Personality and Individual Differences 53 (2012) 660–665
Contents lists available at SciVerse ScienceDirect
Personality and Individual Differences
journal homepage: www.elsevier.com/locate/paid
The interaction between emotional intelligence and cognitive ability
in predicting scholastic performance in school-aged children
Sergio Agnoli a,⇑, Giacomo Mancini b, Tiziana Pozzoli c, Bruno Baldaro b, Paolo Maria Russo b,
Paola Surcinelli b
a
b
c
Marconi Institute for Creativity, MIC, University of Bologna, Italy
Department of Psychology, University of Bologna, Italy
Department of Developmental and Socialization Psychology, University of Padova, Italy
a r t i c l e
i n f o
Article history:
Received 9 March 2012
Received in revised form 14 May 2012
Accepted 15 May 2012
Available online 8 June 2012
Keywords:
Trait emotional intelligence
Emotion recognition ability
Cognitive ability
Scholastic performance
Children
a b s t r a c t
The aim of the present study is to offer an exploration of the predictive validity of cognitive ability and
emotional intelligence (EI) on scholastic achievement in a sample of Italian school-aged children
(8–11 years). In particular, cognitive ability was measured through Raven’s Coloured Progressive Matrices, while trait EI was measured through the Trait Emotional Intelligence Questionnaire–Child Form (TEIQue-CF). In addition to trait EI, we measured also a basic emotional ability, the emotion recognition
ability, through an emotional face recognition task. The results demonstrated an interaction between
trait EI and cognitive ability in predicting academic performance. In particular, trait EI was positively
associated with better language performance in children characterised by low or medium cognitive ability, but not in pupils characterised by high cognitive ability. Moreover, results showed that trait EI had a
unique power to predict math performance. Similarly, the analyses showed an interaction between emotion recognition ability and cognitive ability in predicting both language and math performance. Differences between the two emotional measures were discussed.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
An analysis of the wide literature dealing with the predictors of
academic success indicates that it is generally associated with cognitive intelligence. In particular, past research in this field has primarily focused on cognitive abilities (Colom & Flores-Mendoza,
2007; Farsides & Woodfield, 2003; Neisser et al., 1996) demonstrating the important predictive role of IQ on academic performance
(Neisser et al., 1996). Recently, another construct has attracted a
lot of interest in this area of research: emotional intelligence (EI;
Petrides & Furnham, 2000). EI theoreticians proposed a new perspective in the study of emotions by suggesting that the intelligent
use of emotions is essential for one’s physical and psychological
adaptation (Extremera & Fernández-Berrocal, 2006). The aim of
the present work is to offer an exploration of the predictive validity
of different types of intelligence, in particular of trait EI, on scholastic success.
The literature on EI is mainly characterised by the distinction between trait EI (or trait emotional self-efficacy) and ability EI (or cognitive-emotional ability) (Mavroveli, Petrides, Shove, & Whitehead,
⇑ Corresponding author. Address: Department of Developmental and Socialization Psychology, University of Padova, Via Venezia, 8, 35131 Padova, Italy. Tel.: +39
049 8276551; fax: +39 049 8276511.
E-mail address: [email protected] (S. Agnoli).
0191-8869/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.paid.2012.05.020
2008; Petrides & Furnham, 2000). Trait EI is conceptualised as a
lower-order personality construct and it is defined as a constellation of emotional self-perceptions and behavioural dispositions
(Petrides, Pita, & Kokkinaki, 2007). Ability EI is conceived as an actual ability that comprises emotion-related cognitive abilities.
While trait EI is measured through self-report questionnaires and
is associated with the realm of personality, ability EI is instead measured through maximum-performance tests and is associated to the
realm of cognitive ability (Petrides, Frederickson, & Furnham, 2004;
Petrides, Furnham, & Mavroveli, 2007). According to ability EI theorisation, the emotional abilities are organised along a continuum
from those that are relatively lower level (e.g., capacity to perceive
emotions accurately), to those that are more developmentally complex (e.g., ability to manage emotions properly) (Mayer, Salovey, &
Caruso, 2008).
Several studies have explored whether EI may help explain
academic achievement (Di Fabio & Palazzeschi, 2009; Laborde, Dosseville, & Scelles, 2010; Mavroveli, Petrides, Sangareau, & Furnham,
2009; Mavroveli & Sanchez-Ruiz, 2011; O’Connor & Little, 2003;
Parker et al., 2004; Petrides et al., 2004; Song et al., 2010). On the
basis of trait EI definition, it could be expected that trait EI would
not correlate strongly with cognitive ability, verbal intelligence,
or academic achievement. Indeed, as previously mentioned, trait
EI is considered a personality trait rather than a cognitive ability.
That said, it would not be expected to be highly associated either
S. Agnoli et al. / Personality and Individual Differences 53 (2012) 660–665
with psychometric intelligence or proxies thereof (Petrides et al.,
2004). Nevertheless, although emotion-related self-perceptions
would not be expected to be directly associated with better or
poorer scholastic achievement, it is possible that they may interact
with variables that are (e.g., cognitive abilities). The results of several studies by Mavroveli et al. (2008, 2009), and Petrides et al.
(2004) confirmed these assumptions.
A moderating effect of trait EI on the relation between IQ and
academic performance was demonstrated, so that high trait EI
was associated with better English (but not math or science) performance in low-IQ high school students (Petrides et al., 2004). However, a direct but modest correlation between trait EI and academic
performance in high school and university students has also been
reported (Parker, Creque et al., 2004; Parker, Summerfeldt, Hogan,
& Majeski, 2004). Similarly, Mavroveli and Sanchez-Ruiz (2011) recently showed a significant relationship between trait EI and math
scores in children aged 3 years but not in older pupils (4–6 years).
The different findings that emerged from these studies suggested
that the effects of trait EI may vary across educational levels, age,
and subjects.
Other studies explored the relation between ability EI and academic achievement with mixed results. In one study with university students, ability EI predicted academic performance over and
above general mental abilities (Song et al., 2010). However, another
study with 18- to 32-year-old students showed that emotional
understanding ability, but not overall ability EI, was associated with
academic performance (O’Connor & Little, 2003). The emerging
findings revealed again an inconsistent pattern, probably depending both on how scholastic performance is operationalised in the
different studies and on the characteristics of the sample (e.g., gender, age).
To date, the research on the relationship between EI and academic achievement has been mainly based on samples of college
students and adolescents. In contrast, there is little evidence from
child samples, partly because of a lack of appropriate measures.
Moreover, the results on children are not always fully consistent
(for a review, see Mavroveli & Sanchez-Ruiz, 2011). The current
study aims to fill this gap by analysing the effects of cognitive ability (operationalised through Raven’s Coloured Progressive Matrices) and EI in predicting scholastic achievement in a sample of
school-aged children. Moreover, since the simultaneous contribution of both trait EI and ability EI in explaining scholastic success
has been rarely explored (O’Connor & Little, 2003) and, as far as
we know, it has never been investigated in primary school students,
we tested the effect of EI components on children’s scholastic
achievement by considering both self-report and performance
measures. Instead of seeing the two types of EI as contrasting constructs, we considered them as different ways of approaching the
study of EI, which could bring different and incremental contributions in explaining scholastic achievement. In particular, taking into
account the developmental trajectory followed by emotional abilities, we explored only the effect of the lower-level and fundamental
skill—that is, emotional perception (specifically, the ability to accurately perceive emotions in the face). The most-used ability EI measurement methods (e.g., the Mayer–Salovey–Caruso Emotional
Intelligence Test [MSCEIT]) consider emotional facial expression
recognition as a central task to measure emotional perception ability; for this reason, in the present study we measured emotional
perception ability through an emotional face recognition task. The
choice of using only the recognition task among the eight tasks provided by MSCEIT to assess EI abilities is also due to the lack of published methodologies to measure ability EI in young children1.
1
Although a MSCEIT version for the use with children (MSCEIT-Youth Version) is
available for research purposes, it is still in development and has no validation in
European countries (Windingstad, McCallum, Mee Bell, & Dunn, 2011).
661
Furthermore, emotion recognition, on the basis of a long research
tradition in developmental literature (Camras & Allison, 1985; Herba, Landau, Russell, Ecker, & Phillips, 2006), is, at present, the only
EI ability that can be tested with a reliable task in children (i.e., the
recognition of emotions from prototypical emotional facial expressions). In addition to this basic ability, we measured trait EI
through TEIQue-CF (Mavroveli et al., 2008). We hypothesised that
the two measures would separately predict academic achievement,
interacting with cognitive abilities in predicting children’s performance across different subjects (i.e., mathematics and language).
According to Petrides et al. (2004), the effect of trait EI on academic
performance should be stronger when the demands of the situation outweigh students’ intellectual resources. On the basis of this
assumption, we hypothesised that the effects of trait EI on scholastic performance should be more pronounced in children with low
levels of cognitive ability. We hypothesised that the same trend
should emerge in emotion recognition ability; this ability represents a basic emotional competence, and its impact should emerge
as soon as the situation requires more than intellectual resources.
2. Method
2.1. Participants
Four hundred and forty-seven children ranging in age from 8 to
11 years participated in the study. All participants were recruited
from primary (third- to fifth-grade) state schools in the districts
of Bologna. Pupils with special educational needs (n = 12) were excluded from subsequent analyses; complete data were available for
352 pupils (188 females) ranging in age from 8 to 11 years old
(mean age = 9.35 years; SD = 0.8 years). Children with missing data
were excluded from the analyses.
2.2. Measures
2.2.1. Trait Emotional Intelligence Questionnaire–Child Form
(Mavroveli et al., 2008)
The TEIQue-CF comprises 75 short statements (e.g., ‘It’s easy for
me to show how I feel’) that are responded to on a 5-point Likert
scale, ranging from ‘completely disagree’ to ‘completely agree’. TEIQue-CF has shown satisfactory levels of internal consistency
(alpha > .72) and temporal stability over a 3-month interval
(r = 0.79; Mavroveli & Sanchez-Ruiz, 2011; Mavroveli et al.,
2008). In the present study, we used the Italian version of the
TEIQue-CF (Russo et al., 2012). For each participant, a score for
the global trait EI was computed. The reliability of the global TEIQue-CF score was very high (Cronbach alpha = .88).
2.2.2. Raven’s Progressive Matrices (Raven, 1981)
Raven’s Progressive Matrices are a measure of pure non-verbal
reasoning ability that is thought to be relatively independent of
specific learning acquired in a particular cultural or educational
context. In this study, we administered the Coloured Progressive
Matrices (CPM), a simpler version of the test often used with
children. It consists of 36 items presented in three sets of 12 each.
Research in many different samples and settings has consistently
revealed good psychometric properties for the Raven matrices
(Raven, Raven, & Court, 2000). CPM scores were transformed in percentile scores according to Italian age norms (see Belacchi, Scalisi,
Cannoni, & Cornoldi, 2008).
2.2.3. Emotional facial expression recognition
Forty-eight colour pictures of faces representing five basic emotions (anger, fear, sadness, happiness, and disgust) and neutral
expressions were selected from the Karolinska Directed Emotional
662
S. Agnoli et al. / Personality and Individual Differences 53 (2012) 660–665
Face System (KDEF; Ludqvist, Flykt, & Ohman, 1998). Eight faces,
balanced for gender, were selected for each emotional expression.
The 48 pictures were divided into two sets of 24 facial expressions,
and each set was arranged in four blocks of six expressions, such
that there was one exemplar from each of the six stimulus types
in each block. For each set, four different orders of picture presentation were constructed. Stimulus presentation was conducted
with an Acer laptop computer with a 2.4 GHz processor and a
21-inch monitor. A refresh rate of 60 Hz and a resolution of
1440 900 pixels were used. For each participant, a global emotion recognition ability score was computed on the basis of the
accuracy in the recognition of the six facial expressions.
2.2.4. Scholastic achievement
End-of-year grades in language and math were obtained for
each pupil from school offices and ranged for both language and
math from 4 to 10, with a mean of 7.8 (SD = 1.1) for language
and 7.9 (SD = 1.2) for math grades.
2.3. Procedure
3. Results
Means and standard deviations and correlations between all the
study variables are presented in Table 1. A positive correlation between language and math performance emerged, and both of them
were positively related to cognitive ability, emotion recognition
ability, and trait EI. No relation between trait EI and recognition
ability emerged.
Table 1
Descriptive statistics and correlations among the study variables.
1
*
p < .05.
p < .01.
***
p < .001.
**
Because our data (students nested within classes) are multilevel
in nature (Lee, 2000), two multilevel analyses, also known as hierarchical linear modelling, were performed (HLM 6.04 Software;
Raudenbush & Bryk, 2002). Given that the aim of this study was
to investigate the relations between individual characteristics
and academic performance, we focused only on explaining within-class variability in language and math performance.
Since previous studies reported that age and gender could affect
our dependent variables (e.g., Hyde, Fennema, Ryan, Frost, & Hopp,
1990), these demographic variables were included in the analyses
as control variables together with the three predictors (i.e., cognitive ability, emotion recognition ability, and trait EI). Moreover, on
the basis of our hypotheses regarding the interaction between the
two emotional measurements and cognitive ability, two-way
interaction terms, which were created by mean-centring continuous predictors, were entered. The equations are reported in the
appendix.
3.2. Intraclass correlation coefficients
The purpose of the study was presented to the school principals
and teachers. Informed consent was obtained from parents and all
the participants were asked for their personal assent. Children
were first given Raven’s CPM. It was administered according to
standard instructions (Raven et al., 2000) as a group test and took
place in classrooms without any time limits. Subsequently, all participants were given the emotion recognition task. Pupils were
tested individually in a quiet room arranged for the experimental
procedure. Participants sat approximately 1 m from the computer
screen on which the pictures were presented. Each face was presented on the screen for a 6-s interval, and after each picture offset,
participants were asked to complete a facial expression recognition
task. In particular, participants were asked to select one of six emotion labels (anger, sadness, happiness, fear, disgust, and neutral)
that best described the emotional expression they saw before. Participants were given 10 s to make their selection, and they were
asked to respond as accurately as possible. Finally, the TEIQue-CF
was administered in each classroom. All participants filled out
the questionnaire individually in their classrooms, after brief group
instructions on the answer formats. Administration lasted between
15 and 30 min.
1. Math performance
2. Language performance
3. Cognitive ability
4. Emotion recognition ability
5. Trait EI
Mean
SD
3.1. Multilevel analyses
2
3
4
5
As a preliminary step, for each outcome the unconditional model
was estimated. This step yielded the intraclass correlation coefficients (ICC), which revealed the amount of the variability in each
academic performance that occurred between or within classrooms.
ICC indices showed that 87.8% of the total variance of language performance and 89.6% of the variation of math performance were explained by within-classroom variables.
3.3. Language performance
Results (see Table 2) showed that, controlling for class membership and demographic variables, cognitive ability as well as emotion recognition ability and trait EI positively predicted
participants’ final scores in language performance. The positive
associations of children’s ability to recognise emotions and trait
EI with language performance were specified by the moderating
role played by children’s cognitive ability. In order to interpret
the nature of these interactions, simple slopes were derived for
high (+1 SD), medium (0 SD), and low levels (1 SD) of cognitive
ability (Aiken & West, 1991).
Simple slopes computations revealed that emotion recognition
ability was associated with better language performance in pupils
with low (b = .02, SE = .003, t = 6.38, p < .001) and medium (b = .01,
SE = .003, t = 3.16, p = .002) cognitive ability. In contrast, students’
ability to recognise emotions did not affect language performance
at high levels of cognitive ability. As far as the interaction between
cognitive ability and trait EI is concerned, similar results were obtained. Indeed, students with low (b = .66, SE = .11, t = 5.99,
p < .001) or medium (b = .37, SE = .11, t = 3.36, p = .001) cognitive
ability showed better language performance with the increasing
of trait EI, while the benefit of trait EI was negligible at high levels
of cognitive ability. Overall, individual predictors and interaction
terms explained 20.89% of the variance in participants’ language
performance (out of 87.8% of the total variability attributable to
students’ characteristics).
–
.74***
.40***
.12*
.13*
8.19
1.09
–
.36***
.13*
.18**
8.03
1.00
3.4. Math performance
–
.17**
.13*
61.85
29.28
–
.02
85.99
12.07
–
3.63
.38
Results of the model predicting math performance are reported
in Table 3. Math scores were positively predicted by cognitive ability and trait EI. Moreover, a significant interaction between emotion
recognition and cognitive ability emerged, showing that this emotional ability improved math performance in students with medium (b = .01, SE = .004, t = 2.29, p = .02) or low (b = .02, SE = .004,
663
S. Agnoli et al. / Personality and Individual Differences 53 (2012) 660–665
Table 2
Multilevel modelling predicting language performance.
Language performance
Unconditional model
C
Intercept c00
Agea
Gender (0 = boys; 1 = girls)
Cognitive abilitya
Emotion recognition abilitya
Trait EIa
Cognitive ability emotion recognition abilitya
Cognitive ability trait EIa
Variance components
T
r2c
v2
8.04
Model
SE
t-ratio
d.f.
p
r
C
.08
95.55
25
<.001
.99
.12
.88
70.22
SE
7.91
.02
.25
.01
.01
.37
.0004
.01
.11
.01
.13
.001
.003
.11
.0001
003
t-ratio
d.f.
p
r
68.47
1.85
1.95
6.87
2.78
3.49
4.24
2.38
25
344
344
344
344
344
344
344
<.001
.07
.05
<.001
.01
.001
<.001
.02
.99
.10
.10
.35
.15
.18
.22
.13
.13
.68
88.75
Note: C = coefficient estimate from the population-average models with robust standard errors.
a
Centered around its group mean.
Table 3
Multilevel modelling predicting math performance.
Math performance
Unconditional model
C
Intercept c00
Agea
Gender (0 = boys; 1 = girls)
Cognitive abilitya
Emotion recognition abilitya
Trait EIa
Cognitive ability emotion recognition abilitya
Cognitive ability trait EIa
Variance components
T
r2c
v2
8.22
Model
SE
t-ratio
d.f.
p
r
.09
93.51
25
<.001
.99
.12
1.08
63.98
C
8.29
.03
.12
.01
.01
.31
.0003
.004
SE
t-ratio
d.f.
p
r
.12
.01
.12
.002
.004
.12
.0001
.004
68.41
1.95
1.02
6.83
1.91
2.58
2.28
1.08
25
344
344
344
344
344
344
344
<.001
.05
.31
<.001
.06
.01
.02
.28
.99
.10
.05
.35
.10
.14
.12
.06
.14
.87
78.71
Note: C = coefficient estimate from the population-average models with robust standard errors.
a
Centered around its group mean.
t = 5.05, p < .001) cognitive ability. In contrast, participants with
high cognitive ability showed good performance regardless of the
level of their emotion recognition. No significant interaction between trait EI and cognitive ability emerged. The model predicting
students’ math performance from demographic variables, individual characteristics, and interaction terms explained 19.2% of within-class variability (89.6%).
4. Discussion
The main purpose of the present study was to determine the predictive validity of trait EI on scholastic achievement in a sample of
school-aged children. In addition to trait EI, we explored the impact
of a basic emotional ability, the emotion recognition ability, on scholastic success. As hypothesised, both trait EI and emotion recognition
ability uniquely predicted academic performance and interacted
with cognitive ability in explaining academic performance.
In agreement with Petrides’s results (Petrides et al., 2004), we
found an interaction between trait EI and cognitive ability in predicting academic performance. In particular, trait EI was associated
with better language performance in children characterised by low
or medium cognitive ability. In contrast, trait EI effect did not
emerge in pupils characterised by high cognitive ability. This result
demonstrated that the importance of trait EI in predicting language
performance emerges when the demands of the subject outweigh
the child’s intellectual resources. In contrast to high cognitive ability children, students with lower abilities could be forced to draw
on other resources (e.g., trait EI) in order to cope with the demands
of the subject (Petrides et al., 2004).
We did not find any moderating effect of trait EI on the relation
between cognitive ability and math performance. Consistently with
the findings of Parker, Creque et al. (2004), Parker, Summerfeldt et
al. (2004) and Mavroveli and Sanchez-Ruiz (2011), we found only a
direct effect of trait EI on math scores. These data showed that trait
EI improved math performance regardless of the children’s cognitive ability. Compared with language, math requires higher levels
of mental resources (Downey, Mountstephen, Lloyd, Hansen, &
Stough, 2008). Thus, while in language a high level of cognitive ability may be sufficient to cope with the subject demands, the higher
complexity of math may require more resources. High trait EI could,
for example, help in coping with the anxiety that sometimes is associated with mathematical tasks. Math anxiety is a widely described
phenomenon in literature (Ashcraft, 2002), and it is commonly defined as a feeling of tension, apprehension, or fear that interferes
with math performance. On the basis of our results, we could
hypothesise that, irrespective of students’ cognitive ability level,
664
S. Agnoli et al. / Personality and Individual Differences 53 (2012) 660–665
trait EI could enable children to cope more effectively with the
sources of stress associated with math subjects.
In addition, while previous studies detected a direct influence of
specific emotional ability (O’Connor & Little, 2003) on scholastic
success, the results of this study also showed an interaction effect
between emotion recognition ability and cognitive ability in predicting both language and math performance. Similarly to the
trend that emerged in trait EI results, we found that higher emotion recognition ability was associated with better language and
math performance only in children characterised by low and medium cognitive ability. In contrast, students’ ability to recognise
emotions did not affect language and math performance at high
levels of cognitive ability. On the basis of these results, we could
assume that when the task demands outweigh the student’s cognitive resources, the child could draw both on basic (emotional perception) and on more complex (trait EI) emotional resources.
Consistently with the independence of trait EI and emotional
abilities (i.e., ability EI) constructs, which has been sustained and
demonstrated in previous studies (Petrides, 2011), our data did
not show any relationships between trait EI and recognition ability.
Moreover, an interesting difference between trait EI and recognition ability emerged regarding their effect on math performance.
The present study showed that, differently from emotion recognition ability, trait EI had a unique power to predict math performance. In other words, while recognition ability affected math
performance only at low and medium levels of cognitive ability
and did not have any effect at high levels of cognitive ability, the
predictive value of trait EI on math performance was independent
from cognitive ability. According to the four branches model of EI
(Mayer et al., 2008), the emotional abilities are arranged from more
basic psychological processes to higher, more psychologically integrated processes. The emotion recognition ability is a low-level
skill, and represents a relatively simple ability. Our findings suggested that this basic ability is effective only at low and medium
levels of cognitive ability. Since math is a complex subject and its
performance is often associated with negative feelings (e.g., math
anxiety), high performance in this subject may require more complex resources when also intellectual abilities are insufficient for
facing up to the subject’s problematic requests. From this point
of view, trait EI represents a more complex construct than just
emotion recognition ability, in that it encompasses various dispositions from the personality domain (e.g., empathy, impulsivity,
and assertiveness) as well as elements of social intelligence and
personal intelligence, the latter two in the form of self-perceived
abilities (Petrides & Furnham, 2000; Petrides et al., 2004).
Finally, some limitations should be addressed. Although other
studies have explored, among the various emotional abilities, only
the emotion recognition ability (e.g., Mavroveli et al., 2009), this
emotional process is probably not exhaustive in explaining the
predictability of ability EI on scholastic performance. Further studies are needed to explore the effects of the complex ability EI construct on scholastic achievement during development and their
incremental validity over trait EI. However, a necessary preliminary step is the development of ability EI measurement methods
suitable for use in youths and children and their validation in different countries. Without the diffusion of these instruments, it is
impossible to understand whether ability EI could be considered
an useful predictor of scholastic achievement during childhood.
Moreover, further studies are needed to understand the generalisability of these results across different ages and cultures. Despite
these limitations, the present study represents a first exploration
of the utility of a multi-method approach to explore how emotional dispositions and abilities influence scholastic achievement.
This approach may represent a significant step towards understanding the individual differences emerging during childhood in
scholastic performance.
Acknowledgement
We wish to thank K.V. Petrides for his helpful comments on previous versions of the draft.
Appendix A
Equations to test the effect of emotional intelligence and cognitive ability on language and math performance.
Individual level:
Y ij ¼ b0j þ b1j ðageÞ þ b2j ðgenderÞ þ b3j ðcognitive abilityÞ
þ b4j ðemotion recognition abilityÞ þ b5j ðtrait EIÞ
þ b6j ðemotion recognition ability cognitive abilityÞ
þ b7j ðtrait cognitive abilityÞ þ eij
Class level:
b0j ¼ c00 þ u0j
bij ¼ cij for 1 < i < 7
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. Newbury Park, CA: Sage.
Ashcraft, M. H. (2002). Math anxiety: Personal, educational, and cognitive
consequences. Current Directions in Psychological Science, 11, 181–185. http://
dx.doi.org/10.1111/1467-8721.00196.
Belacchi, C., Scalisi, T. G., Cannoni, E., & Cornoldi, C. (2008). CPM – Coloured
Progressive Matrices. Firenze: Giunti O.S. Organizzazioni Speciali.
Camras, L. A., & Allison, K. (1985). Children’s understanding of emotional facial
expressions and verbal labels. Journal of Nonverbal Behavior, 9, 84–94. http://
dx.doi.org/10.1007/BF00987140.
Colom, R., & Flores-Mendoza, C. E. (2007). Intelligence predicts scholastic
achievement irrespective of SES factors: Evidence from Brazil. Intelligence, 35,
243–251. http://dx.doi.org/10.1016/j.intell.2006.07.008.
Di Fabio, A., & Palazzeschi, L. (2009). An in-depth look at scholastic success: Fluid
intelligence, personality traits or emotional intelligence? Personality & Individual
Differences, 46, 581–585. http://dx.doi.org/10.1016/j.paid.2008.12.012.
Downey, L. A., Mountstephen, J., Lloyd, J., Hansen, K., & Stough, C. (2008). Emotional
intelligence and scholastic achievement in Australian adolescents. Australian
Journal of Psychology, 60(1), 10–17.90. http://dx.doi.org/10.1080/000495307
01449505.
Extremera, N., & Fernández-Berrocal, P. (2006). Emotional intelligence as predictor
of mental, social, and physical health in university students. The Spanish Journal
of Psychology, 9(1), 45–51.
Farsides, T., & Woodfield, R. (2003). Individual differences and undergraduate
academic success: The roles of personality, intelligence, and application.
Personality and Individual Differences, 34, 1225–1243. http://dx.doi.org/
10.1016/S0191-8869(02)00111-3.
Herba, C. M., Landau, S., Russell, T., Ecker, C., & Phillips, M. L. (2006). The
development of emotion-processing in children: Effects of age, emotion, and
intensity. Journal of Child Psychology and Psychiatry, 47, 1098–1106. http://
dx.doi.org/10.1111/j.1469-7610.2006.01652.x.
Hyde, J. S., Fennema, E., Ryan, M., Frost, L. A., & Hopp, C. (1990). Gender comparison of
mathematics attitudes and affect: A meta-analysis. Psychology of Women Quarterly,
14(3), 299–324. http://dx.doi.org/10.1111/j.1471-6402.1990.tb00022.x.
Laborde, S., Dosseville, F., & Scelles, N. (2010). Trait emotional intelligence and
preference for intuition and deliberation: Respective influence on academic
performance. Personality & Individual Differences, 49, 784–788. http://dx.doi.org/
10.1016/j.paid.2010.06.031.
Lee, V. E. (2000). Using hierarchical linear modeling to study social contexts: The
case of school effects. Educational Psychologist, 35, 125–141. http://dx.doi.org/
10.1207/S15326985EP3502_6.
Ludqvist, D., Flykt, A., & Ohman, A. (1998). The Karolinska Directed Emotional Faces.
Stockholm: Karolinska Institutet.
Mavroveli, S., Petrides, K. V., Sangareau, Y., & Furnham, A. (2009). Exploring the
relationship between trait emotional intelligence and objective socio-emotional
outcomes in childhood. British Journal of Educational Psychology, 79, 259–272.
http://dx.doi.org/10.1348/000709908X368848.
Mavroveli, S., Petrides, K. V., Shove, C., & Whitehead, A. (2008). Validation of the
construct of trait emotional intelligence in children. European Child & Adolescent
Psychiatry, 17, 516–526. http://dx.doi.org/10.1007/s00787-008-0696-6.
Mavroveli, S., & Sanchez-Ruiz, M. J. (2011). Trait emotional intelligence influences
on academic achievement and school behaviour. British Journal of Educational
Psychology, 81, 112–134. http://dx.doi.org/10.1348/2044-8279.002009.
S. Agnoli et al. / Personality and Individual Differences 53 (2012) 660–665
Mayer, J. D., Salovey, P., & Caruso, D. R. (2008). Emotional intelligence. New ability
or eclectic traits? American Psychologist, 63(6), 503–517. http://dx.doi.org/
10.1037/0003-066X.63.6.503.
Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., Cesi, S. J., et al.
(1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 77–101.
http://dx.doi.org/10.1037/0003-066X.51.2.77.
O’Connor, R. M., Jr., & Little, I. S. (2003). Revisiting the predictive validity of emotional
intelligence. Self-report versus ability-based measures. Personality and Individual
Differences, 35, 1893–1902. http://dx.doi.org/10.1016/S0191-8869(03)00038-2.
Parker, J. D. A., Creque, S., Barnhart, D. L., Harris, J. I., Majeski, S. A., Wood, L. M., et al.
(2004). Academic achievement in high school: Does emotional intelligence
matter? Personality and Individual Differences, 37, 1321–1330. http://dx.doi.org/
10.1016/j.paid.2004.01.002.
Parker, J. D. A., Summerfeldt, L. J., Hogan, M. J., & Majeski, S. A. (2004). Emotional
intelligence and academic success: Examining the transition from high school
to university. Personality and Individual Differences, 36, 163–173. http://
dx.doi.org/10.1016/S0191-8869(03)00076-X.
Petrides, K. V. (2011). Ability and trait emotional intelligence. In T. ChamorroPremuzic, A. Furnham, & S. von Stumm (Eds.), The Blackwell-Wiley handbook
of individual differences. New York: Wiley.
Petrides, K. V., Frederickson, N., & Furnham, A. (2004). The role of trait emotional
intelligence in academic performance and deviant behavior at school.
Personality and Individual Differences, 36, 277–293. http://dx.doi.org/10.1016/
S0191-8869(03)00084-9.
Petrides, K. V., & Furnham, A. (2000). On the dimensional structure of emotional
intelligence. Personality and Individual Differences, 29, 313–320. http://
dx.doi.org/10.1016/S0191-8869(99)00195-6.
665
Petrides, K. V., Furnham, A., & Mavroveli, S. (2007). Trait emotional intelligence.
Moving forward in the field of EI. In G. Matthews, M. Zeidner, & R. Roberts
(Eds.), Emotional intelligence. Knowns and unknowns (series in affective science).
Oxford: Oxford University Press.
Petrides, K. V., Pita, R., & Kokkinaki, F. (2007). The location of trait emotional
intelligence in personality factor space. British Journal of Psychology, 98,
273–289. http://dx.doi.org/10.1348/000712606X120618.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and
data analysis methods (2nd ed.). Newbury Park, CA: Sage.
Raven, J. (1981). Manual for Raven’s Progressive Matrices and mill hill vocabulary
scales. Oxford: Oxford Psychologists Press.
Raven, J., Raven, J. C., & Court, J. H. (2000). Manual for Raven’s Progressive Matrices
and vocabulary scales. Oxford, UK: Oxford Psychologists Press.
Russo, P. M., Mancini, G., Trombini, E., Baldaro, B., Mavroveli, S., & Petrides, K. V.
(2012). Trait emotional intelligence and the Big Five: A study on Italian children
and preadolescents. Journal of Psychoeducational Assessment, 30, 274–283.
http://dx.doi.org/10.1177/0734282911426412.
Song, L. J., Huang, G., Peng, K. Z., Law, K. S., Wong, C., & Chen, Z. (2010). The
differential effect of general mental ability and emotional intelligence on
academic performance and social interactions. Intelligence, 38, 137–143. http://
dx.doi.org/10.1016/j.intell.2009.09.003.
Windingstad, S., McCallum, R. S., Mee Bell, S., & Dunn, P. (2011). Measures of
emotional intelligence and social acceptability in children: A concurrent
validity study. Canadian Journal of School Psychology, 26, 107–126. http://
dx.doi.org/10.1177/0829573511406510.