Situational Coping and Coping Dispositions in a Stressful Transaction

Abstract

After reporting dispositional coping styles, students reported situational coping and 4 classes of affect (from threat, challenge, harm, and benefit appraisals) 2 days before an exam, after the exam but before grades were posted, and after posting of grades. Coping did not predict lower levels of future distress; indeed, some coping seemed to induce feelings of threat. Feelings of harm before the exam induced several kinds of coping after the exam, mostly dysfunctional. Confidence about one's grade was a better predictor of emotions throughout than was coping. Dispositional coping predicted comparable situational coping at low-moderate levels in most cases. Coping dispositions did not reliably predict emotions, however, with these exceptions: Dispositional denial was related to threat, as was dispositional use of social support; dispositional use of alcohol was related to both threat and harm.

Full text

Journal of Personality and Social Psychology
1994,
Vol. 66, No. 1, 184-195Copyright 1994 by the American Psychological Association, Inc.
0022-3514/94/53.00
Situational Coping and Coping Dispositions in a Stressful Transaction
Charles
S.
Carver and Michael
F.
Scheier
After reporting dispositional coping
styles,
students reported situational coping and
4 classes
of affect
(from threat, challenge, harm, and benefit appraisals) 2 days before an exam, after the exam but
before grades were posted, and after posting of grades. Coping did not predict lower levels of future
distress; indeed, some coping seemed to induce feelings of threat. Feelings of harm before the exam
induced several kinds of coping after the exam, mostly dysfunctional. Confidence about one's grade
was a
better predictor of emotions throughout than
was
coping.
Dispositional coping predicted com-
parable situational coping at low-moderate levels in most
cases.
Coping dispositions did not reliably
predict emotions, however, with these exceptions: Dispositional denial was related to threat, as was
dispositional use of social support; dispositional use of alcohol was related to both threat and harm.
How people cope with difficult or stressful circumstances in
their lives has been the subject of
a
considerable amount of
re-
search over the past decade. Most working in this area have
taken as their conceptual point of departure the model of stress
and coping developed by Lazarus and his colleagues
(e.g.,
Laza-
rus,
1966; Lazarus & Folkman, 1984). In this view, stress con-
sists of
two
appraisal processes (appraisal of threat, challenge,
or loss and appraisal of
how
to respond) plus the execution of
coping responses. Coping is sometimes divided into problem-
focused
coping—efforts
to remove the threatening event or di-
minish its impact—and
emotion-focused coping—efforts
to re-
duce the negative feelings that arise in response to the threat.
Although these two categories are easily distinguished in prin-
ciple, they typically co-occur. Indeed, their effects can be diffi-
cult to disentangle. That is, emotion-focused coping can facili-
tate problem-focused coping by removing some of the distress
that can hamper problem-focused efforts; similarly, problem-
focused coping can render the threat less forbidding, thereby
diminishing distress emotions.
Not all responses to stress produce desired results, of course,
and the idea that some coping responses are dysfunctional has
also received scrutiny. Several responses have been investigated
in that regard, including self-blame (Bolger, 1990; McCrae &
Costa, 1986), wishful thinking
(Bolger,
1990;Felton, Revenson,
& Hinrichsen, 1984; Folkman & Lazarus, 1985), escapism
(Rohde, Lewinsohn, Tilson, & Seeley, 1990), overt efforts to
deny the stressor's reality (Carver et al., 1993), self-distraction
or mental disengagement (Carver, Scheier,
&
Weintraub, 1989),
and even behavioral disengagement, or giving up on the goals
Charles S. Carver, Department of Psychology, University of Miami;
Michael F. Scheier, Department of Psychology, Carnegie Mellon Uni-
versity.
Data collection was supported by National Science Foundation
(NSF) Grants BNS87-17783 and BNS87-06271. We thank Patricia
Wick for her collaboration in obtaining the data. Preparation of the
manuscript was facilitated by NSF Grants BNS90-11653 and BNS90-
10425,
American Cancer Society Grant PBR-56, and National Heart,
Lung, and Blood Institute Grant HL44436-01
Al.
Correspondence concerning the article should be addressed to
Charles
S. Carver, Department of Psychology, University of Miami,
Coral Gables, Florida 33124-2070.
with which the stressor is interfering (Carver et al., 1989). The
cumulative evidence indicates that these "avoidance" types of
coping typically work against people rather than to their advan-
tage (see
also Aldwin
&
Revenson,
1987;
Billings
&
Moos,
1984;
Cronkite & Moos, 1984; Felton et al., 1984; Folkman & Laza-
rus,
1985;
Holahan& Moos,
1985;
Rohde etal., 1990; Vaillant,
1977;
Wills, 1986).
Indeed, the literature on coping seems as a whole to be more
informative about coping that interferes with good outcomes
than about coping that facilitates good outcomes. This
has
even
led some to question whether coping truly does have positive
effects (e.g., Aldwin & Revenson, 1987). This remains an im-
portant question. A picture in which only "errors" in coping
matter, because dysfunctional responses have adverse conse-
quences, would differ greatly from a picture in which effective
coping tactics played a positive role. Although there does exist
evidence that benefits can accrue from task focus (Aldwin &
Revenson,
1987;
Aspinwall
&
Taylor,
1992;
Glyshaw, Cohen, &
Towbes, 1989) and from positive refraining (Menaghan &
Merves, 1984), this evidence is sparse. There are far more find-
ings of positive associations between avoidance coping and dis-
tress.
It is important to gain additional evidence on this question,
perhaps looking at a broader
range
of coping responses than has
been looked at in previous studies. That was one goal of the
study reported here. To assess coping reactions, we used the
COPE Inventory (Carver et al., 1989), which measures several
distinct aspects of both problem-focused and emotion-focused
coping. This instrument assesses several facets of coping that
have not been studied in previous research, and it assesses both
reactions that presumably are helpful and reactions that appear
to be potentially maladaptive.
Ways of Studying Coping
Several more issues should be touched on concerning the
question of how to study the effects of coping. In its early stages,
research on coping tended to use a paradigm in which subjects
were asked to bring to mind the most stressful event of a partic-
ular period of
time
and then indicate the extent to which they
had had a series of responses (e.g., Folkman & Lazarus, 1980;
Folkman, Lazarus, Dunkel-Schetter, DeLongis, &Gruen, 1986;
184

COPING
185
Parkes, 1984). This paradigm, though a useful starting point,
has several problems. First, the stressor differs from person to
person, making it hard to control the characteristics of the event
with which subjects were coping.
Second, in this paradigm, subjects indicate their coping re-
sponses to the situation as a whole (or to some portion of the
event, which may vary among
persons—cf.
Stone, Greenberg,
Kennedy-Moore, & Newman, 1991). Yet Lazarus has repeat-
edly emphasized that coping is a process, a transaction between
person and event that plays out across time and changing cir-
cumstances. Coping reactions that are relevant at one phase of
the transaction may be used less, or may even have different
effects if used, at a different phase of the transaction. The tem-
poral aspect of the stressful transaction is not easy to control
using this paradigm.
Specific Stressors
and
Prospective Prediction
One way of dealing with these two problems is to choose a
particular stressful event to which many individuals are natu-
rally exposed and to examine distinct phases of the unfolding
encounter. An event that is in some ways ideal for this is a major
exam. Exams are evaluative situations that have the possibility
for negative outcomes (threat) as well as positive ones (chal-
lenge).
Although the stress involved is surely not on a par with
the stresses of natural disasters or life-threatening illnesses, the
exam does embody elements of many life stressors. There is
preparation for the event, the event
itself,
a period of uncer-
tainty about the outcome, and a period of dealing with the out-
come. Because an exam can be simultaneous for all reseach sub-
jects,
it permits relatively easy study of all phases of the coping
process. These characteristics led Folkman and Lazarus (1985)
to use an exam as a target event, and others have done the same
(Bolger, 1990; Bolger & Eckenrode, 1991; Roberts & Monroe,
1992;
Smith & Ellsworth, 1987).
The Folkman and Lazarus (1985) study may be viewed as the
prototype of this research. That study found that subjects did
indeed differ in their coping reactions and in their emotions
across the phases of the transaction. It also revealed a number
of relationships between such variables as the individual's per-
sonal stakes in the event and the emotions experienced at vari-
ous phases. The report of that study had a serious limitation,
however, regarding another
issue.
Folkman and Lazarus did not
examine how coping and emotions were interrelated across the
phases of the study. That is, although it was possible that (for
example) problem-focused coping at one phase led to lower lev-
els of negative emotion at the next phase, this possibility was
not assessed. The information that study provided about how
coping relates to emotions was limited to concurrent regres-
sions.
It is highly desirable, however, to have information about
the prospective effects of coping as
well,
because only that infor-
mation can provide clues about causality.
Tests of prospective effects of coping are more rare than many
people may assume. Furthermore, many studies that have used
prospective designs have examined coping with life in general
during the study period rather than coping with particular
stressful events (e.g., Aldwin & Revenson, 1987; Aspinwall &
Taylor, 1992; Glyshaw, Cohen, & Towbes, 1989; Holahan &
Moos,
1987; Menaghan, 1982; Menaghan & Merves, 1984;
Smith, Patterson,
&
Grant,
1990;
Vitaliano, Russo, \bung, Teri,
& Maiuro, 1991). Although the information from such studies
is far from trivial, what is learned from such studies may well
be different from what is learned from prospective analyses of
coping in specific stressful transactions.
In studies of coping with specific stressors, the following
effects have emerged: Bolger (1990) found that reports of wish-
ful thinking were a prospective predictor of increases in anxiety
during the final anticipation of a major exam. Stanton and
Snider (1993) found that cognitive avoidance before a biopsy
predicted higher levels of distress after a positive diagnosis of
breast cancer and after
surgery.
Litt, Tennen, Affleck, and Klock
(1992) found that previous reports of escape as a coping tactic
predicted greater distress among women who experienced fail-
ure in an attempt at in vitro fertilization. Carver et al. (1993)
found that overt denial and behavioral disengagement were
prospective predictors of greater distress among breast cancer
patients and that acceptance and use of humor
were
prospective
predictors of less distress. Once again, then, the available evi-
dence indicates more adverse effects than beneficial ones.
The present research was intended in part to add to knowl-
edge concerning the prospective effects of coping in a transac-
tion with a specific stressor. The study was designed as a repli-
cation and extension of the Folkman and Lazarus (1985) study,
which examined how college students coped with the stress sur-
rounding the first exam in their introductory psychology
course. Unlike Folkman and Lazarus (1985), however, we ex-
amined prospective relations between coping and emotions, as
well as concurrent associations.
Coping Strategies
and
Coping Styles
Another issue that is relevant to the study reported here con-
cerns the difference between situational coping strategies and
dispositional coping styles. Lazarus has emphasized the role
played by various coping responses per
se,
pointing out that cop-
ing reactions can change from moment to moment across the
stages of a stressful transaction. Given this variability in peo-
ple's reactions, he has held that it may not be profitable to try
to predict situational responses from the person's typical style
of coping (e.g., Cohen & Lazarus, 1973). Indeed, the wide range
of variables that can influence coping reactions from one time
to another raises further questions about the usefulness of the
concept of coping style (for a broader discussion, see Silver &
Wortman, 1980).
On the other hand, there is reason to be concerned that clo-
sure may be reached too soon on this question (Carver et al.,
1989;
Miller, 1987). Although we agree that coping can change
from moment to moment, there is also some merit in the argu-
ment that people develop habitual ways of dealing with stress
and that these habits or coping styles can influence their reac-
tions in new situations. There are several ways coping styles
might influence situational coping. The simplest would be a
"main effect" or constant influence, in which a person who re-
ports a tendency to use active coping when under stress displays
active coping at each phase of the transaction, whether relevant
or not. There are also more subtle possibilities: A dispositional
coping style might influence situational coping at one particular
phase of a transaction but not at others.
Because there is very little evidence on this question, it seems
premature to dismiss the possibility that coping styles play a

186CHARLES
S.
CARVER
AND
MICHAEL
F.
SCHEIER
role in situational coping. Gathering evidence on this issue was
another purpose of the study reported here. Subjects reported
their dispositional coping styles, as well as their situational cop-
ing reactions. Styles were denned here not in terms of
a
prefer-
ence for one aspect of coping over others, but in terms of re-
ported tendencies
to use
each of the coping reactions
to a
greater
or lesser degree when under stress. Dispositional styles were
used to predict situational reactions, and both were used to pre-
dict other outcome variables.
Confidence, Coping, and Affect
A final issue behind the study described here concerns the
role played in a stressful transaction by confidence about one's
outcome. Several theorists have proposed that confidence of
eventual success promotes a more complete engagement in
efforts to succeed (e.g., Bandura, 1977, 1986; Carver
&
Scheier,
1983,
1990a). This view suggests that confidence should relate
to aspects of coping—especially inversely to avoidance coping.
It has also been argued that confidence and doubt are related to
the experience of affect, perhaps being products of the same
psychological mechanism (Carver
&
Scheier, 1990b). This view
suggests that confidence should be a predictor of
the
emotions
that surround the stressful transactions. Although we have in
other studies examined generalized confidence about life out-
comes (Scheier
&
Carver, 1992), we focus here instead on sub-
jects'
confidence or doubt about attaining the grade they desired
on the exam.
Method
Subjects and Procedure
Subjects were
125
undergraduates (57
men and 68
women)
at the
University
of
Miami who participated
in
partial fulfillment of a course
requirement. Each subject completed a measure of dispositional coping
styles
at
the beginning
of
the
semester. Each subject also completed
a
number of self-report scales
at
three points
in
the course of
a
stressful
transaction
(the
first exam
in
their introductory psychology course).
The timing
of
the
assessments duplicated the pattern
of
Folkman and
Lazarus (1985), occurring
at
intervals
of
exactly
1
week.
The
first
as-
sessment was
at
the end
of
a class period
2
days before the exam;
the
second was 5 days after the exam and
2
days before posting of grades;
the third
was 5 days
after the grades
were
posted.
To ensure that subjects would all have relatively high personal stakes
in
the exam (thus ensuring that the stressor would
be a
meaningful one),
the data set as analyzed was restricted to students
who
indicated before
the test that they were aiming for either an A or a
B
on the test and who
rated
the
importance
of
doing well
on the
test (beforehand)
as 3 or
higher on a
5-point
scale
(3
=
moderate
importance).
Measures
Coping
styles.
Dispositional coping styles were assessed during the
first week of the semester with
the
dispositional version
of
the COPE
Inventory (Carver
et
al., 1989). The COPE was devised
to
measure
a
wide range
of
potential responses
to
stressors
and to
distinguish each
coping quality
as well as
possible from other coping qualities. Each scale
is brief (four items) and focuses on
a
particular aspect
of
coping.
The
COPE was derived partially through
a
theoretical path,
in
that several
scales chosen for inclusion measure aspects of coping that are of partic-
ular theoretical
interest.
Each item of
the
dispositional COPE
is a
statement worded in the first
person, indicating the use of a particular coping response
(e.g.,
"I
make
a plan
of
action").
The instructions
to
this version
of
the COPE note
that different stressful events obviously bring
out
somewhat different
patterns
of
responses; however, respondents
are to
indicate how they
usually react
to
stressors they encounter. Each item
is
rated on a
4-point
scale:
(1)
/
usually don't do this
at
all,
(2)
I
usually do this a little
bit.
(3)
/
usually
do this
a
medium
amount,
or
(4)
/
usually do
this
a
lot.
The
COPE's dispositional form has good factorial properties
and
moder-
ately high test-retest reliability
(see
Carver et
al.,
1989,
for details).
Dispositional coping tendencies measured
in
this study were
active
coping,
planning,
suppression
of
competing
activities,
restraint
coping,
use
of
instrumental
support,
use
of
emotional
support,
positive
refram-
ing (previously labeled positive
reinterpretation
and
growth),
accep-
tance,
denial,
use
of
religion,
behavioral
disengagement,
and mental
dis-
engagement.
A single item was also included
to
assess habitual use
of
alcohol as
a
dispositional coping
response.1
Although most of the scales
of the COPE measure reactions that are assumed to
be
useful and adap-
tive,
we regard several
of
these
responses
as
potentially dysfunctional.
The scales that
we
would characterize in those terms are denial, mental
disengagement, behavioral disengagement, and use of alcohol.
Situational coping tactics. Subjects' coping reactions
at
various
points in the transaction were measured by the COPE in its situational
format. The instructions for this version ask respondents to indicate the
extent to which they have been engaging in each coping response during
a particular period of
time,
with respect to a particular
stressor,
or both.
The phrasing of
each
item
is
altered (e.g.,
"I
am making
a
plan of
ac-
tion"
or "I
have been making
a
plan
of
action");
the response choices
are similarly altered to indicate
a
current response
(e.g.,
"I am not doing
this at all" or "I have not been doing this at all") instead of a "typical"
response.
In
this study,
the
instructions
for the
situational COPE
re-
ferred explicitly to the exam as the relevant stressor and asked subjects
to indicate how they had been reacting "during the past few days and
presently."
Appraisal-related
emotions.
We followed the lead
of
Folkman
and
Lazarus (1985)
in
measuring the extent to which subjects experienced
several emotions that theoretically reflect the results of anticipatory and
outcome appraisals. At each measurement point, subjects rated the ex-
tent to which they currently
were
experiencing each feeling with respect
to the exam, using
a 1-5
scale.2 Two of the emotions reflect the antici-
patory appraisals
of
threat
and
challenge, respectively.
The
items
for
threat (worried, fearful, and anxious) and challenge (confident, hopeful,
and
eager) were
identical to those used by Folkman and Lazarus (1985).
The threat index had good reliability of measurement (average a
=
.75),
but reliability for challenge
was
poorer (average a
=
.51).
This should be
kept in mind in evaluating the results pertaining to challenge.
The other
two
emotions—harm and benefit—reflect appraisals of the
good and bad outcomes of
one's
coping thus far. The
items
used
to
assess
harm (angry, disappointed,
and
guilty; average
a = .72) and
benefit
1 This study was conducted using the version of the COPE published
in 1989. Since then the single item on alcohol use has been expanded to
an alcohol
and
drug use scale,
and
another scale assessing the use
of
humor
has
been developed.
We have
disregarded in this report one scale
from the 1989
article:
the scale measuring^octw
on
and
venting
of
emo-
tions. We
had
concerns that this scale, which taps both experiencing
distress and expressing distress, too closely resembles a measure of
emo-
tion per
se
to justify its inclusion here.
2 Beck Depression Inventory
and
symptom data were collected
at
each measurement point
as
well,
but
because
the
information they
added was generally redundant with that provided by the measures
of
threat and harm, they are
not
reported here.
We
also assessed feelings
of
being in
control at each point
as
a potential predictor.
The
pattern
for
this variable was generally similar to,
but
weaker than, that
for
confi-
dence. For this reason, discussion of this variable
is
also omitted.

COPING187
(pleased, happy, and relieved; average a =
.76) were
subsets of the items
used by Folkman and Lazarus {1985).
Confidence
and
other
measures.
Subjects were asked several other
questions at the various measurement points. Before the exam, they
were asked how difficult they expected it to be (on a 1-5 scale), what
grade they were honestly aiming for, how confident they were of attain-
ing the grade they were seeking (on a 1-5 scale, with complete verbal
labels),
and how important it
was
to them to do well on the
exam.
Con-
fidence and importance
were
reassessed after the exam
as
well.
Results
The data from this study make several points. In the
first
sec-
tion, we consider the overall course of emotions and coping in
the transaction. Then
we
examine relations between situational
coping and emotions, first concurrently then prospectively. Fi-
nally
we
consider
how
dispositional coping
styles
related
to
both
situational coping and emotion. To reduce the problem of Type
I error (due to the many tests reported), we adopted a signifi-
cance criterion of p
< .01
and describe
as
significant only results
that attained that level.
The
Exam as a
Stressor
The exam appears to have represented a reasonably challeng-
ing event to subjects. The average rating of expected difficulty
was 3.47 on a 1-5 scale (SD = 0.89); the importance of doing
well (the "stakes" in the outcome) averaged
4.81
on a 1-5 scale
(SD
=
0.45); moreover, reports of anticipatory anxiety averaged
just below the midpoint of
the
scale range
(8.53,
SD = 2.78), a
level of anxiety that
was
slightly higher than that found
by
either
Folkman and Lazarus (1985) or Bolger (1990) at comparable
time points. Subjects were, however, moderately confident of
attaining their desired grade (M
=
3.77, SD
=
0.75,
where
3
was
labeled /
have no idea
whether
I'll
get
the grade I
want or not
and
4
was
labeled / am fairly
sure
that I will
get
the
grade
I
want;
by
contrast
5 was
labeled I am
very sure
I will
get
the grade I
want).
The exam was difficult enough that confidence declined af-
terward to 3.35 (SD
=
1.01),
t( 124)
=
4.80,
p < .001.
Changes
in
Emotions Across
the
Transaction
Table
1
displays the levels of anticipatory and outcome emo-
tions,
as reported at the three time points. Comparison of each
set of
scores
was conducted by a repeated measures ANOVA,
followed by contrasts between pairs of measurements if the
overall F was significant. These analyses yielded the following
picture. Threat emotion was relatively high before the exam,
fell off after
the
exam,
t( 124)
=
2.65,
p<
.01,
and subsided much
more after grades were posted, /(124) =
10.24,
p
<
.0001.
Chal-
lenge emotion did not fall off immediately after the test, but fell
from postexam levels after grades were posted, /(124) = 7.35,
p
<
.0001.
Challenge emotions were at higher levels than threat
emotions throughout the transaction
(p <
.005 in each case).
The emotional qualities associated with outcome appraisals
increased rather than decreased. Harm emotions rose steadily
from before the exam onward, although only the increase from
postexam to Time 3 was significant, /(124) = 3.32, p < .001.
This increase in harm reflects the fact that many students did
not receive the grade they sought. Benefit emotion also in-
creased over the transaction. The
big change
in benefit, however,
occurred immediately after the exam was over, /(124) = 9.78,
p <
.001.
Having put aside the uncertainty about what the test
would be like and the effort of preparation, subjects overall be-
came happier. Benefit emotion remained high after grades were
posted. Taken together, these two patterns are just
as
one would
expect. Anticipatory emotions occurred mostly before the
exam, and outcome emotions occurred mostly after the exam.
Another interesting aspect of the difference between anticipa-
tory and outcome emotions concerns relations between the pos-
itive and negative affect of
each
set. Threat and challenge were
positively related at the two postexam points (rs = .34 and .29,
ps < .001). In contrast, harm and benefit emotions were in-
versely related after the exam (r = -.29, p < .001) and particu-
larly after grades
were
posted
(r =
-.60,
p
<
.0001).
This pattern
fits the idea that threat and challenge reflect anticipation of
di-
vergent future outcomes, and that both outcomes seem possi-
ble.
As a result, the emotions can co-occur. In contrast, harm
and benefit emotions reflect the sense that one or the other out-
come has already come to pass. One or the other of
these
out-
come appraisals tends to dominate, and the two outcome emo-
tions are inversely correlated.
Coping
Across
the
Transaction
Table
2
displays the means of the situational coping reactions
reported by subjects at the three measurement points. It is ap-
parent that what
we
characterized earlier as adaptive responses
were reported more commonly than those characterized as po-
tentially dysfunctional. Subjects reported relatively high levels
Table 1
Emotions
Reported
Before the
Exam,
After the
Exam, and
After Posting
of Grades
Scale
Anticipatory appraisal emotions
Threat emotions
Challenge emotions
Outcome appraisal emotions
Harm emotions
Benefit emotions
Before exam
M
8.53
9.39
4.26
5.58
SD
2.78
2.05
1.78
2.32
After exam
M
* 7.91
9.56
4.66
* 8.30
SD
3.02
2.43
2.44
2.78
After grades
M
* 5.02
* 7.38
* 5.45
8.46
SD
2.44
2.71
2.80
3.55
Note. Scale range for appraisals
is
3-15.
* Indicates significant change between measurements (p < .01).

188CHARLES S. CARVER AND MICHAEL F. SCHEIER
Table 2
Situational Coping Responses Reported Before the
Exam,
After the
Exam, and
After Posting
of Grades
COPE scale
Active Coping
Planning
Suppression of Competing Activities
Acceptance
Use of Instrumental Support
Use of Emotional Support
Restraint Coping
Religion
Positive Reframing
Denial
Mental Disengagement
Behavioral Disengagement
Use of Alcohol
Before exam
M
10.94
12.44
10.52
13.58
7.14
6.84
7.54
7.26
10.50
4.27
6.76
4.60
1.13
SD
2.64
2.94
3.00
2.63 <
2.94 i
2.68
2.60
3.45
2.78
0.84
2.15
1.06
0.44
After exam
M
* 8.70
• 9.02
• 6.74
* 12.17
* 6.22
6.94
7.78
7.42
10.84
4.58
6.89
4.85
1.13
SD
3.04
3.56
2.84
3.45
2.80
3.05
2.99
3.76
2.87
1.47
2.13
1.47
0.44
After grades
M
8.75
9.06
6.78
12.39
5.95
* 6.39
* 7.10
6.98
10.55
4.61
* 6.30
4.57
1.10
SD
3.16
3.81
2.70
3.54
2.58
2.89
3.04
3.39
3.38
1.48
2.02
1.06
0.42
Note. Scale range is from 4-16, except Use of Alcohol, which is 1
-4.
* Indicates significant change between measurements (p
<
.01).
of active coping, planning, suppression of competing activities,
positive reframing, and acceptance; they reported low levels of
denial, mental disengagement, behavioral disengagement, and
use of alcohol.
As Table 2 shows, many coping reactions changed from one
stage of the transaction to another. All of the task-focused re-
sponses—active
coping,
planning, suppression of competing
ac-
tivities, and use of instrumental support—along with accep-
tance, were high during the period before the exam, diminished
afterward, and remained stable thereafter. Declines from the
initial to the second measurement were significant for all these
variables
(ps
< .001). Several other reactions occurred before
and after the exam, declining significantly only after exam
scores were available. This pattern occurred for use of emo-
tional support, restraint, and mental disengagement
(ps <
.005).
One coping reaction tended to increase across the transaction
(although it never occurred
widely).
Specifically, reports of overt
denial were lowest before the exam, tended to increase af-
terward, and continued to drift upward to the third measure-
ment, so that the level of denial reported after grades were
posted
was
higher than that before the exam (p< .01).
The extent of change in mean levels of coping across the
stages of the transaction speaks to one kind of consistency. An-
other sort of consistency is reflected in correlations from one
stage to the next. Certain kinds of coping—use of social sup-
port, religion, and alcohol—were highly related from before the
exam to immediately after the exam (average r = .70); corre-
lations for suppression of competing activities and behavioral
disengagement were fairly low (.29 and .27); those for the other
variables
were
moderately strong (ranging from .42 to
.57,
aver-
age = .47). The transition from Time 2 to Time 3 was marked
by greater stability. With the exception of the correlation for
behavioral disengagement (.39), all exceeded .50 and they aver-
aged .66.
Associations of Situational Coping With Emotions
Before the Exam
How
did coping relate to emotions? To simplify the picture,
we
combined certain coping reactions in these analyses (further
reduction of coping variables was prevented by relatively low
correlations among scales). In particular, active coping, plan-
ning, and suppression of competing activities
were
combined as
problem-focused coping (average a =
.84);
use of the two types
of social support was also treated as a single variable (average a
= .74).
Concurrent correlations between coping and negative emo-
tions before the exam are shown in Table
3.
Threat
was
related
to problem-focused coping, to use of
social
support, and to re-
ligious activity. The other anticipatory emotion, challenge
(which
is
not shown in Table
3),
was
related
to
problem-focused
coping (r =
.30,
p
<
.01) and to positive reframing
(r
=
.31,
p <
.01).
Thus, both anticipatory emotions
were
linked to problem-
focused activity, but their other correlates differed. Correlations
between coping and reports of harm emotion also formed an
interesting pattern. Even now, feelings of harm correlated with
use of alcohol, mental disengagement, and
use
of social support.
Benefit emotion was unrelated to any coping reaction.
Table 3
Concurrent Correlations
of
Threat
and Harm Emotions With
Coping Responses
and
Appraisal Perceptions Before
Exam
COPE scale
Problem Focused Coping
Acceptance
Use of Social Support
Restraint Coping
Religion
Positive Reframing
Denial
Mental Disengagement
Behavioral Disengagement
Use of Alcohol
Confidence
Difficulty
Importance
*p<.01,
two-tailed.
**
p <
.001,
Threat
.44**
.19
.41**
.18
.26*
.20
-.01
.01
.10
.09
-.34**
.51**
.33**
two-tailed.
Harm
.05
.02
.29**
.11
.14
-.03
.17
.31**
.11
.39**
-.45**
.25*
.13

COPING
189
Table 4
Concurrent Correlations
of
Threat
and Harm
Emotions
With
Coping Responses
and
Appraisal Perceptions After
Exam
(but
Before Posting
of Grades)
COPE scale
Problem Focused Coping
Acceptance
Use of
Social Support
Restraint Coping
Religion
Positive Reframing
Denial
Mental Disengagement
Behavioral Disengagement
Use of
Alcohol
Confidence
Importance
*p<
.01,
two-tailed.
•*/><.001,
Threat
.22
-.01
.46**
.16
.19
.09
.20
.35**
.25*
.13
-.47**
.35**
two-tailed.
Harm
.27*
.03
.46**
.14
.25*
.12
.26*
.40**
.33**
.34**
-.51**
.06
We also asked subjects to indicate how confident they were
about getting the grade they wanted, how difficult they thought
the exam would be, and how important it was to them to do
well (the strongest relation between pairs of these ratings was
between difficulty and importance, r =
.33).
These ratings were
generally unrelated to coping before the exam, with few excep-
tions:
Importance of the exam (the person's stake in it) related
to problem-focused
coping
(r
=
.36,
p <
.001) as did
perceptions
that the exam would be difficult (r = .24, p < .01). Confidence
related inversely to mental disengagement (r
=
—.23,
p < .01).
These ratings related to emotions more closely than to coping.
As
shown in Table
3,
confidence related to lower levels of threat
and harm, as did the perceptions that the exam would be easy.
Importance of the exam related only to threat.
We
also conducted regression analyses to assess the extent to
which these ratings and coping account for unique variance in
the concurrent emotions. These analyses (using simultaneous
entry) used as predictors all significant correlates of the emo-
tion; unique effects are noted here only if
they
were significant
at
.01.
The analysis for challenge yielded no unique effect. For
threat, four of the six predictor variables contributed uniquely:
low confidence, perceptions that the exam would be difficult,
problem-focused coping, and use of
social
support (overall ad-
justed R2 =
.47).
In the analysis for harm, unique contributions
came from low confidence and from use of alcohol (overall ad-
justed R2
=
.33).
The outcomes of
these
analyses suggest three things: First, a
wide range of variables relate to threat emotion at this stage
of
the
transaction—and do so independently—from aspects of
coping to aspects of subjects' perceptions of
the
situation. Sec-
ond, low confidence was a robust correlate of
negative
feelings,
relating uniquely to both threat and harm emotion. Third, a
certain amount of the prediction of pre-exam feelings comes
from variance shared across predictors.
Associations
of Situational
Coping With
Emotions After
the Exam
Table 4 shows concurrent correlations between coping and
the negative emotions after the exam but before grades were
posted. At this point, significant relations emerged between
threat and both behavioral and mental disengagement. The in-
crease in correlation between threat and mental disengagement
from Time
1
to Time 2 was itself significant at .01 by Z test.
Challenge and benefit emotions were unrelated to coping after
the exam.
As at the first measurement, we asked subjects about their
current levels of confidence and how important the test was
(which was unrelated to confidence). Once again, these ratings
were unrelated to coping, except that confidence was now re-
lated inversely to use of
social
support (r = -.25, p < .01). As
was true before the exam, these ratings were related more to
emotions, especially (though no longer as exclusively) negative
emotions. Confidence related inversely to both threat and harm
emotions (Table 4) and positively to benefit emotion (r = .42,
p <
.001).
Ratings of the exam's importance
were also
positively
related to feelings of threat.
As
before the exam, multiple regressions (with simultaneous
entry) assessed unique contributions to prediction of emotions
after the exam. Confidence, importance, and use of social sup-
port contributed uniquely to prediction of threat (overall ad-
justed R2 = .40). Confidence and use of social support also
made unique contributions to predicting harm (overall ad-
justed R2
=
.43). Benefit emotion was related only to reports of
confidence (adjusted R2
=
.
17).
Results of these analyses suggest two conclusions: First, con-
fidence was again a very robust predictor of emotions, in this
case far more
so
than was any aspect of coping per
se.
Second, a
good
deal
of the contribution to prediction of
negative
emotions
at this stage comes from variance shared among predictors.
Associations
of Situational
Coping With
Emotions After
Posting
of Grades
Table
5 shows
concurrent relations between
coping
and threat
and harm emotions after grades were posted. The patterns for
threat and harm were now quite similar. Both related to prob-
lem-focused coping, use of social support, use of religion, de-
nial, and use of
alcohol.
Correlations with mental and behav-
ioral disengagement, present at Time 2, had now faded. Taken
Table 5
Concurrent Correlations
of
Threat
and Harm
Emotions
With
Coping
Responses,
Grade
Received,
and Appraisal
Perceptions After Posting
of Grades
COPE scale
Problem Focused Coping
Acceptance
Use of
Social Support
Restraint Coping
Religion
Positive Reframing
Denial
Mental Disengagement
Behavioral Disengagement
Use of
Alcohol
Grade received
Threat
.34**
-.10
.43**
.26*
.28*
.06
.42**
.10
.23*
.35**
-.34**
Harm
.32**
-.07
.28**
.01
.25*
.10
.31**
.02
.17
.24*
-.60**
*
p
<
.01,
two-tailed.
**
p
< .001,
two-tailed.

190
CHARLES S. CARVER AND MICHAEL F. SCHEIER
as
a
whole, this pattern suggests
a
split among correlates of the
negative
emotions:
Some coping responses appeared
to
reflect
a
focus
on
moving forward
and
improving future performance,
others reflected instead
a
denial
and
avoidance
of
dealing
with
the situation.
As had been true all along, significant correlations with posi-
tive feelings were scarcer. What relations
did
emerge hint that
positive affect
at
this stage was linked (albeit weakly)
to a ten-
dency to stand back and take
a
breather.
That
is,
feelings of chal-
lenge related
to
positive reframing
(r =
.27,
p < .002) and to
restraint coping
(r =
.28,
p
< .002).
Not surprisingly, the grades students received correlated with
nearly
all
emotions at this
point:
positively with benefit emotion
(r = .63,
p <
.0001)
and
inversely with threat
and
especially
harm emotion (Table
5). The
exception
was
challenge, with
which there
was no
relation. Subjects with higher grades also
reported less problem-focused coping
(r =
.29,
p
< .001),
con-
sistent with the notion that students who had done
well
(and felt
best) were taking a breather.
Multiple regressions indicated that grade received, denial,
and use of social support all made unique contributions to pre-
diction
of
threat emotion
at
this stage (overall adjusted
R2 =
.37).
Neither of the correlates of challenge made
a
unique con-
tribution. Only grades made
a
unique contribution
to
harm
emotions (overall adjusted
R2
=
.44) or to benefit emotions (ad-
justed/?2
= .39).
These results suggest
two
conclusions. First,
as was
true
at
Time 1, threat was uniquely related
to a
wider range
of
coping
variables than were other emotions. Second,
the
results again
indicated that prediction rests
to a
considerable extent
on
vari-
ance shared among predictors.
Prospective Prediction From Coping
to
Emotion
Concurrent correlations
of
coping with appraisal emotions
provide
a
picture
of
subjective experiences
at a
given time,
but
they
do not
tell whether coping influences emotions pros-
pectively.
To
find
out, we
conducted analyses
in
which coping
responses at
one
time were related to an index of emotion
at
the
next time, while controlling
for
the same emotion experienced
at the earlier time.
Coping before
the
exam
was a
poor predictor
of
emotions
after
the
exam. What effects
did
emerge reflected maladaptive
coping. Mental disengagement before the exam predicted more
threat afterward
(r = .37, p <
.001),
and a
similar tendency
emerged
for
use
of
alcohol
before
the
exam, though
not
reach-
ing our significance criterion
(r
=
.21,
p
<
.02).
No coping reac-
tion before
the
exam was
a
significant prospective predictor
of
challenge, harm,
or
benefit emotion after the exam.
Coping with
the
uncertainty that existed after
the
exam
but
before grades were posted was
a
somewhat better predictor
of
emotions that arose after posting
of
grades. Interestingly
enough,
the
emotions
on
which such effects occurred reflected
anticipatory rather than outcome appraisals
(see
Table
6).
Higher levels
of
problem-focused coping activities after
the
exam were followed
by
more of both threat and challenge emo-
tions after posting of grades. Other aspects of coping predicted
only one or the other of these
emotions.
Positive reframing after
the exam predicted more challenge
at
Time
3;
restraint
and
turning
to
religion after
the
exam predicted more threat at Time
Table
6
Prospective Correlations
of
Coping Responses After the
Exam
but Before Posting
of Grades With
Threat and Challenge
Emotions After Posting
of
Grades,
Controlling for
Earlier Levels
of
the
Same Emotion
COPE scale
Problem Focused Coping
Acceptance
Use of Social Support
Restraint Coping
Religion
Positive Reframing
Denial
Mental Disengagement
Behavioral Disengagement
Use of Alcohol
Threat
.30**
.10
.21
.32**
.23*
.13
.20
.18
.15
.15
Challenge
.28*
.20
.09
.19
.19
.29**
.09
.11
.08
.14
*
p <
.01,
two-tailed.
**
p <
.001,
two-tailed.
3.
No aspect of
coping at
Time
2
predicted harm
or
benefit emo-
tions
at
Time 3.3
We
also examined prospective effects
of
subjects'
confidence
about attaining
the
desired grade
on the
exam. Three effects
approached
but did not
attain significance. Confidence before
the exam tended
to
predict more benefit (r
=
.22,
p
< .02)
and
less harm emotion
(r
=
-.20, p < .03)
after
the
exam. Confi-
dence after
the
exam tended
to
predict less threat emotion
at
Time3(r=-.21,/7<.02).
Although
our
focus here
is on
prediction
of
emotions,
we
should note that there is another potential outcome variable
to
consider: exam grades.
Did any
aspect
of
coping before
the
exam predict subjects' grades? Only one COPE scale came even
close. Specifically, reports
of
mental disengagement before
the
exam related inversely
to the
grades obtained
(r
= -.19,
p <
.04).
This tendency, weak
as it
was,
is
consistent with
a
finding
reported earlier
by
Bolger (1990).
To
summarize the results of these prospective analyses, there
was
no
evidence that
the
occurrence before
the
exam
of
what
are generally regarded
as
positive, adaptive aspects
of
coping
prospectively predicted either
low levels
of distress or high levels
of positive emotion after the exam. The only predictors of
emo-
tion after the exam were coping reactions that
we
characterized
earlier
as
potentially dysfunctional.
The
results appear
to con-
firm this characterization.
Later
on,
however, there appeared
to be at
least some benefit
from the more positive aspects of coping, although the benefit was
not unmixed. The pattern later on
suggests
that people who spent
the period after
the
exam gearing
up to
move onward
in
their
course
work
(as
reflected in problem-focused
coping) were
already
beginning
to
feel
the
anticipatory emotions associated with
the
challenge—and
also
the threat—of the next
phase
of
the
course.
Prospective Prediction From Emotion
to
Coping
It
is
also reasonable
to ask
whether emotions prospectively
predict the occurrence of particular coping
reactions.
This pos-
3
It might be argued that the clearest test of coping would entail con-
trolling for all four appraisal emotions. Doing this eliminated only one
of the significant effects described
here:
the prospective effect of turning
to religion (Table 6).

COPING
191
sibility
was
assessed
by
relating emotion at
one stage to coping
at
the subsequent stage, controlling
for
prior levels
of
that coping
response.
Virtually all of the prediction from emotions before the exam
came from
an
outcome emotion: feelings of harm. Harm emo-
tion before the exam predicted the following reactions after
the
exam: more behavioral disengagement
(r
=
.38,
p
<
.001),
more
mental disengagement
(r
=
.28,
p
< .001), more denial
(r
=
.39,
p < .001), less acceptance
(r =
-.23,
p
< .01), more turning
to
alcohol
(r =
.29,
p
< .01),
and
marginally more turning to reli-
gion
(r
=
.22,
p
<
.02).
Threat emotion tended to predict several
of the same reactions, but did so only
at
marginal levels.
In contrast
to
this relatively rich picture
of
affect-induced
coping after the exam, there was
no
relation between emotions
after the exam and coping reported after posting of grades.
Ap-
parently what coping was taking place
at
this final stage largely
reflected
the
nature
of
subjects' grades, which
had now
been
known
for
several days. That is,
as
noted earlier, low grades
re-
lated
to
active coping
and
planning.
In
effect, those
who had
done poorly
on the
exam reported scrambling to
do
something
about making up their performance deficits. Somewhat surpris-
ingly, neither harm emotion
at
Time
2 nor
low test grades
pre-
dicted
the
occurrence
at
Time
3
of the dysfunctional reactions
that had been prompted
at
Time
2
by
harm emotion
at
Time
1
(denial, disengagement,
and
use of alcohol).
To summarize
the
results
of
these
prospective tests, feelings
of harm before
the
exam predicted avoidance coping after
the
exam was over—indeed, every aspect
of
avoidance coping that
the COPE measures. Negative feelings before the exam
did not,
however, predict high levels
of
other aspects
of
coping, other
than
a
marginal effect
for
use of religious activity. Distress after
the exam
did not
affect subsequent coping at
all;
Time
3
coping
seemed to reflect instead responses to grades.
Associations ofDispositional Coping Styles With
Situational Coping
and
Emotion
At
the
beginning
of
the semester, subjects
had
reported
the
extent to which they typically respond to
stress
with the various
coping responses measured by the COPE.
We
turn now to these
dispositional coping styles.
The
first question
is how
well
the
dispositional measure predicted situational
use
of the same
re-
sponse
at
various points
in
the transaction. Table
7
gives corre-
lations between
the
dispositional scales
and the
corresponding
situational scales
at
each time (because predictions here
are so
clearly directional, the
tests
are
one-tailed).
As
shown
there,
dis-
positional scales varied considerably
in
their success
at
predict-
ing situational use of the response.
Two dispositional scales correlated strongly with situational
reports, and did
so
consistently across phases of the transaction:
turning to religion and turning to alcohol. Several dispositional
scales correlated with situational counterparts
at
about
the .3
level:
positive reframing, restraint, use of instrumental support,
denial,
and
mental disengagement.
In
some cases, correlations
were fairly strong
at
one stage
but not
others,
in
intelligible
pat-
terns.
For
example,
dispositional restraint
was
most strongly re-
lated
to
situational restraint after
the
exam (when restraint
is
most relevant)
and
least
so
before
the
exam (when restraint
is
least
adaptive).
Dispositional planning
was
tied most strongly
to
situational planning before
the
exam, when planning
is
most
Table
7
Correlations Between Dispositional Coping Styles and
Situational Reports
of
Using the
Same Coping Responses
Before the
Exam,
After the
Exam, and
After Posting
of Grades
COPE scale
Active Coping
Planning
Suppression of Competing Activities
Acceptance
Use of Instrumental Support
Use of Emotional Support
Restraint Coping
Religion
Positive Reframing
Denial
Mental Disengagement
Behavioral Disengagement
Use of Alcohol
Before
exam
.23*
.30**
.15
.27**
.28**
.23*
.20
.71**
.37**
.22*
.33**
.26*
.40**
After
exam
.25*
.21*
.13
.40**
.25*
.25*
.37**
.68**
.22*
.32**
.35**
.17
.71**
After
grades
22*
.22*
.14
.31**
.24*
.18
.30**
.71**
.31**
.29**
.22*
-.01
.71**
*p<m.
**p<M\.
relevant (and
at its
highest level).
One
dispositional report was
quite poor
as a
predictor
of
situational coping: suppression
of
competing
activities.
This lack of association replicates data re-
ported earlier (Carver et al., 1989).
Do coping dispositions predict variations
in
affective experi-
ence during
the
transaction?
In
most cases,
the
answer
is no.
There were, however, three coping dispositions that correlated
fairly consistently with negative emotions across the transaction
(Table
8),
and another that showed one significant relation.
The
dispositional tendency
to
use instrumental support was consis-
tently related
to
reports
of
threat emotion until grades were
posted,
at
which point the association evaporated. The disposi-
tional tendency toward denial
was
related
to
threat emotions
throughout
the
transaction
and to
feelings
of
harm
as
well
be-
fore
the
exam.
The
dispositional tendency
to use
alcohol
as a
coping tool was related consistently
to
harm emotion.
Further analysis
suggests,
however, that these various associa-
tions are largely attributable
to
broad tendencies
to
experience
a
given
affect
at
relatively consistent
levels
throughout
a
transac-
tion. That
is,
when previous
level
of an affect
was
partialed, only
one subsequent association remained significant:
the
associa-
tion between dispositional
use of
alcohol
and
threat
at
Time
3
(r=.21,p<.003).4
Discussion
The pattern of emotions that unfolded across the transaction
generally resembled that found
by
Folkman
and
Lazarus
(1985).
Threat
and
challenge were high initially
and
fell
off
when grades were known. Where Folkman
and
Lazarus found
threat and challenge to be independent,
we
found they can even
co-occur. As Folkman and Lazarus (1985) noted,
in
conditions
4
Path analyses were conducted to assess whether the coping disposi-
tions operated on affect through situational coping. In general, these
analyses did not strongly support such a hypothesis, which is also un-
dermined by the failure (just noted) to find evidence that coping dispo-
sitions predict later affect after controlling for previous affect.

192CHARLES S. CARVER AND MICHAEL F. SCHEIER
Table 8
Correlations Between Four Dispositional Coping Styles
and
Threat and
Harm Emotions
Before the
Exam,
After the
Exam, and
After Posting
of Grades
COPE scale
Use of Instrumental Support
Use of Emotional Support
Denial
Use of Alcohol
Before exam
Threat
.24*
.21
.24*
.17
Harm
.06
.03
.25*
.44**
After exam
Threat Harm
.29**
.18
.24*
.17
.25*
.18
.19 .41**
After grades
Threat
.08
.06
.23*
.32**
Harm
.06
-.01
.19
.23*
*p<.01.**p<.001.
of high ambiguity it
is
not unusual for
people to
experience both
positive and negative emotions simultaneously. That is, mixed
feelings can result from considering each of
two
contradictory
possibilities at more or less the same time (a position we have
also argued for
elsewhere;
Carver
&
Scheier, 1990b). In contrast
to this pattern for anticipatory emotions, there was an increas-
ingly strong inverse relation between the outcome emotions of
harm and benefit as the event proceeded. That is, as the out-
come became more unambiguous, the outcome emotion expe-
rienced became more specific.
How Do Coping and Emotion Covary?
Folkman and Lazarus (1985) presented only a very sketchy
picture of how coping covaried with the various appraisal-re-
lated emotions. We have given a somewhat broader view of the
phenomenology of how coping related to various emotions at
various stages of the transaction. Here
is
a brief summary: Feel-
ings of threat were inevitably linked to concurrent use of social
support. Threat was also positively related to problem-focused
coping before and after the posting of
grades
(these, of
course,
are the times when problem-focused coping is most useful as a
response to threat). At Time 2, in contrast, threat was linked to
mental disengagement, as threatened subjects apparently tried
to distract themselves from their
worries.
Feelings of threat were
also related to tendencies toward giving up at both measure-
ments after the exam, supplemented at Time
3
by denial, turn-
ing to religion, and use of alcohol. It is clear from these associa-
tions that threat is linked to a very wide range of
coping
quali-
ties—both problem-focused and avoidance coping.
In contrast to this pattern for threat, feelings of challenge
were related to few aspects of
coping.
Challenge before the exam
was related to problem-focused coping and to positive refram-
ing. After the exam, feelings of challenge were unrelated to cop-
ing. After grades had been posted, challenge was again related
to positive reframing and now to restraint coping. As a group,
the correlations suggest that coping is far more responsive to
feelings of threat than to feelings of challenge. In a way, this
should not be surprising. Feelings of challenge are positive—
there is no need to try to reduce them when they arise. Never-
theless, the pattern raises questions about the nature of the role
that challenge plays in stressful transactions. It should be re-
called, though, that the reliability of
this
index was not as high
as for the other
emotions;
thus the pattern obtained may reflect
the
problem
of
differential
measurement reliability.
The patterns of correlations for harm and benefit emotions
were also of interest. In many respects, the pattern for harm
resembled that for
threat:
Harm was linked to use of social sup-
port throughout, with denial at Times
2
and
3,
and with behav-
ioral disengagement at Time 2. In other respects the patterns
differed
in
interesting
ways:
Feelings of harm (in contrast to feel-
ings
of threat) were unrelated to problem-focused coping before
the exam, but did relate to such activities just afterward. Also
unlike feelings of
threat,
feelings of harm were associated with
alcohol use at every measurement point in the study.
What was interesting about feelings of benefit was their lack
of relation to any aspect of coping.
Prospective Effects of Coping and of Emotions
The concurrent associations just described provide a sense of
the phenomenology of the stress and coping process, but they
say nothing about the direction of the causal
flow.
To
gain some
information on causality,
we
conducted prospective tests exam-
ining relations between coping at one time point and emotion
at the next point, controlling for the same emotion at the earlier
point. Similar analyses were conducted to examine prospective
prediction from emotion to coping.
These analyses revealed several effects of coping, all on the
anticipatory emotions of threat and challenge. Adverse effects
tended to predominate, although some positive effects also
emerged. Mental disengagement before the exam (not a very
effective way to prepare for the exam) led to higher levels of
threat afterward. All other significant effects stemmed from
coping after the exam. Problem-focused coping after the exam
predicted higher levels of threat and challenge later. Positive re-
framing also predicted more challenge; restraint and religious
activity predicted more threat. This pattern clearly does not fit
the common stereotype of
coping
as an ameliorator of distress.
On the other hand, the effects on challenge do seem to reflect a
pattern in which task-focused coping promotes enthusiasm
about confronting the next instance of a recurrent stressor.
We also found evidence that emotion induces coping reac-
tions,
but the evidence
was
quite restricted: Only harm emotion
had this effect and only early in the transaction. Specifically,
feelings of harm before the exam were tied to higher levels of
avoidance coping (maladaptive coping) after the exam, that is,
behavioral disengagement, mental disengagement, denial, and
use of alcohol.
Dispositional Coping Styles
We turn now to the role of stable individual differences. To
our knowledge, this is the
first
study in which dispositional cop-

COPING193
ing styles were
examined
as
predictors of situational coping and
appraisal emotions across a specific stressful transaction—al-
though some researchers have examined personality variables
in the same manner
(e.g.,
Bolger,
1990;
Carver et
al.,
1993).
The
findings of
this
aspect of the study
were
mixed. Although coping
dispositions predicted situational coping moderately well, only
a few dispositional qualities related to the
emotions.
In all cases
these aspects of coping were tied to higher rather than lower
distress. Subjects
who
reported that
they
typically react to stress
with overt denial consistently reported more threat than did
subjects less prone to denial. Subjects who reported that they
commonly react to stress by getting help from others also re-
ported more threat before and after the exam.
The findings regarding one of the coping dispositions were
more striking than this, striking enough that they deserve spe-
cial emphasis. Subjects
who
reported that they commonly react
to stress by using alcohol to cope reported higher
levels
of harm
emotions at each measurement point in the
study.
This pattern
strongly suggests that the people who report a dispositional ten-
dency to use alcohol
as
a means of coping are in poorer psycho-
logical shape than people who do not do so (cf. Parker, Brown,
&Blignault, 1986).
What
do
these various
findings
say
about the usefulness of the
concept of dispositional coping styles? It has often been as-
sumed that such styles would have little impact on situational
coping tendencies. In this study, 32 of
the
39 correlations rele-
vant to that question
were
significant, although not
all were
very
strong. On the other hand, the scales were less effective as pre-
dictors of emotion. Perhaps the best conclusion to draw at this
point is that no firm conclusion on the role of coping disposi-
tions
is
yet warranted.
Confidence and
Emotions
We also had interesting findings for one more predictor of
emotions: subjects' confidence about attaining the grade they
were
seeking.
This variable
was a strong
correlate of affect under
stress.
Confidence was related inversely to feelings of threat and
harm both before and after the exam (confidence was not as-
sessed at Time 3, as grades were known by then). Confidence
also was related to benefit emotions after the exam. Multiple
regression
analyses
pitting confidence against other variables
re-
vealed that all these effects involved unique contributions from
confidence, independent of other variables. We also examined
confidence as a prospective predictor. Three strong tendencies
emerged, though none reached our significance level: Confi-
dence before the exam tended to predict more benefit and less
harm emotions after the exam, and confidence after the exam
tended to predict less threat at Time 3. These prospective asso-
ciations are certainly suggestive enough to warrant further ex-
amination of this variable in future work.
As a group, these findings indicate that confidence versus
doubt is an important aspect of the phenomenology of a stress-
ful transaction. The findings stop short of indicating that con-
fidence
plays a
causal role in the emotions experienced, but they
do indicate that confidence is intimately tied to the emotions.
The
findings
thus are consistent with a model in which apprais-
als of threat and of
one's
ability to cope yield both an affective
response and a sense of confidence
versus
doubt.
This pattern of
dual influences is one aspect of
a
model of affect we have pro-
posed elsewhere (Carver
&
Scheier, 1990b). These
findings
also
suggest the potential importance, in any coping transaction, of
people's confidence about attaining desired ends, a point made
in greater detail elsewhere (e.g., Carver & Scheier, 1990a;
Scheier
&
Carver, 1992).
General Conclusions From the Data
The
findings
outlined
above
suggest several
conclusions.
Most
simply, as has been noted by many others (e.g., Bolger, 1990;
Folkman
&
Lazarus,
1985;
Stone et
al.,
1991),
stressful encoun-
ters differ sharply in nature from one stage to another.
To
treat
an entire encounter as a single event, or to assume that coping
is similar throughout, is to greatly oversimplify. People in this
study appeared to be responding to different demands at differ-
ent phases of
the
transaction. Initial coping focused on the up-
coming exam: active coping, planning, suppression of compet-
ing activities, and acceptance predominated. Coping right after
the exam appeared to have been at least partly an effort to deal
with the negative emotions experienced before the exam, al-
though the efforts were channeled primarily into dysfunctional
avoidance coping. By the third time point, the impetus behind
coping had shifted again. Coping now reflected responses to the
grades received on the exam, with subjects
who
had
done
poorly
now reporting higher levels of problem-focused coping.
A second conclusion from the data concerns the effects of
coping. Evidence that coping influenced subsequent affect was
sparse. Moreoever, in no case did coping prospectively predict a
reduction in
negative
emotions.
With
few
exceptions
(all
involv-
ing challenge), all the associations were such that coping led to
increases in negative
emotion.
Mental disengagement before the
exam predicted higher levels of threat after the exam. Reports
of problem-focused coping after the exam predicted more
threat (and more challenge) at Time 3. The latter effects seem
to suggest that people who were remaining task engaged during
the resolution period immediately after the exam (i.e., continu-
ing
their efforts toward the next exam)
were a
week later already
experiencing the appraisal affects that characterize the antici-
pation of the next stressful transaction (the next exam). The
only case in which coping had an unambiguously positive effect
was
that positive reframing after the exam led to feelings of chal-
lenge later (cf. Menaghan
&
Merves, 1984).
Thus,
although some of the coping reactions that subjects re-
ported at Time 2 appeared to reflect efforts to deal with affects
felt earlier, there was no evidence that the coping invoked this
way had a beneficial effect. Indeed, there was a tendency for
coping reactions after the exam induced by harm before the
exam to be prospective predictors of
Time
3 threat (denial and
use of religion). The pattern that emerged from this study is
clearly not one in which coping acts to diminish negative emo-
tions.
These
findings
also serve to illustrate a methodological point:
the importance of examining prospective as well as concurrent
associations. In the absence of prospective effects, the most to
be said would be that aspects of coping correlated with negative
emotions at various phases of the transaction. Such findings
might mean only that distress leads to efforts to cope (which is
the most parsimonious interpretation of most of the concurrent
associations). The prospective tests permit stronger conclusions
about which variable is influencing which.

194CHARLES S. CARVER AND MICHAEL F. SCHEIER
Finally, although the role of dispositions in this study
was
not
as
great as it might have been, there
is
at least one coping dispo-
sition that appears to matter a good deal: the disposition to al-
cohol use. Alcohol use is a response to stress that appears to be
particularly dysfunctional (cf. Cooper, Russell, & George,
1988),
both in situational and in dispositional terms. The re-
sults obtained in this study concerning this variable suggest the
importance of giving it far closer scrutiny in future research.
Relationship to Other Findings: The Importance of the
Situation
How do the effects of
coping
found here compare with those
of other prospective studies? There are certain clear similarities:
Bolger (1990) found that wishful thinking in the preparatory
phase of an exam led to higher levels of anxiety just before the
exam. This resembles our finding that mental disengagement
before the exam led to subsequent threat emotion. Findings of
Stanton and Snider
(1993)
and Litt et
al.
(1992) are also similar
conceptually to this, as are certain aspects of the findings of
Carver etal.(
1993).
There
is
more diversity among studies regarding the effects of
active
coping.
There
is some evidence
that using an
active
coping
approach to life in general promotes well-being (Aldwin & Re-
venson, 1987; Aspinwall & Taylor, 1992; Glyshaw, Cohen, &
Towbes, 1989). On the other hand, several studies have found
that active coping has the opposite effect when one is dealing
with a focal stressor. Bolger (1990) found that problem-focused
coping during exam preparation led to higher anxiety just be-
fore the exam.
We
found that problem-focused coping after the
exam led to higher anxiety after grades were posted. Mattlin,
Wethington, and Kessler (1990) found that active coping (as-
sessed retrospectively) was associated with higher levels of
dis-
tress later on, provided the stressor being coped with was a dis-
crete event rather than a long-lasting problem. This difference
among
findings
appears to suggest that active coping has differ-
ent affective consequences under different circumstances.
A similar point is made by one more striking difference be-
tween the
findings
of Carver et al. (1993) and those of the pres-
ent
study.
Carver
et
al.
examined coping and distress
in a
sample
of breast cancer patients. A key finding of that study was the
important role played by acceptance as a coping reaction. Ac-
ceptance was both a concurrent correlate of
low
distress and a
prospective predictor of low distress. That pattern of findings
contrasts sharply with the absence of any effect of acceptance in
the study reported here.
The most plausible explanation for this difference betwen re-
sults is that it depends on a difference in the situations being
confronted. The exam represents a situation in which steps can
be
taken to diminish the stressor's
impact.
Even a failure on this
exam can be redressed by greater effort later on. The cancer
patients, in contrast, were dealing with a situation that had to
be endured and accommodated to (surgery, a period of
recov-
ery, and a new physical status) before the person could
move
on.
There
is
ample evidence that
these two classes
of situations elicit
different patterns of coping (Folkman et al., 1986; Lazarus &
Folkman, 1984; McCrae, 1984). It may also be that the effects
of
a
given coping response differ across the situations (see also
Mattlin et
al.,
1990;
Pearlin
&
Schooler, 1978).
It should be noted in that regard that the situation examined
in the study reported here
was
characterized more by challenge
than by threat (which was also true for Folkman & Lazarus,
1985,
and Bolger, 1990). Although the potential for a loss was
certainly present, subjects were also in a situation in which they
could create
gains.
This makes the situation very different from
that faced in a health crisis, for example. The broader point
is that this aspect of stressful situations under study should be
systematically assessed and noted as the literature develops. It
may be that links between coping reactions and distress emo-
tions are very different in transactions that involve primarily
threat than in transactions that involve primarily challenge.
This is a dimension of the stressful situation that has not re-
ceived as much careful attention in the stress literature as it
should.
References
Aldwin, C. M., & Revenson, T. A. (1987). Does coping help? A reexam-
ination of the relation between coping and mental health. Journal of
Personality and Social
Psychology,
53, 337-348.
Aspinwall, L. G., & Taylor, S. E. (1992). Modeling cognitive adaptation:
A longitudinal investigation of the impact of individual differences
and coping on college adjustment and performance. Journal of Per-
sonality and Social
Psychology,
63, 989-1003.
Bandura, A. (1977). Self-efficacy theory: Toward a unifying theory of
behavioral change. Psychological
Review,
84, 191-215.
Bandura, A. (1986). Social foundations of thought and action: A social
cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.
Billings, A. G., & Moos, R. H. (1984). Coping, stress, and social re-
sources among adults with unipolar depression. Journal of
Personal-
ity and Social
Psychology,
46,
877-891.
Bolger, N. (1990). Coping as a personality process: A prospective study.
Journal of Personality and Social
Psychology,
59, 525-537.
Bolger, N., & Eckenrode, J. (1991). Social relationships, personality, and
anxiety during a major stressful event. Journal of Personality and
So-
cial
Psychology,
61, 440-449.
Carver, C. S,. Pozo, C, Harris, S. D., Noriega, V., Scheier, M. F., Rob-
inson, D. S., Ketcham, A. S., Moffat, F. L., Jr., & Clark, K. C. (1993).
How coping mediates the effects of optimism on distress: A study of
women with early stage breast cancer. Journal of Personality and So-
cial
Psychology,
65,
375-391.
Carver, C. S., & Scheier, M. F. (1983). A control-theory model of normal
behavior, and implications for problems in self-management. In P. C.
Kendall (Ed.), Advances in cognitive-behavioral research and therapy
(Vol. 2, pp. 127-194). San Diego, CA: Academic Press.
Carver, C. S., & Scheier, M. F. (1990a). Principles of self-regulation:
Action and emotion. In E. T. Higgins & R. M. Sorrentino (Eds.),
Handbook of motivation and cognition: Foundations of social behav-
ior
(Vol.
2, pp. 3-52). New York: Guilford Press.
Carver, C. S., & Scheier, M. F. (1990b). Origins and functions of positive
and negative affect: A control-process view. Psychological
Review,
97,
19-35.
Carver, C. S., Scheier, M. F, & Weintraub, J. K. (1989). Assessing coping
strategies: A theoretically based approach. Journal of Personality and
Social
Psychology,
56, 267-283.
Cohen, F., & Lazarus, R. S. (1973). Active coping processes, coping dis-
positions, and recovery from surgery. Psychosomatic Medicine, 35,
375-389.
Cooper, M. L., Russell, M., & George, W. H. (1988). Coping, expectan-
cies,
and alcohol abuse: A test of social learning formulations. Journal
of Abnormal
Psychology,
97, 218-230.
Cronkite, R. C, & Moos, R. H. (1984). The role of predisposing and
moderating factors in the stress-illness relationship. Journal of
Health and Social
Behavior,
25, 372-393.

COPING195
Felton, B. J., Revenson, T. A., & Hinrichsen, G. A. (1984). Stress and
coping in the explanation of psychological adjustment among chron-
ically ill adults. Social Science
&
Medicine, 18, 889-898.
Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-
aged community sample. Journal of Health and Social Behavior, 21,
219-239.
Folkman, S., & Lazarus, R. S. (1985). If it changes it must be a process:
A study of emotion and coping during three stages of a college exam-
ination. Journal of Personality and Social Psychology, 48, 150-170.
Folkman, S. Lazarus, R. S., Dunkel-Schetter, C, DeLongis, A., &
Gruen, R. (1986). Dynamics of a stressful encounter: Cognitive ap-
praisal, coping, and encounter outcomes. Journal of Personality and
Social
Psychology,
50, 992-1003.
Glyshaw, K., Cohen, L. H., & Towbes, L. C. (1989). Coping strategies
and psychological distress: Prospective analyses of early and middle
adolescents. American Journal of Community Psychology, 17, 607-
623.
Holahan, C. J., & Moos, R. H. (1985). Life stress and health: Personal-
ity, coping, and family support in stress resistance. Journal of Person-
ality and Social
Psychology,
49,139-141.
Holahan, C. J., & Moos, R. H. (1987). Risk resistance and psychological
distress: A longitudinal analysis with adults and children. Journal of
Abnormal
Psychology,
96, 3-13.
Lazarus, R. S. (1966). Psychological stress and the coping process. New
York: McGraw-Hill.
Lazarus, R. S., & Folkman, S. (1984). Stress,
appraisal,
and
coping.
New York: Springer.
Litt, M. D., Tennen, H., Affleck, G., & Klock, S. (1992). Coping and
cognitive factors in adaptation to in vitro fertilization failure. Journal
of Behavioral Medicine, 15, 171-187.
Mattlin, J. A., Wethington, E., & Kessler, R. C. (1990). Situational de-
terminants of coping and coping effectiveness. Journal of Health and
Social
Behavior,
31, 103-122.
McCrae, R. R. (1984). Situational determinants of coping responses:
Loss,
threat, and challenge. Journal of Personality and Social
Psychol-
ogy, 46,919-928.
McCrae, R. R.,
&
Costa,
P.
X, Jr. (1986). Personality, coping, and coping
effectiveness in an adult sample. Journal of Personality, 54, 385-405.
Menaghan, E. (1982). Measuring coping effectiveness: A panel analysis
of marital problems and coping efforts. Journal of Health and Social
Behavior, 23, 220-234.
Menaghan, E., & Merves, E. (1984). Coping with occupational prob-
lems:
The limits of individual efforts. Journal of Health and Social
Behavior, 25,406-423.
Miller, S. M. (1987). Monitoring and blunting: Validation of a question-
naire to assess different styles for coping with stress. Journal of Per-
sonality and Social
Psychology,
52, 345-353.
Parker, G., Brown, L., & Blignault, I. (1986). Coping behaviors as pre-
dictors of the course of clinical depression. Archives of General Psy-
chiatry, 43, 561-565.
Parkes, K. R. (1984). Locus of control, cognitive appraisal, and coping
in stressful episodes. Journal of Personality and Social Psychology,
46, 655-668.
Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal of
Health and Social
Behavior,
19, 2-21.
Roberts, J. E., & Monroe, S. M. (1992). Vulnerable self-esteem and de-
pressive symptoms: Prospective findings comparing three alternative
conceptualizations. Journal of Personality and Social
Psychology,
62,
804-812.
Rohde, P., Lewinsohn, P. M., Tilson, M., & Seeley, J. R. (1990). Dimen-
sionality of coping and its relation to depression. Journal of
Personal-
ity and Social
Psychology,
58,
499-511.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psycholog-
ical and physical well-being: Theoretical overview and empirical up-
date.
Cognitive Therapy and Research, 16, 201-228.
Silver, R. L., & Wortman, C. B. (1980). Coping with undesirable life
events. In J. Garber & M. E. P. Seligman (Eds.), Human helplessness:
Theory and applications. San Diego, CA: Academic Press.
Smith, C. A., & Ellsworth, P. C. (1987). Patterns of appraisal and emo-
tion related to taking an exam. Journal of Personality and Social Psy-
chology, 52, 475-488.
Smith, L. W., Patterson, T. L., & Grant,
1.(1990).
Avoidant coping pre-
dicts psychological disturbance in the elderly. Journal of Nervous and
Mental Disease, 178, 525-530.
Stanton, A. L., & Snider, P. R. (1993). Coping with a breast cancer diag-
nosis:
A prospective study. Health Psychology, 12,
16-23.
Stone, A. A., Greenberg, M. A., Kennedy-Moore, E., & Newman,
M. G. (1991). Self-report, situation-specific coping questionnaires:
What are they measuring? Journal of Personality and Social
Psychol-
ogy, 61, 648-658.
Vaillant, G. E. (1977). Adaptation to life. Boston: Little, Brown.
Vitaliano, P. P., Russo, J., Young, H., Teri, L., & Maiuro, R. D. (1991).
Predictors of burden in spouse caregivers of individuals with Alzhei-
mer's disease. Psychology and
Aging,
6, 392-402.
Wills,
T.
A. (1986). Stress and coping in early adolescence: Relationships
to substance use in urban high schools. Health Psychology, 5, 503-
529.
Received January 6, 1993
Revision received June
4,
1993
Accepted June 9, 1993 •