J Psychopathol Behav Assess (2010) 32:255–263

DOI 10.1007/s10862-009-9142-4

The Affective Style Questionnaire: Development

and Psychometric Properties
Stefan G. Hofmann & Todd B. Kashdan

Published online: 16 June 2009

# Springer Science + Business Media, LLC 2009

Abstract Affective style is an individual difference vari- with existing instruments measuring similar constructs.
able that refers to tendencies for regulating emotions. The Findings were cross-validated in an independent sample
emotion research literature has consistently identified three (n=495). The factor structure and results of psychometric
general strategies to handle emotional reactions: some analyses were replicated. The final 20-item Affective Style
strategies are aimed at re-adjusting affect to adapt success- Questionnaire is a brief instrument to measure individual
fully to situational demands; other strategies are intended to differences in emotion regulation.
conceal or suppress affect; and a third approach is to
tolerate and accept emotions, including unwanted and Keywords Affect regulation . Self-regulation .
aversive reactions. We conducted two studies to develop a Suppression . Cognitive reappraisal . Acceptance .
self-report measure to assess these affective styles. In the Experiential avoidance . Distress tolerance
first study (n=434), a list of 127 items related to this
construct was administered. A factor analysis supported One of the most remarkable features of humans is the
three factors: habitual attempts to conceal or suppress affect capacity to regulate and adjust their emotions depending on
(Concealing subscale; 8 items), a general ability to manage, particular situational demands. It is likely that this capacity
adjust, and work with emotions as needed (Adjusting is evolutionarily adaptive (e.g., Davidson 2003; Ekman
subscale; 7 items), and an accepting and tolerant attitude 2003; Izard 1992; Lazarus 1991) and closely connected to
toward emotions (Tolerating subscale; 5 items). The scale cognitive appraisal processes that distinguish humans from
showed satisfactory internal consistency. Furthermore, the non-humans (e.g., Frijda 1986; Lazarus 1991; Scherer and
respective subscales showed different patterns of relations Ellgring 2007). Emotion regulation refers to the process by
which people influence which emotions they have, when
they have them, and how they experience and express these
emotions. Consistent with previous authors, we define
The contributions of the two authors were equal. Stefan G. Hofmann
is a paid consultant of Organon (Schering-Plough) for work unrelated
affective style as inter-individual differences in the sensi-
to this study. This study was partially supported by NIMH grants tivity to and regulation of emotions (Davidson 1998). Some
MH-078308 and MH-081116 awarded to Dr. Hofmann and affective styles effectively regulate the experience and
MH-73937 to Dr. Kashdan. expression of emotions in ways that increase progress
S. G. Hofmann (*) toward valued aims, whereas other strategies have appar-
Department of Psychology, Boston University, ently unintended, counterproductive effects. For example,
648 Beacon Street, 6th Floor,
attempts to suppress emotions increase physiological
Boston, MA 02215-2002, USA
e-mail: [email protected] arousal (Gross and Levenson 1997), and rumination over
negative emotional events prolongs angry and depressed
T. B. Kashdan affective states (Nolen-Hoeksema and Morrow 1993;
Department of Psychology, George Mason University,
Rusting and Nolen-Hoeksema 1998). In contrast, an
Mail Stop 3F5,
Fairfax, VA 22030, USA accepting stance toward arousing emotional experiences
e-mail: [email protected] without unnecessary attempts to change or avoid them has
256 J Psychopathol Behav Assess (2010) 32:255–263

been linked to increased persistence in challenging sit- for the persistence of ineffective emotion regulation is
uations and reductions in subjective distress (Hayes et al. people’s acceptance and tolerance of particular emotional
2006). experiences (Salovey et al. 1995). Some people respond to
Emotion regulation strategies can be classified based on the onset of emotions by appraising them as intolerable and
the time point at which people engage in these activities subsequently engage in avoidance, concealment, or other
during the emotion generation process and based on the counterproductive response-focused interventions. Recently
efficacy of these efforts (Gross and Levenson 1997). developed treatments for emotional disorders employ
Antecedent-focused regulation occurs before the emotional techniques that target such negative judgments of emotions
response has been fully activated. This includes tactics such and maladaptive emotional control efforts (e.g., Segal et al.
as attention deployment, situation modification, cognitive 2002).
reframing of a situation, and any preparatory action (e.g., In sum, the emotion literature consistently identifies
listening to particular energizing music before a work-out different affective styles for regulating emotions. The first
routine). Response-focused regulation reflects attempts to style includes suppression and other response-focused
alter the experience or expression of ongoing emotions. strategies aimed at concealing and avoiding emotions after
This includes tactics such as suppression and acceptance- they arise. We refer to this affective style as concealing
based attitudes. strategies. A second style characterizes people who are
Laboratory studies suggest that antecedent-focused strate- “more able to access and utilize emotional information in
gies such as reappraisal are relatively effective for regulating adaptive problem solving, and better able to modulate
emotions in the short-term, whereas suppression-based emotional experience and expression according to contextual
response-focused strategies are often unintentionally counter- demands” (Mennin et al. 2002a, b, p. 88). These individuals
productive (e.g., Gross and Levenson 1997). It has further possess the tools to readjust or balance emotions as needed
been shown that people differ in their habitual use of to successfully navigate the rewards and punishments of
antecedent- and response-focused emotion regulation strate- everyday life. We refer to this affective style as adjusting.
gies, and that these individual differences are meaningfully Finally, a third style reflects comfort and non-defensiveness
associated with emotional experiences and psychosocial in response to arousing emotional experiences as they exist
functioning (Gross and John 2003). For instance, people in the present moment. This includes a strong tolerance of
who are more reliant on reappraisal as a regulatory strategy distress. We refer to this affective style as tolerating
experience better social functioning and greater well-being, strategies.
as assessed by self-reports, peer reports, and reactivity to To our knowledge no instrument exists that measures
laboratory stimuli. In contrast, people who are more reliant these three broad affective styles. The most relevant
on suppression as a regulatory strategy experience worse instruments are the 10-item Emotion Regulation Question-
social functioning and well-being (Gross and John 2003). naire (ERQ; Gross and John 2003) measuring individual
The roles of emotion regulation strategies have been differences in expressive suppression and cognitive reap-
discussed for a range of mental disorders, including praisal; the 10-item Acceptance and Action Questionnaire-
substance abuse (Hayes et al. 1996), anxiety and mood II (Bond et al. submitted; Hayes et al. 2004) measuring
disorders (Campbell-Sills and Barlow 2007; Mennin et al. individual differences in the willingness to accept and work
2002a, b), and borderline personality disorder (Linehan with private thoughts and feelings in the pursuit of valued
1993). Furthermore, individual differences in emotion goals (an aggregation of multiple facets); and the 36-item
regulation strategies among non-clinical individuals have Difficulties in Emotion Regulation Scale (DERS: Gratz and
been shown to be associated with subjective well-being Roemer 2004) measuring various ways that people habit-
(Gross and John 2003) and biological correlates (Drabant et ually find themselves unable to successfully regulate
al. 2009). For example, it has been shown that greater use difficult, aversive emotional experiences. Although useful,
of reappraisal in everyday life was related to decreased the existing instruments are limited by the relatively
amygdala activity and increased activity in prefrontal restricted scope of possible affect regulation strategies and
control regions in response to negative emotional stimuli the constraints of a particular theoretical orientation. The
(Drabant et al. 2009). Such individual differences in goal of the present research was to develop a brief but
emotion regulation strategies might predict successful psychometrically sound scale to measure the general
coping with emotional challenges as well as the onset of affective styles of concealing, adjusting, and tolerating. A
emotional disorders. Therefore, an important question reliable, valid scale of these affective styles can be useful
pertaining to emotion regulation concerns the variation for researchers interested in studying people that success-
between people in their habitual tendency to use some fully use their emotions to navigate the shoals of everyday
regulatory strategies over others, especially if the preferred life and people with psychological disruptions characterized
strategy has undesirable outcomes. A potential explanation by emotional difficulties. This particular instrument can
J Psychopathol Behav Assess (2010) 32:255–263 257

also serve the aims of practitioners interested in assessing correlations ranging between .20 and .40. Finally, the
baseline self-regulation and monitoring interventions factorial validity was demonstrated by a confirmatory
addressing psychological, physical, and social well-being. factor analysis that supported the three-factor model.

Difficulties in Emotion Regulation Scale (DERS: Gratz and

Study 1 Roemer 2004) The DERS contains 36 items to assess six
dimensions of self-regulatory difficulties: nonacceptance of
Participants emotional responses (accept), difficulties engaging in goal-
directed behavior (when upset; goals), impulse control
A total of 457 undergraduate students of Boston University difficulties (when upset; impulse), lack of emotional
(BU), 18 years of age or older, participated in this study. awareness (aware), limited access to effective emotion
Due to missing data, the following analyses are based on regulation strategies (strategies), and lack of emotional
434 participants. The mean age of participants was 19.15 clarity (clarity). Gratz and Roemer (2004) reported high
(SD=2.61). The majority of the sample was female (67%) internal consistency of the total DERS (α=.93), adequate
and Caucasian (68.1%). Remaining participants self- internal consistency of all subscales (α’s >.8), and also
identified as being Asian or Asian–American (18.40%), adequate 4–8 weeks test-retest reliability of the total scale
Hispanic, Latino, or Mexican–American (3.7%), and (r = .88). The 4–8 weeks test-retest reliability of the
African American (2.6%). Students most commonly majored subscales ranged from r=.69 (nonacceptance subscale) to
in biology (n=39), psychology (n=36), anatomy/human r=.80 (clarity subscale). The authors further reported
physiology (n=26), business (n=24), education (n=23), evidence for convergent and predictive validity.
communications (n=15), advertising (n=12), biochemistry
and molecular biology (n=11), journalism (n=11) manage- Berkeley Expressivity Questionnaire (BEQ; Gross and John
ment (n=11), marketing (n=10), and public relations 1995) The Berkeley Expressivity Questionnaire consists of
(n=10). Most of the students were undecided (n=63) and 16 items to assess three facets of emotional expressivity:
in 35 cases the major was unknown. The remaining students negative expressivity, positive expressivity, and impulse
majored in one of 37 other subjects. Participants were on strength. Gross and John (1995) reported that the Cronbach’s
average 19.15 years old (SD=1.61). Students attending this α coefficients of the three subscales ranged between 0.71
college typically come from middle to upper middle class and 0.76, and the 2-months test-retest reliability ranged
socioeconomic backgrounds. The study was reviewed and between 0.71 and 0.82. Finally, the 3-factor solution was
approved by the Institutional Review Board of Boston replicated in a separate sample.
University.
Emotion Regulation Questionnaire (ERQ; Gross and John
Instruments 2003) This scale consists of 10 items and assesses
individual differences in two emotion regulation strategies:
In addition to the 127 author-generated emotion items, expressive suppression and cognitive reappraisal. The scale
several validated self-report measures were administered. shows good psychometric properties (Gross and John
2003). Gross and John (2003) reported Cronbach’s alpha
Brief COPE (Carver 1997) The Brief COPE, an abbrevi- coefficients, ranging between .79 (for the Reappraisal
ated version of the COPE (Carver et al. 1989), is a 28-item subscale) and .73 (for the Suppression subscale). The
inventory consisting of 14 subscales. The instrument 3-months test–retest reliability was .69 for both scales.
assesses individual differences in the use of effective and Factor analyses supported the 2-factor, orthogonal factor
non-effective coping strategies. Each subscale has two structure of the measure.
items (α=.50 to .90). Examples of coping scales include
Denial, Active Coping and Behavioral Disengagement. Acceptance and Action Questionnaire-II (AAQ-II; Bond et
al., submitted) The AAQ-II is a 10-item measure, a
Toronto Alexithymia Scale (TAS-20; Bagby et al. 1994) The refinement of the original scale (Hayes et al. 2004), to
TAS-20 is a 20-item scale to measure alexithymia, a assess individual differences in acceptance and experiential
construct reflecting difficulty identifying, describing, and avoidance. The psychometric characteristics are adequate.
being aware of emotions. The scale has become a widely In 7 datasets, the AAQ-II has been shown to possess a
used measure of this construct. Parker et al. (2003) reported unitary factor and adequate internal consistency (α=.81 to
internal consistency estimates (Cronbach’s α) of the three .89) (Bond et al. submitted; McCracken and Zhao-O-Brien
TAS factors to be above .70. The homogeneity of the three in press). Researchers found large positive relations with
scales was further supported by the mean inter-item measures of general health (e.g., r=−.67 with Symptom
258 J Psychopathol Behav Assess (2010) 32:255–263

Checklist-10R) and large negative relations with measures affective styles, concealing, adjusting, or tolerating affect.
of suppression tendencies (e.g., r=−.58 with White Bear The remaining factors could not be easily interpreted. In the
Suppression Inventory) and emotional disturbances (e.g., next step, we selected items loading highly on these factors
r=−.59 with Beck Anxiety Inventory and r=−.75 with and eliminated items with poor item validity.
Beck Depression Inventory-II) (Bond et al. submitted) as
well as construct specificity above and beyond measures of Psychometric Data
pain chronicity and mindful awareness in predicting pain-
related distress and disability in adult patients at a pain The first factor (Concealing), consisting of 8 items,
clinic (McCracken and Zhao-O-Brien in press). accounted for 22.18% of the variance. The second factor,
consisting of 7 items and accounting for 15.81% of the
variance, was interpreted as the Adjusting subscale. The
Procedure third factor, the Tolerating subscale, consisting of 5 items,
accounted for 10.09% of the variance. These three factors
Participants completed a web-based survey (PsychData) were the only ones with Eigenvalues above 1. The scree
that included demographic questions, a pool of emotion plot further confirmed the 3-factor solution.
regulation items for the development of our new scale, and The correlation matrix was subjected to a varimax
several published self-report instruments. The study was rotation. The resulting 3-factor structure is presented in
reviewed and approved by the Institutional Review Boards Table 1 (Study 1 column). Concealing showed a mean of
of Boston University and George Mason University. 23.94 (SD=6.94, median = 24, mode = 24, range: 8–40);
Informed consent was obtained with an initial opening Adjusting showed a mean of 20.36 (SD=5.05, median = 20,
screen providing all of the details about the study and range: 7–34), and Tolerating showed a mean of 15.23 (SD=
potential costs and benefits for participation prior to any 3.42, median = 15, range: 6–25).
survey questions. Potential participants were required to Internal consistency was acceptable for the Concealing
indicate that they had read the consent form prior to (α=.84), Adjusting (α=.80), and Tolerating (α=.66) sub-
continuation. As part of psychology course requirements, scales. The item-total correlation coefficients were r’s >.59
students are asked to participate in research studies to gain for Concealing, r’s >.61 for Adjusting, and r’s >.57 for
direct experience as subjects in research experiments. Tolerating subscales (p’s < .0001). The Concealing and
Typically, students must accumulate 3 h worth of research Adjusting subscales showed a correlation of r=.17, the
credits as part of the course requirements for an introduc- Adjusting and Tolerating subscales correlated at r=.13, and
tory psychology class. If students participate in these the Concealing and Tolerating subscales showed a correlation
experiments, investigators are obligated to provide them of r=−.08.
with credit for these research efforts. In sum, the study was
fully compliant with the ethical guidelines of the institutional Construct Validity
review boards and the ethical guidelines of psychology as a
profession. Table 2 (Study 1 column) shows the correlations between
The two authors generated 127 items assessing different the three subscales of the ASQ and related instruments. As
ways of dealing with emotions. Many items derived were expected, large correlations were observed between the
based on the work by Gross and John (2003), who ASQ-Adjusting subscale and the AAQ-II (r=.47) and
distinguish antecedent- and response-focused strategies, the Reappraisal subscale of the ERQ (r=.54), and between
and the acceptance and mindfulness-based literature (e.g., the ASQ-Concealing subscale and the ERQ suppression
Hayes et al. 2006). The complete item pool can be obtained subscale (r=.60) and BEQ-Negative Expressivity (r=−.68).
from the authors. Interestingly, we also found a moderate negative correlation
between the ASQ-Tolerating subscale and the ERQ
suppression subscale (r=−.34), suggesting that suppressing
Results one’s emotions requires concealment and low distress
tolerance. Also, the ASQ-Concealing and Adjusting sub-
Factor Structure scales were differentially linked to difficulties describing
and identifying feelings, and only the ASQ-Tolerating
The principal component analysis of the entire item pool subscale was linked to being emotionally aware (r=−.34
resulted in 30 factors with Eigenvalues greater than one with externally oriented subscale of TAS-20).
accounting for 67.48% of the total variance. A close As for the DERS subscales, the highest correlations were
inspection of the factor structure suggested that items observed between the Adjusting subscale and the DERS
loading on the first three factors describe three distinct strategies subscale (limited access to effective emotion
J Psychopathol Behav Assess (2010) 32:255–263 259

Table 1 Factor structure from Study 1 (Boston University; N=434) and Study 2 (George Mason University; N=495)

Items Study 1 Study 2

Concealing Adjusting Tolerating Concealing Adjusting Tolerating

People usually can’t tell how I am feeling inside. .65 −.02 −.23 .64 −.02 −.06
I often suppress my emotional reactions to things. .64 −.03 −.03 .66 −.07 −.12
I am good at hiding my feelings. .81 .06 −.03 .78 .05 −.02
People usually can’t tell when I am upset. .72 .19 −.09 .77 .10 −.04
People usually can’t tell when I am sad. .73 .20 −.09 .76 .10 −.02
I can act in a way that people don’t see me being upset. .72 .19 .06 .73 .24 .03
I could easily fake emotions. .60 .05 .14 .58 −.13 .09
I can hide my anger well if I have to. .56 .33 .18 .58 .36 .04
I have my emotions well under control. .21 .63 −.10 .00 .62 .04
I can avoid getting upset by taking a different perspective .17 .63 .05 .10 .65 .06
on things.
I am able to let go of my feelings. .13 .66 −.03 .05 .60 −.10
I can calm down very quickly. .14 .71 .05 .22 .72 .04
I can get out of a bad mood very quickly. .07 .68 .15 .01 .69 .07
I know exactly what to do to get myself into a better mood. −.05 .70 .12 .00 .63 .11
I can get into a better mood quite easily. −.00 .80 .08 −.05 .77 .03
I can tolerate having strong emotions. .05 .20 .66 .00 .06 .72
It’s ok if people see me being upset. −.25 .06 .61 −.27 .10 .61
It’s ok to feel negative emotions at times. .04 −.07 .67 .03 −.07 .63
I can tolerate being upset. .15 .27 .63 .13 .22 .53
There is nothing wrong with feeling very emotional. −.05 −.10 .70 −.05 .04 .74

The Table shows the factor scores (varimax rotation) of the items of the Affective Style Questionnaire items. High-loading items are printed in bold

regulation strategies; r=−.54), goals subscale (difficulties traits provided evidence of convergent and discriminant
engaging in goal-directed behavior; r=−.40), and the validity. To further evaluate the structure and psychometric
impulse subscale (impulse control difficulties; r=−.44). properties of our scale, we conducted a second study with
Moderate correlations were also observed between the an independent sample.
Tolerating subscale and the DERS accept subscale (nonac-
ceptance of emotional response; r=−.31) and the aware
subscale (lack of emotional awareness; r=−.46). The three Study 2
ASQ subscales showed consistently low correlations with
the Brief COPE Subscales, supporting their discriminant Participants
validity (all r’s < .24, except for the correlation between the
Brief COPE planning subscale and the ASQ adjusting A total of 528 undergraduate students of George Mason
subscale, r=−.32). University (GMU) in Virginia participated in this study.
Due to missing data, the following analyses are based on
495 participants.
Discussion of Study 1 Participants had a mean age of 22.02 (SD=5.23). The
majority of the sample was female (78%) and Caucasian
With the goal of developing a short and psychometrically (54.5%). Remaining participants self-identified as being
sound measure of affective style, our results were promis- Asian or Asian–American (17.8%), Hispanic, Latino, or
ing. We found evidence for 3 meaningful and interpretable Mexican–American (7.4%), and African American (8.4%).
factors leading to Concealing, Adjusting, and Tolerating Compared with the Study 1 sample, the GMU sample was
affect subscales. Each of these subscales showed high item significantly older, t (797.57) = −11.23, p<.0001, and there
validity and internal consistency. Inter-correlations among was a greater proportion of women, χ2 (1) = 15.20,
the ASQ subscales with other measures of emotion p<.0001, and racial and ethnic diversity, χ2 (1) = 18.01,
regulation, psychological flexibility, and other personality p<.0001.
260 J Psychopathol Behav Assess (2010) 32:255–263

Table 2 Correlations between affective style questionnaire and other instruments in Study 1 (Boston University) and Study 2 (George Mason
University)

Questionnaires Study 1 Study 2

Concealing Adjusting Tolerating Concealing Adjusting Tolerating

ERQ
Reappraisal .13* .54** .09 .14* .57** .14**
Suppression .60** −.03 −.34** .52** −.05 −.32**
BEQ
Negative Expressivity −.68** −.16* .16* −.70** −.27** .10*
Positive Expressivity −.23** .07 .28** −.33** .07 .28**
Impulse Strength −.15* −.39** .19** −.30** −.38** .21**
AAQ-II −.05 .47** .22** −.03 .48** .18**
Brief Cope
Self-Distraction .10* .12* .09 .17** .19** .07
Active Coping −.11 .08 .29** −.12 .05 .27**
Substance Use .08 .10* .22** .01 .14* .21**
Emotional Support .17** .16** .08 .20** .24** .13**
Instrumental Support .04 −.15** −.06 .06 −.07 −.01
Behavioral Disengagement −.04 .20** .03 −.01 .14** .10*
Venting −.19 .01 .18** −.10* .03 .07
Positive framing .04 −.04 .14** .04 −.02 .07
Planning −.02 −.32** .04 −.07 −.32** .11*
Humor .08 .06 .02 .12** .05 .14**
Acceptance −.02 −.02 −.05 −.07 −.01 −.01
Religion −.19** −.10* .05 −.14** −.14** .06
Self-Blame .05 −.24** −.09 .01 −.20** −.20**
TAS-20
Difficulty Identifying Feelings .18** −.27** −.16** .11* −.25** −.12**
Difficulty Describing Feelings .38** −.18** −.28** .29** −.17** −.25**
Externally Oriented .13 −.02 −.34** .16** −.03 −.28**
DERS
Total .11* −.48** −.32** .02 −.51** −.23**
Clarity .18** −.30** −.28** .09* −.30** −.24**
Aware .15** −.08 −.46** .20** −.13 −.39**
Impulse −.04 −.44** −.13** −.09* −.44** −.08
Accept .10* −.27** −.31** .12** −.27** −.15**
Goals −.06 −.40** −.20 −.08 −.43** −.08
Strategies .06 −.54** −.19** −.04 −.54** −.08**

The Table shows product moment correlations between the ASQ subscales and other questionnaires. AAQ-II Acceptance and Action
Questionnaire, BEQ Berkeley Expressivity Questionnaire, ERQ Emotion Regulation Questionnaire, DERS Difficulties in Emotion Regulation
Scale, TAS-20 Toronto Alexithymia Scale
*p<.05; **p<.001

Students most commonly majored in psychology (n=195), middle to upper middle class socioeconomic backgrounds.
nursing (n=44), biology (n=23), business (n=21), accounting The study was reviewed and approved by the Institutional
(n=19), communication (n=19), administration of justice (n= Review Board of George Mason University.
19), English (n=16), and finance (n=19). Seventeen students
were undecided and in 33 cases the major was unknown. The Instruments and Procedure
remaining students majored in one of 44 other subjects. As in
Study 1, students attending this college typically come from The instruments and procedure were identical to Study 1.
J Psychopathol Behav Assess (2010) 32:255–263 261

Results differences between the two samples in terms of gender,

racial, and ethnic diversity, the results were replicated. The
Factor Structure factor analysis again revealed the identical 3-factor solution
with the same items loading on subscales interpreted as
The principal component analysis with the 20-item ASQ concealing, adjusting or tolerating affect. Based on the
resulted again in 3 factors with Eigenvalues greater than pattern of correlations with other instruments measuring
one accounting for 49.6% of the total variance. As shown emotion regulation, coping, and personality, the results
in Table 1 (Study 2 column), the factor structure of Study 1 support the uniqueness of each ASQ subscale and their
was replicated perfectly. The Concealing, Adjusting, and convergent and discriminant validity. It should be noted that
Tolerating factors accounted for 19.70%, 18.34, and Study 2 essentially replicated Study 1, and that both studies
11.54% of the variance, respectively. The scree plot are limited by the sole reliance on an undergraduate student
confirmed the 3-factor solution. population. Neither sample is representative of the general
population, because both samples comprise young adults
Psychometric Data and predominantly White females. This limitation should
be taken into consideration when interpreting the findings.
Concealing showed a mean of 22.60 (SD=6.31, median =
22, mode = 21, range: 8–40); Adjusting showed a mean of
20.96 (SD=5.15, median = 21, range: 7–35), and Tolerating General Discussion
showed a mean of 15.47 (SD=3.43, median = 15, range:
6–25). Our objective was to develop a short, reliable, and valid
Internal consistency was acceptable for Concealing, α=.84, measure of affective style. Based on a thorough review of
Adjusting, α=.82, and Tolerating, α=.68, subscales. The the emotion and clinical literature, a large pool of items was
item-total correlation coefficients were high for Concealing, generated. Using two large samples of college students at
r’s >.61, Adjusting, r’s >.66, and Tolerating, r’s >.64, separate universities, we were able to create a 20-item scale
subscales (p’s < .0001). The Concealing and Adjusting consisting of 3 subscales: Concealing, Adjusting, and
subscales showed a correlation of r=.30, the Adjusting and Tolerating affect. Given the brevity of the measure, the
Tolerating subscales correlated at r=.17, and the Concealing scale shows excellent psychometric properties. The sub-
and Tolerating subscales showed a correlation of r=−.03. scales map onto existing measures of emotion regulation
and also include additional facets of affective style.
Construct Validity An important limitation of the study is the sole reliance
on undergraduate student samples. Therefore, it is not
Table 2 (Study 2 column) shows the correlations between possible to make generalizations to other samples, such as
the ASQ subscales and other instruments, with patterns adults with evidence of effective emotion regulation (e.g.,
similar to Study 1. The strongest relations with the ASQ- military leaders, parents balancing work and family life)
Concealing subscale were the BEQ-Negative Expressivity and clinical populations. We suggest that future studies
(r=−.70), ERQ-Suppression (r=.52), and BEQ-Positive validate the instrument in clinical populations with affective
Expressivity (r=−.33) subscales. The strongest relations disorders, such as mood and anxiety disorders and
with the ASQ-Adjusting subscale were ERQ-Reappraisal personality disorders characterized by emotion regulation
(r=.57), DERS Strategies (r=−.54), and DERS Goals disturbances. Furthermore, it will be important to study
(r=−.43) subscales and the AAQ-II (r=.48); indicative of changes in affective style during the course of treatment
adaptive regulation and psychological flexibility. The and to examine whether a particular affective style predicts
strongest relations with the ASQ-Tolerating subscale were treatment response. We hypothesize that concealing affect
DERS Aware (r=−.39), ERQ-Suppression (r=−.32), BEQ- and a general inflexibility in adjusting one’s affective style
Positive Expressivity (r=.28), and TAS-20 Externally to the situational demands are particularly maladaptive
Oriented (r=−.28) subscales; indicative of effective use of strategies for coping with negative affect. Future studies
acceptance and mindfulness strategies. will further need to examine the temporal stability of the
measure. Finally, although we found evidence for the
separability of each affective style, we did not evaluate
Discussion of Study 2 whether the ASQ subscales differentially predict how
people respond to aversive and rewarding events in the
The goal of this study was to cross-validate the findings of laboratory and naturalistic environments. It will be important
Study 1 and further validate the ASQ in a sample of college to employ multi-method assessments and investigate the
students from a different university. Although there were temporal course of how people regulate their emotions in
262 J Psychopathol Behav Assess (2010) 32:255–263

future research. Despite the limitations of our two initial

studies, this measure may become a potentially useful tool in
basic and clinical research.

Appendix

ASQ

Instructions We are interested in how you experience and manage your emotions. Obviously, different situations bring out somewhat different
responses, but think about what you usually do. Please try to respond to each item separately in your mind from each other item. Do not indicate
agreement with things that you think you should do or wish you do. Instead, choose your answers thoughtfully, and make your answers about
what is true FOR YOU. Please answer every item. There are no “right” or “wrong” answers, so choose the most accurate answer for YOU−—not
what you think “most people” would say or do. Use the scale below to answer each item

1-------------------------2----------------------3---------------------4--------------------------5
not true of me at all a little bit moderately quite a bit extremely true of me

1. People usually can’t tell how I am feeling inside. 1—2—3—4—5

2. I have my emotions well under control 1—2—3—4—5
3. I can tolerate having strong emotions. 1—2—3—4—5
4. I can avoid getting upset by taking a different perspective on things. 1—2—3—4—5
5. I often suppress my emotional reactions to things. 1—2—3—4—5
6. It’s ok if people see me being upset. 1—2—3—4—5
7. I can calm down very quickly 1—2—3—4—5
8. I am able to let go of my feelings. 1—2—3—4—5
9. I am good at hiding my feelings. 1—2—3—4—5
10. People usually can’t tell when I am upset. 1—2—3—4—5
11. It’s ok to feel negative emotions at times. 1—2—3—4—5
12. I can get out of a bad mood very quickly. 1—2—3—4—5
13. People usually can’t tell when I am sad. 1—2—3—4—5
14. I can tolerate being upset. 1—2—3—4—5
15. I can act in a way that people don’t see me being upset. 1—2—3—4—5
16. I know exactly what to do to get myself into a better mood. 1—2—3—4—5
17. There is nothing wrong with feeling very emotional. 1—2—3—4—5
18. I could easily fake emotions. 1—2—3—4—5
19. I can get into a better mood quite easily. 1—2—3—4—5
20. I can hide my anger well if I have to. 1—2—3—4—5

All items are straight-forward scored. Higher scores reflect a preference for an affective style. Concealing is the sum of items 1, 5, 9, 10, 13, 15, 18, and
20. Adjusting is the sum of items 2, 4, 7, 8, 12, 16, and 19. Tolerating is the sum of items 3, 6, 11, 14, and 17

References mood disorders. In J. Gross (Ed.), Handbook of emotion

regulation (pp. 542–559). New York: Guilford.
Carver, C. S. (1997). You want to measure coping but your protocol’s
Bagby, R. M., Parker, J. D. A., & Taylor, G. J. (1994). The 20-item too long: consider the Brief COPE. International Journal of
Toronto Alexithymia Scale-I. Item selection and cross-validation Behavioral Medicine, 4, 92–100.
of the factor structure. Journal of Psychosomatic Research, 38, Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing
23–32. coping strategies: a theoretically based approach. Journal of
Bond, F. W., Hayes, S. C., Baer, R. A., Carpenter, K. M., Orcutt, Personality and Social Psychology, 56, 267–283.
H. K., Waltz, T., & Zettle, R. D. (submitted). Preliminary psychomet- Davidson, R. J. (1998). Affective style and affective disorders: perspectives
ric properties of the Acceptance and Action Questionnaire—II: A form affective neuroscience. Cognition and Emotion, 12, 307–320.
revised measure of psychological flexibility and acceptance. Davidson, R. J. (2003). Darwin and the neural bases of emotion and
Campbell-Sills, L., & Barlow, D. H. (2007). Incorporating emotion affective style. Proceedings of the New York Academy of
regulation into conceptualizations and treatments of anxiety and Sciences, 1000, 316–336.
J Psychopathol Behav Assess (2010) 32:255–263 263

Drabant, E. M., McRae, K., Manuck, S. B., Hariri, A. R., & Gross, J. J. Lazarus, R. (1991). Emotion and adaptation. New York: Oxford
(2009). Individual differences in typical reappraisal use predict University Press.
amygdala and prefrontal responses. Biological Psychiatry, 65, 367– Linehan, M. M. (1993). Cognitive-behavioral treatment of borderline
373. personality disorder. New York: The Guilford.
Ekman, P. (2003). Emotions revealed. New York: Times Books. McCracken, L. M., & Zhao-O-Brien, J. (in press). General psycho-
Frijda, N. H. (1986). The emotions. Cambridge, England: Cambridge logical acceptance and chronic pain: there is more to accept than
University Press. pain itself. European Journal of Pain.
Gratz, K. L., & Roemer, L. (2004). Multidimensional assessment of Mennin, D. S., Heimberg, R. G., Turk, C. L., & Fresco, D. M. (2002a).
emotion regulation and dysregulation: development, factor Applying an emotion regulation framework to integrative
structure, and initial validation of the difficulties in emotion approaches to generalized anxiety disorder. Clinical Psychology:
regulation scale. Journal of Psychopathology and Behavioral Science and Practice, 9, 85–90.
Assessment, 26, 41–54. Mennin, D. S., Heimberg, R. G., Turk, C. L., & Fresco, D. M. (2002b).
Gross, J. J., & John, O. P. (1995). Facets of emotional expressivity: Commentary on Roemer and Orsillo: applying an emotion regulation
three self-report factors and their correlates. Personality and framework to integrative approaches to generalized anxiety disorder.
Individual Differences, 19, 558–568. Clinical Psychology: Science and Practice, 9, 85–90.
Gross, J. J., & John, O. P. (2003). Individual differences in two Nolen-Hoeksema, S., & Morrow, J. (1993). Effects of rumination and
emotion regulation processes: implications for affect, relation- distraction on naturally-occurring depressed mood. Cognition
ships, and well-being. Journal of Personality and Social and Emotion, 7, 561–570.
Psychology, 85, 348–362. Parker, J. D. A., Taylor, G. J., & Bagby, R. M. (2003). The twenty-
Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: the acute item Toronto Alexithymia scale-III: reliability and factorial
effects of inhibiting negative and positive emotion. Journal of validity in a community population. Journal of Psychosomatic
Abnormal Psychology, 106, 95–103. Research, 55, 269–275.
Hayes, S. C., Luoma, J., Bond, F., Masuda, A., & Lillis, J. (2006). Rusting, C., & Nolen-Hoeksema, S. (1998). Regulating responses to
Acceptance and commitment therapy: model, processes, and anger: effects of rumination and distraction on angry mood.
outcomes. Behaviour Research and Therapy, 44, 1–25. Journal of Personality and Social Psychology, 74, 790–803.
Hayes, S. C., Strosahl, K. D., Wilson, K. G., Bissett, R. T., Pistorello, Salovey, P., Mayer, J. D., Goldman, S., Turvey, C., & Palfai, T.
J., Toarmino, D., et al. (2004). Measuring experiential avoidance: (1995). Emotional attention, clarity, and repair: Exploring
a preliminary test of a working model. The Psychological emotional intelligence using the Trait Meta-Mood Scale. In J. W.
Record, 54, 553–578. Pennebaker (Ed.), Emotion, disclosure, and health (pp. 125–154).
Hayes, S. C., Wilson, K. G., Gifford, E. V., Follette, V. M., & Washington, D.C.: American Psychological Association.
Strosahl, K. (1996). Experiential avoidance and behavioral Scherer, K. R., & Ellgring, H. (2007). Multimodal expression of
disorders: a functional dimensional approach to diagnosis and emotion: affect programs or componential appraisal patterns?
treatment. Journal of Consulting and Clinical Psychology, 64, Emotion, 7, 158–71.
1152–1168. Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002).
Izard, C. E. (1992). Basic emotions, relations among emotions, and Mindfulness-based cognitive therapy for depression: A new
emotion-cognition relations. Psychological Review, 99, 561–565. approach to preventing relapse. New York: Guilford.