Content uploaded by Dale Hample
Author content
All content in this area was uploaded by Dale Hample on Apr 15, 2016
Content may be subject to copyright.
ARTICLES
The Role of Taking Conflict
Personally in Imagined Interactions
about Conflict
Kelly P. Wallenfelsz & Dale Hample
This study examines thoughts and feelings about conflict. A person may use imagined
interactions (IIs) to work through a conflict situation. One factor that may affect the nat-
ure of IIs about conflict is the tendency among some individuals to take conflict person-
ally. Taking conflict personally (TCP) is the feeling that conflict is a negative life event
that is aimed at the self (Hample & Dallinger, 1995). This study examines the relation-
ship between TCP and IIs about conflict. Results indicate that trait TCP is correlated
with rumination, that a variety of significant relationships appear between trait TCP
and II characteristics, and that trait TCP predicts state TCP immediately after a
conflict-oriented II.
People often think about interpersonal conflict in the absence of their relational
partners (Honeycutt, 2003; Honeycutt & Cantrill, 2001). One way that people
work through interpersonal conflict is via imagined interactions (IIs). IIs represent
a ‘‘process of cognition whereby actors imagine themselves in interaction with
others ...they reflect a distinct kind of thinking in which communicators experience
or actually work through cognitive representations of conversation’’ (Edwards,
Honeycutt, & Zagacki, 1988, p. 24). Edwards et al. found that 40%of respondents
reported IIs that focused on conflict when asked to report recent IIs. Zagacki,
Edwards, and Honeycutt (1992) observed that individuals who report IIs that involve
conflict have lower satisfaction with the II, perhaps because they serve to ‘‘review and
Kelly P. Wallenfelsz, Department of Speech, Carl Sandburg College. Dale Hample, Department of Com-
munication, University of Maryland. This article is based on the first author’s M.A. thesis, directed by the second
author, at Western Illinois University. Correspondence to: Dale Hample, Department of Communication,
University of Maryland, 2103 Skinner Building, College Park, MD 20742. E-mail: dhample@umd.edu
Southern Communication Journal
Vol. 75, No. 5, November–December 2010, pp. 471–487
ISSN 1041-794x (print)/1930-3203 (online)
#2010 Southern States Communication Association. DOI: 10.1080/10417940903006057
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
rehearse the negative dimensions of actual experience’’ (p. 65) rather than to help
work through the conflict to a more positive outcome.
Conflict linkage theory (Honeycutt, 2003, 2004) focuses on recurring conflict in
interpersonal relationships. Proactive IIs are those that occur prior to an actual conver-
sation. Retroactive IIs occur after the conversation has taken place (Edwards et al.,
1988). Hence, the linkage can take the form of proactive II to actual conflict to retro-
active II. Specifically, one may anticipate a conflict and have a proactive II about it, the
conflict occurs, and then one may review, re-experience, and revise the conflict by
means of a retroactive II. This linkage of past and future conversations via IIs can help
to explain how one’s predispositions about conflict prefigure and reinforce particular
kinds of conflict experience, either positive or negative (Honeycutt, 2003, p. 68).
While thinking about conflict can lead to productive conflict resolution, it can also
serve as a source of escalation and make a problem seem worse than it is. Cloven and
Roloff (1991) discovered that the more a person thinks about a problem, the more
serious the problem appears and the more likely it is one will blame the relational
partner. This type of repetitive thought about conflict has the label of ‘‘rumination.’’
Rumination occurs when people ‘‘repetitively focus on themselves and on the nature
and implications of their negative feelings’’ (Lyubomirsky, Tucker, Caldwell, & Berg,
1999, p. 1041). Rumination arises under a number of negative affective and decision-
making conditions and, hence, can be counterproductive to effective and positive
conflict resolution.
The outcome of conflicts depends in a large part on how participants perceive
them (Folger, Poole, & Stutman, 2005). One thing that can lead conflict to become
destructive is the participants taking it personally. Hample and Dallinger (1995)
defined taking conflict personally (TCP) as ‘‘a feeling of being personally engaged
in a punishing life event. [The person] feels threatened, anxious, damaged, devalued,
insulted’’ (p. 306). The feelings associated with TCP have been associated with the
tendency to avoid conflicts (Hample & Dallinger, 1995), verbal aggressiveness once
engaged in conflict (Hample & Dallinger, 1995; Hample, Dallinger, & Fofana,
1995), the tendency to use inferior ego-defense mechanisms (Hample, Benson,
Gogliotti, & Jeong, 1997), and diminished communication effectiveness (Barch,
1994). Since individuals who report high levels of TCP view conflict as a negative,
punishing experience, it is likely that trait TCP has some effect on the nature of cog-
nition about conflict communication. People who take conflict personally may
have IIs about conflicts that are less positive and productive than individuals with
low levels of TCP. TCP could help to explain why negative IIs intrude on positive
IIs that surface in the course of therapy (Honeycutt, 2003). Understanding the
relationship between TCP and IIs is an important step in understanding why some
people are unable to think productively about conflict.
Imagined Interactions
There are at least eight characteristics of IIs (Honeycutt, 2003). Frequency refers to
how often people have IIs. Edwards, Honeycutt, and Zagacki (1989) observed that
472 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
women have more IIs than men. In addition, lonely people have fewer IIs than
nonlonely people (Honeycutt, Edwards, & Zagacki, 1989–1990). Greater frequency
of IIs is also associated with better conversational interpretation, which refers to the
ability to detect sarcasm and to paraphrase what others say (Honeycutt et al.,
1992–1993).
Some IIs are proactive; these occur prior to interaction (Honeycutt, 2003).
Retroactive IIs take place after the actual conflict. Proactive IIs are more frequent than
retroactive ones (Edwards et al., 1988; Honeycutt et al., 1989–1990; Zagacki,
Honeycutt, & Edwards, 1992). Zagacki et al. (1992) reported that 83%of IIs entered
into plans for anticipated conversations. Those who have retroactive IIs often
also have proactive IIs about the same encounter (r¼.45, Honeycutt et al.,
1989–1990). These linked IIs often comprise themes in relationships (Honeycutt,
1995) and can explain why conflict in relationships often picks up where it left off
despite a period of separation (Honeycutt, 2003). People tend to recall and review
unpleasant conversations more than pleasant ones (Zagacki et al., 1992).
Variety refers to the diversity of partners and topics in IIs (Honeycutt, 2003).
Edwards et al. (1988) noted that IIs most frequently involve significant others.
Variety correlates with conversational sensitivity, or the ability to interpret verbal
and nonverbal communication, and with conversational alternatives, that is, skill
in stating ideas in different ways (Honeycutt et al., 1992–1993).
Discrepancy has to do with the difference between an II and an actual conversation
(Honeycutt, 2003). This characteristic relates negatively to communication com-
petence (Honeycutt et al., 1992–1993) and may reflect an inability to plan effectively
and to imagine alternative outcomes to communicative actions (Allen & Honeycutt,
1997).
Self-dominance reflects who speaks more in the II, the person who is having the II
or the imagined other (Honeycutt, 2003). Overall, the self talks more in IIs, both in
respect to words and lines of dialogue (Edwards et al., 1988). This is especially true of
less pleasant IIs, such as those dealing with conflict (Honeycutt et al., 1989–1990). In
addition, Zagacki et al. (1992) discovered that the self is more likely to start the
dialogue in IIs.
Valence is the degree of pleasant emotional affect in an II (Honeycutt, 2003). In
addition to having more IIs, females report more pleasant ones than males (Edwards
et al., 1989). IIs relating to conflict are more likely to be ones in which the self
dominates and provide rehearsal for anticipated conflicts, as well as to gain greater
understanding of the situation (Zagacki et al., 1992). People ostensibly remember
unpleasant IIs better than pleasant ones. Finally, individuals tend to have more
retroactive and specific IIs concerning unpleasant communicative experiences than
pleasant ones (Honeycutt et al., 1992–1993).
Specificity denotes the level of ‘‘detail and distinction of images within IIs’’
(Honeycutt, 2003, p. 26). Honeycutt et al. (1992–1993) produced evidence that speci-
ficity predicts conversational sensitivity. Those with specific IIs tend to be skilled at
sensing hidden messages in what others are saying, remembering past conversations,
and expressing the same idea using other words.
Taking Conflict Personally 473
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
Imagined Interactions and Conflict
Individuals in relationships often experience recurring conflict (Mallouk, 1981).
Trapp and Hoff (1985) portray a serial argument as a number of individual
arguments that focus on the same issue. These arguments may occur throughout
the course of a day, a week, or even a lifetime. One reason that serial arguments
continue is that the issue remains unresolved. Johnson and Roloff (1998) reported
that the perception that a serial argument can be resolved relates negatively to the
amount of time spent thinking about the conflict.
Through linked IIs, people are able to replay previous conversations and to antici-
pate future encounters. In this way, conflicts remain intact, even in the absence of the
relational partner, and pick up where they left off despite a hiatus. People can also use
linked IIs to think of new alternatives to resolve the conflict situation. In addition, IIs
may serve as surrogates for the lack of actual communication if one decides that the
issue is not worth pursuing on the basis of the imagined responses of his=her partner
(Honeycutt, 2003).
Linked IIs concerning one’s conflicts may result in a self-fulfilling prophecy. If the
individual imagines that a conflict situation will be unpleasant and unproductive, the
actual interaction may, in fact, turn out that way. On the other hand, the conflict
linkage may distort reality (Honeycutt, 2003). Zagacki et al. (1992) ascertained that
unpleasant IIs, such as those about conflict, are usually more verbal than visual and
that verbal IIs tend to have a higher level of discrepancy. Therefore, IIs about conflict
may be very different from the actual conversation that follows.
Conflict linkage theory (Honeycutt, 2003, 2004; Honeycutt & Cantrill, 2001) pro-
vides an explanation for why serial arguing (Trapp & Hoff, 1985) occurs in interper-
sonal relationships. When a conflict is unresolved, the parties may continue to think
about it, such that it flares up later. Dwelling on problems — rumination — has
unwelcome consequences for the productive resolution of conflicts. Whereas IIs
always involve an imagined other, rumination may or may not include an imagined
interaction partner.
Rumination
Martin and Tesser (1996) define rumination as ‘‘a class of conscious thoughts that
revolve around a common instrumental theme and that recur in the absence of
immediate environmental demands requiring the thoughts’’ (p. 1). It is goal directed
and arises when movement toward a goal does not unfold as the individual expects it
to. Rumination involves the past, present, or future and often continues until the goal
is achieved, movement toward the goal returns to the expected pattern, another goal
becomes predominant, or the person actively thinks about something else.
Rumination often entails feelings of depression and hopelessness (Nolen-
Hoeksema, 1991). Nolen-Hoeksema (1996) noted that when depressed individuals
are induced to ruminate, their moods worsen, and they become more negative in
their thinking. Ruminating among such individuals also contributes a perception
474 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
indicating that problems are more severe than they are and that one’s
problem-solving skills are more deficient than is the case (Lyubomirsky et al.,
1999). In addition, ‘‘self-focused rumination led dysphoric students to mull over
their most troubling problems such as decreasing grades or conflicts with friends
and family members’’ (Lyubomirsky et al., 1999, p. 1051). (Dysphoria is a feeling
of being unwell or unhappy; see Mish, 1995.) In addition, Lyubomirsky et al.
(1999) reported that depressed ruminators tend to be negative and self-critical when
thinking about their problems. Interestingly, more positive, optimistic, and less
problem-focused thoughts appeared when nondepressed people ruminated or when
anyone (depressed or nondepressed) engaged in a distracter task.
Ruminators are more aggressive and have higher levels of anger than nonrumina-
tors (Bushman, 2002). ‘‘[V]enting while ruminating about the source of provocation
kept aggressive thoughts and angry feeling active in memory and only made people
more angry and aggressive’’ (Bushman, 2002, p. 729). Furthermore, McCullough,
Bellah, Kilpatrick, and Johnson (2001) uncovered evidence linking vengefulness with
rumination. Those who seek vengeance may have a number of goals, such as the
desire to get even, to teach the offender a lesson, and to save face. Vengeful people
may use rumination to focus their attention on the attainment of one or more of
these goals. The authors reported a positive correlation between rumination and
the desire for revenge (r¼.40). In addition, people who experienced a reduction
in the amount of rumination over time also experienced a reduction in the desire
for revenge. The authors concluded that rumination interferes with forgiveness,
which they operationalized as a reduction in revenge motivations following an
interpersonal offense.
Cloven and Roloff (1993) noted that mulling about conflict can encompass
intrapersonal communication that may or may not include the presence of the other
person. Reportedly, the more a person thinks about a major conflict, the worse the
problem appears and the more he or she is apt to blame the partner. Moreover,
the amount of thought about the problem predicts perceptions of seriousness in
the case of a major problem, but not in the case of a minor one. Actual communi-
cation seems to attenuate the effect of mulling. Therefore, it would appear that
prolonged thinking about conflict is counterproductive to effective conflict
resolution unless it is accompanied by interpersonal communication.
Honeycutt (2004) investigated the relationship between rumination about conflict
and characteristics of IIs. He found that rumination correlated directly to the speci-
ficity of IIs, prevalence of both proactive and retroactive IIs, discrepancy, self-focus,
and negative valence. One limitation of Honeycutt’s (2004) study is that the measure
of rumination about conflict was measured using various items from the Survey of
Imagined Interaction (SII), not a scale specifically designed to capture it.
Taking Conflict Personally
Taking conflict personally (TCP) is both a stable personality trait and a temporary
state. According to Hample and Dallinger (1995), TCP ‘‘is immediately stimulated
Taking Conflict Personally 475
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
by the present conflict situation’’ (p. 306). Conflict is more likely to be taken
personally if the conflict is stressful or if there is a perception of unfairness or
aggression. In addition, TCP may be contagious since ‘‘it can manifest itself as
aggression or avoidance, and can challenge or frustrate others into reciprocal TCP
reactions’’ (p. 306).
Hample (1999) discussed the life space as it relates to the various periods of a
conflict, which matches up nicely with the ideas of proactive IIs, actual conflict,
and retroactive IIs. In the preconflict life space, a person can see that in achieving
a goal conflict is possible and ‘‘might imagine a life space in the plane of irreality, a
life space whose elements and climate foreshadow what the person expects and
might therefore cause to happen’’ (p. 176). For individuals who tend to personalize
conflict, the preconflict life space is often characterized by climates of avoidance,
anxiety, lack of confidence, and defensiveness. A person’s perceptions of conflict
also play a role while engaged in a conflict. Although a high personalizer will
attempt to avoid conflict, after becoming engaged in the conflict the high persona-
lizer often responds with aggression or reciprocity. In discussing the postconflict
life space, Hample (1999) noted that ‘‘the aftermath of a conflict is more detailed
and stressful for the high personalizer’’ (p. 193). These remarks are consistent with
outcomes in II work. Honeycutt (2004), for instance, found that ruminating about
a conflict led to specific IIs that occurred both before and after the conflict and were
characterized by unpleasant or negative emotions. In addition, Zagacki et al. (1992)
reported that individuals tend to remember negative emotional experiences more
than positive ones.
IIs, Rumination, and TCP
Not all of the main relationships among IIs, rumination, and TCP have yet been
investigated, and this is the primary aim of the present study. All three of these multi-
dimensional concepts have been studied in similar domains, and we have some
reason to expect that tracing out their connections will be theoretically productive.
A key issue in conflict management is the constructiveness of one’s communi-
cation, and messages, of course, arise from people’s cognitions and affect. An other-
wise unpublished study summarized in Honeycutt (2003, pp. 77–81) shows the
relationships between IIs and both verbal aggressiveness and physical coercion. His
path model shows that the latent variable ‘‘II characteristics’’ (composed mainly of
activity, proactivity, and specificity) had a substantial causal effect (a path coefficient
of .44) on verbal aggression, which in turn was the model’s only determinant of
physical coercion (a path coefficient of .52). Argumentativeness, the constructive
counterpart of verbal aggressiveness (Rancer & Avtgis, 2006), had almost nothing
to do with II characteristics, II functions, verbal aggressiveness, or physical coercion
(all the path coefficients were .06 or smaller). So IIs seem to have more to do with
destructive conflict possibilities than constructive ones, perhaps due to rumination.
As reviewed above, incessant thinking about unpleasant conflict experiences is
dysphoric. It may lead to states of depression and anger. If so, this would help us
476 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
understand why IIs have a clearer effect on bad conflict management practice than
constructive action. TCP, of course, indexes a variety of positive or negative feelings
about conflict.
Studying IIs, rumination, and TCP together should enrich our understanding of
why people seek, entertain or experience destructive conflict. Here we share our
specific research objectives and summarize our rationales for them.
Hample and Dallinger (1995) reported that individuals who tend to personalize
conflict wish to avoid conflict situations. However, this is not always possible, since
conflict often occurs when an individual is attempting to reach a goal. Martin and
Tesser (1996) discuss the idea that rumination occurs when goal attainment is
blocked. In the case of a person who is high in trait TCP, the conflict that may result
in the pursuit of a goal can be conceived of as an obstacle to achieving that goal, but,
for a low personalizer, the engagement may be welcome. Hence:
H
1
: Trait personalizing and ruminating about conflict are significantly related.
IIs about conflict have been found to be more unpleasant than IIs about other
topics (Honeycutt, 2004; Zagacki et al., 1992). Those who personalize conflict find
the experience of engaging in conflict to be particularly negative (Hample, 1999).
It follows that a person who is high in TCP would imagine the interaction as
unpleasant. For that reason:
H
2
: II valence is inversely related to trait direct personalization, persecution
feelings, stress reactions, and negative relational effects and is positively
related to positive relational effects and like=dislike valence.
Rumination about conflict has been associated with IIs that are specific, proactive
and retroactive, discrepant from what actually occurs in the conflict and dominated
by the self (Honeycutt, 2004). Since the relationships between TCP and the variables
associated with IIs about conflict have not been explored, it is unclear whether or not
high personalizers have IIs that are characterized by these features. Understanding the
role that personalization plays in IIs could help to explain why some people tend to
take conflict personally. Accordingly:
RQ
1
: What is the relationship between trait TCP and the II characteristics of
activity, specificity, proactivity, retroactivity, discrepancy, self-dominance,
and variety?
The second research question relates to the tendency of IIs to link past and future
conflicts, which enhances the chances of self-fulfilling prophecies, especially in regard
to the battery of feelings measured by TCP. The issue here is whether state persona-
lization is consistent with trait personalization and, if so, to what degree. Scholarly
literature relating to IIs and conflict often mentions the idea that people can create
a self-fulfilling prophecy in conflict scenarios. Since they imagine the interaction to
be unpleasant, it is. Hample (1997), for instance, suggests that people feel what they
expect to feel in their life spaces. Honeycutt (2003) notes that for individuals to have
productive conflict interactions, they need to be able to imagine positive conflict
scenarios. However, this may be difficult for individuals who take conflict personally
Taking Conflict Personally 477
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
since they view it as punishing. It seems likely that their IIs would contain examples
of personalization. However, this is not altogether clear. Therefore:
RQ
2
: Is there a relationship between trait TCP and the state TCP reported in
conjunction with particular IIs?
Method
Respondents
The participants for this study were 210 undergraduate students enrolled in lower
level communication courses at a Midwestern university. They received extra credit
for taking part. The sample included 93 males and 113 females; 4 participants did not
report sex. Age ranged from 18 to 41 years, with a mean of 21 (SD ¼2.05).
Procedures
Data were collected in two sessions to minimize survey fatigue. The participants
received extra credit for their participation. They had to provide an identification
number used to match responses from the first part of the questionnaire to those
from the second. The first part of the survey consisted of 37 items relating to trait
TCP (Hample & Dallinger, 1995), 39 items to the characteristics of IIs (Honeycutt,
2003), and 22 items to trait rumination (Treynor, Gonzalez, & Nolen-Hoeksema,
2003). The second part of the questionnaire contained nine open-ended questions
instructing the participants to describe their most recent II about conflict, 22 items
that addressed the functions of the II (Honeycutt, 2003), 37 relating to state TCP
(Hample et al., 1995), and 11 involving the amount of rumination concerning the
reported conflict in which the respondents engaged (Cloven & Roloff, 1991). Two
hundred eighty participants received the first part of the materials during normal
class time. They were instructed to complete the questionnaire on a Scantron sheet
and bring it to class with them at the next class meeting, and 225 did. Upon return
of the first part of the survey, the participants proceeded to the second part. Some
participants (123) used given class time; the others completed it on their own time
and returned it to the researcher or their instructor. This resulted in a total of 216
complete part two surveys. Using the identification number provided by the parti-
cipants, responses were matched. The procedure resulted in a total of 210 matched
data sets for analysis.
Instrumentation
TCP scales
The taking conflict personally battery (Hample & Dallinger, 1995) consisted of 37 items
on which the participants responded on 5-point scales ranging from strongly disagree
to strongly agree. The TCP instrument has six subscales reflecting its various dimen-
sions. The direct personalization subscale has seven items (a¼.78, e.g., ‘‘Conflict is
478 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
a very personal thing for me’’). The persecution feelings subscale has six items (a¼.67,
e.g., ‘‘I think that people often attack me personally’’). One item ‘‘Conflict situations
make me feel persecuted’’ was dropped to increase reliability to the reported level. Five
items (a¼.67, e.g., ‘‘Sometimes when there are a lot of conflicts in a week, I feel like I
am getting an ulcer’’) served as the measure of stress reaction, and seven did for
positive relational effects (a¼.79, e.g., ‘‘Conflict can really help a relationship’’). The
negative relational effects subscale had five items (a¼.67, e.g., ‘‘Conflicts have a
negative impact on a relationship’’). The final subscale, like=dislike valence, had seven
items (a¼.79, e.g., ‘‘To me, it’s fun to argue’’). The TCP scales have recently been
analyzed with a large-scale (N¼1322) confirmatory factor analysis (Hample & Cionea,
2010), and the battery performs as its generative theory predicts.
The participants reported an II about conflict before completing the state TCP
items (Hample et al., 1995). This scale was modified to reflect the imagined conflict
rather than an actual conflict. The participants received the following directions: ‘‘In
answering the following questions, please think back to the imagined interaction that
you reported.’’ Cronbach’s alpha for each subscale and examples of items are as
follows: direct personalization (a¼.76, ‘‘I imagined that I took the criticisms person-
ally’’; one item, ‘‘When I imagined that the other person criticized something I said I
didn’t take it personally,’’ was dropped to improve reliability to .76), persecution
feelings (a¼.79, ‘‘I imagined that the other person really liked to pick on me’’), stress
reaction (a¼.80, ‘‘Imagining this conflict made me feel like I’m getting an ulcer’’),
positive relational effects (a¼.87, ‘‘I imagined that the conflict will have a positive
impact on the relationship between me and the other person’’), negative relational
effects (a¼.83, ‘‘I imagined that the conflict could really hurt the relationship
between me and the other person’’), and like=dislike valence (a¼.77, ‘‘To me, it
was fun to imagine this argument’’).
Survey of imagined interaction (SII)
The SII (Honeycutt, 2003) was modified to apply to conflict. For example, if the
original item read ‘‘I have imagined interaction many times throughout the week,’’
in the present study, it read ‘‘I have imagined interactions about conflict many times
throughout the week.’’ Honeycutt has reported strong associations between the SII
and data collected from oral history interviews and journal accounts of IIs.
Participants received the standard instructions (Honeycutt, 2003, p. 28).
The respondents assessed eight II characteristics using 5-point scales, with choices
ranging from strongly disagree to strongly agree. Cronbach’s alphas and sample
questions follow: frequency (a¼.71, ‘‘I have imagined interactions about conflict
many times throughout the week’’), proactivity (a¼.77, ‘‘I often have imagined inter-
actions before an argument’’), retroactivity (a¼.71, ‘‘I often have imagined interac-
tions after an argument’’), variety (a¼.63, ‘‘Most of my imagined interactions
about conflict are with different people’’; one item, ‘‘Most of my imagined interactions
about conflict are with the same person,’’ was dropped to yield the reported a),
discrepancy (a¼.72, ‘‘In my real conversations about conflict, I am very different than
Taking Conflict Personally 479
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
in my imagined ones’’; one item, ‘‘When I have a real conversation about conflict that I
have imagined, the actual conversation is very different from what I imagined,’’ was
dropped for the reason above), self-dominance (a¼.55, ‘‘I talk a lot in my imagined
interactions about conflict’’), valence (a¼.73, ‘‘I enjoy most of my imagined interac-
tions about conflict’’), and specificity (a¼.67, ‘‘When I have imagined interactions
about conflict, they tend to be detailed and well developed’’; one item, ‘‘When I have
an imagined interaction about conflict, I often only have a vague idea of what the other
says,’’ was dropped.) In addition to the characteristics of IIs, participants identified the
person to whom their most recent II about conflict applied, the topic of the II, the set-
ting of the II, and some sample lines of dialogue.
Rumination
The measure of trait rumination was an adaptation of the Ruminated Responses Scale
(RRS; Treynor et al., 2003). The modified directions were as follows: ‘‘People think
and do many different things when they are engaged in conflicts with other people.
As you read the following list of possibilities, please use the following scale to indicate
on the Scantron sheet how often you think or do these things when you are engaged
in a conflict with another person.’’ The participants recorded their assessments on
4-point scales (1 ¼never, 2 ¼sometimes, 3 ¼often, 4 ¼always). The item measure
taps three aspects of rumination. The first was depression, as reflected in 12 items
(e.g., ‘‘Think about how alone you feel’’). Cronbach’s alpha for this scale was .90.
The second was reflection, which related to neutrally or positively valenced thoughts
that people use to cope with feelings. The five items (e.g., ‘‘Analyze recent events and
try to understand why you are depressed’’) proved to be reliable (a¼.80). The final
aspect of rumination measured by the RRS was brooding, ‘‘moody pondering’’
(Treynor et al., 2003, p. 251), which was more negatively valenced than reflection,
and captured by five items (e.g., ‘‘Think, ‘What am I doing to deserve this?’ ’’).
Cronbach’s alpha was .78.
Results
Preliminary Findings
As previously mentioned, respondents reported on their most recent conflict II. Only
10 indicated that they did not have or did not recall an II. Of those who did an II
(N¼200), 70%reported that they had an II within the previous week, 16%within
two weeks, and 14%more than two weeks prior to the survey. The IIs concerned
relational conflicts (44%), friend=roommate relationships (36%), and recreation=
activities (20%). The most common parties were boyfriend=girlfriend (42%), friends
(36%), and family members (27%).
Hypothesis 1
The first hypothesis was trait TCP and trait rumination would be positively related.
The hypothesis received partial support (see Table 1). The three core dimensions of
480 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
TCP (direct personalization, persecution feelings, and stress reactions) all signifi-
cantly and consistently correlated with the three aspects of rumination. These
correlations range from .22 to .41 and constitute the clearest support for the hypoth-
esis. Absent or lesser support for the hypothesis appears in the bottom half of the
table. Reflection was not significantly correlated with positive relational effects,
negative relational effects, or like=dislike valence. Brooding was not significantly
correlated with positive relational effects but was significantly correlated with
negative relational effects, and negatively correlated with like=dislike valence. This
was not surprising, inasmuch as brooding tends to be more negatively valenced than
reflection. Finally, depression was not significantly correlated with positive relational
effects or like=dislike valence but was correlated with negative relational effects. It
appears that individuals who believe that conflict has negative effects on relationships
tend to brood about conflict and may become depressed about it.
Hypothesis 2
The second hypothesis posited relationships involving TCP and II valence. The
correlations appear in Table 2. The hypothesis is partially supported. There was no
significant relationship between II valence and direct personalization, persecution
feelings, or negative relational effects, leading to rejection of this part of the hypoth-
esis. There were significant but small correlations between II valence and positive
relational effects (r¼.14) and stress reactions (r¼.15), both in the hypothesized
directions. The strongest relationship was the one for II valence and like=dislike
valence (r¼.40), which is the hypothesized direction. Those who enjoyed real
conflict also tended to enjoy imagined conflict.
Research Question 1
The first research question focused on the relationships between the six dimensions
of trait TCP and the II characteristics of activity, specificity, proactivity, retro-
activity, discrepancy, self-dominance, and variety. Table 3 shows the correlations.
Table 1 Correlations between TCP and Rumination
Reflection Brooding Depression
Direct Personalization .22 .40 .33
Persecution Feelings .33 .37 .41
Stress Reactions .27 .37 .37
Positive Relational Effects .00 .09 .05
Negative Relational Effects .07 .20 .17
Like=Dislike Valence .00 .18 .09
Note: N ¼210.
p<.05. p<.01.
Taking Conflict Personally 481
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
The expectation of positive relational effects significantly correlated with II activity,
specificity, variety, proactivity, and retroactivity. This suggests that individuals who
believe that conflict has a positive impact on relationships have many specific IIs
concerning a variety of topics that occur both before and after the actual communi-
cation. Interestingly, there were also small significant correlations between direct
personalization and the II characteristics of proactivity and activity, which sug-
gested that individuals who tend to personalize conflict sometimes imagine a con-
flict prior to engaging in it, perhaps as a way of rehearsing for the actual encounter.
Discrepant IIs correlated significantly with feelings of persecution, stress reactions,
negative relational effects, and like=dislike valence. This implies that individuals who
find conflict to be an unpleasant experience may have difficulty accurately imagining
a conflict scenario. They may imagine that the conflict will be more unpleasant and
stressful than it actually turns out to be.
Given the number of variables analyzed and their convenient grouping into two
sets, we also conducted a canonical correlation in addressing this research question.
The results appear in Table 4. The first significant root was dominated by the II char-
acteristics of variety, II activity, and specificity, which collectively predicted positive
Table 3 Correlation between TCP and II Characteristics
Act. Spec. Pro. Retro. Disc. Self-Dom. Var.
Direct Personalization .15.09 .16.08 .10 .01 .06
Persecution Feelings .17.09 .13 .08 .16.06 .15
Stress Reactions .12 .08 .11 .12 .21 .00 .02
Positive Relational Effects .34 .31 .15.25 .07 .04 .24
Negative Relational Effects .00 .04 .05 .01 .14.00 .07
Like=Dislike Valence .10 .01 .01 .01 .20 .09 .18
Note: N ¼210. Act. ¼II activity; Spec. ¼II specificity; Pro. ¼proactivity; Retro. ¼retroactivity; Disc. ¼
discrepancy; Self-Dom. ¼self-dominance; Var. ¼variety.
p<.05. p<.01.
Table 2 Correlations between TCP and II Valence
II Valence
Direct Personalization .08
Persecution Feelings .10
Stress Reactions .15
Positive Relational Effects .14
Negative Relational Effects .13
Like=Dislike Valence .40
Note: N ¼210.
p<.05. p<.01.
482 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
relational effects. The second significant root was dominated by the II characteristics
of specificity, retroactivity, and discrepancy and was predictive of like=dislike valence.
The answer to the first research question, then, is that trait TCP and II character-
istics are, in fact, associated. Stress reactions, positive relational effects, and like=
dislike valence have important relationships to all the aspects of IIs except
self-dominance.
Research Question 2
The second research question focused on whether trait TCP predicts state TCP in
reaction to a particular II. Table 5 shows the correlations between trait and state
TCPs. There were numerous significant correlations between dimensions of trait
and state TCP, which indicated that trait predispositions are predictive of state orien-
tations in these circumstances. The diagonal associations (e.g., trait direct personali-
zation compared to state direct personalization) were especially notable. All six
dimensions proved to be predictable, either with the zero-order correlations in
Table 5 or with multiple regressions predicting the state dimensions from all of
the trait measures: state direct personalization (R
2
¼.32, p<.01), state persecution
feelings (R
2
¼.21, p<.01), state stress reaction (R
2
¼.22, p<.01), state positive
Table 4 Loadings and Canonical Correlations Predicting TCP by II Characteristics
TCP Dimensions Root 1 Root 2 R
2
(7, 202)
Direct Personalization .14 .43 .05
Persecution Feelings .30 .29 .07
Stress Reactions .02 .67 .07
Positive Relational Effects .71 .37 .14
Negative Relational Effects .01 .21 .03
Like=Dislike Valence .56 .74 .12
Percentage of Variance 15.7 24.1
II Characteristics R
2
(6, 203)
Activity .72 .34 .12
Proactivity .42 .43 .06
Retroactivity .52 .55 .09
Variety .84 .01 .14
Specificity .63 .60 .12
Self-Dominance .13 .26 .02
Discrepancy .13 .53 .07
Percentage of Variance 30.0 18.8
R
c
.44 .32
R2
c.19 .11
Note: N ¼210.
p<.05. p<.01. p<.001.
Taking Conflict Personally 483
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
relational effects (R
2
¼.09, p<.01), state negative relational effects (R
2
¼.08,
p<.01), and state like=dislike valence (R
2
¼.17, p<.01). To sum up these analyses,
trait TCP does predict the state TCP resulting from an II.
Discussion
Prior to this study, the relationship between taking conflict personally and imagined
interactions about conflict had not been explored. For this reason, some of the
research questions and hypotheses posed in the study were broad. The present study
sheds some light on the relationship between TCP and IIs about conflict and so may
lead to more precise formulations in the future.
IIs about conflict tend to be characterized by a high level of emotion (Zagacki et al.,
1992). They are dominated by the person imagining and frequently involve relational
conflict (Honeycutt et al., 1989–1990). The most frequent type of other parties in IIs
is a relational partner (Edwards et al., 1988). Interestingly, in this sample, the most
recent II about conflict concerned, and was with, a friend or roommate rather than
a romantic partner. Perhaps the discrepancy is attributable to the emotion involved
in romantic relationships that makes these IIs more memorable or the relationship
status in the current sample, which participants most frequently described as casual
dating.
Individuals who believe that conflict has a positive impact on relationships have
more IIs involving conflicts than those who do not. The TCP dimension of positive
relational effects seems to be the one most closely related to IIs concerning conflict.
The belief that conflict has a positive relational effect results in frequent IIs that are
specific and occur both before and after an actual conflict about a variety of topics. It
appears that people who believe that conflict is positive spend more time thinking
about, planning, and rehearsing future conflict scenarios, perhaps in an attempt to
maximize the relational benefits. In addition, such individuals may also spend time
thinking about past conflicts to analyze how best to improve the outcomes of future
conflicts. Those who believe that conflict has a positive effect on relationships also use
Table 5 Correlations between Trait and State TCP
SDP SPF SSR SPRE SNRE SVAL
Direct Personalization .53 .19 .25 .07 .14.17
Persecution Feelings .39 .39 .29 .09 .19 .11
Stress Reactions .38 .31 .43 .02 .24 .33
Positive Relational Effects .06 .08 .08 .25 .04 .14
Negative Relational Effects .18.14.20 .06 .19 .18
Like=Dislike Valence .18 .04 .14 .08 .12 .37
Note: N ¼210. SDP ¼State Direct Personalization; SPF ¼State Persecution Feelings; SSR ¼State Stress Reac-
tions; SPRE ¼State Positive Relational Effects; SNRE ¼State Negative Relational Effects; SVAL ¼
State Like=Dislike Valence. The row variables are the trait measures.
p<.05. p<.01.
484 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
IIs to relieve stress and make themselves feel better. It is interesting that there was no
mirror image for negative relational effects. Possibly individuals who believe that
conflict has a negative effect on relationships imagine it as a negative experience
and, therefore, avoid thinking about it.
One of the most striking findings was the relationship between the core TCP
dimensions of direct personalization, persecution feelings, and stress reactions and
rumination about conflict. People who scored high on these three TCP dimensions
were much more likely to report having recurrent intrusive negative thoughts about
conflict. These thoughts often take the form of IIs. It was beyond the scope of the
current study to determine whether a tendency to take conflict personally leads to
rumination about conflict, or ruminating about conflict leads to a higher degree of
direct personalization, persecution feelings, and stress reactions, but this would be
a suitable topic for further investigation.
Hample et al. (1995) found that trait TCP is predictive of state TCP stimulated by
actual face-to-face conflict. In the present study, we found that trait TCP also predicts
state TCP as it relates to IIs. For example, if an individual is high in direct persona-
lization, his or her IIs about conflict will also be characterized by feelings of direct
personalization. This holds true for all the core dimensions of TCP. It may be difficult
for a person who has a tendency to take conflict personally to imagine a conflict
that is not personal. This has implications for the use of IIs in therapeutic situations.
Honeycutt (2003) notes that in therapeutic situations negative IIs often intrude on
positive IIs. One reason for this may be the tendency to take conflict personally.
Hypothesis 1 received only partial support; however, the three core dimensions
of TCP (direct personalization, persecution feelings, and stress reactions) all signifi-
cantly correlated with the three aspects of rumination. It appears that people who
take conflict personally tend to think about conflicts to an unhealthy degree. In
addition, negative relational effects correlated significantly with the two negatively
valenced aspects of rumination (brooding and depression). Positive relational effects
did not correlate significantly with rumination about conflict. One explanation for
this is that individuals who expect conflict to have a positive effect on their relation-
ships do not spend an inordinate amount of time worrying about it.
Hypothesis 2 also received only partial support. The most significant finding was
that II valence correlated positively with TCP like=dislike valence. Those who enjoy
conflict also tend to enjoy imagining conflict. In addition, there was a small signifi-
cant correlation between positive relational effects and II valence. This indicates that
individuals who believe that conflict has a positive effect on relationships tend to have
more pleasant IIs about conflict than those who do not. Stress reaction was negatively
correlated with II valence, which suggests that individuals who find conflict to be
stressful have unpleasant IIs about conflict. The other three dimensions of TCP
(direct personalization, persecution feelings, and negative relational effects) were
not significantly correlated with II valence. One explanation for this may be that
people scoring high on these dimensions of TCP do not spend a lot of time thinking
about conflict since it is an unpleasant experience for them, and they therefore do
not report either pleasant or unpleasant IIs.
Taking Conflict Personally 485
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
This study, like nearly all we have reviewed for this project, deals with people’s
self-reports of their thoughts and feelings about conflict. While several TCP investiga-
tions have connected trait and state scores to actual behavior (reviewed in Hample &
Cionea, 2010), the II and rumination research have not yet moved solidly into the
behavioral domain. Given the apparent importance of these factors in people’s
self-understandings and relational expectations, this is an important deficiency in
the literature and should be remedied.
Conclusion
Relational conflicts are often recurrent. Although study of each conflict episode is
important, this study focused on several processes that connect serial arguments
and give them a stable historical character within a relationship. Rumination and
imagining an interaction offer immediate bridges from past to present to future
conflict. This study has shown that the degree to which one personalizes conflict is
a correlate of both rumination and imagined conflict interaction. Inclusion of the
personalization concepts in our theories of serial argument and recurrent conflict will
improve our grasp of these phenomena.
References
Allen, T. H., & Honeycutt, J. M. (1997). Planning, imagined interaction, and the nonverbal display
of anxiety. Communication Research,24, 64–82.
Barch, S. L. (1994). Assessing HMO CEOs’ communication competence and propensity to take conflict
personally (Unpublished master’s thesis). Western Illinois University, Macomb.
Bushman, B. J. (2002). Does venting anger feed or extinguish the flame: Catharsis, rumination,
distraction, anger, and aggressive responding. Personality and Social Psychology Bulletin,
28, 724–731.
Cloven, D. H., & Roloff, M. E. (1991). Sense-making activities and interpersonal conflict:
Communicative cures for the mulling blues. Western Journal of Speech Communication,55,
134–154.
Cloven, D. H., & Roloff, M. E. (1993). Sense-making activities and interpersonal conflict, II:
The effects of communicative intentions on internal dialogue. Western Journal of Speech
Communication,57, 309–329.
Edwards, R., Honeycutt, J. M., & Zagacki, K. S. (1988). Imagined interaction as an element of social
cognition. Western Journal of Speech Communication,52, 23–45.
Edwards, R., Honeycutt, J. M., & Zagacki, K. S. (1989). Sex differences in imagined interactions.
Sex Roles,21, 263–272.
Folger, J. P., Poole, M. S., & Stutman, R. K. (2005). Working through conflict: Strategies for relation-
ships, groups, and organizations (5th ed.). Boston: Pearson Education.
Hample, D. (1997). Framing message production research with field theory. In J. O. Greene (Ed.),
Message production: Advances in communication theory (pp. 171–192). Mahwah, NJ:
Lawrence Erlbaum.
Hample, D. (1999). The life space of personalized conflicts. In M. E. Roloff (Ed.), Communication
yearbook 22 (pp. 171–207). Newbury Park, CA: Sage.
Hample, D., Benson, E., Gogliotti, L., & Jeong, J. (1997, November). Ego defense and taking conflict
personally. Paper presented at the annual meeting of the National Communication
Association, Chicago.
486 The Southern Communication Journal
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
Hample, D., & Cionea, I. A. (2010). Taking conflict personally and its connections with
aggressiveness. In T. A. Avtgis & A. S. Rancer (Eds.), Arguments, aggression, and conflict:
New directions in theory and research (pp. 372–387). New York, NY: Routledge=Taylor
and Francis.
Hample, D., & Dallinger, J. M. (1995). A Lewinian perspective on taking conflict personally:
Revision, refinement, and validation of the instrument. Communication Quarterly,43,
297–319.
Hample, D., Dallinger, J. M., & Fofana, J. (1995). Perceiving and predicting the tendency to perso-
nalize arguments. In S. Jackson (Ed.), Argumentation and values (pp. 434–438). Annandale,
VA: Speech Communication Association.
Honeycutt, J. M. (1995). The oral history interview and reports of imagined interactions. The
Journal of Family Psychotherapy,6, 63–69.
Honeycutt, J. M. (2003). Imagined interactions: Daydreaming about communication. Cresskill, NJ:
Hampton.
Honeycutt, J. M. (2004, November). Forgive but don’t forget: Correlates of rumination about conflict.
Paper presented at the National Communication Association Conference, Chicago.
Honeycutt, J. M., & Cantrill, J. G. (2001). Cognition, communication, and romantic relationships.
Mahwah, NJ: Erlbaum.
Honeycutt, J. M., Edwards, R., & Zagacki, K. S. (1989–1990). Using imagined interaction features to
predict measures of self-awareness: Loneliness, locus of control, self dominance, and
emotional intensity. Imagination, Cognition and Personality,9, 17–31.
Honeycutt, J. M., Zagacki, K. S., & Edwards, R. (1992–1993). Imagined interaction, conversational
sensitivity, and communication competence. Imagination, Cognition and Personality,12,
139–157.
Johnson, K. L., & Roloff, M. E. (1998). Serial arguing and relational quality: Determinants and con-
sequences of perceived resolvability. Communication Research,25, 327–343.
Lyubomirsky, S., Tucker, K. L., Caldwell, N. D., & Berg, K. (1999). Why ruminators are poor prob-
lem solvers: Clues from the phenomenology of dysphoric rumination. Journal of Personality
and Social Psychology,77, 1041–1060.
Mallouk, T. (1981). Interpersonal conflict and mental imagery. In E. Klinger (Ed.), Imagery:
Concepts, results, and applications (Vol. 2, pp. 311–326). New York, NY: Plenum.
Martin, L. L., & Tesser, A. (1996). Some ruminative thoughts. In R. S. Wyer (Ed.), Ruminative
thoughts: Advances in social cognition (Vol. IX, pp. 1–49). Mahwah, NJ: Erlbaum.
McCullough, M. E., Bellah, C. G., Kilpatrick, S. D., & Johnson, J. L. (2001). Vengefulness: Relation-
ships with forgiveness, rumination, well-being, and the big five. Personality and Social
Psychology Bulletin,27, 601–610.
Mish, F. C. (Ed.). (1995). Merriam Webster’s collegiate dictionary (10th ed.). Springfield, MA:
Merriam-Webster.
Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of depressive
episodes. Journal of Abnormal Psychology,100, 569–582.
Nolen-Hoeksema, S. (1996). Chewing the cud and other ruminations. In R. S. Wyer (Ed.), Rumi-
native thoughts: Advances in social cognition (Vol. IX, pp. 135–144). Mahwah, NJ: Lawrence
Erlbaum.
Rancer, A. S., & Avtgis, T. A. (2006). Argumentative and aggressive communication: Theory, research,
and application. Thousand Oaks, CA: Sage.
Trapp, R., & Hoff, N. (1985). A model of serial argument in interpersonal relationships. Journal of
the American Forensic Association,22, 1–11.
Treynor, W., Gonzalez, R., & Nolen-Hoeksema, S. (2003). Rumination reconsidered: A psycho-
metric analysis. Cognitive Therapy and Research,27, 247–259.
Zagacki, K. S., Edwards, R., & Honeycutt, J. M. (1992). The role of mental imagery and emotion in
imagined interaction. Communication Quarterly,40, 56–68.
Taking Conflict Personally 487
Downloaded By: [Hample, Dale] At: 15:47 20 November 2010
