teamwork 4955

dominated by an inputprocessoutput model,
mainly because of its simplicity and utility.
Inputs include the task of the team, team com
position (size, functional and demographic
diversity, tenure), and organizational context
(such as culture, support for team working,
structure). Some processes mediate the relation
ships between inputs and outputs such as parti
cipation mediating the effects of diversity upon
innovation, while some inputs such as organiza
tional context directly influence outputs. Pro
cesses include participation (influence over
decision making, interactions, and information
sharing), leadership, conflict, decision making,
interteam processes, and reflexivity. Team out
puts include productivity, innovation, team
member well being, and team learning.

Inputs to Teams

The team task. The task a team performs is a
fundamental influence on the work team, defining
its structural, process, and functional require
ments who is in the team, what their roles are,
how they should work together, and the nature
and processes of the tasks they individually and
collectively perform.
Dimensions for classifying task characteris
tics include task difficulty; solution multiplicity;
intrinsic interest; cooperative requirement tasks
which are unitary versus divisible, conjunctive,
disjunctive, and additive; conflict versus coopera
tion elements; and conceptual versus behavioral
components. These classification systems, devel
oped by social psychologists, have not been fruit
ful for researchers exploring team performance
and innovation in organizational settings, prob
ably because such goals as producing television
programs, battleground training, health care,
product development, and providing financial
services cannot be neatly categorized into dis
crete tasks and subtasks.
Sociotechnical systems theory (STST) pro
vides a powerful framework for examining the
effects of task design upon work team innova
tion. Sociotechnical systems theorists argue that
autonomous work teams provide a structure
through which the demands of the social and
technical subsystems of an organization can be
jointly optimized. The key to effective per
formance is then whether the work team can
control variation in quality and quantity of task
performance at source. The joint optimization
of the two subsystems is more likely when work
teams have the following characteristics:
The team is a relatively independent orga
nizational unit that is responsible for a
whole task.
The tasks of members are related in content
so that awareness of a common task is
evoked and maintained and members are
required to work interdependently.
There is a unity of product and organiza
tion, i.e., the team has a complete task to
perform and team members can identify
with their own product.

The task characteristics that evoke task orien
tation or intrinsic motivation (and therefore
innovation) according to STST are:

completeness (i.e., whole tasks);
varied demands;
opportunities for social interaction;
autonomy;
opportunities for learning;
development possibilities for the task.
Team composition. Team composition used
here to refer to the mix of members making
up a team has been examined in various ways.
One examines the question of whether hetero
geneity is advantageous to groups. The theore
tical perspectives that have guided much of the
research in this area include the attraction
selectionattrition model, similarityattraction
theory, and self categorization theory. A basic
premise of all three is that we are attracted to
those who are similar to us and thus organize,
and evaluate, our social worlds accordingly. In
the second line of research, it is assumed that
heterogeneity is valuable but groups need to
have the right mix of members. This approach
questions which combination of roles, styles, or
skills fits together particularly well and which
types of people are needed within groups.
Research in this area has tended to focus on
heterogeneity in terms of demographic vari
ables, skills, attitudes, cognitive ability, and,
more recently, personality traits. Although
much of the early research on group hetero
geneity examined experimental laboratory based