A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Personality and Social Psychology
This content is subject to copyright. Terms and conditions apply.
Attitudes, Intentions,
and
Behavior:
A
Test
of
Some
Key
Hypotheses
Richard
P.
Bagozzi
Massachusetts
Institute
of
Technology
A
causal modeling methodology
was
used
to
test hypotheses concerning
the
attitude-behavior
relation
in the
context
of a
longitudinal
field
study
of 157
students, faculty,
and
staff. Using measures
of
actual blood donation behavior
at two
points
in
time
as
criteria,
we
found
that
attitudes
influence
behavior
but
do so in an
indirect manner only through their impact
on
intentions,
as
Fishbein
has
long argued. Further,
the
attitude-intentions relationship
was
stronger when
expectancy-value attitude measures were used
as
predictors than when semantic
differential
measures were employed.
However,
for
both attitudinal operation-
alizations,
the
attitude-intention relationship
was
attenuated when
the
extent
of
past behavior
was
included
as an
explanatory variable. Similarly, past behavior
tended
to
lessen
the
impact
of
intentions
on
behavior,
as
Triandis
has
suggested.
With respect
to the
organization
of
attitude,
the
expectancy-value attitude
was
found
to
occur
as a
complex, multidimensional representation, whereas
the se-
mantic
differential
attitude existed
as a
unidimensional response. Finally,
we
examined
a
series
of
additional hypotheses
on the
role
of
correspondence
in
measures,
the
operationalization
of
probabilistic
and
nonprobabilistic intentions,
and
the
paths that attitude, intentions,
and
past behavior take
in
their influence
on
subsequent proximal
and
distal behaviors. Overall,
the
results show that
at-
titude, intentions,
and
behavior articulate
in
somewhat different ways than studies
heretofore have suggested.
The
question
of
whether attitude
and be-
havior
are
causally related
has
been
ad-
dressed
in at
least
two
broad ways.
The first
begins
with
the
premise that
no
necessary
lawlike
connection exists
at
all. Rather,
the
attitude-behavior
relation
is
believed
to be
contingent
on the
context
in
which
it
occurs.
Thus,
among other
findings, the
relation
has
been
found
to
depend
on the
stability
of at-
titudes
(Schwartz,
1979);
the
extent
of
prior
experience
with
the
focal
behavior (Fazio
&
Zanna, 1978a; Regan
&
Fazio,
1977);
the
degree
of
confidence
in
one's attitude (Fazio
The
author
would
like
to
express
his
appreciation
to
the
reviewers
for
comments
made
on an
earlier
version
of
this article,
to D.
MacN.
Surgenor
and R. A.
Boyd
of
the Red
Cross
for the
opportunity
to
conduct
the
research,
and to C.
Goodrich,
W.
Mates,
and P.
Ness
for
assistance
in
data collection.
Requests
for
reprints should
be
sent
to
Richard
P.
Bagozzi,
Sloan School
of
Management, Massachusetts
Institute
of
Technology, E53-339, Cambridge, Massa-
chusetts
02139.
&
Zanna, 1978b);
the
consistency between
affective
and
cognitive responses (Norman,
1975);
and the
interaction between
felt
norms
and
attitudes (Acock
&
DeFleur,
1972;
Andrews
&
Kandel,
1979; Susmilch,
Elliot,
&
Schwartz, 1975; Warner
& De-
Fleur, 1969).
The
alternative perspective begins
with
the
assumption that attitude
and
behavior
are
causally
linked
through nomothetic
mechanisms
but
that previous failures
to
find
such
a
link
have been largely
due to
improper
specification
and
measurement
of
both
variables.
In
this regard, Fishbein
and
Ajzen
identify
three essential aspects
of the
attitude-behavior relation.
The first
con-
cerns
the
parallelism between attitude
and
behavior.
Specifically, their review
of the
lit-
erature reveals that "strong attitude-behav-
ior
relations
are
obtained
only
under high
correspondence between
at
least
the
target
and
action elements
of the
attitudinal
and
Journal
of
Personality
and
Social Psychology,
1981,
Vol.
41,
No. 4,
607-627
Copyright
1981
by
the
American
Psychological Association, Inc.
0022-3514/81/4I04-0607$00.75
607
