Content uploaded by Nathalie Houlfort
Author content
All content in this area was uploaded by Nathalie Houlfort
Content may be subject to copyright.
Passion and Interpersonal Relationships 1
Running head: PASSION AND INTERPERSONAL RELATIONSHIPS
IN PRESS JOURNAL OF PERSONALITY AND SOCIAL PSYCHOLOGY
Passion for an Activity and Quality of Interpersonal Relationships:
The Mediating Role of Emotions
Frederick L. Philippe
McGill University
Robert J. Vallerand
Laboratoire de Recherche sur le Comportement Social
Université du Québec à Montréal
Nathalie Houlfort
École Nationale d’Administration Publique
Geneviève L. Lavigne & Eric G. Donahue
Laboratoire de Recherche sur le Comportement Social
Université du Québec à Montréal
August 28, 2009 Frederick L. Philippe
Human Motivation Lab
Department of Psychology
McGill University
1205 Dr. Penfield Avenue
Montreal, Quebec, Canada
H3A 1B1
Phone: 514-398-8219
e-mail: frederick.philippe@mail.mcgill.ca
Passion and Interpersonal Relationships 2
Abstract
The purpose of this research was to investigate the role of passion (Vallerand et al., 2003)
for a given activity in the quality of interpersonal relationships experienced within the context of
that activity in four studies. Study 1 demonstrated that a harmonious passion was positively
associated with the quality of interpersonal relationships within the context of the passionate
activity, while an obsessive passion was unrelated to it. Furthermore, in line with the Broaden-
and-Build Theory (Fredrickson, 2001), results also showed that positive emotions experienced at
work fully mediated the relation between harmonious passion and quality of interpersonal
relationships. Obsessive passion was not associated with positive emotions. Study 2 replicated
the results from Study 1 while controlling for trait extraversion and also examined the negative
mediating role of negative emotions between obsessive passion and the quality of interpersonal
relationships. Finally, Studies 3 and 4 replicated the results of Study 2 with prospective designs
and with objective ratings of interpersonal relationships quality. Implications for the Dualistic
Model of Passion and the Broaden-and-Build Theory are discussed.
Keywords: Passion, Interpersonal relationships, Emotion, Broaden-and-Build Theory
Passion and Interpersonal Relationships 3
Passion for an Activity and Quality of Interpersonal Relationships:
The Mediating Role of Emotions
People who are passionate for an activity are often depicted as being energetic and
magnetic by others. Their intense focus on the activity, their energy, and the strong positive
emotions that they experience and display often attract and inspire others to connect with them.
Thus, unsurprisingly, people highly involved in an activity are usually seen as very popular and
come to develop positive relationships with the people with whom they partake in the activity. At
the same time, people who are passionate for an activity are also described as putting so much
emphasis on the activity that they come to experience stress and negative emotions. When
displayed, such emotions are likely to be a major turn off for other participants, thereby
undermining the quality of relationships developed within the purview of the activity. We are
thus left with two possibilities: Is passion for an activity benefiting or deteriorating one’s quality
of interpersonal relationships in that activity? Furthermore, do emotions play a mediating role in
the quality of interpersonal relationships that passionate individuals come to develop in that
activity? The purpose of this paper was to examine these issues thus extending knowledge on
passion, relationships, and the social functions of emotions.
A Dualistic Model of Passion
If passion has been an important concept in philosophy throughout centuries (see Rony,
1990 for a review), it has received very little attention in psychology until very recently (see
Vallerand et al., 2003; Vallerand et al., 2007, on this issue). It is only in the last few years that
Vallerand and his colleagues (Vallerand, 2008; Vallerand et al., 2003; Vallerand & Houlfort,
2003) have formulated a Dualistic Model in which passion is defined as a strong inclination or
Passion and Interpersonal Relationships 4
desire toward a self-defining activity that one likes (or even loves), finds important (high
valuation), and in which one invests time and energy. Passion is conceptualized as a moderate-
level construct that is positioned at the contextual level, between the global (trait) and situational
(state) levels (see Vallerand, 1997 for a theoretical account of the these different levels of a
three-level hierarchy). A given construct at the contextual level refers to a relatively stable
person–activity relationship that reflects how a person typically behaves with respect to a
specific activity (and not necessarily with respect to all activities).1
This theoretical model proposes that there exist two types of passion. The first type of
passion is harmonious passion. A harmonious passion refers to a strong desire to engage in the
activity which remains under the person’s control. This type of passion results from an
autonomous internalization of the activity into the person’s identity (Vallerand et al., 2003;
Vallerand, Rousseau, Grouzet, Dumais, & Grenier, 2006). An autonomous internalization occurs
when individuals have freely accepted the activity as important for them without any
contingencies attached to it (Sheldon, 2002; Vallerand, 1997). The activity is thus part of an
integrated self-structure (Hodgins & Knee, 2002). In such a case, the activity occupies a
significant but not overpowering space in the person’s identity and is in harmony with other
aspects of the person’s life. With harmonious passion, the person is in control of the activity and
can freely decide when to engage or not in the activity and even when to stop activity
engagement. The activity is engaged in because of the pleasure that is drawn from it and not
from other extrinsic sources. An example of such a type of passion would be a student with a
passion for basketball who has to study for an important exam the next day. Because this student
is characterized by a harmonious passion for basketball, if his friends call to propose a basketball
Passion and Interpersonal Relationships 5
scrimmage, he will be able to turn down the invitation in order to study for his exam without
experiencing much conflict.
The second type of passion identified by Vallerand and colleagues (2003) is obsessive
passion. This type of passion entails a strong desire to engage in the activity that is not under the
person’s control. It is as if the activity controlled the person. Obsessive passion results from a
controlled internalization of the activity into one’s identity (Vallerand et al., 2003; Vallerand et
al., 2006). Such an internalization originates from intra and/or interpersonal pressure because
certain contingencies are attached to the mental representations of the activity, such as feelings
of social acceptance or self-esteem. Thus, individuals like the activity so much that they cannot
help but to engage in it due to a lack of control over internal contingencies that come to control
the person and preclude the experience of volition in activity engagement. To go back to the
above example, if the student has an obsessive passion for basketball and that his friends call to
propose a basketball scrimmage, he will probably accept the invitation and go play basketball,
while feeling guilty for not studying. This conflicted form of activity engagement should lead to
a narrow range of experiences that can reduce the positive affective outcomes that would be
normally experienced, and even increase negative emotional experiences (e.g., guilt, anxiety).
Research has provided empirical support for the concept of passion. Results from
exploratory and confirmatory factor analyses supported the two-factor structure of the Passion
Scale, an instrument developed to measure harmonious passion and obsessive passion (e.g.,
Vallerand et al., 2003, Study 1; Vallerand, et al., 2006, Study 1). The Passion Scale was also
shown to have high levels of internal consistency as well as predictive, discriminant, construct,
and external validity. Results using this scale revealed that both harmonious passion and
obsessive passion were positively associated with measures of activity valuation, perceptions of
Passion and Interpersonal Relationships 6
the activity as a passion, time and energy investment in the activity, and inclusion of the activity
in the self (Vallerand et al., 2003, Study 1). Both types of passion were also found to be
positively associated with activity perseverance and performance (Vallerand et al., 2008, Study
1; Vallerand et al., 2007, Study 1).
However, each type of passion has also been found to be differently associated with a
variety of outcomes. Indeed, harmonious passion has been found to be positively correlated with
positive experiences such as flow and positive emotions (Mageau, Vallerand, Rousseau, Ratelle,
& Provencher, 2005; Vallerand, 2003, Study 1; Vallerand et al., 2006, Studies 2 and 3), reduced
negative emotions (Mageau et al., 2005; Vallerand et al., 2003, Study 1), and higher levels of
concentration during task engagement (Mageau et al., 2005; Vallerand et al., 2003, Study 1).
Harmonious passion has also been found to be positively associated with psychological
adjustment indices (Rousseau & Vallerand, 2003, 2008; Vallerand et al., 2008, Study 2;
Vallerand et al., 2007, Studies 1 and 2). Conversely, obsessive passion has been found to be
positively associated with negative emotions during task engagement (Vallerand et al., 2003,
Study 1), anxiety (Rousseau & Vallerand, 2003), and rumination when the person is prevented
from engaging in his/her passionate activity (Ratelle et al., 2004; Vallerand et al., 2003, Study 1).
In addition, obsessive passion has been shown to negatively predict psychological adjustment
indices (Rousseau & Vallerand, 2003, 2008; Vallerand et al., 2007, Study 2; Vallerand et al.,
2008, Study 2).
As can be seen from the above, several studies have provided support for the Dualistic
Model of Passion as pertains to a variety of intrapersonal outcomes. However, very little research
has been conducted at the interpersonal level. In fact, only one study to our knowledge has
looked at the role of passion in interpersonal relationships. In this particular study, Séguin-
Passion and Interpersonal Relationships 7
Lévesque, Laliberté, Pelletier, Blanchard, and Vallerand (2003) found that a harmonious passion
for the Internet was associated with better dyadic adjustment in romantic couples, while an
obsessive passion for the Internet was positively correlated with interpersonal conflict in the
couple. However, this study focused on passion for the Internet and its consequences on couple
adjustment and not on the quality of interpersonal relationships with the people with whom the
activity is engaged in. Thus, the results from Seguin-Lévesque and colleagues’ study only
provide indirect support with respect to the role of passion in the quality of interpersonal
relationships experienced within the purview of the passionate activity.
It is hypothesized that the two types of passion should lead to different levels of quality of
interpersonal relationships in the activity. As described above, harmonious passion derives from
an autonomous internalization of the activity. The individual with an harmonious passion is thus
capable of a flexible engagement in the activity that is conducive to a willingness to experience
what is occurring at the present moment (Hodgins & Knee, 2002). This process should thus lead
the person to experience more fully positive emotions during task engagement (Hodgins & Knee,
2002). Past research has indeed found harmonious passion to be positively related to positive
emotions during task engagement (Mageau et al., 2005; Vallerand et al., 2003, Study 1). In
parallel to this, Fredrickson (2001) has proposed through the Broaden-and-Build Theory that
positive emotions have the virtue of broadening people’s thought-action repertoires and self,
leading to the inclusion of other people in one’s self and thus to increased closeness and
relationship satisfaction with these other people (Aron et al., 2000; Fredrickson, 2001; Waugh &
Fredrickson, 2006). Research has shown that positive emotions experienced in a given context
facilitate the broadening of thought-action repertoires and self (Fredrickson & Branigan, 2005;
Waugh & Fredrickson, 2006), leading to an increase in perceived similarities with others
Passion and Interpersonal Relationships 8
(Dovidio, Gaertner, Isen, & Lowrance, 1995; Johnson & Fredrickson, 2005) and in positive
interpersonal relationships (Waugh & Fredrickson, 2006). It is thus posited that the experience of
positive emotions triggered by harmonious passion should promote the quality of interpersonal
relationships (Cappella, 1997; Shiota, Campos, Keltner, & Hertenstein, 2003; Walter & Bruch,
2008).
Conversely, because it originates from a controlled internalization of the activity,
obsessive passion should result in a rigid and conflicted impersonal functioning and in emotional
defensiveness (Hodgins & Knee, 2002). Such a rigid and defensive style should lead to self-
closure from intrapersonal and interpersonal experiences (Aron, Aron, & Smolan, 1992), to a
poor integrative experience during task engagement (Hodgins & Knee, 2002), and thus to a
negative emotional experience, while reducing the positive affective outcomes that would have
been normally experienced (Hodgins & Knee, 2002). Past research has indeed shown that
obsessive passion is either negatively related or unrelated to positive emotions, but positively
associated with negative emotions during task engagement (Mageau et al., 2005; Vallerand et al.,
2003, Study 1; Vallerand et al., 2006, Studies 2 and 3). In line with Fredrickson’s Broaden-and-
Build Theory (2001), negative emotions are posited to lead to a narrowing of the thought-action
repertoires and self (Fredrickson & Branigan, 2005) and thus to poor interpersonal relationships
(Levenson & Gottman, 1983; Waugh & Fredrickson, 2006). Therefore, the different emotional
consequences of each type of passion might mediate the quality of interpersonal relationships
experienced within the purview of the passionate activity.
The Present Research
Emotions are often considered as serving important social functions (Frijda, 2006; Keltner
& Haidt, 2001). Indeed, research has shown that positive emotions represent critical ingredients
Passion and Interpersonal Relationships 9
that facilitate positive interpersonal relationships (e.g., Algoe, Haidt, & Gable, 2008; Walter &
Bruch, 2008; Waugh & Fredrickson, 2006). However, research is lacking on the personal factors
that trigger these positive emotions. As shown above, recent research on passionate activities has
gathered much evidence demonstrating that having a passion for an activity in one’s life may
lead to a host of affective outcomes within the purview of the activity. In line with the Broaden-
and-Build Theory (Fredrickson, 2001), it can be expected that these affective outcomes may, in
turn, influence the quality of interpersonal relationships with the people with whom the activity
is engaged in. However, such a ―Passion-Emotions-Interpersonal Relationships‖ sequence has
remained unexplored. The purpose of the present research was to examine this sequence. Four
studies were thus conducted. Study 1 investigated whether an association existed between
passion and the quality of interpersonal relationships with a sample of middle-age workers
involved in challenging and interesting positions that people can be passionate about. It also
aimed to test the mediating role of positive emotions in the relation between passion and the
quality of interpersonal relationships. The purpose of Study 2 was to replicate the results of
Study 1 and to additionally examine the mediating role of negative emotions, while controlling
for trait extraversion. Finally, Studies 3 and 4 used prospective designs to replicate the results of
Study 2 with a wider range of emotional and interpersonal measures, including objective
interpersonal relationships ratings.
Study 1
Study 1 had two purposes. First, we aimed to investigate if a link existed between passion
and the quality of interpersonal relationships experienced in the course of a passionate activity.
As an initial investigation of this association, work has been selected as an activity that people
can be passionate about and that involves interpersonal relationships. Second, we sought to
Passion and Interpersonal Relationships 10
determine the nature of the psychological processes involved in the relation between passion and
interpersonal relationships. In line with past research with both the Dualistic Model of Passion
and the Broaden-and Build Theory, it was hypothesized that harmonious passion for an activity
(in this case the person’s work) should lead to the experience of positive emotions within the
context of that activity that, in turn, were expected to be conducive to the quality of interpersonal
relationships with work colleagues. Conversely, in line with past research, it was hypothesized
that obsessive passion should be negatively associated or unrelated to positive emotions and,
therefore, should not be conducive to the quality of interpersonal relationships.
Method
Participants
Participants were 195 (97 females, 96 males, and 2 missing values) employees from the
Greater Montreal area working in challenging positions in education as teachers or in
administration as managers. These work positions were selected, as they were very likely to be
passionate profession. Participants were 42 years old on average (SD = 9.68 years). They had
occupied their position for an average of 11.41 years (SD = 8.78 years), and were working on
average 32.87 hours per week (SD = 8.44 hours).
Measures
Passion. The Passion Scale (Vallerand et al., 2003) was administered in order to assess
the type of passion that characterizes participants’ work. The Passion Scale is composed of two
subscales of six items, each assessing a precise type of passion, harmonious and obsessive
passion. Each item is responded to on a 7-point Likert scale ranging from 1 (do not agree at all)
to 7 (completely agree). Sample items for obsessive passion are ―I have difficulties controlling
my urge to do my work‖ or ―I have almost an obsessive feeling for my work‖, while sample
Passion and Interpersonal Relationships 11
items for harmonious passion are ―The new things that I discover in my work allow me to
appreciate it even more‖ or ―My activity is in harmony with other things that are part of me‖.
Alphas in this study were .79 for each of the two subscales.2
Positive Emotions. Two items from the Work Affective and Cognitive Consequences
Scale (WACCS; Blais & Vallerand, 1991) were used to measure positive emotions experienced
at work (happy and in a good mood). Participants were asked to rate the extent to which they
experienced each emotion on a 7-point Likert scale (1 = do not agree at all, 7 = very strongly
agree) according to the stem ―In general, when I do my work I realize that I am…‖. Alpha was
.70.
The Quality of Interpersonal Relationships Scale (QIRS; Sénécal, Vallerand, & Vallières,
1992) is composed of four items responded to on a 5-point Likert scale (0 = not at all, 4 =
extremely). Participants were asked to indicate to what extent the relationships they have with the
people at work are enriching, satisfying, harmonious, and inspire trust. Adequate validity and
reliability indices have been reported for this scale (see Sénécal et al., 1992). Alpha in this study
was .91.
Procedure
Workers were reached through internal mail at their workplace. A written letter asked
them if they would be interested in completing a short questionnaire. Participants were told that
the purpose of the questionnaire was to learn more about what is satisfying and what is not at
work. They were informed that they were free to participate or not and that their answers would
remain anonymous and would never be sent to their supervisor or to anyone in their company or
institution. All questionnaires were returned in a pre-addressed and stamped envelope.
Results and Discussion
Passion and Interpersonal Relationships 12
Table 1 reports the means, standard deviations, and correlations of all study variables. To
assess the relationship between passion and the quality of interpersonal relationships experienced
during activity engagement (work), a multiple regression analysis was conducted. The quality of
interpersonal relationships served as the dependent variable in the regression analysis and the
two types of passion as predictor variables. Finally, because a higher number of hours weekly
worked may lead to better interpersonal relationships, this variable was controlled for in the
analysis. Results of the regression yielded a significant model (F = 6.82, p < .000, R2 = .10).
Results also uncovered that harmonious passion was a significant predictor of quality of
interpersonal relationships ( = .28, p < .001). Obsessive passion was again not associated with it
( = -.02, ns). The number of hours weekly worked was also significant ( = .16, p < .05).
The second hypothesis stated that positive emotions should mediate the relationship
between harmonious passion and the quality of interpersonal relationships. In order to test this
model, a path analysis was conducted with LISREL 8 (Jöreskog & Sörborm, 2003). The
covariance matrix served as the database for the path analysis and the method of estimation was
Maximum Likelihood. All the path analyses conducted in the present research were performed
using this software and method. The present model was composed of three exogenous variables
(harmonious passion, obsessive passion, and number of hours weekly worked) and two
endogenous variables (positive emotions and quality of interpersonal relationships). The paths
were drawn according to the hypotheses presented above. In addition, because the time spent at
work may lead to better interpersonal relationships with work colleagues, a positive path
between the number of hours weekly worked and quality of interpersonal relationships was
added.
Passion and Interpersonal Relationships 13
Results of the path analysis revealed a satisfactory fit to the data,
2 (df = 3, n =
195) = 1.56, p = .67, NFI = 1.00, NNFI = 1.00, CFI = 1.00, RMSEA = .00 [.00; .094], GFI =
1.00, SRMR = .022. As shown in Figure 1, all estimated paths were significant at least at p < .05,
except for the path between obsessive passion and positive emotions that was non-significant (
= -.05, ns). Inspection of the correlation residuals revealed that all were non-significant,
indicating that additional paths would not be significant and would not improve the model fits
(Kline, 2005). In addition, direct paths from harmonious passion and obsessive passion to quality
of interpersonal relationships were not significant and their inclusion in the model did not affect
the fit indices (2[2] = 1.26, ns) or the coefficients of the hypothesized paths, thus suggesting
mediation. Bias-corrected bootstrapped 95% confidence interval estimates of the indirect effect
(see Preacher & Hayes, 2008; Shrout & Bolger, 2002) were then conducted to confirm the
significance of this mediation of positive emotions between harmonious passion and quality of
interpersonal relationships. Bootstrapping is a statistical method that randomly takes a sample
size of n cases from an original sample to estimate and values and replaces these cases into
the original sample. This process is repeated 5000 times, yielding 5000 estimates of the indirect
effect of X on Y. These 5000 estimates thus create a sampling distribution of the indirect effect
and a 95% confidence interval can be extracted from it. Bootstrapped confidence intervals
provide a more accurate estimate of the indirect effect with small-to-moderate samples size
compared to the Sobel tests (see Shrout & Bolger, 2002). The indirect effect is significant at p <
.05 if the 95% confidence intervals do not include the value of zero. As seen in Table 2, this test
confirmed that the mediation effect was significant.
An alternative model tested if the quality of interpersonal relationships mediated the
relationship between harmonious passion and emotions (passion quality of interpersonal
Passion and Interpersonal Relationships 14
relationships positive emotions). Results revealed that the fit indices of this alternative model
were very poor compared to those of the hypothesized model:
2 (df = 3, n = 195) = 50.89, p >
.000, RMSEA = .288 [0.22; 0.36], CFI = .60, NFI = .61, SRMR = .13. In addition, modification
indices suggested the inclusion of a path from harmonious passion to positive emotions. These
are clear evidence that the hypothesized model should be preferred. The model with positive
emotions predicting both passion and interpersonal relationships would represent an equivalent
model (equal fit indices) and inversing the whole original path (interpersonal relationships
positive emotions passion) would also appear plausible. Past research, however, has
consistently shown that passion is a predictor of the consequences of passionate activity
engagement and not an outcome per se (for a review see Vallerand, 2008; see also Carbonneau,
Vallerand, Fernet, & Guay, 2008 for a cross-lagged panel study that provides empirical support
for the present position).
In sum, as predicted, harmonious passion was positively related to positive emotions that
were in turn associated with positive quality of interpersonal relationships at work. Conversely,
obsessive passion was not associated with either positive emotions or quality of interpersonal
relationships at work. Thus, overall, the present findings provide preliminary support for positive
emotions as the psychological process responsible for the positive relation between harmonious
passion and quality of interpersonal relationships.
Study 2
Study 2 had three purposes. First, in order to determine more clearly if the results
obtained in Study 1 were indeed due to passion for an activity and not to the mere result of a
positive interpersonal orientation, we replicated Study 1, this time controlling for trait
extraversion. Controlling for extraversion was deemed important, as it has been shown to be
Passion and Interpersonal Relationships 15
associated with both positive emotional experiences and positive interpersonal relationships
(Costa & McCrae, 1980; Watson & Clark, 1997). In line with Study 1, it was first hypothesized
that harmonious passion would positively predict the quality of interpersonal relationships even
when controlling for trait extraversion, while obsessive passion would be unrelated to the quality
of interpersonal relationships.
Study 1 used only two items to assess positive emotions and those items did not clearly
distinguish emotions from mood. Thus, a second purpose of Study 2 was to replicate the findings
of Study 1 with a more standard measure of positive emotions (PANAS). A third goal was to
assess negative emotions in order to disentangle the role of both positive and negative emotions
in interpersonal relationships. In line with Study 1, it was proposed that harmonious passion for
an activity should lead to the experience of positive emotions within the context of that activity.
These positive emotions were expected, in turn, to be conducive to quality of interpersonal
relationships. Furthermore, in line with past passion research (see Vallerand, 2008), it was
predicted that harmonious passion should be negatively associated with negative emotions. On
the other hand, in line with Study 1, it was hypothesized that obsessive passion for an activity
should be unrelated to positive emotions and to the quality of interpersonal relationships. In
addition, it was also predicted that obsessive passion should be positively associated with
negative emotions (e.g., Mageau et al., 2005; Vallerand et al., 2003, Study 1) that, in turn, should
be negatively related to the quality of interpersonal relationships (Waugh & Fredrickson, 2006).
Finally, it was further predicted that these mediation effects should not be altered by trait
extraversion.
Method
Participants
Passion and Interpersonal Relationships 16
Participants were 266 (174 females, 92 males) French-Canadian undergraduate students
from a university in Quebec, Canada. Because we were interested in the quality of interpersonal
relationships within the context of a passionate activity, participants who reported engaging in
their activity alone were removed from this analysis (n = 77). A total of 12 participants were also
dropped from the analysis because of several missing data. The final sample was composed of
177 participants (M = 26.0 years, SD = 7.43 years) who engaged in their passionate activity on
average for 8.21 hrs/week (SD = 10.62 hrs/week) and had been engaging in it for 57.48 months
on average (SD = 74.50 months).
Measures
The Passion Scale was again administered in order to assess the type of passion that
characterizes a highly valued activity engaged in by the participants. Participants were asked to
list an activity that they love, value, and spend considerable amount of time and energy in (e.g.,
basketball, acting, playing pool). They were then asked to complete the scale while referring to
this activity (e.g., ―I have difficulties controlling my urge to do my activity‖ or ―I have almost an
obsessive feeling for my activity‖). Alphas for the sample of this study were respectively .80 and
.85 for the harmonious passion and the obsessive passion subscales. Similarly to the Vallerand et
al. (2003, Study 1) activity classification, listed activities in this study included team sports (e.g.,
playing basketball, hockey, football: 58%), leisure (e.g., listening to music, gardening, playing
pool: 20%), active arts (e.g., painting, photography, theater: 12%), education/work (e.g., part-
time work, teaching: 5%), and interpersonal activities (e.g., being with friends or family: 5%). 3
Positive Activation and Negative Activation Schedule (PANAS). A short 10-item version
(Mackinnon et al., 1999) of the PANAS (Watson, Clark, & Tellegen, 1988) was used to assess
participants’ emotions while engaging in the activity. Five items pertained to positive emotions
Passion and Interpersonal Relationships 17
(e.g., enthusiastic, determined) and five items assessed negative emotions (e.g., nervous, upset).
All emotion items were responded to on a 5-point Likert scale ranging from 1 (very slightly or
not at all) to 5 (extremely). A factorial analysis with the selected 10 emotions with a Varimax
rotation yielded two factors accounting for 61% of the variance, thus preserving the orthogonal
structure of the original full-scale version (see Mackinnon et al., 1999 for further evidence of
validity and reliability for this short scale). Alphas for this brief version were .80 for the positive
emotions subscale and .84 for the negative emotions subscale.
The Quality of Interpersonal Relationships Scale (QIRS; Sénécal et al., 1992) was also
used in the present study (Alpha = .92).
Extraversion. A short version (12 items) of the extraversion subscale of the NEO-PI-R
(Costa & McCrae, 1992) was also completed by participants on a 5-point Likert scale (1 =
strongly disagree, and 5 = strongly agree). Sample items are ―I like to have a lot of people
around me‖ and ― I laugh easily‖. Alpha in this study was .82.
Procedure
Participants were told that the purpose of the study was to find out more about
participants’ favorite activity. It was also mentioned that participation was voluntary and
anonymous, and that responses would remain confidential. Participants were first asked to
complete the Passion Scale and to mention the number of hours per week they were engaging in
their activity and whether they were practicing it alone or with other people. If so, participants
reported with how many people they were typically engaging in their activity. They then
completed the emotions inventory and the Quality of Interpersonal Relationships Scale. All items
for both scales were referring to the chosen passionate activity. Finally, participants completed
the extraversion scale.
Passion and Interpersonal Relationships 18
Results and Discussion
Table 3 reports the means, standard deviations, and correlations of all study variables. To
assess the relationship between passion and the quality of interpersonal relationships experienced
during activity engagement, a multiple regression analysis was conducted. The quality of
interpersonal relationships served as the dependent variable, and the two types of passion as
predictor variables, controlling for number of hours of weekly involvement in the passionate
activity and extraversion. Results of the regression yielded a significant model (F = 7.95, p <
.000, R2 = .16). Results revealed that harmonious passion was a significant predictor of quality of
interpersonal relationships ( = .19, p < .05). Obsessive passion was not associated with it ( = -
.09, ns), neither was the number of minutes of weekly involvement ( = .09, ns). These results
replicate the findings of Study 1, this time controlling for trait extraversion ( = .29, p < .01).
Our second hypothesis was that positive emotions experienced during the activity would
mediate the relationship between harmonious passion and quality of interpersonal relationships,
while negative emotions would mediate the relationships between obsessive passion and quality
of interpersonal relationships. Based on past research (e.g., Vallerand, et al., 2003, Study 1) and
results of Study 1, obsessive passion was not expected to be associated with positive emotions,
but harmonious passion was hypothesized to be negatively associated to negative emotions. In
order to test this model, a path analysis was conducted. Trait extraversion and the number of
minutes of weekly involvement were added to the model as exogenous variables to serve as
control variables. Paths were also drawn from extraversion to positive emotions and from
extraversion to quality of interpersonal relationships, as research has shown that extraversion is
usually associated with both of these variables (Costa & McCrae, 1980; Watson & Clark, 1997).
Passion and Interpersonal Relationships 19
Positive and negative emotions and quality of interpersonal relationships served as endogenous
variables and a covariance was estimated between positive and negative emotions.
Results of the path analysis revealed a satisfactory fit of the model to the data,
2 (df = 6,
n = 177) = 7.26, p = .30, NFI = .96, NNFI = .97, CFI = .99, RMSEA = .035 [.00; .109], GFI =
.99, SRMR = .033. As shown in Figure 2, all estimated paths were significant at least at p < .05,
except for the path between number of hours weekly involved and quality of interpersonal
relationships that was not significant ( = .09, ns). Inspection of the correlation residuals
revealed that all were non-significant, thus indicating that no additional path was significant
(Kline, 2005). In addition, direct paths from harmonious passion and obsessive passion to quality
of interpersonal relationships were not significant and their inclusion in the model did not affect
the fit indices (2[2] = 1.66, ns) or the coefficients of the hypothesized paths, thus suggesting
mediation. Bias-corrected bootstrapped 95% confidence interval estimates confirmed the
significance of the mediation of positive emotions between harmonious passion and quality of
interpersonal relationships and of negative emotions between both obsessive passion and
harmonious passion and quality of interpersonal relationships (see Table 2).
An alternative model tested if the quality of interpersonal relationships mediated the
relationship between harmonious passion and emotions (passion quality of interpersonal
relationships emotions). Results revealed that the fit indices of this alternative model were
very poor compared to those of the hypothesized model:
2 (df = 7, n = 177) = 66.15, p > .000,
RMSEA = .222 [0.17; 0.27], CFI = .58 NFI = .59, SRMR = .12.
Results of Study 2 supported our hypotheses. First, they showed that harmonious passion
was again positively associated with the quality of interpersonal relationships during activity
engagement, even when controlling for trait extraversion, while obsessive passion was found to
Passion and Interpersonal Relationships 20
be unrelated to quality of interpersonal relationships. Second, results also supported the
mediation hypotheses, as they revealed that the association between harmonious passion and the
quality of interpersonal relationships was totally mediated by the experience of positive emotions
during activity engagement. Such a pattern was not found for obsessive passion, as findings
revealed that it was unrelated to either the experience of positive emotions during activity
engagement or to the quality of interpersonal relationships. Finally, results indicated that
negative emotions were a significant mediator between both obsessive passion and harmonious
passion and interpersonal relationships.
Study 3
There were three purposes to Study 3. First, it aimed at replicating the results of Study 2
while using a prospective design. Specifically, Study 3 investigated how passionate basketball
players attending a 5-day basketball summer camp developed interpersonal relationships with
other people (coach, teammates, and players of the other teams) throughout the camp. One
important limitation of Studies 1 and 2 was that interpersonal relationships ratings were obtained
once the passionate people and their activity partners were accustomed to engaging in the
activity together. Therefore, the prior quality of their relationships might have influenced passion
ratings over time. The prospective design of Study 3 and the fact that basketball players did not
know each other prior to the camp represent two important improvements over Studies 1 and 2 as
they rule out the possibility that passion was determined by the quality of preexisting
relationships. A second purpose of Study 3 was to broaden the assessment of interpersonal
relationships quality. Thus, the self-other overlap (Aron et al., 1992) interpersonal measure was
used with the addition of the positive quality of the relationships (as in Studies 1 and 2). In
addition, all of these relationships elements were examined for each type of people engaging in
Passion and Interpersonal Relationships 21
the activity with the players (coach, teammates, and players of the other teams). Study 3 also
sought to investigate the perspective of other people with whom the passionate people engage in
their activity. Therefore, head coaches of each basketball team at camp (15 teams) rated the
extent to which each of his (all head coaches were males) players engaged in positive
interpersonal relationships with his/her teammates, thus providing an independent and relatively
objective rating of the players’ relationships.
A third goal of Study 3 was to provide an improved examination of positive and negative
emotions at play during passionate activity engagement by using a more extensive assessment of
positive and negative emotions.4 In line with Studies 1 and 2, it was hypothesized that positive
emotions would positively mediate the relationship between harmonious passion and all positive
interpersonal dimensions, including the coaches’ objective ratings. Based on the results of
Studies 1 and 2, it was further predicted that obsessive passion would be unrelated to positive
emotions. Furthermore, it was hypothesized that negative emotions would mediate the
relationship between obsessive passion and harmonious passion and all positive interpersonal
dimensions, including coaches’ ratings.
Method
Participants
Participants were 160 (60 females, 100 males) adolescents aged between 13 and 17 years
(M = 14.50 years; SD = 0.92 year) who were taking part in a summer basketball camp. They
were typically (outside of camp) playing basketball on average for 8.08 hrs/week (SD = 5.96
hrs/week) and had been engaging in it for 3.92 years on average (SD = 1.84 years). One female
participant was removed from the analysis because of incomplete data (analyzed n = 159).
Participants in camp were members of one of 15 teams and were each assigned one coach. All of
Passion and Interpersonal Relationships 22
these fifteen head coaches also participated in the study. Coaches were all males and were aged
between 19 and 31 years (M = 23.67 years; SD = 4.39 years) and had been coaching on average
for 5.27 years (SD = 3.63 years).
Measures
The Passion Scale. The Passion Scale was again administered. Alphas for the sample of
this study were respectively .66 and .83 for the harmonious passion and the obsessive passion
subscales.
Positive and Negative Emotions. Positive and negative emotions were each assessed with
eight items derived from the PANAS and Izard’s differential emotional theory (Izard, 1977). All
emotion items were responded to on a 5-point Likert scale ranging from 1 (very slightly or not at
all) to 5 (extremely). Alphas for the positive and negative emotions subscales were .84 and .81,
respectively.
Positive Quality of Interpersonal Relationships was measured with the Quality of
Interpersonal Relationships Scale used in Studies 1 and 2. It assessed the quality of the
interpersonal relationships with participants’ teammates ( = .86), their coach ( = .89), and
opponent players during camp ( = .85). These items were responded to on a 5-point Likert scale
(0 = not at all, 4 = extremely).
Self-Other Overlap. The Inclusion of Other in Self Scale (IOS; Aron et al., 1992) is a
measure of perceived relationship closeness. It is a single-item designed with seven pairs of
overlapping circles ranging from 1 (no overlap) to 7 (almost complete overlap). For each item,
one circle represented the self (Me) and the other circle represented team players, the coach, or
players of the other teams. Participants were thus asked to rate the extent to which they felt close
Passion and Interpersonal Relationships 23
during the camp to, respectively, their teammates, coach, and players of the other teams by
selecting the appropriate pair of circles.
Coaches’ Ratings. The head coach of each basketball team rated the extent to which each
player appeared to be close to his/her teammates during the camp using the Inclusion of Other in
Self Scale for each rating. The correlation between the players’ ratings and his coach’s ratings
was .27 for perceived interpersonal closeness with teammates.
All six positive interpersonal measures (quality and interpersonal closeness) relative to
teammates, the coach, and other players completed by athletes were averaged together ( = .79)
to yield one index labeled positive interpersonal relationships.5
Procedure
In the early part of the camp, each team composed of 10 to 15 players and the head coach
were met in a classroom. Participants were told that the purpose of the study was to better
understand the attitudes of basketball players and coaches toward their sport. It was also
mentioned that participation was voluntary and anonymous, and that their responses would
remain confidential. Players were then asked to complete the Passion Scale. They were also
asked to indicate how many players of their team and players of the other teams they already
knew prior to the camp and to specify if they knew their coach before the camp began.6 Toward
the end of the camp, participants were met in a classroom. Players were asked to complete the
emotions inventory and the interpersonal relationships measures. The coaches were asked to rate
the extent to which each of their players appeared to be close to the other players of the team
during the camp. Afterwards, participants were fully debriefed and thanked for their
participation.
Results and Discussion
Passion and Interpersonal Relationships 24
Table 4 reports the means, standard deviations, and correlations of all study variables.7 In
order to test the hypothesized model, a path analysis was conducted. The two types of passion
served as exogenous variables in the model. Positive and negative emotions and positive
interpersonal relationships served as endogenous variables, as well as coaches’ ratings for
perceived interpersonal closeness. A positive covariance was also estimated between these two
interpersonal variables, as they were assessing the same construct, but from a different
perspective (self-evaluation vs. observers’ ratings). In addition, in line with Study 2, a covariance
was drawn between positive and negative emotions. Results of the path analysis revealed a
satisfactory fit of the model to the data,
2 (df = 5, n = 159) = 6.12, p = .29, NFI = .96, NNFI
= .98, CFI = .99, RMSEA = .038 [.00; .12], GFI = .99, SRMR = .043. In addition, correlation
residuals were all non-significant. Furthermore, all direct paths were not significant and their
inclusion in the model did not affect the fit indices (direct paths from harmonious passion:
2(2) = 0.96, ns; direct paths from obsessive passion: 2(2) = 1.21, ns) or the coefficients of
the hypothesized paths. Figure 3 shows the final path analysis. As can be seen, all estimated
paths were significant, except for the covariance between positive and negative emotions. Bias-
corrected bootstrapped 95% confidence interval estimates (see Table 2) confirmed the
significance of the mediations of positive emotions between harmonious passion and athletes’
and coaches’ ratings as well as that for negative emotions between harmonious passion and
athletes’ ratings. The mediational role of negative emotions between obsessive passion and
athletes’ ratings was significant, but not that involving harmonious passion or obsessive passion
and coaches’ ratings.
An alternative model tested if the quality of interpersonal relationships mediated the
relationship between harmonious passion and emotions (passion quality of interpersonal
Passion and Interpersonal Relationships 25
relationships emotions). Results revealed that the fit indices of this alternative model were
very poor compared to those of the hypothesized model:
2 (df = 5, n = 159) = 41.87, p > .000,
RMSEA = .217 [0.16; 0.28], CFI = .71, NFI = .71, SRMR = .10. In addition, the prospective
design of Study 3 rules out equivalent models, such as positive emotions predicting both passion
and interpersonal relationships or interpersonal relationships emotions passion, as passion
was assessed prior to all other measures and participants did not know each other prior to the
beginning of the camp.
Overall, results of Study 3 supported most of our hypotheses. First, positive emotions
experienced during the basketball camp mediated (positively) the relationship between
harmonious passion and athletes’ assessment of positive interpersonal relationships developed
throughout the week, as well as between harmonious passion and the coaches’ ratings of
perceived relationship closeness. As in Studies 1 and 2, obsessive passion was unrelated to
positive emotions. Second, there was a significant mediation for negative emotions between
harmonious passion or obsessive passion and athletes’ positive interpersonal relationships. As
hypothesized, obsessive passion and harmonious passion were significantly positively and
negatively associated with negative emotions, respectively, which were, in turn, negatively and
significantly associated with positive interpersonal relationships. However, negative emotions
were not a significant mediator between harmonious passion or obsessive passion and coaches’
positive ratings.
Study 4
There were several purposes to Study 4. The first purpose of Study 4 aimed at replicating
the results of Study 3 while using a longer prospective design. Indeed, Study 3 only examined
the role of passion in the quality of interpersonal relationships in a 5-day prospective design.
Passion and Interpersonal Relationships 26
Replicating Study 3 results within a longer prospective design will facilitate the interpretation of
the stability of the relationship between passion and the development of interpersonal
relationships. Specifically, Study 4 investigated how passion for an academic domain
(management specifically) influenced the development of interpersonal relationships over a full
semester with other work colleagues within an academic work team. A second purpose of Study
4 was to further extend the range of interpersonal relationships measures. In order to improve on
Study 3, Study 4 extended the observers’ ratings to both positive and negative quality of
interpersonal relationships. In addition, Study 4 also looked at two other dimensions of
interpersonal relationships, namely feelings of connectedness and feelings of seclusion (inability
to establish interpersonal relationships), thus assessing one positive and one negative self-
reported measures of interpersonal relationships, respectively. Connectedness (e.g., Deci &
Ryan, 2000) and seclusion (e.g., Russell, 1996) are believed to be important dimensions of the
quality of people’s interpersonal relationships.
In line with Studies 1 to 3, it was hypothesized that positive emotions would positively
mediate the relationship between harmonious passion and feelings of connectedness and work
colleagues’ positive objective-interpersonal ratings. In line with past research (e.g., Barsade,
2002; Vallerand et al., 2003), it was also hypothesized that positive emotions would negatively
mediate the relationship between harmonious passion and feelings of seclusion and work
colleagues’ negative objective-interpersonal ratings. Furthermore, in line Study 3, it was
hypothesized that negative emotions would negatively mediate the relationship between
harmonious passion and all positive interpersonal dimensions. Based on the results of Studies 1
to 3, it was also predicted that obsessive passion would be unrelated to positive emotions. In line
with Studies 2 and 3, it was predicted that obsessive passion would be positively associated with
Passion and Interpersonal Relationships 27
negative emotions, which would be, in turn, negatively associated with the positive interpersonal
relationships variables. Finally, because negative emotions have been shown to lead to
interpersonal conflict (e.g., Barsade, 2002), it was hypothesized that negative emotions would
positively mediate the relationships between obsessive passion and feelings of seclusion and
work colleagues’ negative objective-interpersonal ratings. Thus, taken together, it was expected
that positive and negative emotions would mediate the relationships between harmonious passion
and all four interpersonal relationships variables measured in Study 4 and that negative emotions
would mediate the relationship between obsessive passion and all four interpersonal variables.
Method
Participants
Participants were 187 (102 females, 85 males) university students aged between 19 and
49 years (M = 25.50 years, SD = 6.72 years) who were enrolled in a management program. One
female participant was dropped from the analysis, as her responses on two variables were located
at more than three standard deviations of the distribution.
Measures
The Passion Scale was again administered to tap passion for an academic domain. Alphas
for the sample of this study were respectively .87 and .70 for the harmonious passion and the
obsessive passion subscales.
Positive and Negative Emotions experienced within the work team were each assessed
with five items derived from the PANAS and Izard’s differential emotional theory (Izard, 1977),
used in Study 3. These emotions were selected because of their likelihood to be experienced in
an academic work team (e.g., attentive, contempt). All emotion items were responded to on a 5-
Passion and Interpersonal Relationships 28
point Likert scale ranging from 1 (very slightly or not at all) to 5 (extremely). Alphas for the
positive and negative emotions subscales were .84 and .74, respectively.
Interpersonal Connectedness with Teammates. Participants’ feelings of connectedness
with their teamwork colleagues were assessed with a slightly modified 3-item scale developed by
Sheldon and his colleagues (Sheldon, Elliot, Kim, & Kasser, 2001). A sample item is ―I felt I
was close and connected with the other persons in my team‖. Alpha in this study was .87. All
items were responded to on a 7-point Likert scale ranging from 1 (do not agree at all) to 7
(completely agree).
Interpersonal Seclusion from Teammates. Participants’ feelings of seclusion within their
work team were assessed with a 4-item scale. Items assessed how much participants felt
excluded and isolated within their work team (α = .65). A sample item is ―I felt that I was
isolated from the members of my team work‖. All items were responded to on a 7-point Likert
scale ranging from 1 (do not agree at all) to 7 (completely agree).
Teammates’ Interpersonal Ratings. Participants were asked to rate their perception of
each of their teammate’s quality of interpersonal relationships with the other people in the work
team. Ratings were made on a 7-point Likert scale ranging from 1 (do not agree at all) to 7
(completely agree). Five items assessed positive interpersonal relationships as perceived by the
teammates. A sample item is ―This person had a true interest for the other people in the team‖ (α
= .84). Five other items assessed negative interpersonal relationships as perceived by the
teammates. A sample item is ―This person appeared irritable‖ (α = .66). Because the work teams
were not composed of the same number of participants, teammates’ interpersonal ratings were
made by one to four colleagues. To create the positive and negative interpersonal relationships
rating variables, all existing positive ratings for one participant were averaged together, and the
Passion and Interpersonal Relationships 29
same was done for the negative ratings. ANOVAs revealed that these teammates’ ratings as well
as all other study variables did not differ as a function of the number of colleagues who
evaluated them. In other words, the number of people composing the work teams did not
influence the study variables, all Fs(3, 182) < 1.85, ns.
Procedure
Participants were from management classes for which an important team project had to
be completed over the course of the semester. Participants completed the Passion Scale at the
beginning of the semester. At that time, the work teams were not yet formed. At the end of the
semester, participants completed measures of positive and negative emotions, felt connectedness
and seclusion within their work team, and rated each of their teammates according to their
positive and negative quality of interpersonal relationships with the other people in the work
team. It was clearly stated that the confidentiality of their answers would prevail at all time.
Results
Table 5 presents the correlation matrix with means and standard deviations involving all
study variables. The model was composed of two exogenous variables (i.e., harmonious passion,
obsessive passion) and six endogenous variables (i.e., positive and negative emotions within the
work team, connectedness, seclusion, and participants’ positive and negative ratings of their
teammates’ quality of interpersonal relationships within the work team). The exogenous
variables were Time 1’s variables and the endogenous variables were all Time 2’s variables.
Paths were specified according to the hypotheses, including paths from positive and negative
emotions to all the interpersonal measures. In addition, a positive covariance was estimated
between positive and negative teammates’ ratings and between participant’s felt connectedness
and seclusion self-ratings to account for common measurement methods (i.e., observers’ ratings
Passion and Interpersonal Relationships 30
vs. self-report). In line with Studies 2 and 3, a covariance was also estimated between positive
and negative emotions. Results of this path analysis revealed a satisfactory fit of the model to the
data, χ2 (df = 13, n = 186) = 17.63, p = .17, RMSEA = .044 [.00; .09], NFI = .95, NNFI = .97,
CFI = .98, GFI = .98, and SRMR = .035.
In addition, all correlational residuals in the model were non-significant. Furthermore, all
direct paths were non significant and their inclusion in the model did not affect the fit indices
(direct paths from harmonious passion: 2[4] = 4.80, ns; direct paths from obsessive passion:
2[4] = 9.06, ns) or the coefficients of the hypothesized paths. As shown in Figure 4,
harmonious passion was positively and negatively associated with positive and negative
emotions, respectively, while obsessive passion was positively associated with negative emotions
and unrelated to positive emotions. Positive and negative emotions were both, in turn,
significantly associated to all interpersonal variables. The only non-significant paths were
between positive emotions and teammates’ negative interpersonal ratings ( = -.06, ns) and
between negative emotions and teammates’ positive interpersonal ratings ( = -.07, ns). Finally,
it should be noted that all the above results remained virtually the same when examining only
individuals passionate for management (i.e., 84% of the sample) or when excluding from the
analyses participants who reported to be at least slightly familiar with one of their teammates (n
= 17). An alternative model tested if the quality of interpersonal relationships mediated the
relationship between harmonious passion and emotions (passion quality of interpersonal
relationships emotions). Results revealed that the fit indices for this alternative model were
less adequate:
2 (df = 8, n = 186) = 17.94, p < .02, RMSEA = .082 [.03; .14], CFI = .97, NFI =
.95, NNFI = .89, GFI = .98, SRMR = .06. In addition, the AIC value used to compare non-
hierarchical alternative models (Kline, 2005) revealed that the hypothesized model had a lower
Passion and Interpersonal Relationships 31
value, AICo = 63.63, compared to the alternative model, AICA = 73.94. Therefore, the
hypothesized model with the lowest AIC value should be preferred.
Table 2 provides the bias-corrected bootstrapped estimates examining the significance of
each mediation. As in Studies 1, 2, and 3, positive emotions were a significant mediator between
harmonious passion and interpersonal relationships variables, except for teammates’ negative
ratings. Negative emotions were a significant mediator between obsessive passion and both
positive and negative interpersonal relationships self-report and teammates’ negative ratings, but
not between harmonious passion and interpersonal variables or between obsessive passion and
teammates’ positive ratings. Thus, positive emotions appear to be a stronger mediator of
harmonious passion compared to negative emotions, while obsessive passion appears to be better
mediated by negative than positive emotions. Finally, as mentioned above, positive emotions did
not significantly predict teammates’ negative interpersonal ratings, while negative emotions did
not significantly predict teammates’ positive interpersonal ratings. Therefore, no mediation tests
could be computed for these paths. It should be noted, however, that for both of these
relationships their zero-order correlations were significant (see Table 5). The non-significance of
their paths in the model might be due to a number of factors including the fact that four
interpersonal variables were modeled together as competing covariates (some of these involving
observers’ assessments), thus reducing the strength of their associations to non-significant
results. However, overall, the present study provided support for most of our other hypotheses.
General Discussion
The present research had two major purposes. The first one was to ascertain the role of
passion for an activity in the quality of interpersonal relationships experienced within the context
of that activity. The second purpose of the present research was to examine the psychological
Passion and Interpersonal Relationships 32
processes involved in the relationship between passion and interpersonal relationships. In line
with past research, it was hypothesized that the experience of positive emotions during task
engagement would positively mediate the harmonious passion – interpersonal relationships
relation, but that obsessive passion would be positively associated with negative emotions that
should positively and negatively mediate the relationship between obsessive passion and
negative and positive interpersonal relationships, respectively. The present results provided
support for most of these hypotheses. Overall, the findings of the present research lead to a
number of implications.
Passion and Interpersonal Relationships
A first implication from the present findings is that they support the application of the
Dualistic Model of Passion to interpersonal relationships. It appears that the type of passion one
has for an activity matters with respect to the type of interpersonal relationships that one will
develop during task engagement. Results of four studies showed that harmonious passion was
indirectly positively associated with positive interpersonal relationships within the context of an
activity through the effect of positive emotions (Studies 1 to 4), while obsessive passion was
indirectly negatively related to them through negative emotions experienced within the context
of the activity (Studies 2 to 4). Overall, these findings are very robust as they were obtained with
four different populations (teenager athletes, collegiate students, university students, and
workers), with a variety of activities, and while controlling for trait extraversion. In addition, the
findings of Study 4 further showed that obsessive passion was indirectly conducive to negative
interpersonal relationships through experiences of negative emotions, while harmonious passion
was negatively associated to negative interpersonal relationships through both positive and
negative emotions. However, overall, the indirect effect from harmonious passion to
Passion and Interpersonal Relationships 33
interpersonal relationships variables was stronger through positive emotions, while it was much
weaker through negative emotions. The present results are the first to show that the type of
passion one holds for an activity predicts the quality of interpersonal relationships within the
context of that activity. Such findings highlight the important implications that the type of
engagement in a passionate activity can have on oneself, as well as on other people. These intra
and interpersonal consequences can be important in the context of sports teams, workplaces, or
social groups as they may affect other facets of an activity. For instance, poor interpersonal
relationships triggered by high levels of obsessive passion may negatively influence team unity,
productivity, and performance. Future research is needed on this issue.
A second implication from the findings of the present research is that they highlighted the
importance of emotions as a mediator of the Passion - Quality of Interpersonal Relationships
relation. Specifically, harmonious passion and obsessive passion both appear to influence
interpersonal relationships through different emotional pathways. A first pathway is through the
positive emotions experienced during activity engagement. Findings from four studies showed
that harmonious passion was positively associated with positive emotions experienced during
activity engagement. In addition, as posited by Fredrickson (2001; Waugh & Fredrickson, 2006),
positive emotions were found to be positively associated with positive interpersonal relationships
and negatively associated with negative interpersonal relationships. Obsessive passion was
consistently unrelated to positive emotions across all four studies. Finally, as hypothesized,
positive emotions were found to positively mediate all relationships between harmonious passion
and positive interpersonal relationships and negatively mediate the relationship between
harmonious passion and feelings of seclusion (Study 4).
Passion and Interpersonal Relationships 34
A second emotional pathway facilitated by passion appears to be the negative emotions
experienced during activity engagement. Results showed that obsessive passion was consistently
and positively related to negative emotions (Studies 2 to 4), while harmonious passion was
consistently negatively associated with them. Furthermore, Studies 3 and 4 showed that negative
emotions negatively mediated the relation between obsessive passion and self-reported positive
aspects of interpersonal relationships (but not between obsessive passion and objective positive
interpersonal relationships), while Study 4 showed that negative emotions positively mediated
the relationship between obsessive passion and negative aspects of interpersonal relationships.
Harmonious passion was consistently negatively associated with negative emotions, but negative
emotions were a significant mediator between harmonious passion and interpersonal
relationships variables in only two out of seven tests. Finally, positive emotions were found
throughout Studies 1 to 4 to be a much stronger predictor of positive interpersonal relationships,
while negative emotions seemed to better predict negative aspects of interpersonal relationships.
Therefore, it appears that they are two main pathways that affect emotional experience and
interpersonal relationships with the people with whom the activity is engaged in—one from the
harmonious passion and another from the obsessive passion. Depending of the type of passion at
play, one will experience very different types of emotional experiences and, as a consequence,
interpersonal relationships.
One explanation for these mediation effects is that harmonious passion for an activity is
well integrated within the self and identity, and consequently allows for a flexible engagement in
the task (Hodgins & Knee, 2002). It thus becomes possible to experience task engagement more
fully and with a willingness to experience what is occurring at the moment. Such a flexible form
of activity engagement facilitates the experience of positive emotions while engaging in the
Passion and Interpersonal Relationships 35
activity (Vallerand et al., 2003, Study 1; Vallerand et al., 2006, Study 2; Vallerand & Houlfort,
2003). These positive emotions then sustain the broadening of the thought-action repertoires and
self (Fredrickson, 2001; Waugh & Fredrickson, 2006) leading to positive interactive behaviors
such as smiling, joking, and encouraging others, and to a true interest for other people (self-other
overlap). Indeed, past research has shown that such positive behaviors and attitudes create
conditions that are conducive to positive interpersonal relationships (Cappella, 1997; Frank,
Ekman, & Friesen, 1993; Shiota et al., 2003).
Conversely, an obsessive passion for an activity fosters a rigid and conflicted form of
activity engagement that is unrelated to positive emotions and may even be conducive to
negative emotions (Vallerand et al., 2003, Study 2; also see Vallerand, 2008), thus leading to the
narrowing of the thought-action repertoires and self (Fredrickson, 2001; Waugh & Fredrickson,
2006) and to a closing off from others (Waugh & Fredrickson, 2006). This leaves little space for
the emergence of positive emotions and relating to others. Indeed, results from all four studies
revealed that obsessive passion was not associated with positive emotions during task
engagement. In fact, it was positively related to negative emotions (Studies 2 to 4). These
negative emotional experiences appear to preclude external demonstrations of enjoyment and
feelings of self-other overlap, thereby leading to poor social interaction (Waugh & Fredrickson,
2006). Indeed, negative emotions were found to be negatively associated with positive
interpersonal relationships and positively associated with negative interpersonal relationships.
Therefore, the findings of the present research underscore the fact that passion may lead to two
types of emotional experience—depending of the type of passion at play—and that only positive
emotions triggered by harmonious passion may lead to positive interpersonal relationships
Passion and Interpersonal Relationships 36
through a broaden-and-build process. Conversely, the negative emotions produced by obsessive
passion appear to rather lead to negative interpersonal relationships.
Although most meditational effects were significant, some of them were marginally
significant (see Table 2). In light of these results, and in order to provide clarity on these
mediational effects, a meta-analysis was conducted on the present four studies to examine these
meditational effects with more power. All self-report measures assessing interpersonal
relationships were combined and all objective ratings were combined, holding positive and
negative assessments separate for both self-report and objective ratings. Thus, the model
hypothesized in Study 4 was tested on the aggregation of all four studies data (Hunter &
Schmidt, 2004). Pooled correlations coefficients were calculated for all examined relationships
across all four studies. These correlations were then input in a correlation matrix, which served
as the database for the path analysis (Hunter & Schmidt, 2004). The median sample size among
all examined relationships (n = 345) was used (e.g., Brown & Peterson, 1993).
Fit indices for this meta-analyzed path model were adequate,
2 (df = 13, n =
345) = 22.28, p = .051, RMSEA = .046 [.00; .08], CFI = .98, NFI = .96, NNFI = .96, GFI = .98,
SRMR = .03. Results revealed that all hypothesized paths were significant at p < .05, except for
the covariance between positive and negative emotions. Sobel tests were then conducted to
examine the significance of each mediational effect. Results revealed that they were all
significant (all zs > 2.00). Therefore, it would appear that some meditational effects did not reach
significance in a single study, because their effect size was small and the power was not large
enough to detect them with a p-value below .05. However, when they were examined with more
power, they were statistically significant. However, caution is advised with respect to the
meditational role of both positive and negative emotions between passion and negative quality of
Passion and Interpersonal Relationships 37
interpersonal relationships as these latter interpersonal measures were only used in one study
(Study 4).
Passion, the Social Functions of Emotions, and the Broaden-and-Build Theory
The present findings also lead to important implications for the social functions of
emotions and the Broaden-and-Build Theory (Fredrickson, 2001). The present series of studies
extended past research on these aspects in three ways. First, the present findings suggest that
passion for an activity represents an important igniter of the broaden-and-build (or the
narrowing) process. It appears that harmonious passion would facilitate a broadening process
through the experience of positive emotions and through limiting negative emotions. Obsessive
passion would, however, be conducive to a narrowing process because of the negative emotions
generated by this type of passion. Therefore, passion appears to play a role in triggering the
broadening and narrowing processes. A second extension of the social functions of emotions and
the Broaden-and-Build Theory is that the present findings underscore that various types of
positive and negative emotions, and not only interpersonal emotions such as gratitude or
compassion, experienced in the context of a highly valued activity can lead to important
interpersonal consequences. Indeed, various types of positive and negative emotions were used
across four studies, and each time, results showed that these emotions had an impact on
interpersonal relationships. Of additional interest is that findings of Study 3 even showed that
various emotions of different arousal levels were all associated with positive interpersonal
relationships. Thus, it would appear that various types of emotion might possess social functions.
Finally, the present findings showed that various self-reported emotions predicted
interpersonal relationships as rated by external observers. Studies 3 and 4 showed that players’
various positive emotions were positively associated with the coaches’ ratings of their players’
Passion and Interpersonal Relationships 38
interpersonal closeness with teammates (Study 3) while students’ positive emotions positively
predicted their coworkers’ ratings of positive and negative interpersonal relationships with them.
Although, the present set of studies is not the first to show an association between emotion and
objective interpersonal relationships ratings (e.g., Keltner & Bonano, 1997), the present findings
specifically highlight a potentially important extension of the Broaden-and-Build Theory: Other
people can observe the consequences of the broadening and narrowing effects facilitated by
positive and negative emotions, respectively. Therefore, harmoniously passionate people,
through the positive emotions they experience, might come to display overt behaviors toward
others such as smiles of enjoyment, laughs, or physical closeness which would come to be
perceived by others (Frank et al, 1993). It might even be possible that observers reciprocate such
positive behaviors leading to social bonding with the target (Walter & Bruch, 2008). Future
research is needed to look at the positive external interpersonal demonstrations (verbal and non-
verbal) that accompany emotions and examine their consequences on the development of the
quality of interpersonal relationships.
Some limitations of the present research should be underscored. First, all studies were
correlational in nature. It is thus not possible to establish the causality of the relationships
between passion, emotion, and interpersonal relationship. Although the prospective designs used
in Studies 3 and 4 and the results of the various alternative models in Studies 1 and 2 provide
additional support for the adequacy of the direction of the model Passion Emotion
Relationships, future research is needed in order to test the inverse model, which remains
plausible. Second, emotions in the present research were all assessed with paper-pencil
measures. Future research might do well in extending the present findings to include other types
of assessment such as physiological measures or video recordings of interactions.
Passion and Interpersonal Relationships 39
In sum, to go back to our introductory comments, it would appear that passion for an
activity leads to positive interpersonal relationships within the purview of the passionate activity.
However, such a positive effect takes place only if passion is harmonious. Indeed, an obsessive
passion for an activity may even negatively affect interpersonal relationships. Additional
research on these issues would therefore appear important.
Passion and Interpersonal Relationships 40
References
Algoe, S., Haidt, J., & Gable, S. (2008). Beyond reciprocity: Gratitude and relationships in everyday
life. Emotion, 8, 425-429.
Amiot, C., Vallerand, R. J., & Blanchard, C. (2006). Passion and psychological adjustment: A test of
the Person-Environment fit hypothesis. Personality and Social Psychology Bulletin, 32, 220-229.
Aron, A., Aron, E. N., & Smolan, D. (1992). Inclusion of other in the self scale and the structure of
interpersonal closeness. Journal of Personality and Social Psychology, 63, 596-612.
Aron, A., Norman, C. C., Aron, E. N., McKenna, C., & Heyman, R. E. (2000). Couple’s shared
participation in novel and arousing activities and experienced relationship quality. Journal of
Personality and Social Psychology, 78, 273–284.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173–1182.
Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on group behavior.
Administrative Science Quarterly, 47, 644-675.
Blais, M., & Vallerand, R. J. (1991). The work affective and cognitive consequences scale.
Unpublished manuscript, Université du Québec à Montréal, Montreal, Canada.
Brown, S., & Peterson, R. (1993). Antecedents and consequences of salesperson job-satisfaction –
Meta-analysis and assessment of causal effects. Journal of Marketing Research , 30, 63-77.
Cappella, J. N. (1997). Behavioral and judged coordination in adult informal social interactions :
Vocal and kinesic indicators. Journal of Personality and Social Psychology, 72, 119-131.
Carbonneau, N., Vallerand, R. J., Fernet, C., & Guay, F. (2008). The role of passion for teaching in
intra and interpersonal outcomes. Journal of Educational Psychology, 100, 977-987.
Passion and Interpersonal Relationships 41
Costa, P. T. & McCrae, R. R. (1980). Influence of extraversion and neuroticism on subjective well-
being: Happy and unhappy people. Journal of Personality and Social Psychology, 38, 668-678.
Costa, P. T., & McCrae, R. R. (1992). NEO PI-R Professional Manual. Odessa, FL: Psychological
Assessment Resources.
Deci, E. L. & Ryan, R. M. (2000). The "what" and the "why" of goal pursuits: Human needs and the
self-determination of behavior. Psychological Inquiry, 11, 227-268.
Dovidio, J. F., Gaertner, S. L., Isen, A. M., & Lowrance, R. (1995). Group representations and
intergroup bias: Positive affect, similarity, and group size. Personality and Social Psychology
Bulletin, 21, 856–865.
Frank, M. G., Ekman, P., & Friesen, W. V. (1993). Behavioral markers and recognizability of the
smile of enjoyment. Journal of Personality and Social Psychology, 64, 83-93.
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The Broaden-and-
Build Theory of positive emotions. American Psychologist, 56, 218-226.
Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden the scope of attention and
thought-action repertoires. Cognition and Emotion, 19, 313-332.
Frijda, N. H. The laws of emotion. New York : Lawrence Erlbaum.
Hodgins, H. S. & Knee, R. (2002). The integrating self and conscious experience. In E. L. Deci, &
R. M. Ryan (Eds.), Handbook on self-determination research (pp. 87-100). Rochester, NY:
University of Rochester Press.
Hunter, J. E. and Schmidt, F. L. (2004), Methods of meta-analysis: Correcting error and bias in
research findings, (2nd ed.). Newbury Park, CA: Sage Publications.
Izard, C. E. (1977). Human emotions. New York: Plenum Press.
Johnson, K. J., & Fredrickson, B. L. (2005). ―We all look the same to me:‖ Positive emotions
Passion and Interpersonal Relationships 42
eliminate the own-race bias in face recognition. Psychological Science, 16, 875-881.
Jöreskog, K. G., & Sörbom, D. (2003). LISREL 8.54 for Windows (Computer software).
Lincolnwood, IL: Scientific Software International.
Keltner, D., & Bonanno, G. (1997). A study of laughter and dissociation: Distinct correlates of
laughter and smiling during bereavement. Journal of Personality and Social Psychology , 73,
687-702.
Keltner, D., & Haidt, J. (2001). Social functions of emotions. In T. J. Mayne & G. A. Bonanno
(Eds.), Emotions: Current issues and future directions (pp. 192–213). New York: Guilford Press.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York,
NY: Guilford Press.
Levenson, R., & Gottman J. (1983). Marital interaction - Physiological linkage and affective
exchange. Journal of Personality and Social Psychology, 45, 587-597.
Mageau, G. A., Vallerand, R. J., Rousseau, F. L., Ratelle, C. F., & Provencher, P. J. (2005). Passion
and gambling: Investigating the divergent affective and cognitive consequences of gambling.
Journal of Applied Social Psychology, 35, 100-118.
Mackinnon, A., Jorm, A. F., Christensen, H., Korten, A. E., Jacomb, P. A., & Rodgers, B. (1999). A
short form of the Positive and Negative Affect Schedule: Evaluation of factorial validity and
invariance across demographic variables in a community sample. Personality and Individual
Differences, 27, 405-416.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A
comparison of methods to test mediation and other intervening variable effects. Psychological
Methods, 7, 83–104.
Passion and Interpersonal Relationships 43
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879-
891.
Ratelle, C. F., Vallerand, R. J., Mageau, G. A., Rousseau, F. L, & Provencher, P. J. (2004). When
passion leads to problematic outcomes: A look at gambling. Journal of Gambling Studies, 23,
105-119.
Rousseau, F. L., & Vallerand, R. J. (2003). Le rôle de la passion dans le bien-être subjectif des ainés
[The role of passion in subjective well-being of the elderly]. Revue Québécoise de Psychologie,
24, 197-211.
Rousseau, F. L., & Vallerand, R. J. (2008). An examination of the relationship between passion and
subjective well-being in older adults. International Journal of Aging and Human Development,
66, 195-212.
Russell, D. (1996). The UCLA Loneliness Scale (Version 3): Reliability, validity, and factor
structure. Journal of Personality Assessment, 66, 20-40.
Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology,
39, 1161-1178.
Séguin-Lévesque, C., Laliberté, M.-L., Pelletier, L. G., Blanchard, C., & Vallerand, R. J. (2003).
Harmonious and obsessive passion for the internet: Their associations with the couple’s
relationships. Journal of Applied Social Psychology, 33, 197-221.
Senécal, C. B., Vallerand, R. J., & Vallières, E. F. (1992). Construction et validation de l'Échelle de
la Qualité des Relations Interpersonnelles (EQRI) [Construction and validation of the Quality of
Interpersonal Relationships Scale (QIRS)]. European Review of Applied Psychology, 42, 315-
322.
Passion and Interpersonal Relationships 44
Sheldon, K. M. (2002). The self-concordance model of healthy goal-striving: When personal goals
correctly represent the person. In E. L. Deci & R. M. Ryan (Eds.), Handbook of self-
determination research (pp. 65-86). Rochester, NY: University of Rochester Press.
Sheldon, K. M., Elliot, A. J., Kim, Y., & Kasser, T. (2001). What is satisfying about satisfying
events? Testing 10 candidate psychological needs. Journal of Personality and Social
Psychology, 80, 325-339.
Shiota, M. N., Campos, B., Keltner, D., & Hertenstein, M. (2003). Positive emotion and the
regulation of interpersonal relationships. In P. Philippot & R. S. Feldman (Eds.) The regulation
of emotion (pp. 127-156). Mahwah, NJ : Lawrence Erlbaum.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New
procedures and recommendations. Psychological Methods, 7, 422–445.
Vallerand, R. J. (1997). Toward a hierarchical model of intrinsic and extrinsic motivation. In M. P.
Zanna (Ed), Advances in experimental social psychology (pp. 271-360). New York: Academic
Press.
Vallerand, R. J. (2008). On the psychology of passion: In search of what makes people’s lives most
worth living. Canadian Psychology, 49, 1-13.
Vallerand, R. J., Blanchard, C. M., Mageau, G. A., Koestner, R., Ratelle, C., Léonard, M., Gagné,
M., & Marsolais, J. (2003). Les passions de l’âme: On obsessive and harmonious Passion.
Journal of Personality and Social Psychology, 85, 756-767.
Vallerand, R. J, & Houlfort, N. (2003). Passion at work: Toward a new conceptualization. In D.
Skarlicki, S. Gilliland, & D. Steiner (Eds.), Social issues in management (vol. 3, pp. 175-204).
Greenwich, CT: Information Age Publishing.
Passion and Interpersonal Relationships 45
Vallerand, R. J., Mageau, G. A., Elliot, A., Dumais, A., Demers, M-A., & Rousseau, F. L. (2008).
Passion and performance attainment in sport. Psychology of Sport & Exercise, 9, 373-392.
Vallerand, R. J., Rousseau, F. L., Grouzet, F. M. E., Dumais, A., & Grenier, S. (2006). Passion in
sport: A look at determinants and affective experiences. Journal of Sport & Exercise Psychology,
28, 454-478.
Vallerand, R. J., Salvy, S. J., Mageau, G. A., Elliot, A. J., Denis, P. L., Grouzet, F. M. E., &
Blanchard, C. B. (2007). On the role of passion in performance. Journal of Personality, 75, 505-
533.
Walter, F., & Bruch, H. (2008). The positive group affect spiral: A dynamic model of the emergence
of positive affective similarity in work groups. Journal of Organizational Behavior, 29, 239-261.
Watson, D., & Clark, L. A. (1997). Extraversion and its positive emotional core. In R. Hogan, J.
Johnson, & S. Briggs (Eds.), Handbook of personality psychology (pp. 767-793). San Diego, CA:
Academic Press.
Watson, D., Clark, L. A., Tellegen, A. (1988). Development and validation of brief measures of
positive and negative affect: The PANAS Scales. Journal of Personality and Social Psychology,
54, 1063-1070.
Waugh, C. E., & Fredrickson, B. L. (2006). Nice to know you: Positive emotions, self-other overlap,
and complex understanding in the formation of new relationships. Journal of Positive
Psychology, 1, 93-106.
Yik, M. S. M., Russell, J. A., & Feldman Barrett, L. (1999). Structure of self-reported current affect:
Integration and beyond. Journal of Personality and Social Psychology, 77, 600-619.
Passion and Interpersonal Relationships 46
Footnotes
1. See Vallerand (2008) and Amiot, Vallerand, and Blanchard (2006) for a comprehensive
differentiation between the Dualistic Model of Passion and other related constructs. Also, see
Philippe, Vallerand, & Lavigne (2009) for a differentiation between obsessive passion and
addiction.
2. Participants also indicated on an item ranging from 1 (do not agree at all) to 7 (completely
agree) the extent to which their work was a passion for them. In this study, 61% of the
participants reported to be passionate about their work (a rating of four and up according to
Vallerand and colleagues’ passion criterion (2003). In addition, results were exactly the same
with and without the participants who indicated to be less than moderately passionate for their
work. Thus, all participants were kept in the analyses.
3. Results showed that there were no differences among types of activity and all study variables,
except on positive emotions, F(4, 172) = 4.23, p < .01. Participants passionate for a team sports
activity reported higher levels of positive emotions compared to those passionate for a leisure,
t(137) = 3.45, p < .05. However, team sports and leisure did not differ from any other category.
Therefore, no further analyses were conducted with respect to activity categories. Correlations
between the number of people with whom the activity was engaged in and passion and the
quality of interpersonal relationships were not significant, but it was positively associated with
positive emotions (r = .17, p < .05). Controlling for number of people did not affect the results of
this study.
4. We assessed emotions reflecting the dimensions of valence and activation, as proposed
by Russell and his colleagues (Russell, 1980; Yik, Russell, & Feldman Barrett, 1999).
Specifically, a total of 16 items (four per subscale) were used to measure high positive emotional
Passion and Interpersonal Relationships 47
activation (e.g., active), high negative emotional activation (e.g., angry), medium positive
emotional activation (e.g., joyful), and medium negative emotional activation (e.g., shy). Items
assessing high activation levels were taken from the PANAS, while the items measuring medium
activation levels were derived from the Izard Differential Emotional Theory (Izard, 1977). There
was no difference in the hypothesized model as a function of the level of emotional activation.
Therefore, all emotional items for each valence (positive and negative) were averaged together
and no specific hypotheses were formulated with respect to activation levels.
5. It should be noted, however, that the results of the path analysis were all basically the same
when analyzing players’ interpersonal measures with the coach, teammates, and other players
separately, as well as for each relationship dimensions (i.e., positive quality of interpersonal
relationships and perceived interpersonal closeness).
6. Correlations and results of the path analysis were exactly the same when including the
participants who reported to know their coach and/or one or more of their teammates or players
of other teams prior to camp. Thus, all participants were kept in the analyses.
7. As can be seen in Table 4, there was no significant correlation between harmonious passion
and coaches’ ratings. Although past recommendations (Baron & Kenny, 1986) suggested that for
a mediation to occur the predictor needed to be positively correlated with the outcome, recent
recommendations based on simulation studies proposed that this condition should be dropped
(MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002), especially when the predictor is not
measured at the same time that the outcome (Shrout & Bolger, 2002). Indeed, in these latter
conditions, the direct effect tends to dilute itself over time, while the indirect effect becomes
more preponderant. Important conditions for mediation are that the path (i.e., association
Passion and Interpersonal Relationships 48
between the predictor and the mediator) and the path (i.e., association between the mediator
and the outcome) be both significant (also see Preacher & Hayes, 2008).
Passion and Interpersonal Relationships 49
Table 1
Means, Standard Deviations, and Correlations: Study 1
Mean
SD
1
2
3
4
Harmonious passion (1)
4.69
1.14
--
Obsessive passion (2)
3.14
1.28
.45***
--
Quality of interpersonal relationships (3)
2.46
0.72
.27***
.08
--
Positive emotions (4)
5.32
1.12
.55***
.20**
.36***
--
Hours of weekly involvement (5)
32.87
8.44
-.07
.06
.14*
-.04
Note: n = 195, *p < .05, **p < .01, ***p < .001
Passion and Interpersonal Relationships 50
Table 2
Bias-Corrected Bootstrapped Estimates of the Mediations of Studies 1 to 4
Bias-corrected
bootstrapped
95% confidence
interval estimates
Study 1
HP PE Positive QIR
(.178; .259)*
Study 2
HP PE Positive QIR
(.059; .124)*
HP NE Positive QIR
(.004; .053)*
OP NE Positive QIR
(-.023; -.008)*
Study 3
HP PE Athletes’ positive ratings
(.149; .265)*
HP PE Coaches’ positive ratings
(.064; .273)*
HP NE Athletes’ positive ratings
(.026; .079)*
HP NE Coaches’ positive ratings
(-.002; .142)
OP NE Athletes’ positive ratings
(-.230; -.088)*
OP NE Coaches’ positive ratings
(-.593; .007)
Study 4
HP PE Connectedness
(.117; .166)*
HP PE Seclusion
(-.091; -.055)*
HP PE Teammates’ positive ratings
(.440; .565)*
HP PE Teammates’ negative ratings
--a
HP NE Connectedness
(-.004; .098)
HP NE Seclusion
(-.098; .004)
HP NE Teammates’ positive ratings
--a
HP NE Teammates’ negative ratings
(-.09; .004)
OP NE Connectedness
(-.158; -.011)*
OP NE Seclusion
(.013; .159)*
OP NE Teammates’ positive ratings
--a
OP NE Teammates’ negative ratings
(.001; .014)*
Note. aIn these cases, the path (i.e., the association between the mediator and the outcome) was
not significant, HP = harmonious passion, OP = obsessive passion, PE = positive emotions, NE =
negative emotions, QIR = quality of interpersonal relationships, *p < .05
Passion and Interpersonal Relationships 51
Table 3
Means, Standard Deviations, and Correlations: Study 2
Mean
SD
1
2
3
4
5
6
Harmonious passion (1)
5.50
1.00
--
Obsessive passion (2)
2.29
1.22
.14*
--
Quality of interpersonal relationships (3)
3.23
0.69
.25***
-.06
--
Positive emotions (4)
3.99
0.83
.60***
.15*
.34***
--
Negative emotions (5)
1.77
0.83
-.13*
.19**
-.15*
.05
--
Extraversion (6)
3.76
0.57
.22**
-.06
.33***
.33***
-.11
--
Hours of weekly involvement a (7)
8.21
10.62
.13*
.26**
.06
.06
.16*
-.09
Note: a = non-transformed mean and standard deviation are presented
n = 177, *p < .05, **p < .01, ***p < .001
Passion and Interpersonal Relationships 52
Table 4
Means, Standard Deviations, and Correlations: Study 3
Mean
SD
1
2
3
4
5
Harmonious passion (1)
5.74
0.71
--
Obsessive passion (2)
3.48
1.37
.38***
--
Positive emotions (3)
4.03
0.58
.46***
.18**
--
Negative emotions (4)
1.57
0.53
-.17*
.14
-.06
--
Athletes’ ratings of positive interpersonal
relationships a (5)
0.00
0.70
.24**
.06
.48***
-.20**
--
Coaches’ ratings of interpersonal
closeness (6)
5.24
1.27
.12
.01
.22**
-.18**
.30***
Note: a = The mean and standard deviation are an average of z scores
n = 159, *p < .05, **p < .01, ***p < .001.
Passion and Interpersonal Relationships 53
Table 5
Means, Standard Deviations, and Correlations: Study 4
Mean
SD
1
2
3
4
5
6
7
Harmonious passion (1)
4.93
1.12
--
Obsessive passion (2)
2.52
1.01
.18*
--
Positive emotions (3)
3.94
0.63
.24**
.04
--
Negative emotions (4)
1.55
0.63
-.12
.14*
-.36**
--
Interpersonal connectedness (5)
4.16
1.44
.14
-.01
.56**
-.41**
--
Interpersonal seclusion (6)
2.89
1.08
-.12
.05
-.45**
.55**
-.51**
--
Teammates’ positive QIR ratings (7)
5.12
1.28
.07
-.01
.27**
-.16*
.17*
-.15*
--
Teammates’ negative QIR ratings (8)
2.09
0.88
-.04
.03
-.14*
.24**
-.12
.15*
-.32**
Note: All means and standard deviations pertain to untransformed variables. QIR = Quality of interpersonal relationships
n = 186, *p < .05, **p < .01.
Passion and Interpersonal Relationships 54
Figure Captions
Figure 1. Path analytic model of Study 1 of the relationships among passion, positive emotions,
and quality of interpersonal relationships, controlling for hours of weekly involvement.
Standardized path coefficients are presented. Non-significant direct effect paths are not shown
for the sake of clarity. However, the values of the path coefficients are those of the full mediating
model, including the direct effects of the independent variables to the outcome variables. *p <
.05, **p < .01, ***p < .001
Figure 2. Path analytic model of Study 2 among passion, positive and negative emotions, and
quality of interpersonal relationships, controlling for extraversion and hours of weekly
involvement. Standardized path coefficients are presented. Non-significant direct effect paths are
not shown for the sake of clarity. However, the values of the path coefficients are those of the
full mediating model, including the direct effects of the independent variables to the outcome
variables. *p < .05, **p < .01, ***p < .001
Figure 3. Path analytic model of Study 3 among passion, positive and negative emotions, and
positive and negative interpersonal relationships. Standardized path coefficients are presented.
Non-significant direct effect paths are not shown for the sake of clarity. However, the values of
the path coefficients are those of the full mediating model, including the direct effects of the
independent variables to the outcome variables. *p < .05, **p < .01, ***p < .001
Figure 4. Path analytic model of Study 4 among passion, positive and negative emotions, and
positive and negative interpersonal relationships. Standardized path coefficients are presented.
Non-significant direct effect paths are not shown for the sake of clarity. However, the values of
the path coefficients are those of the full mediating model, including the direct effects of the
independent variables to the outcome variables. QIR = quality of interpersonal relationships.
Passion and Interpersonal Relationships 55
*p < .05, **p < .01, ***p < .001
Passion and Interpersonal Relationships 56
Figure 1.
Passion and Interpersonal Relationships 57
Figure 2.
Passion and Interpersonal Relationships 58
Figure 3.
Passion and Interpersonal Relationships 59
Figure 4.
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Personality and Social Psychology
This content is subject to copyright. Terms and conditions apply.
