ORIGINAL RESEARCH

published: 28 February 2019

doi: 10.3389/fpsyg.2019.00435

An Investigation of the Psychometric

Properties of the Chinese Trait
Emotional Intelligence Questionnaire
Short Form (Chinese TEIQue-SF)
Anita Feher 1* , Gonggu Yan 2 , Donald H. Saklofske 1 , Rachel A. Plouffe 1 and Yan Gao 3
1
Department of Psychology, The University of Western Ontario, London, ON, Canada, 2 Faculty of Psychology, Beijing
Normal University, Beijing, China, 3 School of Education, Beijing Normal University, Zhuhai, China

The present study examined the psychometric properties of the Chinese version of
the Trait Emotional Intelligence Questionnaire Short Form (TEIQue-SF). Analyses were
performed using a sample of undergraduates (N = 585) recruited from four universities
across China. Confirmatory factor analysis of the Chinese TEIQue-SF supported the
one-factor structure of trait emotional intelligence. Measurement invariance analyses
were conducted across the Chinese sample and a sample of Canadian undergraduate
students (N = 638). Although the two samples demonstrated configural and partial
Edited by: metric invariance, scalar invariance was not found. Cross-cultural implications and
Juan-Carlos Pérez-González,
National University of Distance
explanations of the present findings, as well as suggestions for future research
Education (UNED), Spain are discussed.
Reviewed by:
Keywords: trait emotional intelligence, TEIQue-SF, cross-cultural, confirmatory factor analysis,
Heribert Harald Freudenthaler,
measurement invariance
University of Graz, Austria
Wolfgang Schoppek,
University of Bayreuth, Germany
INTRODUCTION
*Correspondence:
Anita Feher
Emotional intelligence (EI) has garnered considerable research interest since its introduction to the
[email protected]
research literature by Salovey and Mayer (1990). EI can be broadly conceived of as an individual
Specialty section:
differences variable describing emotion related distinctions (Petrides et al., 2007a; Petrides et al.,
This article was submitted to 2018). One commonly-cited description of EI, referred to as trait EI (also called trait emotional
Emotion Science, self-efficacy), defines it as a cluster of behavioral dispositions and self-perceptions related to one’s
a section of the journal emotions, positioned at the lower levels of personality hierarchies (Petrides and Furnham, 2001;
Frontiers in Psychology Petrides et al., 2007b, 2016; Petrides, 2011). Trait EI is evaluated using self-report questionnaires
Received: 16 July 2018 (Petrides and Furnham, 2001), and a variety of psychometric measures have been developed to
Accepted: 13 February 2019 assess it (e.g., Bar-On, 1997; Schutte et al., 1998). One prominent measure is the Trait Emotional
Published: 28 February 2019 Intelligence Questionnaire (TEIQue; Petrides, 2009), which is the focus of the present study.
Citation: The TEIQue and its short form, the Trait Emotional Intelligence Questionnaire-Short Form
Feher A, Yan G, Saklofske DH, (TEIQue-SF), were constructed to adequately cover trait EI’s sampling domain in line with trait
Plouffe RA and Gao Y (2019) An EI theory (Pérez et al., 2005; Petrides, 2009, 2011). It operationalizes EI in accordance with the
Investigation of the Psychometric
subjective nature and reporting of emotional experiences, and as a personality trait (Petrides
Properties of the Chinese Trait
Emotional Intelligence Questionnaire
et al., 2007b; Petrides, 2009, 2011). The measure comprises 15 facets which form four correlated
Short Form (Chinese TEIQue-SF). factors, and can be further grouped to produce a global trait EI score (Petrides, 2009). Petrides
Front. Psychol. 10:435. (2009) defines these four factors as: Well-Being (well-being related feelings across time based
doi: 10.3389/fpsyg.2019.00435 around achievements, self-regard, and expectations), Self-Control (regulating and having control

Frontiers in Psychology | www.frontiersin.org 1 February 2019 | Volume 10 | Article 435

Feher et al. Chinese Trait Emotional Intelligence Questionnaire

over emotions, impulses, and stress), Emotionality (ability to version of the TEIQue in Chinese samples demonstrate partial
perceive, express, and connect with emotions in self and replication of the factors. Mavroveli and Siu (2012) found a three-
others, which can be used in creating successful interpersonal factor solution for the adolescent TEIQue, with sociability and
relationships), and Sociability (being socially assertive and aware, emotionality combined to form a single factor. Gökçen et al.
managing others’ emotions, and effectiveness in communication (2014) found a four-factor solution for the TEIQue. However,
and participation in social situations). some of the facets did not load onto factors as expected.
Studies in Western countries utilizing the TEIQue have found There is mixed evidence in the research literature
numerous positive associations with global trait EI amongst regarding cross-cultural comparisons of trait EI levels in
undergraduate students, including higher resiliency (e.g., Vesely more individualistic versus collectivistic countries (e.g., LaPalme
et al., 2014), greater proclivity to use adaptive coping strategies et al., 2016). For example, Gökçen et al. (2014) found that
(e.g., Mikolajczak et al., 2008), and a positive relationship British participants scored higher on global trait EI and on the
with academic performance (e.g., Sanchez-Ruiz et al., 2013; four factors associated with the TEIQue compared to Chinese
Petrides et al., 2018). participants. Studies using other measures of trait EI have
similarly found participants from a more individualistic country
obtaining higher trait EI scores (Koydemir et al., 2013; Nozaki
Cross-Cultural Differences in Trait and Koyasu, 2016). However, another study using the TEIQue-SF
Emotional Intelligence found that the more collectivistic Cape Verdeans scored higher
Cultures vary on multiple dimensions, including individualistic on global trait EI compared to Portuguese participants (Wilks
versus collectivistic societies (Hofstede, 1980, 2001; Triandis, et al., 2015). Using a different trait EI measure, a study comparing
1995). Individualistic societies (e.g., North American) tend United States and Taiwanese academic leaders showed no
toward looser connections between people with more personal significant differences in total EI, though some differences
independence from collectives, and people are more influenced existed on subcomponent measures (Tang et al., 2010).
by personal motivations and goals (Hofstede, 1980; Triandis,
1995; Suh et al., 1998; Hofstede et al., 2010). Collectivistic Present Study
societies (e.g., China) are characterized by tighter connections Trait EI was largely defined and developed in Western contexts,
between people and strong identification with one or more thus raising a need to examine the construct within non-Western
collective in-groups, and valuing and prioritizing the norms and samples (e.g., Gangopadhyay and Mandal, 2008). The present
goals of these groups often over personal goals (Hofstede, 1980; study aims to do so by examining the factor structure of the
Triandis, 1995; Suh et al., 1998; Hofstede et al., 2010). frequently-used trait EI measure, the TEIQue-SF, in a sample
Evidence suggests that emotions and latent personality traits of Chinese undergraduate students. Although the long-form
manifest differently across cultural environments. For example, version of the TEIQue has been previously examined in a Chinese
in line with collectivistic values of preserving interpersonal sample (e.g., Gökçen et al., 2014), to our knowledge, no previous
harmony, Siu and Chang (2011) reported that a Chinese sample study has undertaken this task using the short-form version of
was likely to control feelings of stress related to close relationships a Chinese-translated TEIQue. Short-form scales make important
using avoidance or detachment. Cross-cultural differences have contributions and provide practical benefits to psychological
been found in norms surrounding expression of emotions, research. For example, they prevent participant disengagement
called display rules (Ekman and Friesen, 1969; Caruso, 2008). and are useful for studies including several questionnaires or
Matsumoto (1990) found that negative emotional displays among repeated applications of the same questionnaire over multiple
ingroup members and positive emotions toward outgroups were sessions (Austin et al., 2018). In line with findings of good
considered more appropriate in an American sample compared model fit of the TEIQue-SF factor structure in a Spanish sample
to a Japanese sample. Cross-cultural comparisons of personality (Laborde et al., 2016) and factor analytic findings for the long
traits have found positive associations between individualism version of the TEIQue in a Chinese sample (e.g., Gökçen et al.,
and extraversion, with higher levels of extraversion amongst 2014), we expect to find good model fit for the factor structure
participants from individualistic compared to collectivistic using a Chinese translation of the TEIQue-SF.
countries (Furnham and Cheng, 1999; McCrae, 2001, 2002; Cultural variations in trait EI will also be explored by
Hofstede and McCrae, 2004). assessing the cross-cultural replicability of the TEIQue-SF. Cross-
Measures developed to assess trait EI tend to be factorially cultural measurement invariance will be used to compare the
robust when assessed in different countries and cultures. more collectivistic Chinese sample with a more individualistic
The factor structure defined by trait EI measures other than Canadian sample. Measurement invariance is a means of
the TEIQue show reasonable replicability across collectivistic assessing the psychometric equivalence of a construct (i.e., trait
countries such as China (Li et al., 2012; Kong, 2017), Japan EI as assessed by the TEIQue-SF) across different groups (Putnick
(Fukuda et al., 2011) and Korea (Fukuda et al., 2012). Studies that and Bornstein, 2016). When a construct is invariant across
examine the factor structure of the long version of the TEIQue groups, it indicates that the different groups are attributing
have generally found that the factor structure is replicated the same meaning to that construct (Putnick and Bornstein,
(apart from some minor deviations) in other countries (e.g., 2016). Establishing cross-cultural invariance is important for
Mikolajczak et al., 2007; Freudenthaler et al., 2008; Martskvishvili comparisons across cultures on some construct (Mullen, 1995;
et al., 2013; Aluja et al., 2016). Studies examining the long Libbrecht et al., 2014; LaPalme et al., 2016). While factorial

Frontiers in Psychology | www.frontiersin.org 2 February 2019 | Volume 10 | Article 435

Feher et al. Chinese Trait Emotional Intelligence Questionnaire

equivalence can be demonstrated within multiple cultures, it factors: Well-Being, Self-Control, Emotionality, and Sociability.
does not ensure measurement invariance of the measured As indicated by the TEIQue-SF scoring key, obtained from
construct across cultures (Byrne and Campbell, 1999; Byrne Petrides’ university laboratory website, items 3, 14, 18, and
and Watkins, 2003). Therefore, a separate examination of 29 only contribute to global trait EI, and not to any of the
measurement invariance of the TEIQue-SF is warranted. While four factors. Therefore, these items were only used to calculate
several studies have examined the measurement invariance of global trait EI scores. Participants responded to items using a 7-
other trait EI measures (e.g., Li et al., 2012), to our knowledge, point Likert scale ranging from 1(completely disagree or strongly
no study has previously examined the cultural invariance of the disagree) to 7(completely agree or strongly agree).
TEIQue-SF in a Chinese and Canadian sample. For the Chinese sample, the TEIQue-SF was translated into
One comparative study using the long version of the TEIQue Mandarin following the recommended steps by Hambleton and
found that more individualistic participants scored higher on Lee (2013). Care was taken to maintain content and lexical
global and factor measures of trait EI (Gökçen et al., 2014). equivalence. The translation was done by faculty and graduate
However, other studies making comparisons using participants students who were native Chinese speakers as well as being
from different countries or using different trait EI measures highly fluent in English. Back translations were performed to
have found divergent results (e.g., Wilks et al., 2015). Based on ensure equivalence of the meaning for each item. Specifically,
inconsistencies in previous findings, the present study makes no two psychology graduate students independently translated the
specific hypotheses regarding individualistic-collectivistic group English version into Chinese, and the two translations were
comparisons using the TEIQue-SF. examined for differences. Any differences were discussed, and
final decisions for the scale were made by one of the Chinese
authors of this paper. The Chinese version of the TEIQue-
MATERIALS AND METHODS SF was then translated back into English with the aid of a
professor at a Chinese university who has taught English for over
Participants 30 years. The translation process confirmed proper translation of
The present study included samples of Chinese and Canadian the TEIQue-SF into Mandarin, allowing for distribution of the
undergraduate university students. The Chinese sample (N = 585, TEIQue-SF to Mandarin-speaking participants. The Canadian
89 men, 447 women, and 49 unreported) was recruited from participants completed the English version of the TEIQue-SF as
four Chinese universities. Their ages ranged from 16 to 26 years published by Petrides (2009).
(M = 19.53, SD = 1.01). The Canadian sample (N = 638, 181
males and 456 females, 1 unreported) were recruited from a Data Analytic Strategy
large Canadian university. Their ages ranged from 17 to 43 years The goal of the present study was twofold. The primary aim was
(M = 18.50; SD = 2.14). to assess the factor structure of the TEIQue-SF in the Chinese
The present study followed the ethical guidelines required sample, and the secondary aim was to evaluate cultural variations
by the Canadian and Chinese universities, respectively. Written of the TEIQue-SF using cultural measurement invariance and
informed consent was given by participants in the Chinese mean comparisons. The factor structure for the TEIQue-SF
sample, and explicit informed consent was given by the used by Laborde et al. (2016) and Merino-Tejedor et al. (2018)
Canadian subjects using an online format prior to proceeding to was utilized in the present study. A one-factor model with the
questionnaires. All subjects gave informed consent in accordance four trait EI factors (represented as indicators) loading onto
with the Declaration of Helsinki. There were no formal ethics global trait EI was tested using confirmatory factor analysis
board requirements at the Chinese universities for survey (CFA) in Mplus Version 7.4 (Muthén and Muthén, 1998-
method studies, and rather this was handled internally within 2015). When evaluating model fit, Root Mean Square Error
the department. Ethics approval for the Chinese sample was of Approximation (RMSEA) values of 0.05 or below were
therefore not required as per the Beijing Normal University’s considered good fit, values between 0.05 and 0.08 acceptable fit,
guidelines and national regulations. With respect to the Canadian and values between 0.08 and 0.10 were considered indicative
sample, ethical approval was given by the non-medical Western’s of mediocre fit (Browne and Cudeck, 1993; MacCallum et al.,
Research Ethics Board at the University of Western Ontario. 1996; Hu and Bentler, 1998). In line with Hu and Bentler’s
(1999) suggestions, cut-off values close to 0.95 were considered
Measures demonstrative of good fit concerning Comparative Fit Index
For the Chinese sample, the trait EI data were obtained from (CFI) and Tucker Lewis Index (TLI) values, and values below
a larger resiliency study (Wilson et al., 2018). The trait EI data 0.08 were considered good fit regarding Standardized Root Mean
for the Canadian sample was drawn from a larger personality Square Residuals (SRMR).
study (Plouffe et al., in press). The trait EI measure has not been To assess cultural invariance of the TEIQue-SF, a series of
previously examined for either of these samples. CFA models were tested in hierarchal order using maximum
likelihood estimation in Mplus Version 7.4 (Muthén and
Trait Emotional Intelligence Muthén, 1998-2015). Configural invariance was assessed first in
The Trait Emotional Intelligence Questionnaire-Short Form order to determine whether the basic organization of trait EI
(TEIQue-SF; Petrides, 2009) is a 30-item measure that evaluates assessed with the TEIQue-SF (i.e., four trait EI factors well-
global trait EI, though it can also be used to assess the four trait EI being, self-control, emotionality, and sociability represented as

Frontiers in Psychology | www.frontiersin.org 3 February 2019 | Volume 10 | Article 435

Feher et al. Chinese Trait Emotional Intelligence Questionnaire

indicators loading onto a global trait EI latent factor) is found in to large (α = 0.85) in the Canadian sample. For both samples,
both cultures (Putnick and Bornstein, 2016). Metric invariance the correlations between the factor and global TEIQue-SF scores
was assessed next to ascertain whether there is equivalence were significant and positively related.
of factor loadings (i.e., equivalence in how the four trait EI
factors load onto the global latent trait EI factor) in both
cultural groups. Finally, scalar invariance was investigated to Confirmatory Factor Analysis of the
determine whether there is equivalence of intercepts. Scalar Chinese TEIQue-SF
invariance establishes the connection between observed and A model with the four trait EI indicators (i.e., Well-Being,
latent score findings, such that equal values on latent trait EI Self-Control, Emotionality, Sociability) loading onto one global
will result in the same values on the observed trait EI factors trait EI factor was tested in the present study using CFA with
for both Canadian and Chinese individuals (see Milfont and maximum likelihood robust estimation. When all fit indices
Fischer, 2010). If scalar invariance is not found, a comparison of were considered, the fit for the one-factor model was acceptable:
latent mean scores between groups may not be meaningful. For χ2 (2) = 12.188, RMSEA = 0.096 (90% CI = 0.049–0.151),
example, although the two countries might differ on the observed CFI = 0.980, TLI = 0.939, SRMR = 0.024. The standardized
Sociability factor, this may not be meaningfully associated with factor loadings were generally strong, ranging from 0.58 to
differences between countries in levels of latent trait EI if scalar 0.79 (see Figure 1).
invariance is not satisfied (Putnick and Bornstein, 2016). To
compare the configural, metric and scalar invariance models, χ2
(at p = 0.01 significance level), CFI, and RMSEA difference tests
were utilized. For these tests, 1CFI values less than or equal
to 0.01 in size, and 1RMSEA values less than 0.015 in size,
were utilized to indicate invariant models (Cheung and Rensvold,
2002; Chen, 2007).

RESULTS
Descriptive Statistics and Bivariate
Correlations
Descriptive statistics, Cronbach’s alphas, and bivariate
correlations for the TEIQue-SF factor scores and total scores are
presented for both the Chinese and Canadian data in Table 1.
Means for both the Chinese and Canadian groups are similar
to those found in previous studies (e.g., Herodotou et al., 2011;
Laborde et al., 2016). Across both samples, the alpha coefficient
FIGURE 1 | Confirmatory factor analysis of the Chinese TEIQue-SF (N = 554).
for global trait EI was large (α = 0.88). However, the values
glob_ei, global trait EI; wb, well-being; sc, self-control; emo, emotionality; soc,
for the trait EI factors ranged from small (α = 0.47) to large sociability.
(α = 0.82) in the Chinese sample, and from acceptable (α = 0.67)

TABLE 1 | Descriptive Statistics, Coefficient Alphas, and Correlations for Chinese and English Versions of the TEIQue-SF.

Variable Mean SD α 1 2 3 4 5

Chinese Sample
1 Global trait EI 4.73 0.64 0.88 1.00
2 Well-being 5.10 0.96 0.82 0.83∗ 1.00
3 Self-control 4.53 0.80 0.65 0.81∗ 0.64∗ 1.00
4 Emotionality 4.87 0.74 0.65 0.80∗ 0.55∗ 0.52∗ 1.00
5 Sociability 4.32 0.68 0.47 0.70∗ 0.42∗ 0.44∗ 0.48∗ 1.00
Canadian Sample
1 Global trait EI 4.73 0.69 0.88 1.00
2 Well-being 5.17 1.04 0.85 0.84∗ 1.00
3 Self-control 4.19 0.91 0.67 0.68∗ 0.50∗ 1.00
4 Emotionality 4.79 0.85 0.67 0.73∗ 0.48∗ 0.28∗ 1.00
5 Sociability 4.77 0.87 0.71 0.70∗ 0.48∗ 0.32∗ 0.40∗ 1.00

Listwise deletion used in correlation analysis for Chinese (N = 548) and Canadian (N = 633) data. ∗ p < 0.01.

Frontiers in Psychology | www.frontiersin.org 4 February 2019 | Volume 10 | Article 435

Feher et al. Chinese Trait Emotional Intelligence Questionnaire

Measurement Invariance Across DISCUSSION

Cultures
Overall, the present study had two main purposes: (1) to
To make cross-cultural comparisons and establish the
determine whether the factor structure of the TEIQue-SF
generalizability of the TEIQue-SF, invariance of the factor
(Petrides, 2009) was upheld in a non-Western context, and (2) to
structure of the TEIQue-SF must be established (Reise et al.,
evaluate whether cross-cultural differences exist in the conceptual
2000; Widaman and Reise, 1997) Therefore, nested CFA
interpretation of the TEIQue-SF. Results of the present study
models were compared using both the Chinese and Canadian
demonstrated that while the factor structure of the TEIQue-
samples to determine whether measurement invariance of
SF was replicated, there were differences in the meaning of
the TEIQue-SF was found across cultures (see Table 2).
latent trait EI when assessed in a more collectivistic context in
According to the CFI index, the configural model demonstrated
comparison to a more individualistic sample.
acceptable fit χ2 (4) = 28.24; p < 0.001, RMSEA = 0.100
The present study examined the robustness of the factor
(90% confidence interval [CI] = 0.067–0.137), CFI = 0.981.
structure of the TEIQue-SF (Petrides, 2009) in a Chinese
Examination of the chi-squared change index showed that
undergraduate sample. Trait EI has been primarily defined in a
there was a significant difference between the metric model
Western context, therefore requiring the need for the construct
with constrained factor loadings and the configural model,
to be psychometrically validated in non-Western samples (e.g.,
1χ2 (3) = 19.08, p < 0.01, though large sample sizes may have
Gangopadhyay and Mandal, 2008). This study aimed to do so by
been responsible for this influence (Cheung and Rensvold,
examining the fit of a one-factor model (i.e., global trait EI and
2002). Although examinations of RMSEA change index
Petrides’ four trait EI factors) using the TEIQue-SF.
showed non-significant differences between the models,
Results of the CFA demonstrated that the Mandarin
1RMSEA = −0.002, the CFI change index also showed
translation of the TEIQue-SF had acceptable fit. Previous studies
significant differences between the models, 1CFI = −0.013.
have cross-culturally replicated the factor structure of the TEIQue
To test for partial metric invariance, an examination of the
(e.g., Freudenthaler et al., 2008). While there are fewer studies
modification indices led to the decision to free the self-
similarly examining the factor structure of the TEIQue-SF,
control factor. Subsequent chi square, RMSEA, and CFI
Laborde et al. (2016) and Merino-Tejedor et al. (2018) found
difference tests were non-significant, providing support for
evidence of good model fit for the TEIQue-SF in their Spanish
partial metric invariance, 1χ2 (2) = 7.80, 1RMSEA = −0.009,
samples. Additional studies that have investigated the factor
and 1CFI = −0.005. Scalar invariance was tested only on
structure of the TEIQue-SF have chosen to include items or facets
metric-invariant loadings; therefore, Self-Control was left
in their CFA models (e.g., Jacobs et al., 2015; Snowden et al.,
free to vary (Putnick and Bornstein, 2016). Findings of chi
2015). The current study did not, following on Petrides (2009)
square, RMSEA, and CFI difference tests did not support
suggestion that the TEIQue-SF was not designed to be factor
scalar invariance across cultural groups, 1χ2 (2) = 125.82,
analyzed at the item level or scored at the facet level.
p < 0.01, 1RMSEA = 0.088, and 1CFI = −0.099, which inhibits
Internal reliability analysis of the TEIQue-SF in the Chinese
meaningful assessment of latent mean differences across cultural
sample demonstrated high alpha scores on global trait EI, but
groups, and overall testing of group differences using the
scores were lower when assessed at the TEIQue factor level.
TEIQue-SF (Putnick and Bornstein, 2016).
While internal consistencies for factor scores on the short form
A lack of measurement invariance indicates that trait scores
are expected to be slightly lower (Petrides, 2009), the Sociability
on the TEIQue-SF are not comparable across the Chinese
factor had lower than expected alpha values in the current study.
and Canadian samples assessed in the present study (Reise
Some other studies have also reported low alpha coefficients for
et al., 1993). Therefore, mean differences on this measure or
Sociability on the TEIQue-SF (e.g., Petrides et al., 2010).
correlations using this measure can be potentially misleading
Findings that a measure like the TEIQue-SF has a similar
(Reise et al., 1993). Therefore, further analyses to investigate
factor structure within different cultural contexts do not
group mean differences on the TEIQue-SF between the two
guarantee that the measure will perform equivalently across
groups were not performed.
cultures (Byrne and Campbell, 1999). Analysis of cultural
invariance demonstrated that while configural and partial metric
TABLE 2 | Cultural Measurement Invariance Fit Indices.
invariance for the TEIQue-SF were achieved, scalar invariance
was not achieved across cultural groups. Therefore, the trait
Model χ2 (df) RMSEA RMSEA 90% C.I. CFI EI measure performs differently and has different meanings for
1. Configural 28.236∗ (4) 0.100 0.067–0.137 0.981
Chinese participants in comparison to Canadian participants.
invariance Such a finding is important for moving forward with cross-
2. Metric 47.314∗ (7) 0.098 0.072–0.125 0.968 cultural comparisons using this measure, and inhibits meaningful
invariance interpretation of trait EI comparisons in the present study.
3. Partial metric 36.036∗ (6) 0.091 0.064–0.121 0.976 One potential explanation for lack of invariance is
invariance measurement bias (Hong et al., 2003). Misunderstandings
4. Partial scalar 161.856∗ (8) 0.179 0.155–0.203 0.877 of items in different cultures or potential translational issues
invariance
might serve as potential explanations. Previous studies have
∗p < 0.001. also reported cultural factors having an influence on responses

Frontiers in Psychology | www.frontiersin.org 5 February 2019 | Volume 10 | Article 435

Feher et al. Chinese Trait Emotional Intelligence Questionnaire

to questionnaires, which may affect invariance. For example, Limitations and Future Directions
analysis of multiple countries revealed that individualism was Some limitations with the present study should be considered
negatively related to a middle response style and acquiescent and used to guide future research. The present study’s use
responding (Harzing, 2006). Other studies have also found of undergraduate participants made it possible to compare
evidence for individuals from more collectivistic countries being the Chinese sample to a demographically similar Canadian
more likely to endorse midpoint values, and having less extreme sample. Future studies, however, should also compare the
scores on scales compared to individualists (Chen et al., 1995; psychometric properties of the TEIQue-SF among participants
Takahashi et al., 2002). However, an examination of the data of different age brackets and different groups within a country.
regarding means and standard deviations from both countries The present study also used a predominantly female sample,
suggests these latter points regarding response style are not and future studies should recruit participants with a more
particularly applicable to the current study. equal gender balance. Differences in trait EI levels across
Differences across cultures in whether and/or to what extent different university faculties have been previously observed
emotionality, well-being, self-control, and sociability define (e.g., Sanchez-Ruiz et al., 2010) which might warrant further
latent trait EI across cultures may also serve as a potential study. Future cross-cultural research should include measures
explanation for findings of non-invariance. For example, self- of an individual’s individualism and collectivism to assess these
control factor loadings were not equivalent across Chinese and relationships with trait EI, as well examinations of other cultural
Canadian samples in the present study. A possible explanation factors that might be driving these differences. Finally, our
may be found in greater value ascribed to control over results showed that trait EI manifests differently across the
one’s inner emotions and desires amongst more collectivistic two countries, and future research should examine reasons
individuals (Markus and Kitayama, 1991). In the development for this further.
of their Asian American Values Scale, Kim et al. (2005) listed
emotional self-control as a central value. This potentially suggests
that self-control might be more strongly related to Chinese AUTHOR CONTRIBUTIONS
participants’ self-perceptions of emotion-related competencies.
Findings from the present study demonstrate a need for further All authors listed have made a substantial, direct and
exploration of how cultural differences account for different intellectual contribution to the work, and approved it
interpretations of trait EI. for publication.

REFERENCES Ekman, P., and Friesen, W. V. (1969). The repertoire of nonverbal behavior:
categories, origins, usage, and coding. Semicotica 1, 49–98. doi: 10.1515/semi.
Aluja, A., Blanch, A., and Petrides, K. V. (2016). Psychometric properties of the 1969.1.1.49
Catalan version of the Trait Emotional Intelligence (TEIQue): comparison Freudenthaler, H. H., Neubauer, A. C., Gabler, P., Scherl, W. G., and
between Catalan and English data. Pers. Ind. Diff. 99, 133–139. doi: 10.1016/ Rindermann, H. (2008). Testing and validating the trait emotional intelligence
j.paid.2016.04.098 questionnaire (TEIQue) in a German-speaking sample. Pers. Ind. Diff. 45,
Austin, E., Saklofske, D., and Smith, M. (2018). Development and validation of 673–678. doi: 10.1016/j.paid.2008.07.014
two short forms of the managing the emotions of others (MEOS) scale. Front. Fukuda, E., Saklofske, D. H., Tamaoka, K., Fung, T. S., Miyaoka, Y., and Kiyama, S.
Psychol. 9:974. doi: 10.3389/fpsyg.2018.00974 (2011). Factor structure of Japanese versions of two emotional intelligence
Bar-On, R. (1997). The Emotional Intelligence Inventory (EQ-i): Technical Manual. scales. Int. J. Test. 11, 71–92. doi: 10.1080/15305058.2010.516379
Toronto, ON: Multi-Health Systems. Fukuda, E., Saklofske, D. H., Tamaoka, K., and Lim, H. (2012). Factor structure of
Browne, M. W., and Cudeck, R. (1993). “Alternative ways of assessing model fit,” the Korean version of Wong and Law’s emotional intelligence scale. Assessment
in Testing Structural Equation Models, eds K. A. Bollen and J. S. Long (Newbury 19, 3–7. doi: 10.1177/1073191111428863
Park, CA: Sage), 136–162. Furnham, A., and Cheng, H. (1999). Personality as predictor of mental health and
Byrne, B. M., and Campbell, L. T. (1999). Cross-cultural comparisons and happiness in the East and West. Pers. Ind. Diff. 27, 395–403. doi: 10.1016/S0191-
the presumption of equivalent measurement and theoretical structure: a 8869(98)00250-5
look beneath the surface. J. Cross-Cult. Psychol. 30, 555–574. doi: 10.1177/ Gangopadhyay, M., and Mandal, M. K. (2008). “Emotional intelligence-a universal
0022022199030005001 or culture- specific construct?,” in Emotional Intelligence: Theoretical and
Byrne, B. M., and Watkins, D. (2003). The issue of measurement invariance Cultural Perspectives, eds R. J. Emmerling, V. K. Shanwal, and M. K. Mandal
revisited. J. Cross-Cult. Psychol. 34, 155–175. doi: 10.1177/0022022102250225 (New York, NY: Nova Science Publishers, Inc.), 115–134.
Caruso, D. R. (2008). “Emotions and the ability model of emotional intelligence,” Gökçen, E., Furnham, A., Mavroveli, S., and Petrides, K. V. (2014). A cross-cultural
in Emotional Intelligence: Theoretical and Cultural Perspectives, eds R. J. investigation of trait emotional intelligence in Hong Kong and the UK. Pers.
Emmerling, V. Shanwal, and M. Mandal (Hauppauge, NY: Nova Science Ind. Diff. 65, 30–35. doi: 10.1016/j.paid.2014.01.053
Publishers), 1–16. Hambleton, R. K., and Lee, M. K. (2013). “Methods of translating and adapting
Chen, C., Lee, S. Y., and Stevenson, H. W. (1995). Response style and cross-cultural tests to increase cross-language validity,” in The Oxford Handbook of Child
comparisons of rating scales among East Asian and North American students. Psychological Assessment, eds D. H. Saklofske, C. R. Reynolds, and V. L. Schwean
Psychol. Sci. 6, 170–175. doi: 10.1111/j.1467-9280.1995.tb00327.x (New York, NY: Oxford University Press), 172–181.
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement Harzing, A. (2006). Response styles in cross-national survey research: a 26-country
invariance. Struct. Equat. Model. 14, 464–504. doi: 10.1080/10705510701301834 study. Int. J. Cross Cult. Manage. 6, 243–266. doi: 10.1177/1470595806066332
Cheung, G. W., and Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes Herodotou, C., Kambouri, M., and Winters, N. (2011). The role of trait emotional
for testing measurement invariance. Struct. Equat. Model. Multidiscipl. J. 9, intelligence in gamers’ preferences for play and frequency of gaming. Comput.
233–255. doi: 10.1207/S15328007SEM0902_5 Hum. Behav. 27, 1815–1819. doi: 10.1016/j.chb.2011.04.001

Frontiers in Psychology | www.frontiersin.org 6 February 2019 | Volume 10 | Article 435

Feher et al. Chinese Trait Emotional Intelligence Questionnaire

Hofstede, G. (1980). Culture’s Consequences: International Differences in Work- McCrae and J. Allik (New York, NY: Kluwer Academic/Plenum Publishers),
Related Values. Beverly Hills, CA: Sage. 105–125. doi: 10.1007/978-1-4615-0763-5_6
Hofstede, G. (2001). Cultures Consequences: Comparing Values, Behaviors, Merino-Tejedor, E., Hontangas, P. M., and Petrides, K. V. (2018). Career
Institutions and Organizations Across Nations, 2nd Edn. Thousand Oaks, CA: adaptability mediates the effect of trait emotional intelligence on academic
Sage. engagement. Rev. Psicodidáct. (English ed.) 23, 77–85. doi: 10.1016/j.psicoe.
Hofstede, G., Hofstede, G. J., and Minkov, M. (2010). Cultures and Organizations: 2017.10.002
Software of the Mind, 3rd Edn. New York, NY: McGraw Hill. Mikolajczak, M., Luminet, O., Leroy, C., and Roy, E. (2007). Psychometric
Hofstede, G., and McCrae, R. R. (2004). Personality and culture revisited: linking properties of the trait emotional intelligence questionnaire: factor structure,
traits and dimensions of culture. Cross-Cult. Res. 38, 52–88. doi: 10.1177/ reliability, construct, and incremental validity in a French-speaking population.
1069397103259443 J. Pers. Assess. 88, 338–353. doi: 10.1080/00223890701333431
Hong, S., Malik, M. L., and Lee, M. K. (2003). Testing configural, metric, scalar, Mikolajczak, M., Nelis, D., Hansenne, M., and Quoidbach, J. (2008). If you can
and latent mean invariance across genders in sociotropy and autonomy using regulate sadness, you can probably regulate shame: associations between trait
a non-Western sample. Educ. Psychol. Measure. 63, 636–654. doi: 10.1177/ emotional intelligence, emotion regulation and coping efficiency across discrete
0013164403251332 emotions. Pers. Ind. Diff. 44, 1356–1368. doi: 10.1016/j.paid.2007.12.004
Hu, L. T., and Bentler, P. M. (1998). Fit indices in covariance structure modeling: Milfont, T. L., and Fischer, R. (2010). Testing measurement invariance across
sensitivity to underparameterized model misspecification. Psychol. Methods 3, groups: applications in cross-cultural research. Int. J. Psychol. Res. 3, 111–130.
424–453. doi: 10.1037/1082-989X.3.4.424 doi: 10.21500/20112084.857
Hu, L. T., and Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance Mullen, M. R. (1995). Diagnosing measurement equivalence in cross-national
structure analysis: conventional criteria versus new alternatives. Struct. Equat. research. J. Int. Bus. Stud. 26, 573–596. doi: 10.1057/palgrave.jibs.8490187
Model. Multidiscipl. J. 6, 1–55. doi: 10.1080/10705519909540118 Muthén, L. K., and Muthén, B. O. (1998-2015). Mplus User’s Guide, 7th Edon. Los
Jacobs, I., Sim, C. W., and Zimmermann, J. (2015). The German TEIQue-SF: Angeles, CA: Muthén & Muthén.
factorial structure and relations to agentic and communal traits and mental Nozaki, Y., and Koyasu, M. (2016). Can we apply an emotional competence
health. Pers. Ind. Diff. 72, 189–194. doi: 10.1016/j.paid.2014.09.003 measure to an easternpopulation? Psychometric properties of the profile of
Kim, B. K., Li, L. C., and Ng, G. F. (2005). The Asian American values scale– emotional competence in a Japanese population. Assessment 23, 112–123. doi:
multidimensional: development, reliability, and validity. Cult. Divers. Ethnic 10.1177/1073191115571124
Minor. Psychol. 11:187. doi: 10.1037/1099-9809.11.3.187 Pérez, J. C., Petrides, K. V., and Furnham, A. (2005). “Measuring trait emotional
Kong, F. (2017). The validity of the wong and law emotional intelligence scale in intelligence,” in International Handbook of Emotional Intelligence, eds R.
a Chinese sample: tests of measurement invariance and latent mean differences Schulze and R. D. Roberts (Cambridge, MA: Hogrefe & Huber).
across gender and age. Pers. Ind. Diff. 116, 29–31. doi: 10.1016/j.paid.2017.04. Petrides, K. V. (2009). “Psychometric properties of the trait emotional intelligence
025 questionnaire,” in Advances in the Assessment of Emotional Intelligence, eds C.
Koydemir, S., Şimşek, Ö. F., Schütz, A., and Tipandjan, A. (2013). Differences Stough, D. H. Saklofske, and J. D. Parker (New York, NY: Springer).
in how trait emotional intelligence predicts life satisfaction: the role of affect Petrides, K. V. (2011). “Ability and trait emotional intelligence,” in The Blackwell-
balance versus social support in India and Germany. J. Happiness Stud. 14, Wiley Handbook of Individual Differences, eds T. Chamorro-Premuzic, A.
51–66. doi: 10.1007/s10902-011-9315-1 Furnham, and S. von Stumm (New York, NY: Wiley).
Laborde, S., Allen, M. S., and Guillén, F. (2016). Construct and concurrent Petrides, K. V., and Furnham, A. (2001). Trait emotional intelligence: psychometric
validity of the short- and long-form versions of the trait emotional intelligence investigation with reference to established trait taxonomies. Eur. J. Pers. 15,
questionnaire. Pers. Ind. Diff. 101, 232–235. doi: 10.1016/j.paid.2016.06.009 425–448. doi: 10.1002/per.416
LaPalme, M. L., Wang, W., Joseph, D. L., Saklofske, D. H., and Yan, G. (2016). Petrides, K. V., Furnham, A., and Mavroveli, S. (2007a). “Trait emotional
Measurement equivalence of the Wong and Law emotional intelligence scale intelligence. Moving forward in the field of EI,” in Emotional Intelligence.
across cultures: an item response theory approach. Pers. Ind. Diff. 90, 190–198. Knowns and Unknowns (Series in Affective Science), eds G. Matthews, M.
doi: 10.1016/j.paid.2015.10.045 Zeidner, and R. Roberts (Oxford: Oxford University Press).
Li, T., Saklofske, D. H., Bowden, S. C., Yan, G., and Fung, T. S. (2012). The Petrides, K. V., Pita, R., and Kokkinaki, F. (2007b). The location of trait emotional
measurement invariance of the Wong and Law emotional intelligence scale intelligence in personality factor space. Br. J. Psychol. 98, 273–289. doi: 10.1348/
(WLEIS) across three Chinese university student groups from Canada and 000712606X120618
China. J. Psychoeduc. Assess. 30, 439–452. doi: 10.1177/0734282912449449 Petrides, K. V., Sanchez-Ruiz, M.-J., Siegling, A. B., Saklofske, D. H., and
Libbrecht, N., Beuckelaer, A. D., Lievens, F., and Rockstuhl, T. (2014). Mavroveli, S. (2018). “Emotional intelligence as personality: measurement
Measurement invariance of the Wong and Law emotional intelligence and role of trait emotional intelligence in Education contexts,” in Emotional
scale scores: does the measurement structure hold across far Eastern and Intelligence in Education: Integrating Research Into Practice, eds K. Keefer,
European countries? Appl. Psychol. 63, 223–237. doi: 10.1111/j.1464-0597.2012. J. D. A. Parker, and D. H. Saklofske (New York, NY: Springer.), 49–82.
00513.x Petrides, K. V., Siegling, A., and Saklofske, D. H. (2016). “Theory and measurement
MacCallum, R. C., Browne, M. W., and Sugawara, H. M. (1996). Power analysis of trait emotional intelligence,” in Wiley Handbook of Personality Assessment,
and determination of sample size for covariance structure modeling. Psychol. ed. U. Kumar (Hoboken, NJ: John Wiley), 90–103. doi: 10.1002/97811191
Methods 1, 130–149. doi: 10.1037/1082-989X.1.2.130 73489.ch7
Markus, H. R., and Kitayama, S. (1991). Culture and the self: implications for Petrides, K. V., Vernon, P. A., Schermer, J. A., Ligthart, L., Boomsma, D. I., and
cognition, emotion, and motivation. Psychol. Rev. 98:224. doi: 10.1037/0033- Veselka, L. (2010). Relationships between trait emotional intelligence and the
295X.98.2.224 Big Five in the Netherlands. Pers. Ind. Diff. 48, 906–910. doi: 10.1016/j.paid.
Martskvishvili, K., Arutinov, L., and Mestvirishvili, M. (2013). A psychometric 2010.02.019
investigation of the Georgian version of the trait emotional intelligence Plouffe, R. A., Smith, M. M., and Saklofske, D. H. (in press). A psychometric
questionnaire. Eur. J. Pers. Assess. 29, 84–88. doi: 10.1027/1015-5759/a000135 investigation of the assessment of sadistic personality. Pers. Ind. Diff. doi: 10.
Matsumoto, D. (1990). Cultural similarities and differences in display rules. Motiv. 1016/j.paid.2018.01.002
Emot. 14, 195–214. doi: 10.1007/BF00995569 Putnick, D. L., and Bornstein, M. H. (2016). Measurement invariance conventions
Mavroveli, S., and Siu, A. F. (2012). The factor structure of trait emotional and reporting: the state of the art and future directions for psychological
intelligence in Hong Kong adolescents. J. Psychoeducat. Assess. 30, 567–576. research. Dev. Rev. 41, 71–90. doi: 10.1016/j.dr.2016.06.004
doi: 10.1177/0734282912449596 Reise, S. P., Waller, N. G., and Comrey, A. L. (2000). Factor analysis and scale
McCrae, R. R. (2001). Trait psychology and culture: exploring intercultural revision. Psychol. Assess. 12, 287–297. doi: 10.1037/1040-3590.12.3.287
comparisons. J. Pers. 69, 819–846. doi: 10.1111/1467-6494.696166 Reise, S. P., Widaman, K. F., and Pugh, R. H. (1993). Confirmatory factor
McCrae, R. R. (2002). “NEO-PI-R data from 36 cultures: further intercultural analysis and item response theory: two approaches for exploring measurement
comparisons,” in The Five-Factor Model of Personality Across Cultures, eds R. R. invariance. Psychol. Bull. 114:552. doi: 10.1037/0033-2909.114.3.552

Frontiers in Psychology | www.frontiersin.org 7 February 2019 | Volume 10 | Article 435

Feher et al. Chinese Trait Emotional Intelligence Questionnaire

Salovey, P., and Mayer, J. D. (1990). Emotional intelligence. Imaginat. Cognit. Triandis, H. C. (1995). Individualism and Collectivism. Boulder, CO: Westview
Personal. 9, 185–211. doi: 10.2190/DUGG-P24E-52WK-6CDG Press.
Sanchez-Ruiz, M. J., Mavroveli, S., and Poullis, J. (2013). Trait emotional Vesely, A. K., Saklofske, D. H., and Nordstokke, D. W. (2014). EI training and pre-
intelligence and its links to university performance: an examination. Pers. Ind. service teacher wellbeing. Pers. Ind. Diff. 65, 81–85. doi: 10.1016/j.paid.2014.
Diff. 54, 658–662. doi: 10.1016/j.paid.2012.11.013 01.052
Sanchez-Ruiz, M. J., Pérez-González, J. C., and Petrides, K. V. (2010). Trait Widaman, K. F., and Reise, S. P. (1997). “Exploring the measurement invariance
emotional intelligence profiles of students from different university faculties. of psychological instruments: applications in the substance use domain,” in The
Austr. J. Psychol. 62, 51–57. doi: 10.1080/00049530903312907 Science of Prevention; Methodological Advances from Alcohol and Substance
Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., Cooper, J. T., Golden, Abuse Research, eds K. Bryant, M. Windle, and S. West (Washington,
C. J., et al. (1998). Development and validation of a measure of emotional DC: American Psychological Association), 281–324. doi: 10.1037/
intelligence. Pers. Ind. Diff. 25, 167–177. doi: 10.1016/S0191-8869(98)00001-4 10222-009
Siu, A. F., and Chang, J. F. (2011). Coping styles and psychological distress Wilks, D. C., Neto, F., and Mavroveli, S. (2015). Trait emotional intelligence,
among Hong Konguniversity students: validation of the collectivist coping style forgiveness, and gratitude in Cape Verdean and Portuguese students. S. Afr.
inventory. Int. J. Adv. Counsell. 33, 88–100. doi: 10.1007/s10447-011-9114-8 J. Psychol. 45, 93–101. doi: 10.1177/0081246314546347
Snowden, A., Watson, R., Stenhouse, R., and Hale, C. (2015). Emotional Wilson, C. A., Plouffe, R. A., Saklofske, D. H., Yan, G., Nordstokke, D. W., Prince-
intelligence and nurse recruitment: rasch and confirmatory factor analysis of Embury, S., et al. (2018). A cross-cultural validation of the resiliency scale
the trait emotional intelligence questionnaire short form. J. Adv. Nurs. 71, for young adults in Canada and China. Psych J. [Epub ahead of print]. doi:
2936–2949. doi: 10.1111/jan.12746 10.1002/pchj.256
Suh, E., Diener, E., Oishi, S., and Triandis, H. C. (1998). The shifting basis of
life satisfaction judgments across cultures: emotions versus norms. J. Pers. Soc. Conflict of Interest Statement: The authors declare that the research was
Psychol. 74, 482–493. doi: 10.1037/0022-3514.74.2.482 conducted in the absence of any commercial or financial relationships that could
Takahashi, K., Ohara, N., Antonucci, T. C., and Akiyama, H. (2002). be construed as a potential conflict of interest.
Commonalities and differences in close relationships among the Americans and
Japanese: a comparison by the individualism/collectivism concept. Int. J. Behav. Copyright © 2019 Feher, Yan, Saklofske, Plouffe and Gao. This is an open-access
Dev. 26, 453–465. doi: 10.1080/01650250143000418 article distributed under the terms of the Creative Commons Attribution License
Tang, H. W. V., Yin, M. S., and Nelson, D. B. (2010). The relationship between (CC BY). The use, distribution or reproduction in other forums is permitted, provided
emotional intelligence and leadership practices: a cross-cultural study of the original author(s) and the copyright owner(s) are credited and that the original
academic leaders in Taiwan and the USA. J. Managerial Psychol. 25, 899–926. publication in this journal is cited, in accordance with accepted academic practice. No
doi: 10.1108/02683941011089143 use, distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Psychology | www.frontiersin.org 8 February 2019 | Volume 10 | Article 435