Productivity Loss In Brainstorming Groups: Toward the Solution of a Riddle

Abstract

We conducted four experiments to investigate free riding, evaluation apprehension, and production blocking as explanations of the difference in brainstorming productivity typically observed between real and nominal groups. In Experiment 1, we manipulated assessment expectations in group and individual brainstorming. Although productivity was higher when subjects worked under personal rather than collective assessment instructions, type of session still had a major impact on brainstorming productivity under conditions that eliminated the temptation to free ride. Experiment 2 demonstrated that inducing evaluation apprehension reduced productivity in individual brainstorming. However, the failure to find an interaction between evaluation apprehension and type of session in Experiment 3 raises doubts about evaluation apprehension as a major explanation of the productivity loss in brainstorming groups. Finally, by manipulating blocking directly, we determined in Experiment 4 that production blocking accounted for most of the productivity loss of real brainstorming groups. The processes underlying production blocking are discussed, and a motivational interpretation of blocking is offered. (PsycINFO Database Record (c) 2012 APA, all rights reserved)

Full text

Journal
of
Personality
and
Social
Psychology
1987,
Vat.
53, No.
3.497-509
Copyright
1987
by the
American
Psychological
Association,
Inc.
0022-3514/87/500.75
Productivity
Loss
In
Brainstorming
Groups:
Toward
the
Solution
of a
Riddle
Michael
Diehl
and
Wolfgang
Strpebe
Universitat
Tubingen,
Tubingen,
Federal Republic
of
Germany
We
conducted
four
experiments
to
investigate
free
riding,
evaluation
apprehension,
and
production
blocking
as
explanations
of the
difference
in
brainstorming
productivity
typically
observed
between
real and
nominal
groups.
In
Experiment
1, we
manipulated
assessment
expectations
in
group
and
individual
brainstorming.
Although
productivity
was
higher
when
subjects
worked
under
personal
rather
than
collective
assessment
instructions,
type
of
session
still
had a
major
impact
on
brainstorm-
ing
productivity
under
conditions
that
eliminated
the
temptation
to
free
ride.
Experiment
2
demon-
strated
that
inducing
evaluation
apprehension
reduced
productivity
in
individual
brainstorming.
However,
the
failure
to find an
interaction
between
evaluation
apprehension
and
type
of
session
in
Experiment
3
raises
doubts
about
evaluation
apprehension
as a
major
explanation
of the
productiv-
ity
loss
in
brainstorming
groups.
Finally,
by
manipulating
blocking
directly,
we
determined
in
Exper-
iment
4
that
production
blocking
accounted
for
most
of the
productivity
loss
of
real
brainstorming
groups.
The
processes
underlying
production
blocking
are
discussed,
and a
motivational
interpreta-
tion
of
blocking
is
offered.
In
his
influential
book,
Osborn
(1957)
suggested brainstorm-
ing
as a
method
of
group problem solving that considerably
in-
creases
the
quality
and
quantity
of
ideas produced
by
group
members. Brainstorming groups
are
traditionally
given
instruc-
tions designed
to
free
the
individual members
from
the
inhibit-
ing
effects
of
self-criticism
and the
criticism
by
others during
the
problem-solving session.
The
rules behind brainstorming
are as
follows:
keep
in
mind that
the
more ideas
the
better
and
the
wilder
the
ideas
the
better,
improve
or
combine
ideas
al-
ready
suggested;
and do not be
critical.
Osborn
(1957)
claimed
that
if
these rules
are
followed
"the
average person
can
think
up
twice
as
many ideas when
working
with
a
group than
when
working
alone"
(p.
229).
Taylor,
Berry,
and
Block
(1958)
were
the first to
test
Osborn's
claim
in a
study
in
which subjects
were
asked
to
brainstorm
for
a
period
of
12
min
either
individually
or in
4-person
groups.
To
allow
for a
statistical
comparison between
results
from
individ-
ual and
group
sessions,
nominal groups
were
formed
from
sub-
jects
who had
brainstormed
individually.
For
each nominal
group
the
ideas
of 4
subjects were combined, eliminating
re-
dundant ideas
by
counting only once
any
idea that
had
been
suggested several
times.
Thus,
the
scores
of
nominal groups
rep-
resent the
level
of
productivity
one
would expect
if
group inter-
action
neither facilitated
nor
inhibited group productivity. Con-
trary
to
Osborn's
claim,
Taylor
et
al.
found
that nominal groups
produced nearly twice
as
many
different
ideas
as the
real
groups. This
finding
has
since been
frequently
replicated.
Of
The
authors
are
indebted
to
Thomas
Ostrom,
Kenneth
Gergen,
and
Margaret
Stroebe
for
helpful
comments
on an
earlier
draft
of
this
paper.
We
are
also
grateful
to
Riidiger
Arnscheid,
Claudia
Brand),
and
Christ!
Fischer
for
their
help
in
collecting
some
of the
data.
Correspondence
concerning
this
article
should
be
addressed
to Mi-
chael
Diehl,
Psychologisches
Institut,
Universitat
Tubingen,
D-74
Tu-
bingen,
Federal
Republic
of
Germany.
the
22
experiments listed
in
Table
1,18
reported
the
perfor-
mance
of
nominal groups
to be
superior
to
that
of
real groups,
and
only
4, all
involving
2-person
groups (Cohen,
Whitmyre,
&
Funk,
1960; Pape
&
Bolle,
1984;
Torrance,
1970,
Experiments
1
and 2),
reported
no
difference.
Results
have
been more equivocal
with
regard
to
quality
of
ideas.
Of the few
studies that assessed
quality,
most
have
re-
ported
a
measure
of
lota!
quality
(i.e.,
the sum of the
quality
ratings
of the
ideas produced
by a
given
subject
or
group).
Be-
cause
the
total
quality
is
highly
related
to the
number
of
ideas,
some authors
have
preferred
to use
average
quality.
However,
as
brainstorming
is
assumed
to
increase
the
production
of
good
ideas,
the
number
of
good ideas appears
to be a
more
appropri-
ate
measure
of
quality.
Consequently,
in
these studies that
re-
ceived
a
score above
a
chosen
cutoff
point
on a
scale
of
quality
ratings
was
classified
as
"good."
Finally, some studies
have
as-
sessed
the
number
of
unique
or
original ideas,
having
used
the
frequency
with
which
the
idea
is
suggested
as a
criterion.
The findings for
quality
appear
to be
heavily
dependent
on
the
type
of
measure used:
In all six
studies that assessed
total
quality,
nominal groups performed better than real groups did.
No
consistent pattern emerged
for the
other measures. Among
those studies,
findings
were
not
only
inconsistent between stud-
ies
but
even
within
the
same
study,
if
several topics,
subject
groups,
or
experimental
conditions
had
been used.
Theories
of
Productivity
Loss
in
Brainstorming
Groups
In
view
of the
accumulation
of
evidence
for the
superior
pro-
ductivity
of
nominal groups,
at
least
in
terms
of the
quantity
of
ideas produced,
it is
surprising that
the
reasons
for
their superi-
ority
have
so far not
been
explained.
The
three major
interpre-
tations that
have
been
offered
to
account
for the
lower produc-
tivity
of
real groups
are
production
blocking,
evaluation
appre-
hension,
and
free
riding.
In the first
part
of
this
article,
we
discuss these interpretations
in
light
of
existing
evidence.
In the
497
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.