would be a negative relationship
between
demographic differences and creativity when there was a larger
difference in
technical experience between virtual collaborators and a
positive relationship
when there was a smaller difference in technical experience.
This hypothesis
was supported (p < .05) for differences in age (Table 2, Model
3d; F = 4.26,
p < .01), and to some extent for differences in nationality
(Table 2, Model 4d;
F = 5.14, p < .01), but not for differences in race or sex.
Difference in age was
negatively related to creativity when there was a large
difference in technical

59. experience between virtual collaborators but was slightly
positively related
when there was a smaller difference in technical experience.
Difference in
nationality was more negatively related to creativity when there
was a greater
difference in technical experience than when there was a
smaller difference.
Discussion
The pattern of our findings is relatively consistent when looked
at from the
perspective of the various dimensions of difference we
examined. We found
that our hypotheses were consistently supported for differences
in age. The
effect of the difference in age between virtual collaborators on
creativity was
contingent on various aspects of their interaction and on their
difference in
technical experience. All the interaction process factors we
examined moder-
ated the effects of age difference on creativity. We found that a
difference in
age led to greater creativity when virtual collaborators had
spent some time
establishing rapport, when they had equal participation in the
discussion, and
when process conflict was low. These findings support the idea
that when
demographically different virtual collaborators are able to
utilize effective
processes, they are better able to deal with interaction
difficulties that may
arise from their differences and virtual interaction, and thus,
benefit from the

60. differences in perspectives associated with their age
differences. In contrast,
when such processes are not in place, the interaction difficulties
caused by
age differences and virtual interactions may lead to lower
creative perfor-
mance. We found that the difference in technical experience
between virtual
552 Small Group Research 42(5)
collaborators was an important moderator of the effects of
difference in age
on creativity, such that it exacerbated the negative effect of
difference in age
on creativity. Difference in technical experience likely cr eated
communica-
tion problems that further increased the difficulty of interaction
beyond that
already caused by a difference in age and the virtual interaction
medium,
thereby reducing creativity. Overall, our findings for age
differences are consis-
tent with the suggestion in prior research that contextual
conditions may be
important in determining the effects of age differences on
outcomes (Ferris,
Judge, Chachere, & Liden, 1991; Shore, Cleveland, & Goldberg,
2003).
Given the strong effects for age differences in our model, we
conducted
post hoc analyses to examine alternative explanations for our
findings. To

61. determine whether the effects of differences in age were really
due to differ-
ences in experience between virtual collaborators, we reran the
regression
analyses using difference in work experience in place of
difference in age.
We found that whereas difference in work experience
functioned similarly in
the analysis to difference in age, the results were much stronger
for differ-
ence in age. This suggests that difference in work experience
may explain the
effects of difference in age to a large extent but that other
factors associated
with difference in age may also play a part in generating the
effects obtained.
Another argument that can be made to explain the effect of age
difference is
that the older of the virtual collaborators brought to the task
greater life and
work experience, thus increasing the creativity of the
collaborative product.
However, we found that creativity did not correlate significantly
with the
virtual collaborators’ average age (r = -.07, p = ns) or average
amount of
work experience (r = .02, p = ns). Taken together, our post hoc
analyses sug-
gest that the effects we found were more likely due to
differences between
the ages of the two collaborators (which partly reflects
difference in work
experience) rather than due to higher- or lower-average age of
the two virtual
collaborators.

62. For differences in nationality, we found only one significant
interaction
effect, and that was not entirely in the predicted direction.
Essentially, for vir-
tual collaborators from different nationalities, even relatively
equal technical
experience did not produce a positive effect of difference in
nationality on
creativity, though it did reduce the strength of the negative
effect experienced
by the virtual collaborators with unequal technical experience.
In addition,
though we did not predict a direct effect of a difference in
nationality on
creativity, we did find a relatively strong direct effect. The
finding is consis-
tent with prior research that has found that national differences
create com-
munication problems in cross-national teams (e.g., Kayworth &
Leidner,
2000) and points out that these problems may be amplified by
differences in
Martins and Shalley 553
technical experience between employees working virtually.
Furthermore, the
negative effect for difference in nationality on creativity was
not moderated
by the interaction processes we examined. This lack of
moderation suggests
that the interaction difficulties encountered in cross-national
virtual teams
may not be easily overcome in short-term virtual interactions,

63. such as those
examined in this study. Given that short-term problem-solving
interactions
among globally distributed employees are becoming
increasingly common in
organizations, our finding suggests that much more research
needs to be con-
ducted in order to understand and manage such interactions.
For differences in race and sex we did not find any significant
effects. In
general, the effects of demographic differences hinge on the
cognitive avail-
ability of demographic characteristics that feed into social
categorization
processes (e.g., van Knippenberg et al., 2004). Since virtual
communication
technologies are low in the richness of information carried, this
reduces the
cognitive availability of demographic differences in virtual
interactions (e.g.,
Nowak, 2003; Sproull & Kiesler, 1986). Thus, rather than react
to surface-
level characteristics using social category stereotypes,
demographically dif-
ferent collaborators may instead focus on differences in virtual
interaction
patterns that may accompany their demographic differences.
Furthermore,
since researchers have found that there are no readily detectable
differences
in the behavior of men and women in a virtual context, such as
in the extent
of interaction (Gefen & Straub, 1997), it is understandable that
there were no
significant effects for difference in sex.

64. In contrast, differences in age and nationality have been found
to affect
interaction patterns in virtual teams. Individuals from different
nationalities
may encounter difficulties in interacting virtually due to cross -
national dif-
ferences in communication styles and differences in word
connotations even
when communicating in the same language (e.g., Maznevski &
Chudoba,
2000; Zakaria et al., 2004). Furthermore, difficulties may arise
in virtual
interactions between individuals from different nationalities due
to differ-
ences in reliance on body language, facial expressions, gestures,
and physical
distance in communication in their respective countries (Farmer
& Hyatt,
1994). Also, national differences in usage of the English
language may cause
difficulties in communication in virtual teams (e.g., Zakaria et
al., 2004). In
a similar vein, prior research has found that an individual’s age
affects his or
her attitude toward, and comfort with the use of, information
technologies
such as e-mail (e.g., Agarwal & Prasad, 1999; Burton-Jones &
Hubona,
2005). Also, individuals of different ages may communicate
differently, in
terms of the formalness of their communication style, their use
of slang or

65. 554 Small Group Research 42(5)
certain terms, and norms regarding communication in general
(e.g., Lancaster
& Stillman, 2002).
Limitations, Directions for Future
Research, and Contributions
An obvious limitation of this study is the sample size, which
was largely a
consequence of the logistical and financial difficulty in setting
up a large
number of virtual collaborations with the type of students we
used in the
study—graduate business school students—and coding hour-
long chat logs.
As a consequence, we were unable to examine all demographic
differences
and moderator variables together in the same regression
equation. Although
the small sample may reduce confidence in the nonsignificant
findings, it
does make us more confident about the significant effects we
found. In addi-
tion, it is possible that the context in which our study was
conducted (an MBA
program) may have made differences in age salient. However,
given that the
participants on average had more than 5 years of work
experience, this con-
cern is mitigated to some extent. But the effects we found
should be tested in
other samples, in particular in field settings, to establish
broader generaliz-
ability of the findings.

66. Since the effects of demographic diversity can be different
depending on
time and the type of task, future research should examine these
effects using
different types of tasks that vary in complexity and need for
consensus, as
well as examine these effects over the entire lifecycle of a team.
Also, we
focused on short-lived virtual interactions, which can contribute
to our under-
standing of short-term virtual teams as well as early interactions
in longer-
term virtual teams. On average, virtual teams tend to have a
shorter lifecycle
than face-to-face teams (Jarvenpaa & Leidner, 1999); therefore,
our findings
should apply well to the average virtual team. However, future
research
should examine longer lifecycle virtual teams that will enable
an examina-
tion of different stages in a team’s lifecycle; such an
examination is impor-
tant because how demographic differences affect outcomes may
vary by the
amount of time that a team has spent working together (Harrison
et al., 1998).
In addition, we looked at only two-party virtual collaborations.
If more peo-
ple were involved, the need for effective processes would be
expected to be
even stronger, but this remains an empirical question. Finally, it
should be
pointed out that we examined the effect of demographic
differences on cre-
ativity for CMC exclusively. Although this medium represents a
large part of

67. virtual work, other ways of working virtually also need to be
examined.
Martins and Shalley 555
This study contributes to research and practice in several ways.
It is one
of the first studies to examine how process and input factors
influence the
relationship between demographic differences and creativity in
a virtual
work context. As such, it contributes to research on creativity,
virtual work,
and diversity. Shalley, Zhou, and Oldham (2004), in their
integration of the
creativity literature, called for more research on team creativity
since prior
research on creativity had tended to focus on individual
creativity, with
only a few studies having examined team creativity (Gilson &
Shalley,
2004; Taggar, 2002). Furthermore, since most employees are
now working
virtually, at least to some extent in their collaboration with
coworkers on
projects (Griffith & Neale, 2001), research is needed that
explores what
aspects are most important for creativity in virtual teams
(Martins et al.,
2004). In this study we start to address this topic by providing
insights into
how demographic differences may affect the performance of
team members
working together virtually on a problem-solving task. In

68. addition, our study
contributes to developing an understanding of the circumstances
(i.e., mod-
erators) that enable demographically different coworkers to
overcome
interaction difficulties resulting from their differences and from
the limita-
tions of virtual collaboration, and consequently, to improve the
quality of
their performance.
Our findings indicate that demographic differences can be
effectively
used to tease out creative contributions as long as organizations
focus on
important team input and process factors. Given that working
virtually
requires a certain level of technical expertise, attention should
be paid on the
front end to making sure that employees are comfortable with
the technology
and can easily use it to interact with others in their team. This is
particularly
important, as our findings indicate that for virtual teams that are
working
across national boundaries differences in technical abilities may
cause inter-
action difficulties that worsen the interaction difficulties teams
face while
working across nationalities. Also, developing routines that
encourage the
formation of rapport early on in virtual interactions may benefit
performance
when there are large age differences between individuals
working together
virtually. This could be done by encouraging employees to

69. initially make
time in a virtual interaction to chat and get to know each other.
Finally,
knowledge of our results can lead to the design of process
interventions to
improve creativity. For example, when individuals work
virtually, managers
should pay attention to facilitating the process by providing
training up front
in communication and process management skills. Overall, our
findings also
suggest that developing more elaborate research models is
necessary in order
to better understand the dynamics and outcomes of virtual
teams.
556 Small Group Research 42(5)
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interests with
respect to the authorship
and/or publication of this article.
Financial Disclosure/Funding
The author(s) received no financial support for the research
and/or authorship of this
article.
References
Agarwal, R., & Prasad, J. (1999). Are individual differences
germane to the acceptance

70. of new information technologies? Decision Sciences, 30, 361-
391. doi:10.1111/
j.1540-5915.1999.tb01614.x
Aiken, L., West, S., & Reno, R. (1991). Multiple regression:
Testing and interpreting
interactions. Newbury Park, CA: SAGE.
Allport, G. (1954). The nature of prejudice. Reading, MA:
Addison-Wesley.
Amabile, T. M. (1988). A model of creativity and innovation in
organizations. In
B. M. Staw & L. L. Cummings (Eds.), Research in
organizational behavior (Vol. 10,
pp. 123-167). Greenwich, CT: JAI Press.
Amabile, T. M. (1996). Creativity in context. Boulder, CO:
Westview.
Ancona, D. G., & Caldwell, D. F. (1992). Demography and
design: Predictors of
new product team performance. Organization Science, 3, 321-
341. doi:10.1287/
orsc.3.3.321
Burton-Jones, A., & Hubona, G. (2005). Individual differences
and usage behavior:
Revisiting a technology acceptance model assumption. ACM
SIGMIS Database,
36(2), 58-77. doi:10.1145/1066149.1066155
Byrne, D. E. (1971). The attraction paradigm. San Diego, CA:
Academic Press.
Coutu, D. L. (1998). Trust in virtual teams. Harvard Business
Review, 76, 20-21.

71. Dew, R., & Hearn, G. (2009). A new model of the learning
process for innovation
teams: Networked nominal pairs. International Journal of
Innovation Manage-
ment, 13, 521-535. doi:10.1142/S136391960900239X
Edmondson, A. (1999). Psychological safety and learning
behavior in work teams.
Administrative Science Quarterly, 44, 350-383.
doi:10.2307/2666999
Farmer, S. M., & Hyatt, C. W. (1994). Effects of task language
demands and task
complexity on computer-mediated work groups. Small Group
Research, 25, 331-366.
doi:10.1177/1046496494253001
Ferris, G. R., Judge, T. A., Chachere, J. G., & Liden, R. C.
(1991). The age context
of performance-evaluation decisions. Journal of Applied
Psychology, 6, 616-622.
Gefen, D., & Straub, D. W. (1997). Gender differences in the
perception and use
of e-mail: An extension to the technology acceptance model.
MIS Quarterly, 21,
389-400. doi:10.2307/249720
Martins and Shalley 557
Giambatista, R. C., & Bhappu, A. D. (2010). Diversity’s
harvest: Interactions of diver-
sity sources and communication technology on creative group

72. performance. Orga-
nizational Behavior and Human Decision Processes, 111, 116-
126. doi:10.1016/
j.obhdp.2009.11.003
Gilson, L. L., & Shalley, C. E. (2004). A little creativity goes a
long way: An exami-
nation of teams’ engagement in creative processes. Journal of
Management, 30,
453-470. doi:10.1016/j.jm.2003.07.001
Griffith, T. L., & Neale, M. A. (2001). Information processing
in traditional, hybrid,
and virtual teams: From nascent knowledge to transactive
memory. Research in
Organizational Behavior, 23, 379-421.
Griffith, T. L., Sawyer, J. E., & Neale, M. A. (2003).
Virtualness and knowledge in
teams: Managing the love triangle of organizations, individuals,
and information
technology. MIS Quarterly, 27, 265-287. doi:10.2307/30036531
Hansen, M. T., Nohria, N., & Tierney, T. (1999). What’s your
strategy for managing
knowledge? Harvard Business Review, 77(2), 106-116.
Hargadon, A. B., & Bechky, B. A. (2006). When collections of
creatives become creative
collectives: A field study of problem solving at work.
Organization Science, 17,
484-500. doi:10.1287/orsc.1060.0200
Harrison, D. A., Price, K. H., & Bell, M. P. (1998). Beyond
relational demography:
Time and the effects of surface- and deep-level diversity on

73. work group cohesion.
Academy of Management Journal, 41, 96-107.
Hoffman, L. R., & Maier, N. R. F. (1961). Quality and
acceptance of problem solu-
tions by members of homogeneous and heterogeneous groups.
Journal of Abnor-
mal and Social Psychology, 62, 401-407. doi:10.1037/h0044025
Jackson, S. E., & Joshi, A. (2004). Diversity in social context:
A multi-attribute, mul-
tilevel analysis of team diversity and sales performance. Journal
of Organiza-
tional Behavior, 25, 675-702. doi:10.1002/job.265
Jarvenpaa, S. L., & Leidner, D. E. (1999). Communication and
trust in global virtual
teams. Organization Science, 10, 791-815.
doi:10.1287/orsc.10.6.791
Jehn, K. A., & Mannix, E. A. (2001). The dynamic nature of
conflict: A longitudi-
nal study of intragroup conflict and group performance.
Academy of Management
Journal, 44, 238-251. doi:10.2307/3069453
Kayworth, T., & Leidner, D. (2000). The global virtual
manager: A prescription
for success. European Management Journal, 18, 183-194.
doi:10.1016/S0263-
2373(99)00090-0
Kruempel, K. (2000). Making the right (interactive) moves for
knowledge-producing
tasks in computer-mediated groups. IEEE Transactions on
Professional Commu-

74. nication, 43, 185-195. doi:10.1109/47.843645
Lancaster, L. C., & Stillman, D. (2002). When generations
collide. San Francisco,
CA: Jossey-Bass.
558 Small Group Research 42(5)
Magadley, W., & Birdi, K. (2009). Innovation labs: An
examination into the use of
physical spaces to enhance organizati onal creativity. Creativity
and Innovation
Management, 18, 315-325. doi:10.1111/j.1467-
8691.2009.00540.x
Majchrzak, A., Malhotra, A., Stamps, J., & Lipnack, J. (2004).
Can absence make a
team grow stronger? Harvard Business Review, 82(5), 131-137.
Martins, L. L., Gilson, L. L., & Maynard, M. T. (2004). Virtual
teams: What do
we know and where do we go from here? Journal of
Management, 30, 805-835.
doi:10.1016/j.jm.2004.05.002
Maznevski, M. L., & Chudoba, K. M. (2000). Bridging space
over time: Global
virtual team dynamics and effectiveness. Organization Science,
11, 473-492.
doi:10.1287/orsc.11.5.473.15200
McDonough, E. F., Kahn, K. B., & Barczak, G. (2001). An
investigation of the use of
global, virtual, and colocated new product development teams.

75. Journal of Product
Innovation Management, 18, 110-120. doi:10.1111/1540-
5885.1820110
McLeod, P. L., & Lobel, S. A. (1992). The effects of ethnic
diversity on idea generation
in small groups. Academy of Management Best Paper
Proceedings, pp. 227-231.
Milliken, F. J., Bartel, C. A., & Kurtzberg, T. (2003). Diversity
and creativity in work
groups: A dynamic perspective on the affective and cognitive
processes that link
diversity and performance. In P. Paulus & B. Nijstad (Eds.),
Group creativity
(pp. 32-62). New York, NY: Oxford University Press.
Milliken, F. J., & Martins, L. L. (1996). Searching for common
threads: Understand-
ing the multiple effects of diversity in organizational groups.
Academy of Manage-
ment Review, 21, 402-433. doi:10.2307/258667
Montoya-Weiss, M. M., Massey, A. P., & Song, M. (2001).
Getting it together: Tem-
poral coordination and conflict management in global virtual
teams. Academy of
Management Journal, 44, 1251-1262. doi:10.2307/3069399
Nemeth, C. J. (1986). Differential contributions of majority and
minority influence.
Psychological Review, 93(1), 23-32. doi:10.1037/0033-
295x.93.1.23
Nemiro, J. E. (2002). The creative process in virtual teams.
Creativity Research Journal,

76. 14(1), 69-83. doi:10.1207/S15326934CRJ1401_6
Nowak, K. L. (2003). Sex categorization in computer mediated
communication
(CMC): Exploring the utopian promise. Media Psychology, 5,
83-103. doi:10.1207/
S1532785XMEP0501_4
Ocker, R. J. (2005). Influences on creativity in asynchronous
virtual teams: A qualita-
tive analysis of experimental teams. IEEE Transactions on
Professional Commu-
nication, 48(1), 22-39. doi:10.1109/TPC.2004.843294
Paulus, P. (2000). Groups, teams, and creativity: The creative
potential of idea-
generating groups. Applied Psychology, 49, 237-262.
doi:10.1111/1464-0597.00013
Payne, R. (1990). The effectiveness of research teams: A
review. In M. West &
J. Farr (Eds.), Innovation and creativity at work: Psychological
and organiza-
tional strategies (pp. 101-122). New York, NY: John Wiley.
Martins and Shalley 559
Pelled, L. H., Eisenhardt, K. M., & Xin, K. R. (1999). Exploring
the black box: An
analysis of work group diversity, conflict, and performance.
Administrative Sci-
ence Quarterly, 44, 1-28. doi:10.2307/2667029
Pirola-Merlo, A., & Mann, L. (2004). The relationship between

77. individual creativity
and team creativity: Aggregating across people and time.
Journal of Organiza-
tional Behavior, 25, 235-257. doi:10.1002/job.240
Sarker, S., Lau, F., & Sahay, S. (2001). Using an adapted
grounded theory approach
for inductive theory building about virtual team development.
Database for
Advances in Information Systems, 32(1), 38-56.
doi:10.1145/506740.506745
Sarker, S., & Sahay, S. (2002). Information systems
development by US-Norwegian
virtual teams: Implications of time and space. Proceedings of
the thirty-fifth
annual Hawaii international conference on system sciences
(January), pp. 1-10.
Saunders, C. S. (2000). Virtual teams: Piecing together the
puzzle. In R. W. Zmud
(Ed.) Framing the domain of IT management: Projecting the
future through the
past, pp.29-50. Cincinnati, OH: Pinnaflex.
Shalley, C. E. (1991). Effects of productivity goals, creativity
goals, and personal
discretion on individual creativity. Journal of Applied
Psychology, 76, 179-185.
doi:10.1037/0021-9010.76.2.179
Shalley, C. E. (1995). Effects of coaction, expected evaluation,
and goal setting
on creativity and productivity. Academy of Management
Journal, 38, 483-503.
doi:10.2307/256689

78. Shalley, C. E., & Perry-Smith, J. E. (2001). Effects of social-
psychological factors on
creative performance: The role of informational and controlling
expected evalu-
ation and modeling experience. Organizational Behavior and
Human Decision
Processes, 84, 1-22. doi:10.1006/obhd.2000.2918
Shalley, C. E., & Perry-Smith, J. E. (2008). The emergence of
team creative cogni-
tion: The role of diverse outside ties, socio-cognitive network
centrality, and team
evolution. Strategic Entrepreneurship Journal, 2, 23-41. doi:
10.1002/sej
Shalley, C. E., Zhou, J., & Oldham, G. R. (2004). The effects of
personal and con-
textual characteristics on creativity: Where should we go from
here? Journal of
Management, 30, 933-958. doi:10.1016/j.jm.2004.06.007
Shore, L. M., Cleveland, J. N., & Goldberg, C. B. (2003). Work
attitudes and deci-
sions as a function of manager age and employee age. Journal of
Applied Psychol-
ogy, 88, 529-537. doi:10.1037/0021-9010.88.3.529
Sproull, L., & Kiesler, S. (1986). Reducing social context cues:
Electronic mail in
organizational communication. Management Science, 32, 1492-
1512. doi:10.1287/
mnsc.32.11.1492
Straus, S. G., & McGrath, J. E. (1994). Does the medium
matter? The interaction of

79. task type and technology on group performance and member
reactions. Journal of
Applied Psychology, 79, 87-97. doi:10.1037/0021-9010.79.1.87
560 Small Group Research 42(5)
Stroebe, W., & Diehl, M. (1994). Why groups are less effective
than their members:
On productivity losses in idea-generating groups. European
Review of Social Psy-
chology, 5, 271-303. doi:10.1080/14792779543000084
Suchan, J., & Hayzak, G. (2001). The communication
characteristics of virtual teams:
A case study. IEEE Transactions on Professional
Communication, 44, 174-186.
doi:10.1109/47.946463
Taggar, S. (2002). Individual creativity and group ability to
utilize individual creative
resources: A multilevel model. Academy of Management
Journal, 45, 315-330.
doi:10.2307/3069349
Tajfel, H. (1981). Human groups and social categories: Studies
in social psychology.
Cambridge, UK: Cambridge University Press.
Townsend, A. M., DeMarie, S. M., & Hendrickson, A. R.
(1998). Virtual teams:
Technology and the workplace of the future. Academy of
Management Executive,
12(3), 17-29.

80. Tsui, A. S., & O’Reilly, C. A., III. (1989). Beyond simple
demographic effects: The
importance of relational demography in superior-subordinate
dyads. Academy of
Management Journal, 32, 402-423. doi:10.2307/256368
Tsui, A. S., Xin, K. R., & Egan, T. D. (1995). Relational
demography: The missing
link in vertical dyad linkage. In S. E. Jackson & M. N.
Ruderman (Eds.), Diver-
sity in work teams: Research paradigms for a changing
workplace (pp. 97-129).
Washington, DC: American Psychological Association.
van Knippenberg, D., De Dreu, C. K. W., & Homan, A. C.
(2004). Work group diver-
sity and group performance: An integrative model and research
agenda. Journal of
Applied Psychology, 89, 1008-1022. doi:10.1037/0021-
9010.89.6.1008
van Knippenberg, D., & Schippers, M. C. (2007). Work group
diversity. Annual Review
of Psychology, 58, 515-541.
doi:10.1146/annurev.psych.58.110405.085546
Watson, W. E., Kumar, K., & Michaelsen, L. K. (1993).
Cultural diversity’s impact
on interaction process and performance: Comparing
homogeneous and diverse
task groups. Academy of Management Journal, 36, 590-602.
Webber, S. S., & Donahue, L. M. (2001). Impact of highly and
less job-related diversity
on work group cohesion and performance: A meta-analysis.
Journal of Manage-

81. ment, 27, 141-162. doi:10.1177/014920630102700202
Williams K. Y., & O’Reilly, C. A. (1998). Demography and
diversity in organiza-
tions: A review of 40 years of research. In B. M. Staw & L. L.
Cummings (Eds.),
Research in organizational behavior (Vol. 20, pp. 77-140).
Greenwich, CT: JAI
Press.
Woodman, R. W., Sawyer, J. E., & Griffin, R. W. (1993).
Toward a theory of organiza-
tional creativity. Academy of Management Review, 18, 293-
321. doi:10.2307/258761
Zakaria, N., Amelinckx, A., & Wilemon, D. (2004). Working
together apart? Build-
ing a knowledge-sharing culture for global virtual teams.
Creativity & Innovation
Management, 13, 15-29. doi:10.1111/j.1467-8691.2004.00290.x
Martins and Shalley 561
Zhou, J. (1998). Feedback valence, feedback style, task
autonomy, and achievement
orientation: Interactive effects on creative performance. Journal
of Applied Psy-
chology, 83, 261-276. doi:10.1037/0021-9010.83.2.261
Bios
Luis L. Martins is an associate professor at the McCombs
School of Business, Uni-
versity of Texas at Austin, USA. He received his PhD from the

82. Stern School of
Business, New York University. His research examines the
dynamics of diversity,
particularly in the context of virtual teams and global virtual
work.
Christina E. Shalley is the Thomas R. Williams-Wachovia and
ADVANCE profes-
sor at the College of Management at the Georgia Institute of
Technology, USA. She
received her PhD in business administration from the University
of Illinois, Urbana–
Champaign. Her research examines the effects of social and
contextual factors on
creativity and innovation.
Research Alternative Assignment:
This option involves writing a review and critique of an
academic research article. A review should include:
· A statement in your own words of what you thought the author
hoped to discover.
· A brief description of the research approach that was used
including the nature of the sample, the way key variables were
measured, and the type of analysis used.
· A summary of the key results of the study.
· A description of what you think the results of the article
suggest for how organizations and/or their managers might
change the way they operate in the future.
· Your overall evaluation of the article. What did you like about
the article? Why? What did you not like? Why?