Professional Documents
Culture Documents
Spaapenetal 2013
Spaapenetal 2013
net/publication/256982512
CITATIONS READS
60 8,997
5 authors, including:
Some of the authors of this publication are also working on these related projects:
Subjective Cognitive Decline: A new scale and potential for psychological interventions View project
All content following this page was uploaded by Romola S Bucks on 19 May 2014.
CITATION
Spaapen, D. L., Waters, F., Brummer, L., Stopa, L., & Bucks, R. S. (2013, September 23). The
Emotion Regulation Questionnaire: Validation of the ERQ-9 in Two Community Samples.
Psychological Assessment. Advance online publication. doi: 10.1037/a0034474
Psychological Assessment © 2013 American Psychological Association
2013, Vol. 25, No. 4, 000 1040-3590/13/$12.00 DOI: 10.1037/a0034474
The 10-item Emotion Regulation Questionnaire (ERQ) was developed by Gross and John (2003) to
measure the habitual use of 2 emotion regulation strategies: reappraisal and suppression. Several studies
using student samples have provided validation for the ERQ, although the only article (Wiltink et al.,
2011) that evaluated the ERQ in a community sample was unable to replicate the original factor structure.
Before using the ERQ in non-student samples, it is important to validate the scale in a sample broadly
representative of the adult population and to determine the influence of demographic variables. The
current study examined the psychometric properties of the ERQ in 2 community samples (Australia, N ⫽
550; United Kingdom, N ⫽ 483; 17–95 years of age) using confirmatory analysis. The original ERQ
factor structure was not supported by either the Australian or United Kingdom samples. However, with
the removal of 1 item, a strong model fit was obtained for both samples (9-item ERQ [ERQ-9]). Using
measurement invariance tests, the revised ERQ-9 was found to be equivalent across the samples and
demographics (age, gender, and education). Gender, depression, anxiety, and stress were the only factors
that were significantly associated with reappraisal and suppression use. Overall, the ERQ-9 provides
better fit of the data than the 10-item ERQ. The utility of this measure is enhanced by the provision of
normative data for males and females.
Emotion regulation is a key mechanism for our survival, and is Schweizer, 2010; Beevers & Meyer, 2004; Matheson & Anisman,
at the core of successful social interactions. Emotion regulation is 2003; Wenzlaff & Luxton, 2003), decreased quality of life and
a controlled process that is used to change a person’s spontaneous well-being (Gross & John, 2003), impaired memory (Nielson &
emotional response (Koole, 2009). Indeed, it is required to attain Lorber, 2009; Richards & Gross, 2000), and increased sympathetic
desired affective states and intentional goals (Gross, 2002). Dys- nervous system activity (Demaree et al., 2006; Gross, 1998a).
functional cognitive emotion regulation is linked to increased There are several ways people may modulate their emotions,
psychopathology and negative affect (Aldao, Nolen-Hoeksema, & commonly referred to as emotion regulation strategies (Gross &
John, 2003). Contemporary models propose that, at the broadest
level, emotion regulation strategies are distinguished by whether
they are “antecedent” or “response-focused” (Gross & John,
Devon L. Spaapen, School of Psychiatry and Clinical Neurosciences,
2003), referring to when these cognitive events occur along the
University of Western Australia, Perth, Western Australia, Australia; Fla- timeline of information processing. The use of antecedent and
vie Waters, School of Psychiatry and Clinical Neurosciences, University of response-focused strategies are considered to differ in their con-
Western Australia, Perth, Western Australia, Australia, and Clinical Re- sequences to health and well-being (Gross, 1998a). To examine
search Centre, Graylands Hospital, Perth, Western Australia, Australia; these differences, studies commonly compare two types of emo-
Laura Brummer and Lusia Stopa, School of Psychology, University of tion regulation strategies: cognitive reappraisal (antecedent) and
Southampton, Southampton, England; Romola S. Bucks, School of Psy- expressive suppression (response-focused; Gross & John, 2003).
chology, University of Western Australia, Perth, Western Australia, Aus- The Emotion Regulation Questionnaire (ERQ; Gross & John,
tralia.
2003) was originally designed to measure the habitual use of
We thank Mark A. Griffin and Patrick D. Dunlop for statistical analysis
guidance.
reappraisal and suppression strategies. It is now a widely accepted
Correspondence concerning this article should be addressed to Romola and commonly used measure of emotion regulation. ERQ studies
S. Bucks, M304, School of Psychology, University of Western Australia, have consistently found that habitual suppression is associated
35 Stirling Highway, Crawley WA 6009, Australia. E-mail: romola with increased depression, anxiety, and stress symptoms (negative
.bucks@uwa.edu.au affect), while habitual reappraisal is associated with greater posi-
1
2 SPAAPEN, WATERS, BRUMMER, STOPA, AND BUCKS
tive emotion and less negative emotion (Gross & John, 2003; (Gross et al., 1997; Helson & Klohnen, 1998). According to some
Joormann & Gotlib, 2010; Moore, Zoellner, & Mollenholt, 2008; theories (including the socioemotional selectivity theory [SST];
Wiltink et al., 2011). Compared to the use of suppression, reap- Charles & Carstensen, 2007), this may be due to the increased use
praisal is also associated with better interpersonal functioning, of antecedent strategy in older adults, specifically selective atten-
positive well-being (Gross & John, 2003), better social adjustment, tion, situation selection, and situation modification to modulate
and better decision making (Heilman, Crisan, & Houser, 2010; emotion. In conjunction with this theory, Gross (1998b) suggested
Magar, Phillips, & Hosie, 2008). The ERQ is therefore useful in that this increased use of antecedent strategies also includes cog-
identifying groups vulnerable to negative physical and psycholog- nitive reappraisal. This is consistent with results from Folkman,
ical outcomes, as well as factors with an important influence on Lazarus, Pimley, and Novacek (1987), who found that older par-
emotional well-being. ticipants reported greater positive reappraisal and distancing, and
When Gross and John (2003) first developed the 10-item ERQ less confrontational coping, compared to younger participants.
scale, a principal-components analysis (PCA) on the scores of These results, however, stand in contrast to Wiltink et al.’s (2011)
1,483 students (across four samples) revealed a two-factor solu- German study, in which age was not a significant predictor of
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
tion. Together, the two factors (represented by reappraisal and reappraisal or suppression use, after controlling for other demo-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
suppression) accounted for over 50% of the variance in all four graphic and clinical variables (e.g., gender, depression, and anxi-
samples. Further studies in student samples have replicated this ety).
two-factor solution using exploratory (Chen, 2010; D’Argembeau There is limited research into how education is related to emo-
& Van der Linden, 2006) and confirmatory factor analyses (CFAs; tion regulation strategy use. Wiltink et al. (2011) found that
Balzarotti, Gross, & John, 2010; Matsumoto, Yoo, & Nakagawa, individuals with a tertiary degree reported lower suppression
2008; Melka, Lancaster, Bryant, & Rodriguez, 2011). Few studies, scores (but not reappraisal) compared to those without. In contrast,
however, have examined the psychometric properties of the ERQ Moore et al. (2008) did not find any relationship between educa-
outside student samples. Two exceptions include Moore et al. tion and reappraisal or suppression. However Moore et al.’s sam-
(2008), who, in addition to a student sample (N ⫽ 289), recruited ple was largely homogenous with respect to education, possibly
a small (N ⫽ 67) all-female, trauma-exposed group, and Wiltink et explaining the lack of significant association. Further research
al. (2011), who tested a German sample using a German transla- using an independent community sample is needed to clarify this
tion of the ERQ (Abler & Kessler, 2009; N ⫽ 2,524, 55.5% relationship.
female). The association between the 10-item ERQ, gender differences,
Moore et al. (2008) found an overall good fit using the original and clinical symptoms has been examined previously. Studies
ERQ factor structure when combining their student and commu- show that males report a greater use of expressive suppression
nity samples. Their results however, may have been heavily compared to females (Balzarotti et al., 2010; Chen, 2010; Gross &
skewed by the inclusion of both a student sample (N ⫽ 289) and John, 2003; Melka et al., 2011; Wiltink et al., 2011). For reap-
a clinical group in the analysis. In contrast, Wiltink et al. (2011) praisal, no gender differences have been observed in any ERQ
failed to replicate the original ERQ factor structure. Instead, they validation studies, with the exception of Chen (2010), who found
proposed a modification of the scale in which Item 8 (“I control that female students use reappraisal more often than males.
my emotions by changing the way I think about the situation I’m As mentioned above, with regards to psychological symptoms,
in”) was allowed to load onto both reappraisal and suppression individuals who habitually use reappraisal tend to have lower
factors equally, and was freed from equality constraint with other levels of depression and anxiety (with greater positive functioning;
item factor loadings. This raises the question whether this devia- e.g., life satisfaction), whereas individuals who habitually use
tion in factor structure stems from cultural differences, the use of suppression show the reverse pattern with decreased positive func-
a German translation of the original scale, or demographic differ- tioning (Gross & John, 2003; Wiltink et al., 2011).
ences between student and community samples (e.g., age, educa- In summary, this study (a) examined the psychometric proper-
tion). These questions are important, because the ERQ is increas- ties of the ERQ in two separate community samples (Australia,
ingly used in research with non-undergraduate samples (Henry, N ⫽ 550; United Kingdom, N ⫽ 483; 17–95 years of age); (b)
Rendell, Green, McDonald, & O’Donnell, 2008; Perry, Henry, & conducted invariance tests by demographic variables; and (c) ex-
Grisham, 2011) when it has not yet been validated for such use. amined the relationships between the ERQ scales, demographics,
Establishing the psychometric properties of an instrument is an and clinical variables. We hypothesized that habitual reappraisal
important step in assessing the responsiveness of measurement would be associated with reduced psychological symptoms and
instruments, particularly for measures that are frequently used. would be more prevalent in older individuals. In contrast, we
The primary objectives of the current study were to test the predicted that suppression would be associated with increased
original two factor structure fit of the ERQ (Gross & John, 2003) psychological symptoms and would show greater prevalence in
on two large community samples (Australian and United King- male, younger, and less (formally) educated individuals.
dom) and to determine the influence of demographic variables.
The focus was on variables that are generally homogeneous in Method
student samples—age and education—as well as other variables
such as gender, depression, anxiety, and stress. Participants
There are mixed findings in the literature about the relationship
between emotion regulation strategies and demographic variables. Participants were drawn from Australian (AUS) and United
With regard to age, some studies show a link between aging and a Kingdom (UK) community samples (17–95 years of age). Sample
decrease in negative emotions and an increase in positive emotions characteristics are presented in Table 1. Questionnaires were dis-
ERQ-9 VALIDATION 3
tributed via broad community advertising that asked for volunteers Data Analysis Plan
to take part in an “Emotions Across the Lifespan” study. Volun-
AMOS 20 (Arbuckle, 2011) was used to conduct the CFA. The
teers were given the choice to complete the questionnaires online
current study’s multivariate critical ratio (22.97), which calculates
or on hardcopy (mailed out with a reply-paid envelope). This
overall non-normality across all items, exceeded Bentler’s (1990)
approach was considered acceptable due to evidence that online
recommended critical value of 5, suggesting that the data were not
and printed versions of questionnaires give similar results (Ritter, normally distributed. To address this issue, asymptotic distribution
Lorig, Laurent, & Matthews, 2004). Participation was voluntary, free (ADF) estimation was adopted for single group CFAs. The
and completion of the questionnaires was taken to indicate con- ERQ’s factor structure was examined using chi-square, the stan-
sent. This project was granted approval from the University of dardized root-mean-square residual (SRMR; Jöreskog & Sörbom,
Western Australia Human Research Ethics Committee (Project 1993), the comparative fit index (CFI; Bentler, 1990), the Tucker–
No. RA/4/3/1247) and from the University of Southampton School Lewis Index (TLI; Tucker & Lewis, 1973), and the root-mean-
of Psychology-Research Ethics Committee (Project No. ST/03/ square error of approximation (RMSEA; Steiger & Lind, 1980).
90). Chi-square non-significance, SRMR values below .08, TLI and
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
CFA values above .95, and RMSEA values below .06 are recom-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Table 1
Sample Characteristics
AUS UK Total
Item M SD Range M SD Range M SD Range
change in CFI (⌬CFI ⬍ .002) was below the cutoff (Meade, (metric and scalar invariance). Results are presented in Table 3. As
Johnson, & Braddy, 2008). Effect size was compared to Cohen’s strong, between-group invariance was established between the
(1988) guidelines (.20 ⫽ small, .50 ⫽ medium, .80 ⫽ large). It samples, it was considered viable to combine the samples for
should be noted, however, that these values are a rough measure of subsequent analyses.
practical significance, and thus that the results should be evaluated In summary, data from our AUS and UK community samples
based on the substantive context (Cohen, 1988). For analyses of did not adequately fit the original 10-item, two-factor structure,
the relationship between ERQ subscales and gender, a multivariate ERQ (Gross & John, 2003). However, a revised, nine-item ERQ
analysis of variance (MANOVA) was used. Correlation analysis was a good fit for both samples. These findings raise the question,
was used to examine the relationships between demographic vari- why do the current community samples differ from the many
ables (age and education), clinical variables (depression, anxiety, student samples previously used successfully to validate the orig-
and stress), and the ERQ subscales. Missing data by variable inal 10-item ERQ? One hypothesis is that variations in demo-
ranged from 0.2% to 3.7%. Little’s Missing Completely at Ran- graphic variables (e.g., age, education, and gender) between stu-
dom (MCAR) test revealed that the data were missing completely dent and community samples influence the model.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 2
Single Group CFA—Australian and United Kingdom Samples
Table 3
Multi-Group Analysis (Nine-Item)—Invariance Test Between Australian and United Kingdom Community Samples
Unconstrained 56 107.308 ⬍.001 2.064 .030 .973 .980 .032 .023 .041
Equal factor loading 49 113.897 ⬍.001 1.930 .031 .976 .980 .030 .022 .038
Equal means/intercepts 40 128.041 ⬍.001 1.883 .031 .977 .979 .029 .021 .037
Equal covariance 37 134.961 ⬍.001 1.901 .037 .977 .977† .030 .022 .037
Equal error 28 174.793 ⬍.001 2.185 .038 .970 .966† .034 .027 .041
Note. Unconstrained ⫽ the standard nine-item model; SRMR ⫽ standardized root-mean-square residual; TLI ⫽ Tucker–Lewis Index; CFI ⫽ comparative
fit index; RMSEA ⫽ root-mean-square error of approximation; LO 90 and HI 90 ⫽ lower and upper boundary of 90% confidence interval for RMSEA.
Maximum likelihood estimation was used because it is a more thorough method to examine the extent of invariance (compared to asymptotic distribution
free).
†
⌬CFI ⬎ .002.
6 SPAAPEN, WATERS, BRUMMER, STOPA, AND BUCKS
Table 4
Multi-Group Analysis (Nine-Item ERQ)—Invariance Test for Age, Education, and Gender With the Combined Australian and United
Kingdom Community Sample
Age
Unconstrained 52 121.913 ⬍.001 2.344 .032 .965 .975 .036 .028 .045
Equal factor loadings 59 133.145 ⬍.001 2.257 .037 .968 .973 .035 .027 .043
Equal means/intercepts 68 157.736 ⬍.001 2.320 .036 .966 .968 .036 .029 .043
Equal covariance 71 160.223 ⬍.001 2.257 .040 .968 .968 .035 .028 .042
Equal error 80 174.617 ⬍.001 2.183 .039 .970 .966 .034 .027 .041
Education
Unconstrained 52 122.978 ⬍.001 2.365 .038 .965 .974 .037 .029 .045
Equal factor loadings 59 139.7314 ⬍.001 2.361 .042 .965 .971 .037 .029 .045
Equal means/intercepts 68 168.728 ⬍.001 2.481 .042 .962 .964 .038 .031 .046
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Equal covariance 71 182.901 ⬍.001 2.576 .049 .959 .960 .040 .033 .047
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Equal error 80 210.497 ⬍.001 2.631 .044 .958 .953 .040 .034 .047
Gender
Unconstrained 52 121.109 ⬍.001 2.329 .049 .965 .974 .036 .028 .045
Equal factor loadings 59 130.884 ⬍.001 2.218 .053 .968 .973 .035 .027 .043
Equal means/intercepts 68 165.679 ⬍.001 2.436 .053 .962 .964 .038 .030 .045
Equal covariance 71 169.908 ⬍.001 2.393 .058 .963 .963 .037 .030 .044
Equal error 80 201.935 ⬍.001 2.524 .057 .959 .955 .039 .032 .046
Note. ERQ ⫽ Emotion Regulation Questionnaire; SRMR ⫽ standardized root-mean-square residual; TLI ⫽ Tucker–Lewis Index; CFI ⫽ comparative fit
index; RMSEA ⫽ root-mean-square error of approximation; LO 90 and HI 90 ⫽ lower and upper boundary of 90% confidence interval for RMSEA.
Maximum likelihood was used because it is a more thorough method to examine the extent of invariance (compared to asymptotic distribution free).
education, and gender differences were present in reappraisal and current results, however, are consistent with Wiltink et al. (2011),
suppression assessed with the nine-item ERQ. who also found no age differences.
Contrary to expectation, no significant age differences were Similarly to age, and again not as predicted, there were no
observed for reappraisal or suppression strategy use. This stands in differences in reappraisal and suppression use based on level of
contrast to the literature that suggests that older people shift toward education. These findings are consistent with Moore et al. (2008),
using more antecedent strategies (including reappraisal) than their in a student sample using the 10-item ERQ, but are divergent from
younger counterparts (Charles & Carstensen, 2007; Folkman, Wiltink et al.’s (2011) community study, which indicated that
Lazarus, Pimley, & Novacek, 1987; Gross et al., 1997). Differ- people with a college/university (tertiary) degree reported lower
ences between studies may be due to differences in methodologies, suppression scores (but not reappraisal) compared to those with-
sample characteristics, or methods of analysis. For example, the out. Again, translation effects may account for this.
differences between the current study and that of Folkman et al.
Gender Differences
(1987) could be due to a different measure of reappraisal (a
one-item measure in Folkman et al.’s, 1987, study) or sample Males reported greater use of suppression compared to females:
characteristics (only married couples of middle to older age). The a finding consistent with previous studies (Balzarotti et al., 2010;
Table 5
Correlations Between ERQ, Demographics, and Negative Affect—Combined Sample
Reappraisal
Measure Education (5-item) Reappraisal Suppression Depression Anxiety Stress
ⴱⴱ ⴱⴱ ⴱⴱ
Age ⫺.149 .054 .043 ⫺.014 ⫺.223 ⫺.260 ⫺.241ⴱⴱ
Education — .023 .021 ⫺.109ⴱⴱ ⫺.061 ⫺.053 .022
Reappraisal (5-item) — .973ⴱⴱ ⫺.098ⴱⴱ ⫺.136ⴱⴱ ⫺.087ⴱⴱ ⫺.086ⴱⴱ
Reappraisal — ⫺.091ⴱⴱ ⫺.126ⴱⴱ ⫺.069ⴱ ⫺.065ⴱ
Suppression — .256ⴱⴱ .157ⴱⴱ .118ⴱⴱ
Depression — .538ⴱⴱ .611ⴱⴱ
Anxiety — .617ⴱⴱ
Note. Values reflect Spearman’s rho correlation coefficients. ERQ ⫽ Emotion Regulation Questionnaire; Education ⫽ total years of education;
Depression, Anxiety, and Stress were derived from the Depression, Anxiety, and Stress Scale—21; Reappraisal (five items) and Suppression (four items)
were derived from the nine-item ERQ.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01 (two-tailed).
ERQ-9 VALIDATION 7
10 1.1 2.2 36.3 28.0 establishing test–retest reliability, predictive validity, and concur-
11 1.6 3.2 44.6 32.6
This document is copyrighted by the American Psychological Association or one of its allied publishers.
DiStefano, C., & Motl, W. R. (2006). Further investigating method effects (DASS) with the Beck Depression and Anxiety Inventories. Behaviour
associated with negatively worded items on self-report surveys. Struc- Research and Therapy, 33, 335–343. doi:10.1016/0005-
tural Equation Modeling: A Multidisciplinary Journal, 13, 440 – 464. 7967(94)00075-U
doi:10.1207/s15328007sem1303_6 Magar, E. C., Phillips, L. H., & Hosie, J. A. (2008). Self-regulation and
Folkman, S., Lazarus, R. S., Pimley, S., & Novacek, J. (1987). Age risk-taking. Personality and Individual Differences, 45, 153–159. doi:
differences in stress and coping processes. Psychology and Aging, 2, 10.1016/j.paid.2008.03.014
171–184. doi:10.1037/0882-7974.2.2.171 Matheson, K., & Anisman, H. (2003). Systems of coping associated with
Gross, J. J. (1998a). Antecedent- and response-focused emotion regulation: dysphoria, anxiety and depressive illness: A multivariate profile per-
Divergent consequences for experience, expression, and physiology. spective. Stress, 6, 223–234. doi:10.1080/10253890310001594487
Journal of Personality and Social Psychology, 74, 224 –237. doi: Matsumoto, D., Yoo, S. H., & Nakagawa, S. (2008). Culture, emotion
10.1037/0022-3514.74.1.224 regulation, and adjustment. Journal of Personality and Social Psychol-
Gross, J. J. (1998b). The emerging field of emotion regulation: An inte- ogy, 94, 925–937. doi:10.1037/0022-3514.94.6.925
grative review. Review of General Psychology, 2, 271–299. doi:10.1037/ Meade, A. W., Johnson, E. C., & Braddy, P. W. (2008). Power and
1089-2680.2.3.271 sensitivity of alternative fit indices in tests of measurement invariance.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Gross, J. J. (2002). Emotion regulation: Affective, cognitive, and social Journal of Applied Psychology, 93, 568 –592. doi:10.1037/0021-9010
consequences. Psychophysiology, 39, 281–291. doi:10.1017/ .93.3.568
This document is copyrighted by the American Psychological Association or one of its allied publishers.
S0048577201393198 Melka, S. E., Lancaster, S. L., Bryant, A. R., & Rodriguez, B. F. (2011).
Gross, J. J., Carstensen, L. L., Pasupathi, M., Skorpen, C. G., Tsai, J., & Confirmatory factor and measurement invariance analyses of the Emo-
Hsu, A. Y. (1997). Emotion and aging: Experience, expression, and tion Regulation Questionnaire. Journal of Clinical Psychology, 67,
control. Psychology and Aging, 12, 590 –599. doi:10.1037/0882-7974 1283–1293. doi:10.1002/jclp.20836
.12.4.590 Moore, S. A., Zoellner, L. A., & Mollenholt, N. (2008). Are expressive
Gross, J. J., & John, O. P. (2003). Individual differences in two emotion suppression and cognitive reappraisal associated with stress-related
regulation processes: Implications for affect, relationships, and well- symptoms? Behaviour Research and Therapy, 46, 993–1000. doi:
being. Journal of Personality and Social Psychology, 85, 348 –362. 10.1016/j.brat.2008.05.001
doi:10.1037/0022-3514.85.2.348 Nielson, K. A., & Lorber, W. (2009). Enhanced post-learning memory
Heilman, R. M., Crisan, L. G., & Houser, D. (2010). Emotion regulation consolidation is influenced by arousal predisposition and emotion reg-
and decision making under risk and uncertainty. Emotion, 10, 257–265. ulation but not by stimulus valence or arousal. Neurobiology of Learning
doi:10.1037/a0018489 and Memory, 92, 70 –79. doi:10.1016/j.nlm.2009.03.002
Helson, R., & Klohnen, E. C. (1998). Affective coloring of personality Nye, C. D., & Drasgow, F. (2011). Effect size indices for analysis of
from young adulthood to midlife. Personality and Social Psychology measurement equivalence: Understanding the practical importance of
Bulletin, 24, 241–252. doi:10.1177/0146167298243002 differences between groups. Journal of Applied Psychology, 96, 966 –
Henry, J. D., & Crawford, J. R. (2005). The short-form version of the 980. doi:10.1037/a0022955
Depression Anxiety Stress Scales (DASS-21): Construct validity and Perry, Y., Henry, J. D., & Grisham, J. R. (2011). The habitual use of
normative data in a large non-clinical sample. The British Journal of emotion regulation strategies in schizophrenia. British Journal of Clin-
Clinical Psychology, 44, 227–239. ical Psychology, 50, 217–222. doi:10.1111/j.2044-8260.2010.02001.x
Henry, J. D., Rendell, P. G., Green, M. J., McDonald, S., & O’Donnell, M. Richards, J. M., & Gross, J. J. (2000). The cognitive costs of keeping one’s
(2008). Emotion regulation in schizophrenia: Affective, social, and cool. Journal of Personality and Social Psychology, 79, 410 – 424.
clinical correlates of suppression and reappraisal. Journal of Abnormal doi:10.1037/0022-3514.79.3.410
Psychology, 117, 473– 478. doi:10.1037/0021-843X.117.2.473 Ritter, P., Lorig, K., Laurent, D., & Matthews, K. (2004). Internet versus
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance mailed questionnaires: A randomized comparison. Journal of Medical
structure analysis: Conventional criteria versus new alternatives. Struc- Internet Research, 6, e29. doi:10.2196/jmir.6.3.e29
tural Equation Modeling: A Multidisciplinary Journal, 6, 1–55. doi: Steiger, J. H., & Lind, J. C. (1980). Statistically based tests for the number
10.1080/10705519909540118 of factors. Paper presented at the annual meeting of the Psychometric
Joormann, J., & Gotlib, I. H. (2010). Emotion regulation in depression: Society, Iowa City, IA.
Relation to cognitive inhibition. Cognition & Emotion, 24, 281–298. Tucker, J. S., & Lewis, C. (1973). The reliability coefficient for maximum
doi:10.1080/02699930903407948 likelihood factor analysis. Psychometrika, 38, 1–10. doi:10.1007/
Jöreskog, K., & Sörbom, D. (1993). LISREL 8: Structural equation mod- BF02291170
eling with the SIMPLIS command language. Chicago, IL: Scientific Wenzlaff, R. M, & Luxton, D. D. (2003). The role of thought suppression
Software International. in depressive rumination. Cognitive Therapy and Research, 27, 293–
Koole, S. L. (2009). The psychology of emotion regulation: An integrative 308. doi:10.1023/A:1023966400540
review. Cognition & Emotion, 23, 4 – 41. doi:10.1080/ Wiltink, J., Glaesmer, H., Canterino, M., Wolfling, K., Knebel, A., Kessler,
02699930802619031 H., . . . Buetel, M. E. (2011). Regulation of emotions in the community:
Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative Suppression and reappraisal strategies and its psychometric properties.
emotional states: Comparison of the Depression Anxiety Stress Scales Psycho-Social Medicine, 8, 1–12. doi:10.3205/psm000078
(Appendix follows)
ERQ-9 VALIDATION 9
Appendix
Revised Nine-Item Emotion Regulation Questionnaire
Scale Items
Reappraisal 1. When I want to feel more positive (such as joy or amusement), I change what I’m thinking about.
5. When I’m faced with a stressful situation, I make myself think about it in a way that helps me calm down.
7. When I want to feel more positive emotion, I change the way I’m thinking about the situation.
8. I control my emotions by changing the way I think about the situation I’m in.
10. When I want to feel less negative emotion, I change the way I’m thinking about the situation.
Note. Adapted from “Individual Differences in Two Emotion Regulation Processes: Implications for Affect, Relationships, and Well-Being,” by J. J.
Gross and O. P. John, 2003, Journal of Personality and Social Psychology, 85, p. 351. Copyright 2003 by the American Psychological Association.