APA NLM

tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

Psychological Bulletin
© 2019 American Psychological Association 2019, Vol. 1, No. 999, 000
ISSN: 0033-2909 [Link]

AQ: 1 Emotional Intelligence Predicts Academic Performance: A Meta-Analysis

AQ: au Carolyn MacCann, Yixin Jiang, Kit S. Double

and Luke E. R. Brown University of Oxford
The University of Sydney

Micaela Bucich Amirali Minbashian

AQ: 2 The University of Sydney University of New South Wales Sydney
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Schools and universities devote considerable time and resources to developing students’ social and
emotional skills, such as emotional intelligence (EI). The goals of such programs are partly for personal
development but partly to increase academic performance. The current meta-analysis examines the
degree to which student EI is associated with academic performance. We found an overall effect of ␳ ⫽
.20 using robust variance estimation (N ⫽ 42,529, k ⫽ 1,246 from 158 citations). The association is
significantly stronger for ability EI (␳ ⫽ .24, k ⫽ 50) compared with self-rated (␳ ⫽ .12, k ⫽ 33) or mixed
EI (␳ ⫽ .19, k ⫽ 90). Ability, self-rated, and mixed EI explained an additional 1.7%, 0.7%, and 2.3% of the
variance, respectively, after controlling for intelligence and big five personality. Understanding and manage-
ment branches of ability EI explained an additional 3.9% and 3.6%, respectively. Relative importance analysis
suggests that EI is the third most important predictor for all three streams, after intelligence and conscien-
tiousness. Moderators of the effect differed across the three EI streams. Ability EI was a stronger predictor of
performance in humanities than science. Self-rated EI was a stronger predictor of grades than standardized test
scores. We propose that three mechanisms underlie the EI/academic performance link: (a) regulating academic
emotions, (b) building social relationships at school, and (c) academic content overlap with EI. Different
streams of EI may affect performance through different mechanisms. We note some limitations, including the
lack of evidence for a causal direction.

Public Significance Statement

This meta-analysis shows that emotional intelligence has a small to moderate association with
academic performance, such that students with higher emotional intelligence tend to gain higher
grades and achievement test scores. The association is stronger for skill-based emotional intelligence
tasks than rating scales of emotional intelligence. It is strongest for skill-based tasks measuring
understanding emotions and managing emotions.

Keywords: academic performance, emotional intelligence, intelligence, meta-analysis, personality

AQ: 3 Supplemental materials: [Link]

Emotional intelligence (EI) has captured the public imagination, Jin, Newman, & O’Boyle, 2015; Joseph & Newman, 2010;
and rightly so. Recent meta-analyses clearly demonstrate that O’Boyle, Humphrey, Pollack, Hawver, & Story, 2011), and have
emotionally intelligent people perform better in their jobs (Joseph, better health and wellbeing outcomes (Martins, Ramalho, &
Morin, 2010; Schutte, Malouff, Thorsteinsson, Bhullar, & Rooke,
2007). In education, there is a growing consensus among educa-
X Carolyn MacCann, Yixin Jiang, and Luke E. R. Brown, School of
tors, researchers, and policymakers that EI is an important skill for
Psychology, The University of Sydney; Kit S. Double, Department of students to develop, both for their future wellbeing as well as their
Education, University of Oxford; Micaela Bucich, School of Psychology, future workplace success. Although there is evidence that social
The University of Sydney; Amirali Minbashian, School of Management, and emotional learning programs in school are effective (Durlak,
AQ: 15 UNSW Business School, University of New South Wales Sydney. Weissberg, Dymnicki, Taylor, & Schellinger, 2011) and that non-
This research was funded in part by an Australian Research Council cognitive constructs are powerful predictors of academic perfor-
Discovery Grant (DP150101158) awarded to Carolyn MacCann and Ami-
mance (Poropat, 2009; Richardson, Abraham, & Bond, 2012),
rali Minbashian.
Correspondence concerning this article should be addressed to Carolyn there is not yet a large-scale meta-analysis examining the extent to
MacCann, School of Psychology, The University of Sydney, Paramatta Road, which EI correlates with academic performance. The current arti-
Sydney, NSW 2037, Australia. E-mail: [Link]@[Link] cle provides the first comprehensive large-scale meta-analyses

1
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

2 MACCANN ET AL.

estimating the extent to which EI predicts academic performance. rated EI assessments based on this model), and the major mixed
We consider all major conceptualizations of EI, all stages of models of EI (and the mixed model assessments based on these).
education (from elementary school through to university), and the
different facets of EI. We also examine the incremental validity of Ability EI: A Four-Branch Hierarchical Model of
EI above and beyond the traditional psychological characteristics Emotional Skills
known to predict academic performance (intelligence and the five
major personality domains). There is general agreement on a single theoretical model that
describes the component abilities of EI. The hierarchical four-
branch model was first described by Mayer and Salovey in 1997.
AQ: 4 Emotional Intelligence
This model outlines four key branches of emotion-related abilities
Emotional intelligence (EI) is a relatively new construct com- that range in complexity from low-level information processing to
pared with intelligence or personality, with the first academic strategic and deliberative use of emotional information to meet
article appearing in 1990 (Salovey & Mayer, 1990). The concept personal goals. These four branches are: (a) perceiving emotions
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

was relatively unknown until it was popularized by science jour- accurately, (b) using emotions to facilitate decision-making,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

nalist Daniel Goleman in his 1995 book Emotional Intelligence: (c) understanding emotions, and (d) managing emotions to up-
Why It Can Matter More Than IQ. This book sparked massive regulate positive emotions and down-regulate negative emotions.
interest from researchers and the general public in the late 1990s. We describe these in detail below. The best-known assessment of
One effect of this sudden widespread popularity was that research these four branches is the ability-based Mayer-Salovey-Caruso
teams commenced their work in parallel, creating their own theo- Emotional Intelligence Test (MSCEIT, Mayer, Salovey, Caruso, &
ries and assessments rather than building on existing research. For Sitarenios, 2003), which has two subtests for each of the four
much of the 1990s there was little agreement on how to define or branches. The MSCEIT is the only commercially available ability
measure EI, leading to many different theories and measures that EI measure and is the most commonly used ability measure in
were often quite dissimilar from each other (Davies, Stankov, & research. Although there are several noncommercial alternative
Roberts, 1998). To bring some clarity to the field, researchers ability EI assessments, these tend to measure only one or two of
suggested that a distinction should be made between two kinds of the four branches (e.g., Freudenthaler & Neubauer, 2007; Mac-
measurement models—ability scales and rating scales (Mayer, Cann & Roberts, 2008; Matsumoto et al., 2000). Earlier research
Caruso, & Salovey, 2000). Ability scales require test-takers to on the four-branch model also used the Multifactor Emotional
demonstrate knowledge or to process emotion-related information Intelligence Scale (MEIS), the precursor to the MSCEIT (Mayer,
to provide a response. Rating scales require test-takers to rate their Caruso, & Salovey, 1999). Youth versions of the MSCEIT and
agreement with a series of statements about themselves (e.g., “I am MEIS have often been used for research in schools (Rivers et al.,
able to handle most upsetting problems”; Brackett, Rivers, Shiff- 2012).
man, Lerner, & Salovey, 2006). Evidence to date suggests that In addition to the ability-based assessments, there are several
rating scales and ability scales of EI capture different constructs assessments that use rating-scales to assess self-rated ability EI.
and are only weakly related to each other (Brackett & Mayer, One of the earliest measures of this kind was the 33-item Assess-
2003; Brackett et al., 2006). ing Emotions Scale (AES; Schutte et al., 1998). The AES was
Paralleling the distinction between two measurement models is based on an earlier definition of EI that predated the four-branch
a similar distinction of two theoretical models—mixed model and hierarchical model. The early definition included perceiving, us-
ability model theories of EI. Mixed model conceptualizations of EI ing, and managing emotions, but did not include understanding
include a broad mix of constructs that lead to emotionally intelli- emotions (Salovey & Mayer, 1990). Because the AES was avail-
gent behavior, including emotion-related abilities, character traits, able early in the public domain, it was frequently used in EI
and motivational elements (Bar-On, 2006; Petrides, Pita, & Kok- research. Another EI measure that used this early definition is
kinaki, 2007). In contrast, ability models of EI conceptualize EI as Wong’s Emotional Intelligence Scale (WEIS), which contains four
a cognitive ability of a similar type to verbal ability or quantitative subscales that assess perceiving one’s own emotions, perceiving
ability, with the content domain as emotions rather than words or others’ emotions, using emotions, and managing emotions (i.e., it
numbers (MacCann, Joseph, Newman, & Roberts, 2014). Ashka- does not include emotion understanding, in line with the earlier
nasy and Daus (2005) distinguished between rating scales based on definition of EI; Law, Wong, & Song, 2004). A rating-scale
ability theories and those based on mixed-model theories. They instrument designed specifically after the four-branch model is the
refer to three ‘streams’ of EI measures: (a) ability scales, (b) Self-Rated Emotional Intelligence Scale (SREIS; Brackett et al.,
ratings of EI abilities (self-perceptions of EI, sometimes referred to 2006), which contains 19 items that assess five subscales (perceiv-
as emotional self-efficacy; Qualter, Gardner, Pope, Hutchinson, & ing emotions, using emotions, understanding emotions, managing
Whiteley, 2012), and (c) ratings of mixed model EI (often referred one’s own emotions, and managing others’ emotions). The four
to as trait EI, after the major mixed model conceptualization; branches of EI are described in detail below.
Furnham & Petrides, 2003; Petrides & Furnham, 2000, 2003). In Emotion perception is the ability to “identify emotional content
the current meta-analysis, we separately consider results for these in faces, voices, and designs and ability to accurately express
three different types of assessments, based on theoretical and emotions” (Mayer, Caruso, & Salovey, 2016). Theoretically, this
empirical evidence that these are three separate constructs (Joseph branch includes several related abilities, including (a) the ability to
& Newman, 2010; O’Boyle et al., 2011). We refer to these as identify emotions in external stimuli, (b) the ability to identify
ability EI, self-rated EI, and mixed EI. The paragraphs below one’s own emotions (i.e., internal stimuli), (c) the ability to express
describe the major ability model of EI (and the ability and self- one’s own emotions accurately, (d) the ability to distinguish be-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 3

tween genuine emotion expressions and deceptive or forced ex- emotion management (the emotion management branch is some-
pressions, and (e) knowledge of display rules for emotion expres- times also referred to as emotion regulation; e.g., Joseph & New-
sion in different cultures and contexts (Mayer et al., 2016; Mayer man, 2010; Mayer & Salovey, 1997).
& Salovey, 1997). However, this branch has been operationalized Emotion understanding encompasses one’s knowledge base re-
solely as the first of these—tests assess the capacity to identify the garding emotions and emotion-related phenomena. It is the “cen-
type and extent of emotion present in external stimuli such as tral locus of abstract processing and reasoning about emotions and
facial expressions, microexpressions, tone-of-voice, body pos- emotional information” (Mayer et al., 2001, p. 235). It includes the
tures, landscapes and evocative art (Matsumoto et al., 2000; Mayer following types of emotion knowledge: the vocabulary of emotion
et al., 2003; Nowicki & Duke, 1994; Schlegel, Grandjean, & terms, the antecedents and consequences of emotions, the way
Scherer, 2014). As such, the empirical basis for what is known emotions may combine or change over time, and the likely effect
about emotion perception is solely defined as individual differ- of a specific situation on one’s emotions now or in the future
ences in identifying emotions in others (and not the wider array of (Mayer et al., 2016; Mayer & Salovey, 1997). Emotion under-
abilities that may theoretically be included). Once emotions are standing may be considered as domain-specific knowledge for the
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

perceived, this emotion information acts as input for the cognitive content domain of emotions. Emotion understanding shows the
This document is copyrighted by the American Psychological Association or one of its allied publishers.

system (Mayer, Salovey, Caruso, & Sitarenios, 2001). strongest links to conventional cognitive abilities of the four EI
Emotion facilitation of thought involves the use of emotions and branches, with meta-analytic estimates ranging from ␳ ⫽ .39 to .42
emotional information as input or guidance in cognitive tasks or (Joseph & Newman, 2010; MacCann, 2010; Olderbak, Semmler,
decisions. It has been defined as the ability to “facilitate thinking & Doebler, 2019; Roberts, Schulze, & MacCann, 2008). AQ: 17
by drawing on emotions as motivational and substantive inputs” Emotion management is the ability to manage emotions in
(Mayer et al., 2016, p. 296). Both the theory and measurement of oneself and others by upregulating positive emotions and down-
this branch involve two key elements: (a) using existing emotions regulating negative emotions to achieve a desired outcome such as
to guide task selection or approaches to tasks, and (b) generating personal growth (Mayer et al., 2001, 2016; Mayer & Salovey,
new emotions to aid performance on a specific task. When using 1997). There are four key elements to emotion management. First,
existing emotions, a person uses their current emotional state as a this branch involves managing both one’s own and others’ emo-
critical task parameter to guide the strategies or processes used in tions (termed intrinsic and extrinsic emotion regulation in the
problem solving in two ways. First, emotions can direct attention process model of emotion regulation; Gross, 2008; Gross &
to critical information through the action tendencies associated Thompson, 2007). Second, this branch includes both (a) knowl-
with each emotion. For example, positive affect relates to a broad edge of emotion management and (b) metacognitive strategies
rather than narrow outlook and may lead to creative exploration pertaining to emotion management, such as the ability to “monitor
whereas anxiety is associated with hypervigilance to threat (Bar- emotional reactions” and to “evaluate strategies to maintain, re-
Haim, Lamy, Pergamin, Bakermans-Kranenburg, & van IJzen- duce, or intensify an emotional response” (Mayer et al., 2016, p.
doorn, 2007; Fredrickson, 2001). Second, tasks can be selected to
294). Third, emotions are managed with respect to personal goals.
take advantage of a mood state that might help performance. For
That is, up- or downregulation of emotion is undertaken strategi-
example, one could choose to write an enthusiastic welcome
cally to achieve goals such as personal growth. As such, emotion
e-mail when in a happy mood but wait to counsel a disgruntled
management represents not only the knowledge of how emotions
employee until one is feeling more serious (i.e., the emotion
are managed, but the motivational elements that determine when
regulation strategy of situation selection). The MSCEIT contains
and why emotions are managed. That is, emotion management
two subtests assessing facilitation. The Sensations test assesses the
“interface(s) with personality and personal goals” (Mayer et al.,
generation of emotions—test-takers must generate an emotion and
2001, p. 235). Of the four EI branches, emotion management
rate the similarity of their sensory experience to sensations such as
shows the strongest relationships to personality traits, particularly
hot, red, or quiet. The Facilitation test assesses knowledge of
agreeableness (␳ ⫽ .29; Joseph & Newman, 2010).
which mood states will be most helpful in different types of tasks.
Given the known relationships of ability EI with both intelli-
The facilitation branch has been criticized on both empirical and
gence and personality, some researchers have argued that ability EI
theoretical grounds. Empirically, factor analyses have generally
shows little incremental prediction of key outcomes above and
not supported the inclusion of a clear and distinct facilitation
branch, possibly because of the dual nature of facilitation as both beyond the effects of personality (e.g., Schulte, Ree, & Carretta,
emotion generation and situation selection (Fan, Jackson, Yang, 2004). For this reason, it is essential to provide evidence of
Tang, & Zhang, 2010; Palmer, Gignac, Manocha, & Stough, 2005; incremental validity when considering whether EI predicts aca-
Rossen, Kranzler, & Algina, 2008). Theoretically, emotion facil- demic performance. We will therefore control for personality and
itation seems like a subset of emotion management (the fourth intelligence in the prediction of academic performance from EI.
branch). Emotion generation is a core concept for emotion man-
agement, as managing emotions involves the ability to generate the Mixed Models of EI
desired emotion to match the task at hand (mostly, but not always,
the upregulation of positive emotions and downregulation of neg- While there are many mixed models of EI, the three major
ative emotions; Joseph & Newman, 2010; MacCann et al., 2014; conceptualizations are Goleman’s emotional competence (Gole-
Mestre, MacCann, Guil, & Roberts, 2016). The other element of man, 1998), Bar-On’s emotional and social competence (Bar-On,
emotion facilitation—situation selection—is a well-known emo- 2006), and Petrides and Furnham’s trait emotional intelligence
tion regulation strategy (Gross & Thompson, 2007; Werner & (Petrides, Perez-Gonzalez, & Furnham, 2007; Petrides, Pita, et al.,
Gross, 2009), and thus might also be considered a key element of 2007). We describe these below.
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

4 MACCANN ET AL.

Emotional competence. Goleman’s model, first outlined in tators of social and emotional competence (Bar-On, 2000, 2006).
his 1998 book Working with Emotional Intelligence, consists of A youth version of the EQ-i is also available, and the EQ-i has
four major competencies: (a) self-awareness (being aware of one’s frequently been used in peer-reviewed research linking EI to
emotions, accurate in one’s self-assessments, and self-confident); academic performance. EQ-i scores correlate very highly with
(b) self-management (being conscientious, trust-worthy, adaptable, personality traits. For example, EQ-i total scores correlate at ⫺.62
achievement-oriented, and able to control one’s emotions and to ⫺.72 with neuroticism (Dawda & Hart, 2000), .52 to .56 with
behaviors); (c) social awareness (showing empathy to others, and extraversion, and ⫺.76 with anxiety (a facet of neuroticism;
having a service orientation and organizational awareness); and (d) Dawda & Hart, 2000; O’Connor & Little, 2003). Correlations of
social skills (being skilled in leadership, communication, influ- this magnitude have led some researchers to claim that EI (espe-
ence, conflict management, building relationships, showing good cially mixed-model conceptualizations) represents “old wine in a
teamwork and collaboration skills, as well as the ability to mentor new bottle”—that is, a rebranding of personality rather than a new
others; Boyatzis, Goleman, & Rhee, 2000; Goleman, 1998). That and distinct construct (Matthews et al., 2002).
is, the model distinguishes between awareness (the tendencies and Trait EI. Trait EI is the most comprehensive mixed model of
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

abilities to detect essential emotional information in oneself and EI, consisting of 15 facets taken from both the ability model of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

one’s environment) and management (being able to change or emotion as well as the two models of emotional competence
regulate the social and emotional content of oneself and one’s described above (Petrides, Pita, et al., 2007; Petrides, 2009). The
surroundings) as they are applied to the self and to others. This four ability facets in this model are (a) accurately perceiving
emotional competence model is the basis for the emotional com- emotions in oneself and others, (b) expressing and communicating
petence inventory (ECI), a rating-scale inventory commercially emotions clearly, (c) managing others’ emotions, and (d) regulating
available to assess EI. There is relatively little peer-reviewed one’s own emotions. The nonability facets include adaptability, as-
empirical evidence evaluating the ECI, and it has been criticized sertiveness, low impulsivity, fulfilling personal relationships, self-
based on a lack of content validity and predictive validity evidence esteem, self-motivation, social awareness, stress management, trait
(Landy, 2005; Matthews, Zeidner, & Roberts, 2002). Nevertheless, empathy, trait happiness, and trait optimism. The trait EI model is
the theoretical model has been very influential in both business and assessed with the Trait Emotional Intelligence Questionnaire
education settings. For instance, this model formed the basis for (TEIQue; Petrides, Pita, et al., 2007), which also has a short form,
the social and emotional learning (SEL) competency model used adolescent form, child form, and adolescent and child short forms.
by the Collaborative for Academic, Social, and Emotional Learn- The TEIQue is very frequently used in EI peer-reviewed research.
ing (CASEL, 2003; Durlak et al., 2011). CASEL’s SEL model TEIQue scores show very high correlations with the five major
includes the four Goleman competencies as well as a fifth com- domains of personality (Matthews et al., 2002; Mayer, Roberts, &
petency of responsible decision making (being able to make con- Barsade, 2008). Some researchers have argued that this is a prob-
structive choices about personal behavior and social interactions). lem for the discriminant validity of trait EI—that trait EI is in fact
This model is widely used internationally to guide educational
indistinguishable from personality. A recent meta-analysis pro-
interventions designed to increase student, class and school-level
vides empirical justification for this idea, showing that trait EI
socioemotional competencies.
correlates at .85 with a general personality factor derived from the
Social and emotional competence. Bar-On’s model of social
five major domains of personality, suggesting that these two con-
and emotional competence was developed to represent “key com-
structs are “very similar, perhaps even synonymous” (p. 36, van
ponents of effective emotional and social functioning that lead to
der Linden et al., 2017). Given that personality is known to predict
psychological well-being” (Bar-On, 2000, p. 364). There are five
academic performance, the question of whether trait EI can add
major domains of emotional and social competence—interpersonal
anything to this prediction is therefore important. In the current
competence, intrapersonal competence, stress management, adapt-
article, we thus examine whether trait EI predicts academic per-
ability, and general mood (Bar-On, 2006). The Emotional Quotient
formance over and above the effects of personality.
Inventory (EQ-i) instrument is based on this model, and contains
15 subscales that are unevenly distributed across these five do-
mains: (a) intrapersonal competence (self-awareness and self- Emotional Intelligence and Academic Performance
expression) contains five subscales—self-regard, emotional self-
awareness, assertiveness, independence, and self-actualization; (b) There is ample evidence to suggest that EI has a positive
interpersonal competence (social awareness and interpersonal re- association with academic performance. Social and emotional
lationships), contains three subscales— empathy, social responsi- learning programs (which are broadly based on Goleman’s model
bility, and interpersonal relationship; (c) stress management (emo- of EI) are known to increase academic performance, with Durlak
tion management and regulation) contains two subscales—stress et al.’s (2011) meta-analysis showing that such programs result in
tolerance and impulse control; (d) adaptability (change manage- an 11-percentile improvement in academic performance. Social
ment) contains three subscales—reality testing, flexibility, and and emotional learning focuses on developing five key competen-
problem solving; and (e) general mood (self-motivation) contains cies that overlap substantially with Goleman’s emotional compe-
two subscales— optimism and happiness (Bar-On, 2006). Explor- tencies (self-awareness, social awareness, self-management, rela-
atory and confirmatory factor analysis of the items provided sup- tionship skills, and responsible decision making; CASEL, 2003).
port for only 10 of the 15 subscales. Both of the general mood Programs were more effective if they followed a sequenced, step-
subscales, two of the five intrapersonal competence subscales and by-step approach, used active forms of learning, allowed adequate
the social responsibility subscale of interpersonal competence time for skill development, and had explicit learning goals (Durlak
were not supported, but were retained in the instrument as facili- et al., 2011). The effect of SEL programs on academic perfor-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 5

mance was stronger when teachers ran the programs (d ⫽ .34) ing a correlation matrix of EI, intelligence, personality and aca-
compared with nonschool personnel (d ⫽ .12). demic performance. The correlations in this matrix are drawn from
There is also some direct evidence that EI is positively associ- (a) our original meta-analyses (EI/academic performance and in-
ated with academic performance. Three meta-analyses to date have telligence/academic performance correlations) and (b) previously
examined this question, and all have found a positive association. published meta-analyses (personality/EI, personality/performance,
First, Van Rooy and Viswesvaran (2004) estimated a corrected intelligence/performance, intelligence/personality, and the rela-
correlation of .10 between EI and academic performance (k ⫽ 10). tions among personality domains; Joseph & Newman, 2010;
This analysis did not distinguish between the different streams of Judge, Jackson, Shaw, Scott, & Rich, 2007; Poropat, 2009; van der
EI. Second, Perera and DiGiacomo (2013) examined rating scales Linden et al., 2017; van der Linden, te Nijenhuis, & Bakker, 2010).
of EI, finding a corrected correlation of .20 with academic perfor- Moreover, we use relative importance analysis to compare the
mance (k ⫽ 48). Effects were stronger for younger students and at relative contribution of intelligence, personality, and EI to explain-
earlier levels of education. Other moderators were examined were ing differences in academic performance (Johnson, 2000; LeBre-
not significant. These included (a) the gender composition of the ton & Tonidandel, 2008). That is, we are building a more com-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

sample, (b) the instrument used, and (c) whether the sample was in prehensive picture of the extent to which EI predicts academic
This document is copyrighted by the American Psychological Association or one of its allied publishers.

a transition year (i.e., first year of high school or university). performance than ever before, including a consideration of the
Perera and DiGiacomo did not include ability scales in their relative importance of EI compared with other well-known pre-
analysis nor distinguish between mixed EI and self-rated EI. They dictors of academic performance. Our major hypothesis is that EI
also did not assess the incremental prediction of EI above the will be positively associated with academic performance (Hypoth-
known effects of personality and intelligence. Third, Richardson et esis 1).
al. (2012) examined the relationship between EI and academic
performance as part of a wide-sweeping meta-analytic review of Moderators of the EI/Academic Performance
42 noncognitive correlates of academic performance. They re-
Relationship
ported a slightly smaller relationship between EI and academic
performance (␳ ⫽ .17) but included only 14 studies and did not EI Stream. Although evidence suggests a positive association
differentiate between ability scales and rating scales. of EI with academic performance, it is not clear whether this
The current comprehensive meta-analysis expands on previous relationship differs for ability versus self-rated versus mixed EI.
work in five ways. First, we cover all relevant research whereas Meta-analyses predicting workplace performance and wellbeing
previous studies included only a small subset (i.e., we located 162 outcomes have found ifferent effects for these three streams (Harms
relevant citations, such that Richardson’s k of 14 studies represents & Credé, 2010; Joseph & Newman, 2010; Martins et al., 2010; Miao,
less than 10% of the available data). Second, we include ability- Humphrey, & Qian, 2017; O’Boyle et al., 2011; Sánchez-Álvarez,
based EI assessments as well as rating scales. The relation between Extremera, & Fernandez-Berrocal, 2016; Schutte et al., 2007). These
ability EI and academic performance has never previously been findings, summarized in Table 1, cut across multiple outcomes: T1
reported in meta-analyses, despite being the most objective and workplace performance, organizational citizenship behaviors,
arguably most valid assessments of EI (Matthews et al., 2002; counterproductive workplace behaviors, leadership, subjective
Mayer et al., 2008). Third, we use the now-standard categorization wellbeing, and health outcomes. Across all major meta-analyses
of EI scales into ability EI, self-rated EI, and mixed EI to sepa- linking EI to positive life outcomes, there are two consistent
rately examine the effects of EI on academic performance across findings. First, the relationship of EI to positive outcomes is
these three different constructs (cf. Ashkanasy & Daus, 2005; uniformly positive, showing that high EI confers benefits to those
Joseph & Newman, 2010; O’Boyle et al., 2011). Fourth, we who possess it. Second, ability EI consistently shows the lowest
examine a range of moderators of the effects (described in more relationships with criteria of the three streams of EI. This differ-
detail in the sections below), including EI stream, EI facet, sample ence among streams may relate to method effects. To date, the
age, gender composition, and publication type. Fifth, and perhaps criteria examined in meta-analyses are dominated by rating scales.
most importantly, we examine the incremental validity of EI above With little objective data for criteria such as job performance,
and beyond the effects of personality and intelligence by construct- health outcomes, or wellbeing, higher criterion-correlation with

Table 1
Meta-Analytic Correlations of EI With Positive Life Outcomes, by EI Stream

Outcome Ability EI Self-rated EI Mixed EI

Workplace performance (Joseph & Newman, 2010) .18 .23 .47

Workplace performance (O’Boyle, Humphrey, Pollack, Hawver, & Story, 2011) .24 .30 .28
Organizational citizenship behaviors (Miao, Humphrey, & Qian, 2017) .17 .57 .48
Counter-productive workplace behaviors (Miao et al., 2017) .01 ⫺.38 ⫺.42
Transformational leadership (Harms & Credé, 2010) .24 .66a
Subjective well-being (Sánchez-Álvarez, Extremera, & Fernandez-Berrocal, 2016) .22 .32 .38
Health (Schutte, Malouff, Thorsteinsson, Bhullar, & Rooke, 2007) .11 .32a
Health (Martins, Ramalho, & Morin, 2010) .17 .34a
Note. EI ⫽ emotional intelligence.
a
These analyses did not distinguish between rating scales based on mixed models and rating scales representing self-ratings of ability EI.
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

6 MACCANN ET AL.

rating scales of EI (both mixed models and self-rated EI) compared subjective criteria such as participation marks, teachers may award
with ability scales is consistent with a mono-method bias in better marks to students who have formed better relationships with
measurement. That is, rating scales correlate more highly with them). This may also indirectly influence grades through providing
rating scales, irrespective of content. There is some support for this the student with a social support network that protects them in
idea of a method-effect component to the EI/outcome associations times of stress. Developing social relationships would require
in Harms and Credé’s (2010) meta-analysis of EI and leadership. emotion management (particularly managing social relation-
They found that effects were vastly stronger when EI and leadership ships—a subscale of the MSCEIT; Mayer et al., 2016). If this
were both rated by the same source (self, subordinate, peer, or super- mechanism accounts for the EI/academic performance relation-
visor) compared with different sources—␳ ⫽ .59 versus ␳ ⫽ .12. The ship, then we would again expect emotion management branch to
pattern of correlations in Table 1 also supports this argument. show stronger effects than the other branches.
Self-rated EI measures have much higher association with criteria Third, there may be an overlap between emotional competencies
as compared to ability EI measures. As both types of assessments and intellectual competencies. For example, knowing a lot of
are meant to measure the same underlying constructs, the differ- emotion words and being able to communicate one’s feelings
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ence must be attributable to method effects rather than construct could be conceptualized as a subset of vocabulary/verbal ability.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

AQ: 5 effects. Teaching emotional competencies may in fact result in teaching

Unlike the other outcomes in previous meta-analyses, academic academic competencies—learning the language of emotions in-
performance is rarely assessed with rating scales (either self- or creases vocabulary and spelling more generally, and interventions
observer-rated) but rather is an objective composite composed of involving written text may also improve reading comprehension.
course work, examination results, and participation (in the case of That is, one possibility is that there is nothing particularly special
course grades) or objective performance on standardized assess- about emotional skills—it is the overlap between emotional skills
ments (in the case of standardized test results). Although neither and other academic skills that mean students who are emotionally
course grades nor standardized test scores are problem-free as intelligent are also likely to be intelligent in other ways (e.g., good
valid measures of academic performance, their issues are qualita- with language). MacCann et al. (2014) proposed that EI may in
tively different to rating-scale measures of critical criteria. If some fact be one element of intelligence—a student can be smart with
of the predictive superiority of EI rating scales over EI ability emotions in the same way that they can be good with numbers or
scales is due to method bias (rather than substantive differences), good with words. Under this conceptualization, the overlap be-
then ability EI would show a stronger prediction of an objective tween EI and academic skills is the purported mechanism by which
criteria such as course grades, as compared to self-rated EI or EI would affect academic performance. If this mechanism ac-
mixed EI. We thus propose a directional hypothesis that ability EI counts for the relationship between EI and academic performance,
will show stronger prediction of performance as compared to we would expect the EI/academic performance association to (a)
self-rated or mixed EI (Hypothesis 2). be strongest for ability EI and weakest for mixed EI (providing
Facet of EI. There are three possible mechanisms that may further justification for Hypothesis 2), (b) decrease in size after
account for the relationship between EI and academic perfor- controlling for intelligence, and (c) be strongest for the knowledge-
mance. First, emotionally intelligent students may be able to deal based branch of EI (understanding emotions). In combination with
more easily with negative emotions elicited by academic settings. the previously described mechanisms regarding regulating aca-
The prototypical academic emotion is test anxiety, but there are a demic emotions and building social relationships, we thus hypoth-
variety of other emotions specific to academic settings (Pekrun, esize that effects will be strongest for the understanding and
Elliot, & Maier, 2009; Pekrun, Goetz, Titz, & Perry, 2002). For management branches of EI than the perception and facilitation
example, students need to regulate the disappointment of lower- branches (Hypothesis 3).
than-expected test scores or negative feedback, or the boredom Peer-reviewed journal article status. The recent replication
involved in learning concepts and subject matter that are of instru- crisis in psychology and other sciences based on inferential statis-
mental rather than intrinsic interest (e.g., learning a tax code to tics demonstrates that published articles may represent selective
pass an accountancy exam; Pekrun, Goetz, Daniels, Stupnisky, & reporting, where effects reported in peer-reviewed publications are
Perry, 2010). Students with the emotion management skills to systematically higher than studies that remained unpublished
down-regulate their anxiety, disappointment or boredom will be (Open Science Collaboration, 2015). Along with other question-
able to achieve better exam results, and to learn more from able research practices (such as selective analysis), selective re-
negative feedback or boring subject matter. If this mechanism porting may overestimate effect sizes to such an extent that pub-
accounts for the EI/academic performance association, the emotion lished estimates are up to twice as large as the real effects (Open
management branch should show the strongest association with Science Collaboration, 2015). To assess whether there is a file-
academic performance. drawer problem where nonsignificant results remain unpublished,
Second, social demands are present at all stages of education, we will test for any differences between peer-reviewed publica-
from sharing crayons in kindergarten, resisting peer pressure and tions and unpublished data (e.g., unpublished data sets, dissertation
managing group projects in high school, to adjusting to moving out abstracts, conference proceedings). If published findings are bi-
of home when starting university. For learning and achievement to ased toward significant results, we would expect a stronger asso-
take place, students must first be able to manage their lives so that ciation for peer-reviewed journal articles than other sources of
they show up to class in a fit state to concentrate on the subject data. The fourth hypothesis is thus a test of the file-drawer prob-
matter. Students with higher EI may be better able to manage the lem, where we hypothesize that the EI/academic performance
social world around them, forming better relationships with teach- association will be significantly larger in published than unpub-
ers, peers, and family. This may directly influence grades (e.g., in lished sources (Hypothesis 4).
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 7

Student age. Perera and DiGiacomo (2013) reported a stron- size that EI will show a stronger relationship to grades than to
ger effect of EI on academic performance in elementary school standardized test scores (Hypothesis 6).
students than in university students (r ⫽ .28 vs. r ⫽ .18) but did Gender composition of the sample. Although Perera and
not examine ability-based EI. Similarly, Poropat (2009) found DiGiacomo (2013) found that gender composition of the sample
significantly stronger associations between conscientiousness and did not moderate the effect of EI on academic performance, we
academic performance at earlier stages of education. This implies revisit this moderator for two reasons. First, we examine ability EI
that self-regulatory processes (such as EI) may be a more critical (which Perera and DiGiacomo did not). Second, we have a much
determinant of educational outcomes at earlier ages and stages of larger sample and therefore greater power to detect an effect.
education. We will therefore test whether the EI/academic perfor- There are two main ways in which the gender composition of
mance relationship differs across different across age groups. Our the sample could moderate the EI/academic performance associa-
expectation is that EI may prove more important at earlier ages. tion. First, moderation could occur as a result of the gender effects
This is because when the average level of emotional skills is at the individual level of analysis. To the extent that say males are
relatively low (as for 5- to 7-year-olds), being below average more likely to benefit academically from higher levels of EI than
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

constitutes a lower absolute level of skill that might prove more females, having a greater proportion of males in the sample will
This document is copyrighted by the American Psychological Association or one of its allied publishers.

detrimental to performance overall. For example, a 5-year-old with result in a stronger relationship between EI and academic perfor-
lower-than-average emotion management skills may spend the day mance. In this regard, there is some evidence that the relationship
crying, hitting other children, and be completely unable to focus on of EI to social and emotional outcomes is stronger for males than
the tasks at hand. A university student with lower-than-average females, suggesting that EI may confer greater benefit to males
emotion management skills may have a less satisfying a time at (Brackett et al., 2006; Brackett, Warner, & Bosco, 2005). There is
university (Brackett, Mayer, & Warner, 2004; Brackett et al., evidence that males and females experience different kinds of
2006) but is nevertheless likely be able to self-regulate sufficiently emotions—females experience greater internalizing emotions such
well so as not to spend the day crying and hitting others. Moreover, as anxiety whereas males experience greater externalizing emo-
there is likely to be selective attrition based on emotional skills, tions such as anger (Chaplin & Aldao, 2013; Fischer, Rodriguez
resulting in a restriction of range at university compared to ele- Mosquera, van Vianen, & Manstead, 2004). To the extent that
mentary schools and high schools. Students with very low emo- uncontrolled anger may be more damaging to school engagement
tional skills are more likely to have anxiety-based school refusal, and mastery goals as compared with anxiety (Pekrun et al., 2002,
conduct problems leading to suspensions or expulsion, and/or 2009), this may mean that higher EI is more beneficial to males
more difficulties resisting peer pressure (resulting in early drug than females, as the consequences of failing to regulate are differ-
and alcohol use or unintended pregnancy), and are therefore less ent for the different kinds of emotions.
likely to obtain entrance to university. Thus, we hypothesize that Second, moderation could occur as a result of contextual effects.
the EI/academic performance association will be stronger at The gender composition of the sample defines the broader context
younger ages (Hypothesis 5). in which academic performance takes place, which in turn could
Type of academic performance: Grades versus standardized influence the beneficial value of high EI, regardless of whether an
test scores. Academic performance is commonly measured in individual is male or female. As described above, the beneficial
two ways; as standardized test scores or as end-of-semester grades. value of EI may be increased by contexts that place greater
Standardized tests are usually developed and administered by state demands on dealing with social relationships, managing negative
or national bodies (or even international bodies, as in the case emotions, or learning academic material with emotional or social
of the OECD’s Program for International Student Assessment content. Coeducational schools place greater emphasis on affilia-
[PISA]). They are often considered high-stakes either for the tion and social relationships than same-sex schools (e.g., Schneider
individual student or for their institution. School funding, reputa- & Coutts, 1982). Furthermore, academic contexts with a greater
tion, or position within publicly available league tables may de- proportion of males can create an intellectually threatening envi-
pend on test score results, and students’ entrance into selective ronment and heightened text anxiety among females, thus ham-
secondary schools or university programs may depend on their pering their performance (see Inzlicht & Ben-Zeev, 2000).
individual test scores. Standardized tests would normally be ad- Taken together, the above arguments suggest that samples that
ministered in large groups with a set time limit for completion, and are comprised of largely female participants will benefit less from
classroom teachers and school administrators have no direct input the effects of EI than samples that are more mixed or male
on the content of these tests. In contrast, end-of-semester grades dominated. We thus hypothesize that samples with a higher pro-
are composite measures of multiple forms of assessment, which portion of females will show weaker effects of EI on academic
may include formal examinations, written assignments, regular performance (Hypothesis 7).
home-work, contribution to class discussion, group projects, or Student minority status. SEL programs are of interest to
other assessment tasks. Classroom teachers and school or univer- educators and policymakers as a pathway to address issues of
sity administrators have much more control over the content of equity for students from traditionally disadvantaged backgrounds
these assessments. That is, one of the primary ways in which EI or at disadvantaged schools. For SEL programs to reduce achieve-
could contribute to high scores on standardized tests is through the ment gaps between groups: (a) programs must be at least as
regulation of test anxiety. Grades (being a more diffuse criterion) effective for minority/disadvantaged populations, and (b) the EI/
would also be influenced via the ability to maintain relationships achievement association must be at least as large for minority/
with one’s instructors and other students, to manage group proj- disadvantaged groups (because increasing EI will increase
ects, and to manage the emotions that motivate procrastination as achievement only in so much as EI and achievement are related).
well as the ability to manage test anxiety. We therefore hypothe- Evidence is mixed for point (a). Taylor, Oberle, Durlak and Weiss-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

8 MACCANN ET AL.

berg’s (2017) recent meta-analysis demonstrates that there are no to humanities performance as compared to mathematics/science
significant differences in the effectiveness of SEL programs for performance (Hypothesis 9).
different levels of SES or for majority versus minority groups.
However, there is research to suggest that SEL programs are less The Incremental Validity and Relative Importance of
effective in disadvantaged schools (Bierman et al., 2010). Such
EI for Academic Performance
differences may be attributable to the challenges of implementing
SEL programs in schools serving disadvantaged students (who are Other predictors of academic performance: Intelligence.
disproportionately ethnic minorities). The dominant view for much of the 20th century was that intelli-
Our moderation analysis will test point (b). We hypothesize that gence is the critical ingredient for success at school. While differ-
samples with larger proportions of minorities may have a stronger ent theorists have defined intelligence in slightly different ways,
EI/achievement association. Our logic is that low EI constitutes a the American Psychological Association Taskforce defined it as
greater barrier in a disadvantaged environment (where EI is re- the “ability to understand complex ideas, to adapt effectively to the
quired to deal with daily obstacles) than in more supportive envi- environment, to learn from experience, [and] to engage in various
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ronment. Ethnic minority status is associated with social and forms of reasoning” (Neisser et al., 1996, p. 77).
This document is copyrighted by the American Psychological Association or one of its allied publishers.

educational disadvantage. In addition, ethnic minority students Since the dawn of intelligence testing, one of the key tenets of
face additional daily obstacles in the form of microaggressions and intelligence tests is that scores should correspond to student’s
racism which produce negative emotions (such as anger and anx- achievement at school. In fact, intelligence tests were originally
iety) that require emotion regulation. devised by Alfred Binet to assess students’ ability to succeed at
Because our arguments are partly based on advantaged versus school. Prominent economists and psychologists have tried to
disadvantaged environments, they would also hold for compari- gauge the size of the intelligence/achievement association, with
sons of high versus low socioeconomic status (irrespective of estimates ranging from around r ⫽ .40 to .70 (e.g., Deary, Strand,
ethnic minority status). We do not examine this directly because Smith, & Fernandes, 2007; Jencks, 1979; Kaufman & Lichten-
sample socioeconomic status is rarely if ever reported in correla- berger, 2005). Early meta-analyses estimated this relationship as
tional research on EI. We use the sample percentage who are of a .34, .43, and .48, respectively, but were focused only on achieve-
minority ethnicity (i.e., non-White) as a moderator of the EI/ ment in science in primary and secondary school (Boulanger,
academic performance association. Because of the difficulty of 1981; Fleming & Malone, 1983; Steinkamp & Maehr, 1983).
defining minority across different countries, we have restricted this Three recent meta-analyses have estimated the size of the intelli-
analysis to samples from the United States (where sample ethnicity gence/academic performance correlation more generally but have
is commonly reported using consistent categories of White, Black, come to different conclusions on the size of this effect. Roth et al.
Asian, and/or Hispanic in most educational and psychological (2015) found a corrected correlation of .54 across 240 studies of
research). We hypothesize that the EI/academic performance as- primary and secondary students. Poropat (2009) found a corrected
sociation will be larger for samples with higher proportions of correlation of .25 across 47 studies of primary, secondary and
minority students (Hypothesis 8). university students. Richardson et al. (2012) found a corrected
Subject area: Mathematics/Science versus Humanities. correlation of .21 across 35 studies of university students.
Many of the arguments mentioned above assume that EI affects the There are two possible reasons for this large discrepancy—stage
learning process, where high EI students learn more and perform of education and the time-period covered in the meta-analysis.
better by regulating their anxiety or boredom, or managing the Roth et al.’s (2015) samples were predominantly preadolescent
social relationships of the classroom, playground or campus. How- children (no university students were included, and only 1/3 of the
ever, it is also possible that the EI/academic performance associ- samples were from secondary school). In contrast, Poropat (2009)
ation is attributable to the content of what is learned. Emotional included primary, secondary, and tertiary students with 55% of
content is self-evidently more relevant to a performing arts grade studies from university samples, and less than 10% from primary
(where accurately portraying and evoking emotion form part of the school samples. Richardson et al. (2012) included university stu-
assessable content) as compared with mathematics (where the dents only and had the lowest estimate of the three meta-analyses.
content is unrelated to emotions). Similarly, analyzing the univer- Although no single meta-analysis has tested for moderation across
sal themes of a poem or canonical play requires an understanding tertiary versus earlier levels of education on the intelligence/
of the emotions and motivations of the characters, such that emo- academic performance relationship, this comparison of different
tions may be a required knowledge base for humanities-based but meta-analyses suggests that effects may be smaller in tertiary
not science/mathematics-based content domains. Understanding education than they are in primary or secondary education. Jensen
human motivations and emotions may be a required skill for (1998) also believed that the importance of intelligence for aca-
interpreting the meaning of some texts. This is true from the demic achievement dropped from primary to secondary to tertiary
earliest levels of education (e.g., the dramatic tension of The Cat to postgraduate education.
in the Hat derives from understanding both the protagonist’s initial Second, more recent studies show smaller effects. Roth et al.
boredom and the anxiety caused by the cat’s chaos) through to the (2015) reported that older studies (pre-1983) showed a signifi-
higher ones (e.g., analyzing the role of charismatic leaders in cantly stronger relationship than newer studies. All studies in-
creating a totalitarian state when studying modern history). To test cluded in Poropat’s (2009) meta-analysis were published after
whether the EI/performance association derives from process ver- 1990, all but four after 1995, and the majority after 2000. As such,
sus content, we will test whether the effect is stronger for discrepancies between Roth et al. and Poropat may be attributable
humanities-based subjects versus mathematics and science-based to the time-periods examined. Roth et al. propose that the reduced
subjects. We hypothesize that EI will show a stronger relationship effect of intelligence on academic performance over time is attrib-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 9

utable to increasing grade inflation (which reduces the variance in 1998). However, research on the importance of these personality
academic performance in more recent studies, depressing correla- traits for academic performance did not follow for another 20
tions). There is abundant evidence of continued grade inflation years. Poropat (2009) provided the first large scale meta-analysis
since 1983 (Bachan, 2017; Nata, Pereira, & Neves, 2014; Rojstac- linking five domains of personality to academic performance.
zer & Healy, 2012). Given that conceptualizations and study of EI Across all levels of education, he found a conscientiousness/
did not commence until the 1990s, and much of the research academic performance relationship of ␳ ⫽ .22, which was of
occurred only in the last 10 years, we believe it is important to similar magnitude to his estimate of intelligence/performance re-
control for intelligence/academic performance estimates obtained lationship (␳ ⫽ .25). Moreover, conscientiousness predicted aca-
in the current era, ideally from similarly constituted samples (i.e., demic performance independently of the effects of intelligence.
from tertiary samples as well as primary and secondary). For this Other recent meta-analyses found similar estimates of .23 and .24
reason, we conduct a secondary meta-analysis of studies in our for the conscientiousness/academic performance corrected corre-
EI/academic performance citations, examining the intelligence/ lation in tertiary academic performance (O’Connor & Paunonen,
academic performance relationship reported in these citations. This 2007; Richardson et al., 2012). Across these three meta-analyses,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

effectively controls for both the time period of the study, and the agreeableness also showed a significant (but small) positive rela-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

stage of education (i.e., estimates of the intelligence/academic tionship with academic performance. As the five factors of per-
performance association are then drawn from the same time period sonality are known to predict academic performance, we will
and same samples as the estimates of EI/academic performance). examine whether EI provides incremental prediction above and
We will then use our intelligence/academic performance estimate beyond the effects of personality as well as intelligence.
(along with other meta-analytic correlations) to construct a corre-
lation matrix to test for the incremental validity of EI on academic The Combined Effect of EI, Intelligence, and
performance, over and above the effect of intelligence and the five
Personality on Academic Performance
major domains of personality.
Other predictors of academic performance: Noncognitive Given that intelligence and personality (particularly conscien-
constructs and personality. A recent zeitgeist led by behavioral tiousness) are well-established predictors of academic perfor-
economist James Heckman has focused on noncognitive factors mance, it is important to examine the incremental validity of EI
such as personality traits as critical predictors of education and alongside the contribution of these traditional predictors. As re-
workplace outcomes (Heckman & Rubinstein, 2001; Heckman, viewed in earlier sections, EI has known relationships with both
Stixrud, & Urzua, 2006). Although earlier canonical writings on intelligence and personality. This has led some researchers to
the impact of intelligence stressed that noncognitive, motivational, argue that EI is redundant and has little incremental validity above
or conative factors were also likely to be impactful for life success and beyond the effects of intelligence and personality (Schulte et
(Jencks, 1979; Wechsler, 1943), these lacked an organizing frame- al., 2004). Indeed, in a few studies where EI has been related to
work. positive outcomes, the positive effect of EI disappears after con-
One possible organizing framework for noncognitive factors is trolling for intelligence and personality (see Walter, Cole, &
the Big Five model of personality, which gained acceptance among Humphrey, 2011, for a review). Therefore, researchers have ar-
psychologists from the 1990s onward (Digman & Inouye, 1986; gued for the need to establish not only the main effect of EI, but
Goldberg, 1981; John, 1989; Tupes & Christal, 1992). Factor also the incremental validity of EI relative to intelligence and
analysis of trait adjectives produced a five-factor solution (the Big personality (Antonakis, Ashkanasy, & Dasborough, 2009; Miao et
Five) that was largely similar to the Five Factor Model (FFM) of al., 2017). Moreover, as meta-analyses arguably provide the best
personality that emerged from factor analysis of questionnaire estimates for effect sizes of the relationship between different
data. These five personality factors are extraversion (positive variables, this article is well-placed to determine the incremental
affect, high energy levels, and sociability), agreeableness (sympa- validity of EI (ability EI, self-rated EI, and mixed EI) for academic
thy, kindness, and submissiveness), conscientiousness (hardwork- performance. Given the high correlations between mixed EI and
ing, detail-minded, and organized), neuroticism (easily and fre- personality (e.g., van der Linden et al., 2017) and the high corre-
quently feeling stress and negative affect), and openness to lations between ability EI and conventional nonemotional mea-
experience (open to new ideas, enjoyment of arts, culture, and sures of intelligence (e.g., MacCann et al., 2014), it is reasonable
aesthetics; referred to as Intellect in the Big-Five and Openness in to expect that correlations of EI with academic performance will
the Five-Factor Model). These five major domains of personality shrink after accounting for these variables. However, we hypoth-
provide a conceptual framework for grouping the multiple differ- esize that overlap with personality and intelligence will not be
ent noncognitive constructs that have emerged from different entirely responsible for the effect of EI on academic performance,
subdisciplines of psychology and education. For example, consci- such that all three streams of EI will predict academic perfor-
entiousness provides an overarching framework for the conceptu- mance above and beyond the contributions of intelligence and
ally similar psychological constructs of delay of gratification, ego personality (Hypothesis 10).
control, effortful control, self-control, self-regulation, and grit Whereas incremental validity examines the prediction of a vari-
(Heckman & Kautz, 2012; Roberts, Lejuez, Krueger, Richards, & able to a criterion above and beyond the contribution of other
Hill, 2014). variables, relative importance refers to the contribution each pre-
A series of meta-analyses demonstrates that these five person- dictor makes to the total criterion variance in combination with
ality domains, and particularly conscientiousness, predict work- other predictors (Johnson & LeBreton, 2004). The goal of relative
place performance independently from the effects of intelligence weights analysis is to “partition explained variance among multi-
(Barrick & Mount, 1991; Salgado, 1998; Schmidt & Hunter, ple predictors to better understand the role played by each predic-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

10 MACCANN ET AL.

tor in a regression equation” (Tonidandel & LeBreton, 2011, p. 1). Inclusion and Exclusion Criteria
This is a separate question to incremental prediction, as the order
in which variables are entered into the analysis does not affect the Citations were included in the meta-analyses if they met the
relative weights. Relative weights can be expressed as a percentage following inclusion criteria: (a) written in English; (b) the measure
of the R-squared value that each variable independently contrib- of EI was published in either a test manual or journal article; (c)
utes. We will examine not only the incremental prediction of EI included academic performance indices appraised either directly
but also the relative importance of EI as a predictor when consid- (e.g., GPA, SAT results) or self-reported (self-reported GPA has
ered in conjunction with intelligence and personality. Previous been found to correlate at r ⫽ .90 with actual GPA for college
meta-analyses have used relative weights analysis to compare the students and r ⫽ .82 for high school students; Kuncel, Crede, &
relative contribution of EI to the prediction of workplace outcomes Thomas, 2005); (d) referred to original data not reported in any
(Miao et al., 2017; O’Boyle et al., 2011). Many of the original other citation; (e) reported an effect size between EI and academic
claims in the 1990s popular science books on emotional intelli- performance, or reported data from which an unbiased estimate of
gence proposed that EI was more important than intelligence— effect size could be calculated; and (f) reported the sample size.
Conversely, studies were excluded from the meta-analyses based
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

that “it can matter more than IQ” (Goleman, 1995, p. 1). A
on the following exclusion criteria: (a) measured clinical or work-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

comparison of the relative contribution of EI and intelligence to

the prediction of academic performance would thus test whether related training performance; (b) used degree attainment as an
there is any merit to these original claims. Given the ubiquitous index of academic performance (e.g., Kapp, 2000; Kashani, Azimi,
influence of intelligence and conscientiousness on academic per- & Vaziri, 2012); (c) used an in vivo laboratory achievement test
formance, we hypothesized that EI will be among the top three (e.g., Shao, Yu, & Ji, 2013); (d) were publications based on the
predictors of academic performance, along with intelligence and same data as another citation in the meta-analysis (e.g., Ahmad,
conscientiousness (Hypothesis 11). 2010); or (e) the EI constructs were not based on established
ability, trait, or mixed EI theory (e.g., social intelligence).
Most citations included multiple correlations estimating the
Method EI/academic performance association (e.g., correlations for several
different EI tests with both university and high school grade point
Search Strategy average, for several different subscales of EI, or several different
First, we searched the following nine databases for relevant subject areas of achievement). We recorded all possible combina-
studies on EI and academic performance: ERIC, Google Scholar, tions of EI/achievement as within-citation effects. After exclusion
ISI Web of Science, Medline, ProQuest Dissertations and Theses, criteria, there were n ⫽ 162 different citations (188 samples), with
PsycINFO, PubMed, ScienceDirect, and Scopus. For academic k ⫽ 1,276 correlations available for analysis. The final file for
performance we used the search string “(academic OR education analysis is included as an excel sheet in the online supplementary
OR university OR school) AND (grade OR GPA OR performance material, and a summary of citations is provided in the Appendix.
OR achievement)” in line with Poropat (2009). For emotional Citations represented 27 different countries, with most data from
intelligence we used the search string “(emotional intelligence) English-speaking countries (76.5%, k ⫽ 974) such as the United
OR EI OR (emotion perception) OR (emotion understanding) OR States (43.9%, k ⫽ 560), U.K. (8.2%, k ⫽ 105), or Australia
(emotion facilitation) OR (emotion recognition) OR (emotion man- (7.8%, k ⫽ 99). The largest number of observations from non-
agement) OR MSCEIT OR MEIS OR TEIQue OR SREIS OR English speaking countries were from Iran (k ⫽ 74), Portugal (k ⫽
WLEIS OR (Mayer-Salovey-Caruso Emotional Intelligence Test) 54), and Spain (k ⫽ 35).
OR (Multifactor Emotional Intelligence Scale) OR (Trait Emo-
Coding
tional Intelligence Questionnaire) OR (Schutte Self Report Emo-
tional Intelligence Test) OR (Wong and Law Emotional Intelli- Each of the 1,276 effects was coded on the criteria described
gence Scale).” Combining these yielded an initial search result of below. The first set of coding was conducted by the second and
6139 citations. All citations up until November 1, 2016 were third authors, who both hold postgraduate degrees in psychology.
considered, with no lower limit to the publication year (unpub- A second coder (the fourth author, who holds an undergraduate
lished data or studies were labeled with the year they were lo- psychology degree) coded 73 randomly selected citations to test
cated). After removing duplicates, the title, abstract, and, where the reliability of coding. Errors were resolved by the first author
necessary, full-text, of the remaining citations were reviewed. going back to the original paper.
Ancestry searches of the reference lists yielded an additional 14 Effect size. Pearson’s correlation (r) was used as the metric of
citations. After reviewing the reference section of relevant existing effect size. In the few samples where r was not reported, the
meta-analytic reviews (Perera & DiGiacomo, 2013; Richardson et available statistic (e.g., ␹2, t value, F value) was transformed into
al., 2012) an additional 10 relevant citations were added. We then r using effect size calculators (Lenhard & Lenhard, 2016; Lyons &
conducted a search for unpublished data on this question by (a) Morris, 2017). Coder agreement was 93%.
searching the American Education Research Association database Sample size. Sample size was ranged from 18 to 2,195 with a
(which resulted in no additional relevant citations), and (b) email- median of 180 and a mean of 256 (25% of observations were based
ing 23 prolific emotional intelligence researchers and putting out a on 99 or fewer participants and 25% of observations were based on
call on the EMONET listserv to request unpublished data, and (c) 291 or more participants). Coder agreement was 90%.
contacting major testing companies and educational research com- Reliabilities of measures. The reliability of both EI and GPA
panies. Together, (b) and (c) resulted an additional seven citations. were coded. For measures of EI, if a study did not provide
F1 This search strategy is outlined in the PRISMA chart in Figure 1. reliability information, the average reliability from all studies that
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 11

Records idenfied through Addional records idenfied through

database searching other sources
(n = 6,139) (n = 29)
Ancestry searches; n = 14
Review of other meta-analyses; n = 10
Requests for unpublished data; n = 7

Records aer duplicates removed

(n = 4,959)
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Records screened Records excluded

(n = 4,959) (n = 4,682)

Full-text assessed for Full-text excluded

eligibility (n = 115)
(n = 277)

Records included in meta-

analysis
(n = 162, k = 188)

Figure 1. PRISMA flowchart for the identification, screening, and inclusion of publications in the meta-
AQ: 16 analyses.

used the same measure in the current meta-analysis was imputed. Age. Mean sample age of participants was coded as a contin-
If no other studies used the same measure, the reliability estimate uous variable, and ranged from 7.68 to 40 years, with a median of
was obtained from the test manual or original psychometric vali- 19.52. Coder agreement (to the nearest whole number of age in
dation study. If either of these were not available, then a reliability years) was 98%. If the study did not report mean age, but gave an
estimate was obtained from a published, large sample study. Coder age range, the median age in the range was used. For studies that
agreement for EI was 80%. For academic performance indexes, did not report an age mean or range, the mean age was imputed
very few studies reported reliabilities. Therefore, the reliability based on an average of the mean ages in other studies that included
estimates were obtained from published studies on the reliability of participants at a similar academic level (Poropat, 2009).
different academic performance measures, as follows: GPA ob- EI stream. EI stream was coded into three categories: Ability
tained from academic records (Westrick, 2017); self-reported GPA EI (31%, k ⫽ 399), Self-rated EI (13%, k ⫽ 161), and Mixed EI
(Kuncel et al., 2005); self-reported SAT or other equivalent stan- (56%, k ⫽ 716). Coder agreement was 100%. The most commonly
dardized tests scores (Kuncel et al., 2005); SAT (Ewing, Huff,
used tests were the EQ-i (31.6%, k ⫽ 403) and the MSCEIT
Andrews, & King, 2005); and American College Test (ACT,
(24.1%, k ⫽ 307).
2014).
EI facet. Effects were coded as representing one of the four
Standard deviation of EI. Standard deviations were recorded
ability facets (Perception, Facilitation, Understanding, and Man-
for EI measures to correct for range restriction. Coding agreement
agement), five Bar-On EQ-i facets (Intrapersonal, Interpersonal,
was 80%.
Gender. Gender was coded as a continuous variable using Stress Management, Adaptability, General Mood), or Overall EI.
percentage (%) of female participants in each citation. This value Other facets of EI were not coded. Coder agreement was 95%.
ranged from 0% to 100%, with a median of 61%. Coder agreement Educational level. Educational level was coded as a categor-
was 95%. ical moderator with three levels: primary (7.6%, k ⫽ 97), second-
Ethnicity. Ethnicity was coded as a continuous variable using ary (32.3%, k ⫽ 412), or tertiary (59.6%, k ⫽ 761). One study
percentage (%) of White participants in each citation. This was (Alumran & Punamaki, 2008) used a mixed sample of secondary
only coded for samples from the United States, where ethnicity and tertiary students, and so was excluded from the relevant
was commonly reported using the same categories. Most observa- moderator analyses. Coder agreement was 99%.
tions (k ⫽ 520 of 558) reported this information. Percentage White Type of academic performance measure. This variable was
ranged from 0% to 100% with a median of 52% and a mean of coded as a categorical moderator with two levels: course grade
51%. (e.g., GPA, semester course grade, and school subject mark;
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

12 MACCANN ET AL.

83.9%, k ⫽ 1071) and standardized test score (e.g., SAT and GRE; sizes (i.e., citation level clustering). We use RVE with a correlated
16.1%, k ⫽ 205). Coder agreement was 96%. effects structure, which stipulates that effect sizes from the same
Publication format. This variable was coded as a categorical citation are likely to be correlated with each other because of
moderator with 4 levels: peer-reviewed articles (54.5%, k ⫽ 696), erroneous study characteristics. RVE has been shown to accurately
conference proceedings (3.7%, k ⫽ 47), dissertations (38.1%; k ⫽ estimate effect sizes even when the precise nature of the correla-
486), and unpublished data (3.7%, k ⫽ 47). Coder agreement was tion between effects is unknown (Moeyaert et al., 2017). We
100%. specified a correlation rho of .80 as is typical when the correlation
Subject area. The subject area was coded as general (for between effect sizes unknown (Hedges et al., 2010). However, we
overall GPA or standardized test covering both numeracy and also performed a sensitivity analysis which suggested none of our
literacy areas; 54.3%, k ⫽ 694), as mathematics/science (21.1%, findings different substantially as a function of the selected rho.
k ⫽ 267), or as humanities/arts (23.4%, k ⫽ 299). The multilevel random effect meta-analysis with RVE estimation
were performed using the ‘robumeta’ package (Fisher, Tipton, &
Statistical Analyses Hou, 2016) in R Version 3.4.3 (R Core Team, 2017). The R-code
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

is provided as online supplemental material. To test for the effect

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Prior to calculating mean sample-weighted correlations, effect sizes

of publication bias we looked at funnel plots (see Figure 2 for F2
were corrected for: (a) unreliability in academic performance, (b)
range restriction in EI, and (c) unreliability in EI. We used the overall funnel plot) and Egger’s test.
formulae and processes outlined in Hunter, Schmidt, and Le (2006) To test Hypotheses 2 through 9 (testing for moderators), we
for direct range restriction. This equation uses mu (the ratio of the conducted metaregressions using RVE. We conducted multiple
sample to population standard deviation). For citations where mu regressions simultaneously controlling for publication type,
could not be calculated (because sample or population standard de- stream, sample mean age, gender composition, subject area, and
viation was not available), we imputed a value of mu equal to the performance type. Because there was a lot of missing data for EI
mean mu for other studies in the same stream and stage of education. facet (only available for a subset of Stream 1 studies) and ethnicity
Outliers for each meta-analysis were detected and removed (only available for U.S. studies), these were conducted as separate
using Hoaglin and Iglewicz’s (1987) outlier labeling rule, using the simple metaregressions (otherwise we would have a substantially
more conservative multiplier of g ⫽ 2.2. When interpreting the reduced k and reduced power for multiple regressions). For cate-
AQ:6, 7 effect size, we used Cohen’s (1962, 1988) benchmarks of .10, .30, gorical moderators (EI Stream, EI facet, publication status,
and .50 for small, medium, and large. education stage, and subject area of achievement), we also
We used Robust Variance Estimation (RVE) to control for conducted subgroup analyses so that the relative magnitude of
dependencies between effect sizes (Hedges, Tipton, & Johnson, the difference could be easily communicated. For categorical
2010). Most citations we located report more than one effect size. moderators, we used contrast coding in the metaregressions.
Including multiple effect sizes from the one citation can be prob- Because age and stage of education are dependent, we only entered
lematic due to the potential for dependencies caused by theoreti- age into the multiple regressions, to avoid collinearity.
cally irrelevant variables, such as lab specific procedures, similar To test Hypothesis 10 (the incremental validity of EI in predict-
participants, and so forth. RVE is a multilevel approach that ing academic performance controlling for cognitive ability and
calculates standard errors that are adjusted for clustering of effect personality), we created a correlation matrix of meta-analytic

Figure 2. Funnel plot for overall analysis of emotional intelligence (EI) with academic achievement.
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 13

correlations. These were drawn from: (a) the current meta-analysis within each of the subgroups. Given the systematic variability
(EI/academic performance and intelligence/academic performance between samples within each of the EI streams, and the differing
cells) and (b) other published meta-analyses. We used this matrix theoretical meaning of the three streams, the remaining moderators
to test hierarchical regression models where intelligence was en- were examined separately for each EI stream, as well as overall.
tered in Step 1, the Big Five personality domains in Step 2, and EI
in Step 3. The harmonic mean of the sample size across all cells in Hypothesis 3: Moderating Effects of EI Facet
the matrix was the input sample size. Regression models were run
separately for the three streams of EI, and for the four branches of For ability EI, there were moderate effect sizes for understand-
ability EI. It was not possible to conduct a meta-analysis of ing emotions (␳ ⫽ .35; 95% CI [.28, .43]) and emotion manage-
beta-weights, because only four citations had conducted regres- ment (␳ ⫽ .26; 95% CI [.16, .35]), a small to moderate effect size
sions where academic performance was regressed on all three of for emotion facilitation (␳ ⫽ .18; 95% CI [.09, .27]), and a small
intelligence, big five personality, and EI. SPSS syntax for inputting effect size for emotion perception (␳ ⫽ .09; 95% CI [.01, .18]; see
these matrices and running these analyses is provided as online Table 2). Follow-up tests of moderation using multilevel metare-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

supplementary material. gressions with contrasts coded as (⫺.5 ⫺.5 .5 .5), (0 0 ⫺1 1), and
This document is copyrighted by the American Psychological Association or one of its allied publishers.

To test Hypothesis 11 (the relative importance of EI in predict- (⫺1 1 0 0) for Perception, Facilitation, Understanding, and Man-
ing academic performance as compared to intelligence and per- agement, respectively. Because only a subset of studies reported
sonality), we conducted relative importance analysis using the R branch-level effects, this analysis was run separately to the regres-
code provided by Tonidandel and LeBreton (2011). This was sion shown in Table 3 (so as not to reduce power). Results showed
conducted for each regression described above. that the effect size was significantly larger for strategic EI (under-
standing and management), compared with experiential EI (perceiv-
Results ing and facilitation; b ⫽ .16, SE ⫽ .04, 95% CI [.08, .24], p ⬍ .001).
This regression also controlled for publication type, age, gender
composition, subject area, and performance type. This supports Hy-
Hypothesis 1: Overall Correlation Between EI and pothesis 3. The effect was not significantly different for understanding
Academic Performance versus management branches (b ⫽ ⫺.03, SE ⫽ .04, 95% CI [⫺.07,
An initial multilevel random effects meta-analysis with RVE .01], p ⫽ .157), nor for perception versus facilitation branches (b ⫽
was conducted on the measurement and range corrected correla- .03, SE ⫽ .04, 95% CI [⫺.05, .12], p ⫽ .408).
tions to estimate the mean true correlation between overall EI
(across all EI streams) and academic performance. Prior to con- Hypothesis 4: Moderating Effect of Publication Status
ducting the analysis, 30 effect sizes were removed using the outlier
labeling rule (Hoaglin & Iglewicz, 1987). All 30 of these effect As can be seen in Tables 2 and 3, publication format (published
sizes had corrected correlations in excess of ␳ ⫽ .73 (such that vs. unpublished) did not significantly moderate the effect of EI on
their inclusion would have increased the overall size of the effect). academic performance. The effect size was larger for published
There was a significant positive correlation between overall EI and than unpublished research in the case of ability and self-rated EI,
academic performance, of small to moderate effect size (␳ ⫽ .20, the differences were mostly very small (⌬␳ ⫽ .01, .07, .03, and .01
95% CI [.17, .22]), supporting hypothesis 1. The test for hetero- for overall EI, ability EI, self-rated EI, and mixed EI, respectively).
geneity of effect sizes was statistically significant, Q(1245) ⫽ The metaregressions indicated no significant difference between
18,050.20, p ⬍ .001, indicating true differences in effect sizes published versus unpublished citations for all streams and for
across samples. Moreover, the very large I2 value of 91.41% overall EI. That is, there is no evidence for the file drawer effect
suggests that an overwhelming amount of observed variation be- (selective publication) for included studies. In addition, we further
tween samples was due to systematic between-samples variability. explored the possibility of publication bias by inspecting funnel
As such, the planned subgroup analyses and metaregressions using plot asymmetry (see Figure 2). No visual signs of asymmetry were
RVE were conducted, testing Hypotheses 2 through 9 (possible noted, and an Eggers test suggested that there was no significant
T2-T3 moderators of the effect). Tables 2 and 3 present the results of the asymmetry (z ⫽ ⫺1.5788, p ⫽ .11).1 Collectively, the results do Fn1

F3-F6 moderator analyses, which are also illustrated in Figures 3 to 6. not support Hypothesis 4 (that there would be selective publication
such that published articles showed a higher effect size than
unpublished studies).
Hypothesis 2: Moderating Effects of EI Stream
Subgroup analyses and metaregressions with RVE were used to Hypothesis 5: Moderating Effect of Age
examine the moderating effect of EI stream on the association
between EI and academic performance (see Tables 2 and 3). As shown in Table 2, effect sizes tended to be smaller for
Studies using ability EI measures obtained the largest average tertiary samples as compared to primary or secondary samples.
effect size (␳ ⫽ .24; 95% CI [.18, .30]), followed by studies using This was true for overall EI (␳ ⫽ .16, .23, and .22 for tertiary,
mixed EI measures (␳ ⫽ .19; 95% CI [.15, .22]), and studies using secondary, and primary respectively), ability EI (␳ ⫽ .18, .30, and
self-rated EI measures (␳ ⫽ .12; 95% CI [.07, .18]). Metaregres- .29), and mixed EI (␳ ⫽ .17, .20, and .20) but not self-rated EI (␳ ⫽
sions showed that the ability EI showed a stronger effect than the
other two streams (b ⫽ .07, p ⫽ .006), supporting Hypothesis 2. 1
Because Egger’s test has not been implemented with robust variance
However, the significant heterogeneity Q statistics and high I2 estimation, we therefore calculated Egger’s test using a meta-regression
index (ranging from 85.28% to 92.39%) suggest heterogeneity with the standard error of the effect size as a predictor variable.
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

14 MACCANN ET AL.

Table 2
Meta-Analytic Results of the Overall Link Between EI and Academic Performance and Subgroup Analyses—Overall EI, and by
Streams 1, 2, and 3

95% CI
Subgroup n k N r ␳ Lower Upper I2 p

Overall 158 1,246 42,529 .14 .20 .17 .22 91.41% ⬍.001
Stream
Stream 1 50 383 8,826 .16 .24 .18 .30 92.39% ⬍.001
Stream 2 33 160 8,492 .10 .12 .07 .18 85.28% ⬍.001
Stream 3 90 703 27,939 .13 .19 .15 .22 91.49% ⬍.001
Education level
Primary 13 91 2,329 .14 .22 .10 .34 91.81% .001
Secondary 60 507 16,139 .16 .23 .18 .28 91.93% ⬍.001
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Tertiary 103 643 28,596 .11 .16 .13 .20 90.64% ⬍.001
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Achievement measure
Course grade 141 1,050 38,759 .14 .20 .17 .23 91.24% ⬍.001
Stand. test 37 196 8,178 .12 .17 .10 .23 93.53% ⬍.001
Publication type
Journal article 82 684 23,798 .14 .19 .15 .23 91.35% ⬍.001
Unpublished 76 562 18,731 .14 .20 .16 .24 91.57% ⬍.001
Subject area
General 125 687 33,941 .13 .19 .16 .22 89.60% ⬍.001
Humanities 35 285 7,231 .17 .24 .18 .29 91.94% ⬍.001
Math/Science 40 258 9,750 .11 .16 .10 .22 92.27% ⬍.001
Stream 1 (Ability) 50 383 8,826 .16 .24 .18 .30 92.39% ⬍.001
Facet
Perception 29 68 3,919 .07 .09 .01 .18 87.86% .025
Facilitation 25 53 3,234 .12 .18 .09 .27 90.52% .001
Understanding 29 85 4,815 .22 .35 .28 .43 94.01% ⬍.001
Management 31 67 4,898 .16 .26 .16 .35 93.54% ⬍.001
Education level
Primary 3 11 199 .15 .29 ⫺.86 1.43 96.92% .394
Secondary 25 176 5,177 .20 .30 .22 .38 90.99% ⬍.001
Tertiary 35 196 5,515 .11 .18 .11 .24 89.13% ⬍.001
Achievement measure
Course grade 46 291 8,406 .15 .24 .17 .30 92.07% ⬍.001
Stand. test 19 92 3,135 .17 .24 .12 .36 94.84% .001
Publication type
Journal article 25 183 5,443 .17 .28 .20 .36 92.52% ⬍.001
Unpublished 25 200 3,383 .14 .21 .11 .30 92.45% ⬍.001
Subject area
General 40 207 6,608 .14 .22 .16 .28 90.25% ⬍.001
Humanities 14 71 3,221 .26 .38 .28 .49 93.34% ⬍.001
Math/Science 15 98 3,205 .13 .21 .05 .38 96.00% .014
Stream 2 (Self-rated) 33 160 8,492 .10 .12 .07 .18 85.28% ⬍.001
Education level
Primary 2 10 479 .05 .07 .03 .12 90.35% .033
Secondary 10 92 3,777 .12 .14 .03 .26 88.35% .022
Tertiary 24 58 4,676 .08 .10 .03 .18 83.32% .006
Achievement measure
Course grade 31 146 8,013 .11 .13 .07 .19 83.89% ⬍.001
Stand. test 5 14 1,087 ⫺.03 ⫺.03 ⫺.18 .13 84.29% .67
Publication type
Journal article 24 134 6,757 .10 .13 .06 .20 84.50% ⬍.001
Unpublished 9 26 1,735 .08 .10 ⫺.04 .24 88.27% .135
Subject area
General 27 73 5,337 .10 .13 .06 .19 82.01% ⬍.001
Humanities 7 45 1,683 .07 .10 .00 .20 83.51% .056
Math/Science 9 37 3,722 .05 .07 ⫺.08 .21 90.92% .324
Stream 3 (Mixed) 90 703 27,939 .13 .19 .15 .22 91.49% ⬍.001
Education level
Primary 9 70 1,723 .15 .20 .14 .27 79.50% ⬍.001
Secondary 33 239 9,035 .15 .20 .13 .27 90.64% ⬍.001
Tertiary 53 389 19,715 .11 .17 .12 .21 92.67% ⬍.001
Achievement measure
Course grade 78 613 24,996 .14 .20 .16 .24 91.58% ⬍.001
(table continues)
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 15

Table 2 (continued)
95% CI
Subgroup n k N r ␳ Lower Upper I2 p

Stand. test 19 90 5,088 .08 .10 .04 .16 88.18% .004

Publication type
Journal article 48 336 14,585 .13 .19 .14 .24 91.76% ⬍.001
Unpublished 42 367 13,354 .13 .18 .13 .23 91.25% ⬍.001
Subject area
General 68 407 23,439 .13 .19 .15 .23 90.29% ⬍.001
Humanities 22 169 4,406 .12 .17 .11 .22 79.62% ⬍.001
Math/Science 22 123 4,488 .10 .13 .07 .19 89.16% ⬍.001
Note. EI ⫽ emotional intelligence.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

.10, .14, and .07), although confidence intervals overlapped in all Hypothesis 6: Moderating Effect of
cases. When we tested the effect of sample mean age using Type of Achievement
metaregressions, the effect of age was not significant for overall EI
or for any of the three streams. Taken together, results do not As shown in Table 3, the metaregressions revealed showed that
support Hypothesis 5, showing no moderating effect of age. achievement type (course grade vs. standardized test) was a sig-

Table 3
Meta-Regressions Testing Moderators of EI/Academic Performance Link (Multiple Regressions
With All Moderators in Same Equation

95% CI
Moderator b SE Lower Upper p

Overall EI (k ⫽ 1,057)
Publication type ⫺.013 .031 ⫺.073 .048 .680
Age ⫺.004 .003 ⫺.009 .002 .191
Gender (% female) ⫺.002 .001 ⫺.003 .000 .026
Streama
Ability vs. others .070 .024 .021 .118 .006
Self-rated vs. mixed ⫺.017 .019 ⫺.054 .021 .372
Subject areab
General vs. others .016 .019 ⫺.022 .055 .401
Math/science vs. humanitiesb .034 .017 ⫺.001 .068 .057
Performance type ⫺.071 .034 ⫺.140 ⫺.002 .045
Stream 1: Ability (k ⫽ 331)
Publication type .062 .058 ⫺.057 .180 .296
Age ⫺.006 .008 ⫺.023 .010 .428
Gender (% female) .000 .002 ⫺.004 .004 .852
Subject areab
General vs. others .090 .036 .013 .167 .025
Math/science vs. humanities .082 .037 .003 .162 .043
Performance type ⫺.086 .065 ⫺.221 .050 .202
Stream 2: Self-rated (k ⫽ 140)
Publication type .008 .078 ⫺.165 .181 .918
Age ⫺.010 .007 ⫺.025 .006 .188
Gender (% female) ⫺.003 .001 ⫺.006 .001 .088
Subject areab
General vs. others .003 .046 ⫺.104 .110 .950
Math/science vs. humanities ⫺.005 .038 ⫺.094 .084 .896
Performance type ⫺.219 .070 ⫺.384 ⫺.053 .017
Stream 3: Mixed (k ⫽ 586)
Publication type ⫺.015 .039 ⫺.094 .064 .702
Age ⫺.002 .003 ⫺.009 .005 .530
Gender (% female) ⫺.002 .001 ⫺.004 .001 .161
Subject areab
General vs. others ⫺.015 .024 ⫺.065 .034 .527
Math/science vs. humanities .020 .016 ⫺.013 .052 .222
Performance type ⫺.072 .044 ⫺.163 .019 .115
Note. EI ⫽ emotional intelligence.
a
Contrasts were (1 ⫺.5 ⫺.5) and (0 1 ⫺1) for ability, self-rated, and mixed EI, respectively. b
Contrasts were
(⫺1 .5 .5) and (0 ⫺1 0) for general, math/science and humanities, respectively.
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

16 MACCANN ET AL.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 3. Forest plot for overall analysis of emotional intelligence (EI) with academic achievement, showing
separate effects by education level, performance measure, publication status, and subject area.

nificant moderator for overall EI (b ⫽ ⫺.07, p ⫽ .045) and for small effect such that samples with larger proportions of females
self-rated EI (b ⫽ ⫺.22, p ⫽ .02), but not for ability EI or mixed showed weaker effects than more gender-diverse samples when all
EI. The effect was in the same direction mixed EI, but the differ- streams were considered together.
ence was not significant. There were stronger effects for grades
than standardized tests for overall EI (␳ ⫽ .20 vs. .17), self-rated Hypothesis 8: Moderating Effect of
EI (␳ ⫽ .13 vs. ⫺.03), and mixed EI (␳ ⫽ .20 vs. .10) but not
Ethnic Composition
ability EI (␳ ⫽ .24 vs. .24). Taken together, Hypothesis 6 received
partial support. A simple metaregression (k ⫽ 522) suggested that the ethnic AQ: 8
composition of the sample (percentage White) was not statistically
Hypothesis 7: Moderating Effect of Gender significant, either for overall EI (b ⫽ .001, p ⫽ .219), ability EI
Composition (b ⫽ ⫺.001, p ⫽ .289), self-rated EI (b ⫽ .002, p ⫽ .258), or mixed
EI (b ⫽ .000; p ⫽ .863). Therefore, Hypothesis 8 was not supported.
Table 3 shows the results of moderator analyses using metaregres- The EI/academic performance association does not appear to be
sions for gender composition. In interpreting results, we note that affected by the proportion of minority students in the sample.
gender composition was not distributed evenly across the studies—
studies tended to have more females than males. The median propor-
Hypothesis 9: Moderating Effect of Subject Area
tion of females was 0.61, and only 17.6% of studies had more males
than females. For overall EI, the simple metaregression (b ⫽ ⫺.002, As shown in Tables 2 and 3, subject area (math/science vs.
p ⫽ .026) showed a significant effect of gender, with a higher humanities) is a significant moderator of the EI/performance rela-
percentage of females associated with lower effect size. Gender com- tionship for ability EI only (b ⫽ .08, p ⫽ .04), where effects are
position was not significant moderator for any of the streams individ- stronger for humanities (␳ ⫽ .38) than for math/science (␳ ⫽ .21).
ually. This result provides partial support for hypothesis 7, showing a The effect was not significantly different for math/science versus
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 17

This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 4. Forest plot for analysis of Stream 1 emotional intelligence (EI) with academic achievement, showing
separate effects by facet, education level, achievement measure, publication type, and subject area.

humanities for overall EI, self-rated EI, or mixed EI. Results mental validity of EI over-and-above intelligence and big five
therefore provide partial support for Hypothesis 9. personality in the analyses below.
Incremental prediction of EI to academic performance.
Hypothesis 10: Incremental Prediction of EI to Table 4 shows the meta-analytic correlation matrix we derived from T4
Academic Performance our original meta-analyses and previously published estimates. We
used this matrix to conduct multiple regressions predicting academic
The relationship between IQ and EI. In addition to the
performance from intelligence, personality, and EI for each of the
above metaregressions, we examined the relationship between
three EI streams. We checked multicollinearity using tolerance values
intelligence and academic performance from the studies included
in each case. Tolerance values were acceptable for all three streams,
under the current search criteria, to draw comparisons with EI. For
the intelligence/academic performance effect, there were 23 stud- ranging from .70 to .91 for ability EI, .70 to .92 for self-rated EI, and
ies with 84 effect sizes and 4,801 participants. Intelligence showed .58 to .92 for mixed EI. Intelligence, personality, and EI predicted
a moderate to large association with academic performance (␳ ⫽ 24.0% of the variation in academic performance for the ability EI
.39, CI [.31, .46], p ⬍ .001, I2 ⫽ 89.98%). This value is larger than analysis, 22.9% of the variance in academic performance for the
that reported by Poropat (2009) where there was no correction for self-rated EI analysis, and 24.6% of the variance in academic perfor-
range restriction, but smaller than that reported by Roth et al. mance for the mixed EI analysis (see Table 5). Ability EI and mixed T5
(2015) where only primary and secondary education samples were EI showed modest incremental prediction (⌬R2 ⫽ .017 and .023,
included. We will use this value of ␳ ⫽ .39 to assess the incre- respectively) and self-rated EI showed negligible incremental predic-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

18 MACCANN ET AL.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 5. Forest plot for analysis of Stream 2 emotional intelligence (EI) with academic achievement, showing
separate effects by education level, achievement measure, publication type, and subject area.

tion (⌬R2 ⫽ .007). Partial correlations of EI with academic perfor- management. They also demonstrate that the strong effect of
mance were significant at p ⬍ .001 in all cases, and were .148 for emotion understanding on academic performance cannot be ex-
ability EI, .092 for self-rated EI, and .248 for mixed EI. These results plained solely by its overlap with intelligence.
provide support for Hypothesis 10, although the size of the incremen-
tal prediction was very small.
We conducted these regressions separately for each of the four Hypothesis 11: The Most Important Predictors of
branches of ability EI. Results of these analyses are shown in Table Academic Performance Will Be Intelligence,
T6 6. The understanding emotions branch showed the strongest incre- Conscientiousness, and EI
mental prediction (⌬R2 ⫽ .039), followed by emotion management
(⌬R2 ⫽ .036). Both emotion perception and emotion facilitation Tables 5 and 6 show the relative weights and the corresponding
showed negligible incremental prediction of academic perfor- percentage of R2 explained by each variable in the regressions, as
mance (⌬R2 ⫽ .000 for emotion perception and .009 for emotion determined by relative weights analysis. For all three streams, (a)
facilitation). Partial correlations of the ability EI branches with intelligence was the most important variable (accounting for between
academic performance were significant for facilitation, r ⫽ .108, 58% and 69% of the explained variance), (b) conscientiousness the
p ⬍ .001, understanding, r ⫽ .223, p ⬍ .001, and management, second-most important (accounting for between 20% and 21% of the
r ⫽ .216, p ⬍ .001, but not for perception, r ⫽ ⫺.017, p ⫽ .074. explained variance), and (c) EI the third-most important variable.
These results demonstrate that the active ingredients of EI for The relative importance of EI was greater for mixed EI (RW% ⫽
academic performance are emotion understanding and emotion 15%) than self-rated (RW% ⫽ 10%) than ability EI (RW% ⫽ 4%).
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 19

This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 6. Forest plot for analysis of Stream 1 emotional intelligence (EI) with academic achievement, showing
separate effects by education level, achievement measure, publication type, and subject area.

However, results were very different across the four branches of effects, for stronger effects in younger students, nor that effects
ability EI. Whereas intelligence remained the most important pre- differ depending on the proportion of ethnic minority students in
dictor in all cases, EI was the second most important predictor for the sample. For the other moderators, effects were mixed or
both emotion understanding (RW% ⫽ 31% for EI vs. 19% for limited. There is limited evidence that the effect is stronger: (a) for
conscientiousness) and emotion management (RW% ⫽ 20% for EI less female-dominated samples (this effect was significant for total
vs. 19% for conscientiousness). Relative importance was lower for EI, but not for any of the three streams), (b) for grades than
emotion facilitation (RW% ⫽ 9%) and negligible for emotion standardized test scores (this was significant for total EI and
perception (RW% ⫽ 1%). It is clear that the strategic branches of Stream 2 only), and (c) for humanities versus mathematics/science
EI (emotion understanding and management) are more important performance (this was significant for ability EI only).
than the experiential branches (perception and facilitation). There was evidence of incremental validity of EI over intelligence
Taken together, these results provide support for Hypothesis 11, and personality, but this was largely restricted to mixed EI (which
showing that EI is among the top three most important predictors explained an additional 2.3% of the variance) and the understanding
of academic performance. However, this analysis also highlights and management branches of ability EI (which explained an addi-
that results are quite different for different streams and different tional 3.9% and 3.6% of the variance respectively). That is, self-rated
branches of EI. EI, total ability EI, and the lower two branches of ability EI (emotion
perception and facilitation) provide little to no explanatory power for
Discussion academic performance over intelligence and personality. These dif-
ferences across the three streams suggest that the underlying mecha-
Results from these meta-analyses demonstrate that EI shows a nisms accounting for the EI/performance relationship may differ for
small to moderate relationship with academic performance, of ability EI, self-rated EI, and mixed EI.
similar effect size to well-known noncognitive predictors (e.g.,
␳ ⫽ .20 for EI vs. ␳ ⫽ .22 for conscientiousness, based on the Why Does EI Predict Academic Performance? Insights
current meta-analysis and Poropat [2009]). Ability EI was a sig- Based on Moderators
nificantly stronger predictor than self-report or mixed EI, as hy-
pothesized. Within ability EI, understanding and management In our introduction, we suggested there were three reasons why
branches had a stronger effect than perception or facilitation EI may predict academic performance. First, students with higher
branches. There is no evidence for selective publication of larger EI may be more able to regulate the negative emotions such as
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

20 MACCANN ET AL.

anxiety, boredom, and disappointment involved in academic per-

Note. EI ⫽ emotional intelligence; g ⫽ intelligence; Perf ⫽ academic performance; O ⫽ openness to experience; C ⫽ conscientiousness; E ⫽ extraversion; A ⫽ agreeableness; N ⫽ neuroticism;

Original meta-analysis. b Joseph and Newman (2010). c van der Linden et al. (2017). d Poropat (2009). e Judge, Jackson, Shaw, Scott, & Rich (2007). f van der Linden, te Nijenhuis, & Bakker
n.r./39,595
⫺.34f
Meta-Analytic Correlation Matrix of EI (Stream 1 Total and Four Branches, Stream 2, and Stream 3), Academic Performance, Cognitive Ability, and Big Five Personality

formance. If this is true, emotion management would be respon-

A
sible for the effects. Second, students with higher EI may be better
able to manage the social world around them, forming better
relationships with teachers, peers, and family. If this was the case,

n.r./39,595

n.r./39,595
emotion management would again be responsible for the effect,

.33f

⫺.37f
E

and the effect would be stronger for grades than for standardized
tests. Third, EI competencies may overlap with the academic
competencies required for humanities subjects like history and

n.r./39,595

n.r./39,595

n.r./39,595
language arts (e.g., understanding human motivations and emo-
.27f

.46f

⫺.36f
C

tions). In this case (a) understanding—the knowledge base of

EI—would show the strongest effect and (b) the effect would be
bigger for humanities than sciences. Based on the significant
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

moderations, there is some support for each of these effects, with

n.r./39,595

n.r./39,595

n.r./39,595

n.r./39,595
This document is copyrighted by the American Psychological Association or one of its allied publishers.

.24f

.39f

.30f

⫺.18f
slightly different results for different streams of EI. We discuss the
O

significance moderations below, with respect to these three pro-

posed mechanisms.
46/13,182

56/15,429

61/21,602

38/11,190

61/21,404
.22e

⫺.04e

.02e

.00e

⫺.09e
g

Evidence for Mechanism 1: Is Emotion Management

the Key Ingredient in EI?
Joseph and Newman (2010) proposed a “key conceptual role” of
113/60,442

138/70,926

113/59,986

109/58,552

114/59,554
27/4,801
.12d

.22d

⫺.01d

.07d

⫺.02d
.39a

emotion management for predicting job performance (p. 69), pro-

Perf

Perc ⫽ emotion perception; Fac ⫽ emotion facilitation; Und ⫽ understanding emotions; Man ⫽ managing emotions.
From Current Meta-Analysis (EI/Academic Performance and Intelligence/Academic Performance Cells)

posing emotion management as the proximal predictor of perfor-

mance. The facet-level moderation for ability EI provides partial
support for this assumption, finding that management and under-
Stream 3 EI

90/27,939

standing are jointly the strongest predictors of academic perfor-

19/2,880

30/5,386

31/5,591

30/5,552

30/5,386

30/5,386
.11b

.29b

.38b

.46b

.43b

⫺.53b
.19a

mance. Both these branches (management and understanding)

showed significantly stronger effects than the two other branches.
The effect was larger for understanding than management, but not
significantly so (␳ ⫽ .35 vs. ␳ ⫽ .26) and was equal after account-
Stream 2 EI

ing for the effects of intelligence and personality (partial ␳ ⫽ .22

33/8,492

16/2,158

26/8,479

27/8,566

26/8,479

23/8,479

26/8479
.00b

.29b

.38b

.32b

.31b

⫺.04b
.12a

in both cases). That is, both emotion understanding and emotion

management are active ingredients in the prediction of aca-
demic performance. We believe this is consistent with an in-
terpretation that EI affects academic performance through the
31/4,898

19/4,277

22/3,374

22/3,374

22/3,374

22/3,374

22/3,374
.16b
.26a

.16c

.16c

.18c

.30c

⫺.17c
Man

regulation of academic emotions, but also due to the relevance

of emotion content knowledge for academic performance in the
humanities.
29/4,815

20/4,581

22/3,374

22/3,374

22/3,374

22/3,374

22/3,374

The critical role of emotion understanding for academic perfor-

.39b
.35a

.18c

.09c

.07c

.12c

⫺.09c
Und

mance has implications for comparing ability EI with self-rated EI.

For self-rated EI, many of the effects in our meta-analysis used
instruments that did not include emotion understanding content
Stream 1 EI

25/3,234

18/3,971

23/3,582

23/3,582

23/3,582

23/3,582

23/3,582
.18b
.18a

.10c

.11c

.10c

.17c

⫺.11c

(because they were based on an older definition of EI that did not

Fac

have emotion understanding in the definition). Specifically, 50%

of the Stream 2 citations used the Schutte Self-Report Scale, Trait
(2010), university student population.

Meta-Mood Scale, or the Wong-Law Emotional Intelligence Scale,

29/3,919

21/4,710

23/3,582

23/3,582

24/3,696

23/3,582

24/3,696
.10b
.09a

.07c

.28c

.09c

.15c

⫺.12c
Perc

which do not include a subscale assessing emotion understanding.

Given that ability EI shows the strongest relationship for emotion
understanding, the difference in effect size between ability EI and
self-rated EI measures may in fact represent a difference in content
47/10,258

47/10,258

47/10,258

47/10,258

47/10,258
50/8,826

28/5,383
.25b
.24a

.14c

.09c

.05c

.16c

⫺.09c
Total

(i.e., prediction is greater for tests that include emotion understand-

ing content) rather than a difference in method (ability scales vs.
rating-scales). Many of the more recent self-rated EI tests do
include an emotion understanding component (e.g., Anguiano-
Table 4

Variable

Perf

Carrasco, MacCann, Geiger, Seybert, & Roberts, 2015; Brackett et

O

N
C

E
g

al., 2006).
a
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 21

Table 5
Regressions Predicting Academic Performance From Intelligence (Step 1), Personality (Step 2), and EI (Step 3) Based on the
Correlation Matrix Shown in Table 3 – Ability EI Versus Self-Rated EI Versus Mixed EI

Stream 1: Ability EI Stream 2: Self-rated EI Stream 3: Mixed EI

(Nharmonic ⫽ 18,367) (Nharmonic ⫽ 14,647) (Nharmonic ⫽ 12,939)
Step ␤ ⌬R2 RW RW% ␤ ⌬R2 RW RW% ␤ ⌬R2 RW RW%
ⴱⴱ ⴱⴱ ⴱⴱ
Step 1 .152 .152 .152
Intelligence .38ⴱⴱ .152 69.42% .42ⴱⴱ .143 62.01% .40ⴱⴱ .135 58.01%
Step 2 .071ⴱⴱ .071ⴱⴱ .071ⴱⴱ
Openness .00 .005 2.49% ⫺.01 .006 2.42% ⫺.01 .005 2.29%
Conscientiousness .29ⴱⴱ .044 20.20% .27ⴱⴱ .045 19.52% .27ⴱⴱ .049 21.05%
Extraversion ⫺.05ⴱⴱ .004 1.87% ⫺.07ⴱⴱ .006 2.75% ⫺.10ⴱⴱ .003 1.32%
Agreeableness ⫺.03ⴱⴱ .003 1.17% ⫺.02ⴱ .003 1.36% ⫺.05ⴱⴱ .003 1.24%
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Neuroticism .10ⴱⴱ .002 .88% .12ⴱⴱ .004 1.65% .16ⴱⴱ .002 .81%
.017ⴱⴱ .007ⴱⴱ .023ⴱⴱ
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Step 3
EI .14ⴱⴱ .009 3.96% .09ⴱⴱ .024 10.28% .20ⴱⴱ .036 15.28%
Total R2 .240ⴱⴱ .230ⴱⴱ .246ⴱⴱ
Note. EI ⫽ emotional intelligence; RW ⫽ relative weight.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01.

Evidence for Mechanism 2: Are EI Competencies as put forward by the National Council of Teachers of English
Required for Academic Content? (1996) states that the purpose of reading texts is to “build an
understanding of . . . themselves and the cultures of the United
Moderation analyses largely support the idea that performance States and the world” and the second standard states that the
on academic tasks require some EI competencies. First, academic purpose is “to build an understanding of the many dimensions
performance related significantly more strongly to ability EI than (e.g., philosophical, ethical, aesthetic) of human experience” (p.
to the other two streams. This finding differs from meta-analyses 19). That is, broad statements of content for achievement in
predicting job performance, where ability EI is consistently the
language arts inherently involve an understanding of oneself and
weakest predictor of the three streams (Joseph & Newman, 2010;
of others in terms of the intangible nature of being human—which
Miao et al., 2017; O’Boyle et al., 2011). This difference may relate
we would argue is essentially emotions and social interactions.
to assessment methods. Academic performance is mainly assessed
That is, understanding human emotions and the social and situa-
with objective tasks (i.e., evaluations of a product, such as an
tional causes appear to be an underlying component of achieve-
essay, lab report, speech, worksheet or test), whereas job perfor-
ment in language arts.
mance is most often assessed via supervisor ratings. Similarly,
In addition, the fact that the emotion understanding showed the
ability EI is assessed with objective tasks and self-rated and
strongest relationship to academic performance (as compared to
mixed EI are assessed with rating scales. We would expect
the other four branches) supports the interpretation mentioned
stronger predictor-criterion relationships when predictor and
above, where understanding emotional content is a key part of the
criterion have the same method. As such, a higher relationship
for ability EI (as compared to the other streams) may represent content of language arts education. It is possible to view emotion
method bias rather than content overlap of academic and emo- understanding as a kind of domain-specific knowledge, where the
tional knowledge. content domain is emotions. Content knowledge of emotion words,
However, ability EI (but not self-rated or mixed EI) relates more as well as the causes and consequences of emotions, appear highly
strongly to performance in humanities than sciences. This is one of relevant for understanding character motivations in literature as
the larger differences we found, where the effect was nearly twice well as other academic subject matter relating to people and how
as large for humanities as sciences (␳ ⫽ .38 vs. .21). Objective they shape societies, countries and history (i.e., history, geography,
measurement of performance is similar across humanities and psychology, sociology).
sciences. The academic processes (social context and the student’s One possible interpretation is that the ability EI/academic per-
emotions and emotion regulation in the classroom) are also similar formance association may be due to a third variable—reading
for different subdisciplines. Although subdisciplines differ in the comprehension. Because ability EI tests involve interpreting writ-
degree of social interaction involved, the degree of social interac- ten text, reading comprehension ability may constitute construct-
tion does not align with the humanities versus science categoriza- irrelevant variance on such tests (AERA, APA, & NCME, 2014)
tion (e.g., science frequently involves lab partners or group work, that may partially explain the relationship between EI and aca-
whereas this is rare for mathematics). As such, we interpret this demic performance. This particularly affects understanding and
difference in subject areas to be largely attributable to a difference management tests, which involve more and more complex text
in academic content, and specifically the relevance of emotion (e.g., most management tests involve a paragraph of text in each
knowledge to subjects requiring an understanding of people and item stem). However, the fact that emotion understanding and
their interactions, motivations, and emotions (i.e., literature, his- management predicted academic performance over-and-above the
tory, geography, drama and other humanities subjects). The first effect of intelligence suggests that this confound does not account
standard of The Standards for the English Language Arts (1996), for the entirety of the relationship between ability EI and academic
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

22 MACCANN ET AL.

performance. Nevertheless, the relationship was greatly reduced,

54.98%

2.23%
19.28%
1.73%
1.41%
.83%

19.54%
RW%
Regressions Predicting Academic Performance From Intelligence (Step 1), Personality (Step 2), and EI (Step 3) Based on the Correlation Matrix Shown in Table 3 for the

particularly for emotion understanding. Because the partial corre-

lations remained of small to moderate size after accounting for
Emotion management
(Nharmonic ⫽ 10,736)
intelligence, our interpretation is that the bulk of the content

.138

.006
.048
.004
.004
.002

.049
RW
overlap represents more than a reading comprehension method
effect, particularly for emotion management. Taken together, re-
sults support the suggested mechanism whereby EI predicts aca-
ⴱⴱ

.071ⴱⴱ

.036ⴱⴱ

.258ⴱⴱ
⌬R2

.152

demic performance because of the emotional content required in

academic subjects.
.38ⴱⴱ

.29ⴱⴱ
⫺.07ⴱⴱ
⫺.06ⴱⴱ
.10ⴱⴱ

.20ⴱⴱ
.00
␤

Evidence for Mechanism 3: Does EI Affect Academic

Performance Through Interpersonal Processes?
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

45.48%

1.88%
18.79%
1.26%
1.12%
.76%

30.72%
RW%
This document is copyrighted by the American Psychological Association or one of its allied publishers.

If EI exerts an influence on academic performance via the

ability to develop social relationships in the educational con-
Emotion understanding

text, then EI should have a stronger effect on grades than

(Nharmonic ⫽ 10,786)

standardized tests (as the social networking and relationship

.115

.005
.048
.003
.003
.002

.078
RW

building with other students and teachers should have a stronger

effect on grades than on standardized tests). This difference was
ⴱⴱ

.071ⴱⴱ

.039ⴱⴱ

.261ⴱⴱ
⌬R2

significant only for self-rated EI and the three streams com-

.152

bined (not for ability or mixed EI). Self-rated EI did not relate
to standardized test scores at all (␳ ⫽ ⫺.03). In contrast, ability
.28ⴱⴱ
⫺.06ⴱⴱ
⫺.03ⴱⴱ
.09ⴱⴱ

.22ⴱⴱ

EI and mixed EI related to both grades and standardized tests.

.33

⫺.01
␤

This suggests that academic performance relates to self-rated EI

through relationship building only. In contrast, academic per-
formance relates to ability EI and mixed EI through both
63.28%

2.63%
21.76%
1.56%
1.27%
.84%

8.67%
RW%

relationship building and mechanisms related to regulating ac-

ademic emotions.
(Nharmonic ⫽ 10,586)
Emotion facilitation

For all three streams of EI, there is evidence that higher EI

.142

.006
.049
.003
.003
.002

.019
RW

relates to building social relationships in a school environment.

Ability EI relates to peer-nominations of reciprocal friendship
in college students and to higher-quality of social interactions
ⴱⴱ

.071ⴱⴱ

.009ⴱⴱ

.232ⴱⴱ
⌬R2

.152

with others (Lopes et al., 2004; Lopes, Salovey, Coté, Beers, &
Petty, 2005). Self-rated EI predicts greater social support in
both high school and university students (Ciarrochi, Chan, &
.39ⴱⴱ

.29ⴱⴱ
⫺.06ⴱⴱ

.10ⴱⴱ

.10ⴱⴱ

Bajgar, 2001; Kong, Zhao, & You, 2012). Mixed EI is associ-

.00

⫺.03
␤

ated with peer reports of cooperative behavior (Mavroveli,

Petrides, Rieffe, & Bakker, 2007; Petrides, Sangareau, Furn-
69.62%

2.83%
22.50%
1.53%
1.36%
.86%

1.30%

ham, & Frederickson, 2006). There is also evidence that both

RW%

ability EI and mixed EI relate to using more effective strategies

to regulate negative emotions (Peña-Sarrionandia, Mikolajczak,
(Nharmonic ⫽ 11,049)
Emotion perception

& Gross, 2015).

.150

.006
.049
.003
.003
.002

.003
RW

Taken together with these findings, we propose that differ-

ences between the three streams of EI relate to the number of
mechanisms that underlie the EI/performance relationship. Spe-
ⴱⴱ

.071ⴱⴱ

.222ⴱⴱ
⌬R2

.152

.000

cifically, (a) self-rated EI predicts academic performance only

through a relationship building pathway (students with higher
Note. EI ⫽ emotional intelligence.

emotional self-efficacy can build better relationships with

.41ⴱⴱ

.30ⴱⴱ
⫺.06ⴱⴱ

.10ⴱⴱ
Four Branches of Ability EI

.00

⫺.01

⫺.02
␤

teachers and peers), (b) mixed EI predicts academic perfor-

mance through both relationship building and the regulation of
academic emotions, and (c) ability EI predicts academic per-
Conscientiousness

formance through relationship building, regulation of academic

Agreeableness

emotions, and also through emotion content knowledge require-

Extraversion

Neuroticism
Intelligence
Step

ments of some academic areas. This explanation accounts for

Openness

p ⬍ .01.

the relatively greater prediction of academic performance by

Table 6

Total R2
Step 1

Step 2

Step 3

ability EI than mixed EI than self-rated EI and is consistent with

EI

the pattern of moderators we found.

ⴱⴱ
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 23

The Relative Importance of EI to other students occurs through in-person conversation, with access
Academic Achievement to multiple channels of information (e.g., facial and vocal expres-
sion, body language, and real-time clarification of misunderstand-
One of the critical drivers of EI’s early popularity was the idea ings). Online communication is more often based on text (e.g.,
that emotional skills are more important than intelligence in pre- discussion boards, emails, or computer chat). Most neurotypical
dicting life success. Indeed, the title of Daniel Goleman’s first people find it more difficult to detect another person’s emotions
book, the catalyst for EI’s snowballing popularity, was Emotional and social needs from text rather than face-to-face contact. As
intelligence: Why It Can Matter More Than IQ. The 1995 cover such, greater emotional skills are required to build relationships
story of TIME magazine made similar claims, stating that “emo- with the instructor or other students in an online environment.
tions, not IQ, may be the true measure of human intelligence” Thus, social and emotional skills (both self-regulation and inter-
AQ: 9 (Gibbs, 1995, p. 60). These early claims were generally not borne personal skills) may become increasingly important as tertiary
out by research on job performance. Although EI predicts better education involves a greater amount of online content.
job performance (Joseph & Newman, 2010; O’Boyle et al., 2011),
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

a critical mass of research indicates that intelligence is a much

This document is copyrighted by the American Psychological Association or one of its allied publishers.

stronger predictor and is in fact the single best predictor of job Practical Implications
performance (Ree & Earles, 1992; Salgado et al., 2003; Schmidt &
Hunter, 1998). We found largely similar results for academic One of the major findings of this meta-analysis is that different
performance. Although EI predicts academic performance, intelli- parts of EI are differentially important for academic performance.
gence was a much stronger predictor, with relative importance Any applied uses of EI in education seem limited to the three parts
analysis indicating that cognitive ability was the single most im- of EI with nontrivial incremental validity: mixed EI, emotion
portant predictor of academic performance. management ability, and emotion understanding ability. There are
Although the popular press hype about EI was not substantiated, three broad applications that might be considered: (a) identifying
we nevertheless believe that demonstrating a small to moderate students at risk for failure, attrition, or underperformance; (b)
effect size is informative for research and practice. Moreover, selection decisions for high-stakes educational opportunities; and
some of the recent changes occurring in education and assessment (c) policy decisions about the relative cost versus benefit of im-
practices may increase the importance of noncognitive qualities, plementing SEL or EI training programs in schools.
including EI. The first two applications (identifying at risk students, and
The first such change to modern assessment and learning prac- high-stakes selection) require careful consideration of response
tice is the increasing use of group activities, including collabora- distortion issues. Particularly in a high-stakes selection context,
tive group assessments (Ahles & Bosworth, 2004). Managing the test-takers are motivated to gain high scores and will distort their
social relationships and interpersonal conflicts of the group may responses on rating scales to ‘fake high’ (Birkeland, Manson,
thus become more and more reflected in students’ end of semester Kisamore, Brannick, & Smith, 2006; Viswesvaran & Ones, 1999).
grades. A second change to education practices is the extent to Faking is a consequential issue with personality scales, which use
which graduate attributes (also referred to as 21st century skills or self-report or observer-report ratings that test takers can fake.
noncognitive constructs) are emphasized by schools and universi- Observer-reports do not necessarily solve this problem, as the
ties (e.g., Clarke, Double, & MacCann, 2017). Graduate attributes observers are often not impartial, but may be school staff with a
often include social-emotional skills such as leadership, commu- vested interest in their students gaining entrance to prestigious
nication, teamwork, and intercultural competencies, with some colleges or programs. Faking is a problem for rating-scale mea-
institutions explicitly including EI as a graduate attribute. For sures of EI, but not for ability scales (Day & Carroll, 2008; Grubb
example, Australia uses Goleman’s model of EI as the basis for its & McDaniel, 2007; Tett, Freund, Christiansen, Fox, & Coaster,
national K–10 curriculum of personal and social competencies that 2012). The current meta-analysis is the first to demonstrate that the
students should be developing as they progress through primary relationship between EI and academic performance holds for
and secondary education (ACARA, 2017). Schools and universi- ability-based tests as well as rating scales (in fact, the relationship
ties are increasingly attempting to embed these graduate qualities is actually higher for ability-based EI tests as compared to rating
within the content that is taught and assessed. As such, high grades scales). As such, we demonstrate a pathway that might provide
might increasingly reflect skill development in these areas. modest increments in high-stakes education selection decisions—
A third change to the classroom is the extent to which computers using ability-based EI assessments of understanding and managing
and technology are now an integral part of education. In tertiary emotions (based on current results, other parts of ability EI are not
education, there are a large and increasing number of online only important). Ability-based EI tests are already used for selection
courses or courses that have at least some online-only content into medical schools in several countries, and evidence supports
(most famously, the Massive Open Online Courses [MOOCs]). their use for selecting better candidates (Libbrecht, Lievens,
There are two main differences between traditional face-to-face Carette, & Côté, 2014; Lievens & Sackett, 2006). However, such
learning and online learning. First, in a traditional face-to-face tests are rarely used in other broader education selection contexts.
university course, the schedule of learning is set by the schedule of If EI is considered as a selection procedure (perhaps as an add-on
the face-to-face lectures. In contrast, an online only course requires to intelligence and personality assessment), we suggest that ability
the learner to manage their own schedule of accessing online tests of understanding and managing emotions be preferred over
content, such that students with poor time management will not rating scales (due to response distortion) or tests of facilitation or
succeed (MacCann, Fogarty, & Roberts, 2012). Second, in a tra- management (due to low incremental prediction over intelligence
ditional face-to-face course, communication with teachers and and personality).
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

24 MACCANN ET AL.

A national and international focus on standardized tests to universities report similar findings regarding the largest increases
measure academic performance and milestones has lead schools, for emotion understanding. For example, Pool and Qualter (2012)
districts, states, and countries to focus on achievement in the conducted a training study in university students and found the
narrow range of academic content that such tests focus on. Along- largest increase in ability EI was for emotion understanding (and
side this, classroom teachers face increasing challenges to their the second-largest for emotion management). Moreover, evidence
workload, including adapting the curriculum to individual stu- from the RULER Feeling Words curriculum (an EI development
dents’ needs, the mainstreaming of students with special educa- program for secondary school students) shows that EI training
tional requirements, and adapting to rapidly changing curriculum programs increase grades as well as social and emotional compe-
and policy (Skaalvik & Skaalvik, 2007). Against this background, tencies. Specifically, students completing the RULER showed
devoting resources to teaching children EI skills can be seen as improved school grades as well as improved teacher ratings of
taking teacher resources and classroom time away from more social and emotional competencies compared to control groups
critical activities that will increase test scores and achievement. (Brackett, Rivers, Reyes, & Salovey, 2012). In fact, the relative
What our meta-analysis shows is that EI skills are in fact associ- increase in school grades was a larger effect than the relative
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ated with higher academic performance. This implies that time increase in social and emotional competencies. That is, EI training
This document is copyrighted by the American Psychological Association or one of its allied publishers.

spent teaching EI skills may not necessarily detract from student programs are likely to increase academic performance as well as
achievement, given that higher EI students also show higher social and emotional outcomes, such that education decision-
achievement. Again, we highlight the different importance of the makers and policymakers are not faced with a decision of whether
four EI abilities as a guide for where to focus skills training—a to invest in social/emotional wellbeing at the expense of student
focus on perceiving emotions is likely to be less useful than a focus achievement— evidence suggests that these programs likely do
on understanding and managing emotions. both. This is a critical piece of information for schools deciding
Our meta-analysis also has implications for the effects of the where to best allocate their resources.
such training programs (or a focus on EI more generally) on the
known achievement gaps between ethnic groups and between
Limitations
males and females. Although there is evidence that the Black-
White achievement gap is slowly closing, differences in the Our results demonstrated only that EI and academic perfor-
achievement for minority students compared with White students mance are significantly associated, but not that higher EI causes
remain substantial, at around 0.75 standard deviations for Black higher achievement. Only three of the citations reported a longi-
students and around 0.60 standard deviations for Hispanic students tudinal design, such that the empirical evidence for EI causing later
(Hansen, Mann Levesque, Quintero, & Valant, 2018). There is also achievement is very weak (Costa & Faria, 2015; Qualter et al.,
increasing evidence that males are falling behind females in terms 2012; Stewart & Chisholm, 2012). This association could occur
of the grades they receive and their participation in higher educa- because (a) higher EI causes increased academic performance, (b)
tion (Fortin, Oreopoulos, & Phipps, 2015). Against this back- higher achievement causes increased EI, or (c) there are one or
ground, it is important to note that the effect of EI on academic more variables that influence both EI and academic performance.
performance does not appear to differ for minority students versus In the introduction, we outlined the reasons we believe theoreti-
White students, and that gender differences are negligible and cally that EI could cause later achievement. However, there are
when significant effect favor males (who currently show lower also feasible pathways by which greater academic performance
achievement). These results imply, at the very least, that efforts to could cause higher EI. Greater academic performance could fea-
improve EI are unlikely to widen the achievement gaps. sibly result in increased self-esteem, greater opportunities for
The key role of emotion understanding and management is also social and emotional development, and higher expectations for
important to consider in terms of EI training programs. Three social skills and emotion regulation. High academic performance
recent meta-analyses on the effectiveness of EI training have may act as a gateway for gifted and talented programs, streaming
reported significant increases in EI, with effect sizes of .45, .46, into enrichment activities, and a culture of high expectations from
AQ: 10 .51, and .61 (Hodzic, Scharfen, Ripoll, Holling, & Zenasni, 2018; teachers, parents, and communities that permeate social and emo-
AQ: 11 Mattingly & Kraiger, 2019; Schutte, Malouff, & Thorsteinsson, tional behaviors as well as academic ones via the halo effect
AQ: 12 2013). Hodzic et al. found that programs based on the ability (Nisbett & Wilson, 1977). Conversely, low academic performance
model were significantly more effective than those based on mixed may act as a barrier to opportunities to develop social and emo-
models (g ⫽ .60 vs. .31), and that emotion understanding showed tional skills through loss of privileges for academic failures (e.g.,
the largest increase of all the ability EI branches—significantly losing recess playtime or evening socialization to complete work
more than emotion facilitation (g ⫽ .69 vs. .42). That is, it seems or denied extracurricular activity participation because of course
that programs are effective for increasing ability EI, and particu- failure), the development of strong negative emotions surrounding
larly its emotion understanding facet. This is highly relevant for school and schoolwork, and the correspondingly low expectations
our own meta-analysis, where ability EI (and specifically emotion for social and emotional behaviors. It seems likely that the reality
understanding) showed the highest association with academic per- is complex, with bidirectional effects of academic and emotional
formance. That is, EI training seems to produce the strongest development, particularly in the earlier years of school.
increases in exactly those competencies that are most relevant for One further limitation of the current article concerns the use of
academic performance. the meta-analytic correlation matrix (used to test Hypotheses 10
Although Hodzic et al. did not distinguish between EI training and 11). This was composed of estimates taken from different
programs for workplace applications and EI training programs for journal articles from different research teams, and therefore did not
schools and universities, several studies conducted in schools and use the same methods for estimation nor the same samples. Al-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 25

though we used RVE in the current study, all other sources for students. Pakistan Journal of Psychological Research, 19, 105–121.
effect sizes were obtained by aggregating multiple effect sizes Retrieved from [Link]
ⴱ
from the same study. All studies except for Poropat (2009) cor- Abel, N. R. (2014). Trait emotional intelligence, perceived discrimina-
rected for unreliability as well as range restriction. The personality/ tion, and academic achievement among African American and Latina/o
academic performance estimation was not corrected for range high school students: A study of academic resilience (Unpublished
restriction of measurement in either the predictor or criterion doctoral dissertation). Minnesota State University, Mankato, MN.
ACARA. (2017). Australian curriculum. Personal and social capability
(Poropat, 2009). The possible effect of this would be to underes-
learning continuum: Retrieved December 6, 2017, from [Link]
timate the prediction and relative importance of personality traits
.[Link]/resources/General_capabilities_-PSC_-_learning_
such as conscientiousness. [Link]
ACT. (2014). American College Test (ACT). Iowa City, IA: Research
Future Research and Recommendations Services, ACT.
ⴱ
Adeyemo, D. A. (2007). Moderating influence of emotional intelligence
One obvious future direction for further research is to test our on the link between academic self-efficacy and achievement of univer-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

three proposed mechanisms of the EI/performance relationship: (a) sity students. Psychology and Developing Societies, 19, 199 –213. http://
This document is copyrighted by the American Psychological Association or one of its allied publishers.

social relationship building, (b) regulation of academic emotions, [Link]/10.1177/097133360701900204

and (c) content overlap between EI and academic subject matter. AERA, APA, & NCME. (2014). Standards for educational and psycho-
For point a, an analysis of content overlap between the competen- logical testing. Washington, DC: AERA.
ⴱ
cies of EI and the different processes required for success in Afolabi, O. A., Ogunmwonyi, E., & Okediji, A. (2009). Influence of
different disciplines could be undertaken by a panel of educators. emotional intelligence and need for achievement on interpersonal rela-
Longitudinal research involving all three streams could test tions and academic achievement of undergraduates. Educational Re-
whether all three mediate ability EI, (a) and (b) mediate mixed EI, search Quarterly, 33, 60 –72. Retrieved from [Link]
ⴱ
Agnoli, S., Mancini, G., Pozzoli, T., Baldaro, B., Russo, P. M., &
and (a) along mediates self-rated EI, as we proposed. As we
Surcinelli, P. (2012). The interaction between emotional intelligence and
mention above, there is a paucity of long-term longitudinal re-
cognitive ability in predicting scholastic performance in school-aged
search on EI and academic performance. As such, examining
children. Personality and Individual Differences, 53, 660 – 665. http://
mediators of the link as well as conducing lagged panel models to [Link]/10.1016/[Link].2012.05.020
tease apart the direction of causation is important. ⴱ
Ahammed, S., Abdullah, A. S., & Hassane, S. H. (2011). The role of
Although there is ample evidence that training EI works (e.g., emotional intelligence in the academic success of United Arab Emirates
Hodzic et al., 2018; Mattingly & Kraiger, 2019; Schutte et al., 2013), university students. International Education, 41, 7–25. Retrieved from
we are not aware of experimental studies on EI training that examine [Link]
the effects of training different branches of EI. Such designs would Ahles, C. B., & Bosworth, C. C. (2004). The perception and reality of
isolate which facets of EI are most relevant for the improvement of student and workplace teams. Journalism & Mass Communication Ed-
which types of outcomes and would also provide stronger evidence ucator, 59, 41–59. [Link]
for the causal direction from EI to academic performance. Ahmad, I. (2010). Trait-emotional intelligence as predictor of academic
performance. Pakistan Journal of Social and Clinical Psychology, 8,
54 – 65. Retrieved from [Link]
Conclusion ⴱ
Ahmad, I. (2011). Psychological predictors of college students’ perfor-
Although we know that intelligence and conscientiousness are mance. Pakistan Journal of Psychological Research, 26, 87–103. Re-
trieved from [Link]
collectively the most important psychological characteristics ⴱ
Ahmad, I., & Rana, S. (2012). Affectivity, achievement motivation, and
needed for academic performance, this article highlights that there
academic performance in college students. Pakistan Journal of Psycho-
is a third broad psychological characteristic that may help students logical Research, 27, 107–120. Retrieved from [Link]
succeed—EI. The different varieties of EI most likely predict .pk/pjpr/[Link]/pjpr
academic performance via different pathways. Emotional self- ⴱ
Alhashemi, S. E. (2014). Measuring emotional intelligence of university
efficacy (the self-beliefs about one’s emotional skills captured by students: A comparison between China and Bahrain. International Jour-
self-rated EI) is the least important. Knowledge about the causes nal of Social and Organizational Dynamics in IT, 3, 59 –76. [Link]
and consequences of emotions and a vocabulary of emotions .[Link]/10.4018/ijsodit.2013100104
ⴱ
words, along with knowing how to manage emotional situations Alumran, J. I., & Punamaki, R.-L. (2008). Relationship between gender,
are potentially the most important parts of EI for academic per- age, academic achievement, emotional intelligence, and coping styles
formance. It is not enough to be smart and hardworking—to have among Bahraini adolescents. Individual Differences Research, 6, 104 –
the added edge for success, students must also be able to under- 119. Retrieved from [Link]
AQ: 13 stand and manage emotions to succeed at school. individ-differ-res/
ⴱ
Amdurer, E., Boyatzis, R. E., Saatcioglu, A., Smith, M. L., & Taylor,
S. N. (2014). Long term impact of emotional, social and cognitive
References intelligence competencies and GMAT on career and life satisfaction and
career success. Frontiers in Psychology, 5, 1447. [Link]
References included in the meta-analysis are marked with an asterisk.
.3389/fpsyg.2014.01447
ⴱ ⴱ
Abdo, N. (2012). Academic performance and social/emotional compe- Anand, A. F. J., Ramgopal, C. N., & Swaminathan, V. D. (2016).
tence in adolescence (Unpublished doctoral dissertation). Yeshiva Uni- Emotional intelligence and academic performance of graduate engineer-
versity, New York City, NY. ing students. Research Journal of Pharmaceutical, Biological and
ⴱ
Abdullah, M. C., Elias, H., Mahyuddin, R., & Uli, J. (2004). Emotional Chemical Sciences, 7, 2540 –2551. Retrieved from [Link]
intelligence and academic achievement among Malaysian secondary .com/
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

26 MACCANN ET AL.

Anguiano-Carrasco, C., MacCann, C., Geiger, M., Seybert, J. M., & faking on personality measures. International Journal of Selection and
Roberts, R. D. (2015). Development of a forced-choice measure of Assessment, 14, 317–335. [Link]
typical-performance emotional intelligence. Journal of Psychoeduca- .00354.x
ⴱ
tional Assessment, 33, 83–97. [Link] Borin, E. B. (2013). The relationship between emotional intelligence and
4550387 college success for students with learning disabilities (Unpublished
Antonakis, J., Ashkanasy, N. M., & Dasborough, M. T. (2009). Does doctoral dissertation). Capella University, Minneapolis, MN.
leadership need emotional intelligence? The Leadership Quarterly, 20, Boulanger, F. D. (1981). Ability and science learning: A quantitative
247–261. [Link] synthesis. Journal of Research in Science Teaching, 18, 113–121. http://
ⴱ
Arbabisarjou, A., Maede-Sadat, R., Narges, M., & Shekoofeh-Sadat, R. [Link]/10.1002/tea.3660180203
ⴱ
(2013). Relationship between different types of intelligence and student Bowman, J. K. (2007). The utility of emotional intelligence as a predictor
achievement. Life Science Journal, 10, 128 –133. of school psychologists’ clinical competence (Unpublished doctoral dis-
ⴱ
Aremu, O. A., Tella, A., & Tella, A. (2006). Relationship among emo- sertation). St. John’s University, New York, NY.
tional intelligence, parental involvement and academic achievement of Boyatzis, R. E., Goleman, D., & Rhee, K. S. (2000). Clustering compe-
secondary school students in Ibadan, Nigeria. Electronic Journal of tence in emotional intelligence: Insights from the Emotional Compe-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Research in Educational Psychology, 5, 163–180. Retrieved from http:// tence Inventory. In R. Bar-On & J. D. A. Parker (Eds.) The handbook of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

[Link]/revista/new/english/[Link] emotional intelligence: Theory, development, assessment, and applica-

Ashkanasy, N. M., & Daus, C. S. (2005). Rumors of the death of emotional tion at home, school, and in the workplace (pp. 343–362). San Francisco,
intelligence in organizational behavior are vastly exaggerated. Journal of CA: Jossey-Bass.
ⴱ
Organizational Behavior, 26, 441– 452. [Link] Boyce, D. A. (2002). The correlation of emotional intelligence, academic
.320 success, and cognitive ability in master’s level physical therapy students
Bachan, R. (2017). Grade inflation in U. K. higher education. Studies in (Unpublished doctoral dissertation). Spalding University, Louisville,
Higher Education, 42, 1580 –1600. [Link] KY.
ⴱ
.2015.1019450 Brackett, M. A., & Mayer, J. D. (2003). Convergent, discriminant, and
ⴱ
Barchard, K. A. (2003). Does emotional intelligence assist in the predic- incremental validity of competing measures of emotional intelligence.
tion of academic success? Educational and Psychological Measurement, Personality and Social Psychology Bulletin, 29, 1147–1158. [Link]
63, 840 – 858. [Link] .[Link]/10.1177/0146167203254596
ⴱ
Bar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J., & Brackett, M. A., Mayer, J. D., & Warner, R. M. (2004). Emotional
van IJzendoorn, M. H. (2007). Threat-related attentional bias in anxious intelligence and its relation to everyday behaviour. Personality and
and nonanxious individuals: A meta-analytic study. Psychological Bul- Individual Differences, 36, 1387–1402. [Link]
letin, 133, 1–24. [Link] S0191-8869(03)00236-8
ⴱ
Barisonek, E. S. (2005). The relationship between emotional intelligence, Brackett, M. A., Rivers, S. E., Reyes, M. R., & Salovey, P. (2012).
academic achievement and academic production among third- and Enhancing academic performance and social and emotional competence
sixth-grade students (Unpublished master’s thesis). State University of with the RULER Feeling Words Curriculum. Learning and Individual
New York at Buffalo, New York, NY. Differences, 22, 218 –224. [Link]
Bar-On, R. (2000). Emotional and social intelligence: Insights from the Brackett, M. A., Rivers, S. E., Shiffman, S., Lerner, N., & Salovey, P.
Emotional Quotient Inventory. In R. Bar-On & J. D. A. Parker (Eds.) (2006). Relating emotional abilities to social functioning: A comparison
The handbook of emotional intelligence: Theory, development, assess- of self-report and performance measures of emotional intelligence. Jour-
ment, and application at home, school, and in the workplace (pp. nal of Personality and Social Psychology, 91, 780 –795. [Link]
363–388). San Francisco, CA: Jossey-Bass. .org/10.1037/0022-3514.91.4.780
Bar-On, R. (2006). The Bar-On model of emotional-social intelligence Brackett, M. A., Warner, R. M., & Bosco, J. S. (2005). Emotional intelli-
(ESI). Psicothema, 18, 13–25. gence and relationship quality among couples. Personal Relationships,
Barrick, M. R., & Mount, M. K. (1991). The Big 5 Personality Dimensions 12, 197–212. [Link]
ⴱ
and Job-Performance - a Metaanalysis. Personnel Psychology, 44, 1–26. Bradshaw, F. B. (2008). Exploring the relationship between emotional
[Link] intelligence and academic achievement in African American female
ⴱ
Bastian, V. A., Burns, N. R., & Nettelbeck, T. (2005). Emotional intelli- college students (Unpublished doctoral dissertation). College of Notre
gence predicts life skills, but not as well as personality and cognitive Dame of Maryland, Baltimore, MD.
ⴱ
abilities. Personality and Individual Differences, 39, 1135–1145. http:// Brouzos, A., Misailidi, P., & Hadjimattheou, A. (2014). Associations
[Link]/10.1016/[Link].2005.04.006 between emotional intelligence, socio-emotional adjustment, and aca-
ⴱ
Berenson, R., Boyles, G., & Weaver, A. (2008). Emotional Intelligence as demic achievement in childhood: The influence of age. Canadian Jour-
a predictor for success in online learning. The International Review of nal of School Psychology, 29, 83–99. [Link]
Research in Open and Distance Learning, 9, 1–17. [Link] 0829573514521976
ⴱ
.19173/irrodl.v9i2.385 Carrothers, R. M., Gregory, S. W., Jr., & Gallagher, T. J. (2000). Mea-
Bierman, K. L., Coie, J. D., Dodge, K. A., Greenberg, M. T., Lochman, suring emotional intelligence of medical school applicants. Academic
J. E., McMahon, R. J., . . . the Conduct Problems Prevention Research Medicine, 75, 456 – 463. [Link]
Group. (2010). The effects of a multiyear universal social-emotional 200005000-00016
learning program: The role of student and school characteristics. Journal CASEL. (2003). Safe and sound: An educational leader’s guide to
of Consulting and Clinical Psychology, 78, 156 –168. [Link] evidence-based social and emotional learning (SEL) programs. Chi-
10.1037/a0018607 cago, IL: CASEL in collaboration with the Mid-Atlantic Regional Ed-
ⴱ
Billings, C. E. W., Downey, L. A., Lomas, J. E., Lloyd, J., & Stough, C. ucational Laboratory The Laboratory for Student Success. Retrieved
(2014). Emotional Intelligence and scholastic achievement in pre- from [Link]
ⴱ
adolescent children. Personality and Individual Differences, 65, 14 –18. Castro-Johnson, M., & Wang, A. Y. (2003). Emotional intelligence and
[Link] academic performance of college honors and non-honors freshmen.
Birkeland, S. A., Manson, T. M., Kisamore, J. L., Brannick, M. T., & Journal of the National Collegiate Honors Council, 4, 105–115. Re-
Smith, M. A. (2006). A meta-analytic investigation of job applicant trieved from [Link]
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 27

ⴱ
Catalina, C. C., Stanescu, D. F., & Mohorea, L. (2012). Academic self- Day, A. L., & Carroll, S. A. (2008). Faking emotional intelligence (EI):
efficacy, emotional intelligence and academic achievement of Romanian Comparing response distortion on ability and trait-based EI measures.
students: Results from an exploratory study. Journal of Educational Journal of Organizational Behavior, 29, 761–784. [Link]
Sciences and Psychology, 2, 41–52. Retrieved from [Link] .1002/job.485
[Link]/ Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007). Intelligence and
ⴱ
Cavins, B. J. (2006). The relationship between emotional-social intelli- educational achievement. Intelligence, 35, 13–21. [Link]
gence and leadership practices among college student leaders (Unpub- .1016/[Link].2006.02.001
ⴱ
lished doctoral dissertation). Bowling Green State University, Bowling Di Fabio, A., & Palazzeschi, L. (2009). An in-depth look at scholastic
Green, OH. success: Fluid intelligence, personality traits or emotional intelligence?
Chaplin, T. M., & Aldao, A. (2013). Gender differences in emotion Personality and Individual Differences, 46, 581–585. [Link]
expression in children: A meta-analytic review. Psychological Bulletin, 10.1016/[Link].2008.12.012
ⴱ
139, 735–765. [Link] Di Fabio, A., & Palazzeschi, L. (2015). Beyond fluid intelligence and
ⴱ
Chapman, B. P., & Hayslip, B., Jr. (2005). Incremental validity of a personality traits in scholastic success: Trait emotional intelligence.
measure of emotional intelligence. Journal of Personality Assessment, Learning and Individual Differences, 40, 121–126. [Link]
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

85, 154 –169. [Link] .1016/[Link].2015.04.001

ⴱ
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Cheshire, M. H., Strickland, H. P., & Carter, M. R. (2015). Comparing Digman, J. M., & Inouye, J. (1986). Further specification of the five robust
traditional measures of academic success with emotional intelligence factors of personality. Journal of Personality and Social Psychology, 50,
scores in nursing students. Asia-Pacific Journal of Oncology Nursing, 2, 116 –123. [Link]
ⴱ
99 –106. [Link] Ditterline, J. W. (2012). The predictive validity of self-reported emotional
ⴱ
Chew, B. H., Zain, A. M., & Hassan, F. (2013). Emotional intelligence intelligence in children (Unpublished doctoral dissertation). University
and academic performance in first and final year medical students: A of Florida, Gainesville, FL.
ⴱ
cross-sectional study. BMC Medical Education, 13, 44–54. [Link] Doring, D. P. (2006). Social and academic predictors of success for gifted
.org/10.1186/1472-6920-13-44 children (Unpublished doctoral dissertation). St. John’s University, New
Ciarrochi, J., Chan, A. Y., & Bajgar, J. (2001). Measuring emotional York City, NY.
ⴱ
intelligence in adolescents. Personality and Individual Differences, 31, Downey, L. A., Lomas, J., Billings, C., Hansen, K., & Stough, C. (2014).
1105–1119. [Link] Scholastic success: Fluid intelligence, personality, and emotional intel-
ⴱ
Clark, P. R. (2004). An analysis of emotional intelligence: Its relationship ligence. Canadian Journal of School Psychology, 29, 40 –53. [Link]
to student performance and the implications for education (Unpublished .[Link]/10.1177/0829573513505411
ⴱ
doctoral dissertation). Arkansas State University, Jonesboro, AR. Downey, L. A., Mountstephen, J., Lloyd, J., Hansen, K., & Stough, C.
Clarke, I. E., Double, K., & MacCann, C. (2017). Rethinking how we (2008). Emotional intelligence and scholastic achievement in Australian
prepare students for the workforce. In J. Burrus, K. Mattern, B. Naemi, adolescents. Australian Journal of Psychology, 60, 10 –17. [Link]
& R. Roberts (Eds.), Building better students: Preparation for the .org/10.1080/00049530701449505
ⴱ
workforce (pp. 229 –243). New York, NY: Oxford University Press. Drago, J. M. (2005). The relationship between emotional intelligence and
ⴱ
Codier, E., & Odell, E. (2014). Measured emotional intelligence ability academic achievement in nontraditional college students (Unpublished
and grade point average in nursing students. Nurse Education Today, 34, doctoral dissertation). Walden University, Minneapolis, MN, United
608 – 612. [Link] States.
ⴱ
Cohen, J. (1962). The statistical power of abnormal-social psychological Drati, B. A. (2010). Closing the achievement gap: The relationship
research: A review. The Journal of Abnormal and Social Psychology, 65, between emotional intelligence, race identity attitudes, and academic
145–153. [Link] achievement with African American adolescents (Unpublished doctoral
Cohen, J. (1988). Statistical power analysis for the behavioral sciences dissertation). California State University, Long Beach, CA.
(2nd ed.). Hillsdale, NJ: Erlbaum. Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., &
ⴱ
Collins, S. B. (2012). The emotional intelligence profiles and cognitive Schellinger, K. B. (2011). The impact of enhancing students’ social and
measures of nurse anaesthesia students in the South-Eastern United emotional learning: A meta-analysis of school-based universal interven-
States (Unpublished doctoral dissertation). Andrews University, Berrien tions. Child Development, 82, 405– 432. [Link]
Springs, MI. .1467-8624.2010.01564.x
ⴱ ⴱ
Colston, R. (2008). The relationship between emotional intelligence and Edison, B. R. (2003). Emotional intelligence as a mediating variable in
academic achievement: Implications of birth order based on social rank Astin’s I-E-O model of higher education (Unpublished doctoral disser-
for nontraditional adult learners (Unpublished doctoral dissertation). tation). The College of William and Mary, Williamsburg, VA.
ⴱ
Capella University, Minneapolis, MN. Evenson, E. A. (2008). Examining the relationship between emotional
ⴱ
Costa, A., & Faria, L. (2015). The impact of emotional intelligence on intelligence and college success (Unpublished doctoral dissertation).
academic achievement: A longitudinal study in Portuguese secondary Capella University, Minneapolis, MN.
school. Learning and Individual Differences, 37, 38 – 47. [Link] Ewing, M., Huff, K., Andrews, M., & King, K. (2005). Assessing the
.org/10.1016/[Link].2014.11.011 reliability of skills measured by the SAT (Research Notes, RN-24). New
ⴱ
Cyr, J. (2007). Emotional intelligence as a predictor of performance in York, NY: College Board.
ⴱ
college courses (Unpublished doctoral dissertation). Rutgers, The State Fahim, M., & Pishghadam, R. (2007). On the role of emotional, verbal,
University of New Jersey, New Brunswick, NJ. and psychometric intelligences in the academic achievement of students
Davies, M., Stankov, L., & Roberts, R. D. (1998). Emotional intelligence: majoring in language learning. Asian EFL Journal, 9, 240 –253. Re-
In search of an elusive construct. Journal of Personality and Social trieved from [Link]
ⴱ
Psychology, 75, 989 –1015. [Link] Fallahzadeh, H. (2011). The relationship between emotional intelligence
.989 and academic achievement in medical science students in Iran. Proce-
Dawda, D., & Hart, S. D. (2000). Assessing emotional intelligence: Reli- dia: Social and Behavioral Sciences, 30, 1461–1466. [Link]
ability and validity of the Bar-On Emotional Quotient Inventory (EQ-i) 10.1016/[Link].2011.10.283
in university students. Personality and Individual Differences, 28, 797– Fan, H. Y., Jackson, T., Yang, X. G., Tang, W. Q., & Zhang, J. F. (2010).
812. [Link] The factor structure of the Mayer-Salovey-Caruso Emotional Intelli-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

28 MACCANN ET AL.

gence Test V 2.0 (MSCEIT): A meta-analytic structural equation mod- Goleman, D. (1995). Emotional intelligence: Why it can matter more than
eling approach. Personality and Individual Differences, 48, 781–785. IQ. London, UK: Bloomsbury.
[Link] Goleman, D. (1998). Working with emotional intelligence. London, UK:
ⴱ
Fannin, B. E. (2002). The contributions of emotional intelligence to Bloomsbury.
ⴱ
academic achievement and production (Unpublished doctoral disserta- Goodwin, W. N. (2016). Assessing the link between emotional intelligence
tion). Walden University, Minneapolis, MN. and online student achievement (Unpublished doctoral dissertation).
ⴱ Texas A&M University, Corpus Christi, TX.
Fatum, B. A. (2008). The relationship between emotional intelligence and
academic achievement in elementary-school children (Unpublished doc- Gross, J. J. (2008). Emotion regulation. In M. Lewis, J. M. Haviland-Jones,
toral dissertation). University of San Francisco, San Francisco, CA. & L. F. Barrett (Eds.), Handbook of emotions (3rd Edition ed., pp.
ⴱ 497–513). New York, NY: Guilford Press.
Fayombo, G. A. (2012). Relating emotional intelligence to academic
achievement among university students in Barbados. The International Gross, J. J., & Thompson, R. A. (2007). Emotion regulation: Conceptual
Journal of Emotional Education, 4, 43–54. foundations. In J. J. Gross (Ed.), Handbook of emotion regulation (pp.
ⴱ 3–24). New York, NY: Guilford Press. AQ: 14
Feldman, J. M. (2004). The relationship among college freshmen’s cog-
nitive risk tolerance, academic hardiness, and emotional intelligence Grubb, W., III, & McDaniel, M. A. (2007). The fakability of Bar-On’s
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

and their usefulness in predicting academic outcomes (Unpublished Emotional Quotient Inventory Short Form: Catch me if you can. Human
This document is copyrighted by the American Psychological Association or one of its allied publishers.

doctoral dissertation). Temple University, Philadelphia, PA. Performance, 20, 43–59. [Link]
ⴱ
Fischer, A. H., Rodriguez Mosquera, P. M., van Vianen, A. E., & Man- Hall, C. P., & West, J. H. (2011). Potential predictors of student teaching
stead, A. S. (2004). Gender and culture differences in emotion. Emotion, performance: Considering emotional intelligence. Issues in Educational
4, 87–94. [Link] Research, 21, 145–161. Retrieved from [Link]
Fisher, Z., Tipton, E., & Hou, Z. (2016). Robumeta: Robust variance Hansen, M., Mann Levesque, E., Quintero, D., & Valant, J. (2018). Have
meta-regression. R package (Version 2.0) [Computer software]. Re- we made progress on achievement gaps? Looking at evidence from the
trieved from [Link] new NAEP results. Retrieved from [Link]
[Link] brown-center-chalkboard/2018/04/17/have-we-made-progress-on-
Fleming, M. L., & Malone, M. R. (1983). The relationship of student achievement-gaps-looking-at-evidence-from-the-new-naep-results
Harms, P., & Credé, M. (2010). Emotional intelligence and transforma-
characteristics and student performance in science as viewed by meta-
tional and transactional leadership: A meta-analysis. Journal of Leader-
analysis research. Journal of Research in Science Teaching, 20, 481–
ship & Organizational Studies, 17, 5–17. [Link]
495. [Link]
1548051809350894
Fortin, N. M., Oreopoulos, P., & Phipps, S. (2015). Leaving boys behind: ⴱ
Hassan, A., Sulaiman, T., & Ishak, R. (2009). Philosophy underlying
Gender disparities in high academic achievement. The Journal of Human
emotional intelligence in relation to level of curiosity and academic
Resources, 50, 549 –579. [Link]
achievement of rural area students. Journal of Social Sciences, 5, 95–
Fredrickson, B. L. (2001). The role of positive emotions in positive
103. [Link]
psychology. The broaden-and-build theory of positive emotions. Amer-
Heckman, J. J., & Kautz, T. (2012). Hard evidence on soft skills. Labour
ican Psychologist, 56, 218 –226. [Link]
Economics, 19, 451– 464. [Link]
.56.3.218
.014
Freudenthaler, H., & Neubauer, A. C. (2007). Measuring emotional man-
Heckman, J. J., & Rubinstein, Y. (2001). The importance of noncognitive
agement abilities: Further evidence of the importance to distinguish
skills: Lessons from the GED testing program. The American Economic
between typical and maximum performance. Personality and Individual
Review, 91, 145–149. [Link]
Differences, 42, 1561–1572. [Link]
Heckman, J. J., Stixrud, J., & Urzua, S. (2006). The effects of cognitive and
.031 noncognitive abilities on labor market outcomes and social behavior.
Furnham, A., & Petrides, K. V. (2003). Trait emotional intelligence and Journal of Labor Economics, 24, 411– 482. [Link]
happiness. Social Behavior and Personality, 31, 815– 823. [Link] 504455
.org/10.2224/sbp.2003.31.8.815 Hedges, L. V., Tipton, E., & Johnson, M. C. (2010). Robust variance
ⴱ
Garg, R., Levin, E., & Tremblay, L. (2016). Emotional intelligence: estimation in meta-regression with dependent effect size estimates. Re-
Impact on post-secondary academic achievement. Social Psychology of search Synthesis Methods, 1, 39 – 65. [Link]
Education, 19, 627– 642. [Link] Hoaglin, D. C., & Iglewicz, B. (1987). Fine-tuning some resistant rules for
Gibbs, N. (1995). The EQ factor. TIME, 146, 60 – 67. outlier labeling. Journal of the American Statistical Association, 82,
ⴱ
Gibson, G. J. (2005). Emotional intelligence and performance in a grad- 1147–1149. [Link]
uate school counseling program (Unpublished doctoral dissertation). Hodzic, S., Scharfen, J., Ripoll, P., Holling, H., & Zenasni, F. (2018). How
University of North Dakota, Grand Forks, ND. efficient are emotional intelligence trainings: A meta-analysis. Emotion
ⴱ
Gilbert, T. (2003). Emotional perception as a component of emotional Review, 10, 138 –148. [Link]
intelligence: Its role in academic achievement, social interaction, and ⴱ
Hogan, M. J. (2003). Prediction of grade point average and retention in
focus of attention (Unpublished doctoral dissertation). Middlesex Uni- first-year students by social and emotional competencies (Unpublished
versity, London, UK. master’s thesis). Trent University, Peterborough, Ontario, Canada.
ⴱ ⴱ
Gil-Olarte Márquez, P., Palomera Martín, R., & Brackett, M. A. (2006). Hogan, M. J., Parker, J. D. A., Wiener, J., Watters, C., Wood, L. M., &
Relating emotional intelligence to social competence and academic Oke, A. (2010). Academic success in adolescence: Relationships among
achievement in high school students. Psicothema, 18, 118 –123. http:// verbal IQ, social support and emotional intelligence. Australian Journal
[Link]/ of Psychology, 62, 30 – 41. [Link]
ⴱ
Glickman-Rogers, A. (2010). How do social and emotional competencies 903312881
ⴱ
influences the grades that male students receive? (Unpublished doctoral Holt, S. (2007). Emotional intelligence and academic achievement in
dissertation). Hofstra University, Hempstead, NY. higher education (Unpublished doctoral dissertation). Pepperdine Uni-
Goldberg, L. R. (1981). Language and individual differences: The search versity, Malibu, CA.
ⴱ
for universals in personality lexicons. Review of Personality and Social Humphrey-Murto, S., Leddy, J. J., Wood, T. J., Puddester, D., & Moineau,
Psychology, 2, 141–165. G. (2014). Does emotional intelligence at medical school admission
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 29

predict future academic performance? Academic Medicine, 89, 638 – Kaufman, A. S., & Lichtenberger, E. O. (2005). Assessing adolescent and
643. [Link] adult intelligence (3rd ed.). Boston, MA: Allyn & bacon.
ⴱ
Hunter, J. E., Schmidt, F. L., & Le, H. (2006). Implications of direct and Khajehpour, M. (2011). Relationship between emotional intelligence,
indirect range restriction for meta-analysis methods and findings. Jour- parental involvement and academic performance of high school students.
nal of Applied Psychology, 91, 594 – 612. [Link] Procedia: Social and Behavioral Sciences, 15, 1081–1086. [Link]
0021-9010.91.3.594 .org/10.1016/[Link].2011.03.242
ⴱ
Inzlicht, M., & Ben-Zeev, T. (2000). A threatening intellectual environ- Killen, M. A. (2016). The relationship between parents’ perceptions of
ment: Why females are susceptible to experiencing problem-solving their child’s emotional intelligence and academic performance among
deficits in the presence of males. Psychological Science, 11, 365–371. students with autism spectrum disorder (Unpublished doctoral disserta-
[Link] tion). Capella University, Minneapolis, MN, United States.
ⴱ Kong, F., Zhao, J., & You, X. (2012). Emotional intelligence and life
Izaguirre, R. (2008). The relationship among emotional intelligence, ac-
ademic achievement, and demographic characteristics in first-year com- satisfaction in Chinese university students: The mediating role of self-
munity college students (Unpublished doctoral dissertation). University esteem and social support. Personality and Individual Differences, 53,
of the Incarnate Word, San Antonio, TX. 1039 –1043. [Link]
ⴱ
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ⴱ Kracher, C. M. (2009). Academic and non-academic factors associated

Jaeger, A. J. (2003). Job competencies and the curriculum: An inquiry into
This document is copyrighted by the American Psychological Association or one of its allied publishers.

emotional intelligence in graduate professional education. Research in with retention of undergraduate college students (Unpublished doctoral
Higher Education, 44, 615– 639. [Link] dissertation). Lynn University, Boca Raton, FL, United States.
ⴱ
119724265 Kumar, A., Puranik, M. P., & Sowmya, K. R. (2016). Association between
Jencks, C. (1979). Who gets ahead? The determinants of economic success dental students’ emotional intelligence and academic performance: A
in America. New York, NY: Basic Books. study at six dental colleges in India. Journal of Dental Education, 80,
Jensen, A. R. (1998). The g factor: The science of mental ability (Vol. 648). 526 –532. [Link]
ⴱ
Westport, CT: Praeger. Kumar, V. V., Mehta, M., & Maheshwari, N. (2013). Effect of emotional
John, O. P. (1989). Towards a taxonomy of personality descriptors. In intelligence on the achievement motivation, psychological adjustment
D. M. Buss & N. Cantor (Eds.), Personality psychology (pp. 261–271). and scholastic performance of secondary school students. Journal of the
New York: Springer. [Link] Indian Academy of Applied Psychology, 39, 60 – 67. Retrieved from
ⴱ [Link]
Johnson, G. D. K. (2008). Learning styles and emotional intelligence of
Kuncel, N. R., Crede, M., & Thomas, L. L. (2005). The validity of
the adult learner (Unpublished doctoral dissertation). Auburn Univer-
self-reported grade point averages, class ranks, and test scores: A meta-
sity, Auburn, AL.
analysis and review of the literature. Review of Educational Research,
Johnson, J. W. (2000). A heuristic method for estimating the relative
75, 63– 82. [Link]
weight of predictor variables in multiple regression. Multivariate Be- ⴱ
Kvapil, L. (2007). The impact of emotional intelligence on the academic
havioral Research, 35, 1–19. [Link]
performance of at-risk high school students (Unpublished doctoral dis-
906MBR3501_1
sertation). University of the Incarnate Word, San Antonio, TX.
Johnson, J. W., & LeBreton, J. M. (2004). History and use of relative ⴱ
Laborde, S., Dosseville, F., & Scelles, N. (2010). Trait emotional intelli-
importance indices in organizational research. Organizational Research
gence and preference for intuition and deliberation: Respective influence
Methods, 7, 238 –257. [Link]
ⴱ
on academic performance. Personality and Individual Differences, 49,
Jones, D. P. (2013). The relationship between emotional intelligence and
784 –788. [Link]
literacy achievement of secondary students (Unpublished doctoral dis- ⴱ
Lanciano, T., & Curci, A. (2014). Incremental validity of emotional
sertation). Walden University, Minneapolis, MN.
ⴱ
intelligence ability in predicting academic achievement. The American
Jordan, J. A., McRorie, M., & Ewing, C. (2010). Gender differences in the Journal of Psychology, 127, 447– 461. [Link]
role of emotional intelligence during the primary-secondary school amerjpsyc.127.4.0447
transition. Emotional & Behavioural Difficulties, 15, 37– 47. [Link] Landy, F. J. (2005). Some historical and scientific issues related to research
.[Link]/10.1080/13632750903512415 on emotional intelligence. Journal of Organizational Behavior, 26,
Joseph, D. L., Jin, J., Newman, D. A., & O’Boyle, E. H. (2015). Why does 411– 424. [Link]
self-reported emotional intelligence predict job performance? A meta- ⴱ
Lasser, M. H. (1997). A study of possible correlates of emotional intelli-
analytic investigation of mixed EI. Journal of Applied Psychology, 100, gence: Gender and grade point average (Unpublished master’s thesis).
298 –342. [Link] Kean University, Union, NJ.
Joseph, D. L., & Newman, D. A. (2010). Emotional intelligence: An Law, K. S., Wong, C. S., & Song, L. J. (2004). The construct and criterion
integrative meta-analysis and cascading model. Journal of Applied Psy- validity of emotional intelligence and its potential utility for manage-
chology, 95, 54 –78. [Link] ment studies. Journal of Applied Psychology, 89, 483– 496. [Link]
Judge, T. A., Jackson, C. L., Shaw, J. C., Scott, B. A., & Rich, B. L. (2007). .org/10.1037/0021-9010.89.3.483
Self-efficacy and work-related performance: The integral role of indi- ⴱ
Lawrence, S. A., & Deepa, T. (2013). Emotional intelligence and aca-
vidual differences. Journal of Applied Psychology, 92, 107–127. http:// demic achievement of high school students in Kanyakumari district.
[Link]/10.1037/0021-9010.92.1.107 International Journal of Physical and Social Sciences, 3, 101–107.
ⴱ
Kaliská, L. (2015). Three types of intelligences and their relationship to Retrieved from [Link]
students’ school performance. The New Educational Review, 41, 275– LeBreton, J. M., & Tonidandel, S. (2008). Multivariate relative impor-
286. [Link] tance: Extending relative weight analysis to multivariate criterion spac-
Kapp, C. A. (2000). Emotional intelligence (EQ) and success in post- es. Journal of Applied Psychology, 93, 329 –345. [Link]
graduate studies: A pilot study. South African Journal of Higher Edu- .1037/0021-9010.93.2.329
ⴱ
cation, 14, 151–160. Leddy, J. J., Moineau, G., Puddester, D., Wood, T. J., & Humphrey-
Kashani, F. L., Azimi, A. L., & Vaziri, S. (2012). Relationship between Murto, S. (2011). Does an emotional intelligence test correlate with
emotional intelligence and educational achievement. Procedia: Social traditional measures used to determine medical school admission? Ac-
and Behavioral Sciences, 69, 1270 –1275. [Link] ademic Medicine, 86, S39 –S41. [Link]
.sbspro.2012.12.061 .0b013e31822a6df6
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

30 MACCANN ET AL.

Lenhard, W., & Lenhard, A. (2016). Calculation of effect sizes. Retrieved American and Chinese business students. College Student Journal, 46,
from [Link] 776 –787. Retrieved from [Link]
ⴱ
Lewis, E. S. (2004). A study of emotional intelligence, cognitive intelli- college_student_journal
ⴱ
gence and clinical performance of physical therapy students (Unpub- Martin, T. R. (2011). The relationship between emotional intelligence and
lished doctoral dissertation). University of Massachusetts Lowell, Low- academic success of 6th grade students (Unpublished doctoral disserta-
ell, MA. tion). Wilmington University, Newcastle, DE.
ⴱ
Libbrecht, N., Lievens, F., Carette, B., & Côté, S. (2014). Emotional Martins, A., Ramalho, N., & Morin, E. (2010). A comprehensive meta-
intelligence predicts success in medical school. Emotion, 14, 64 –73. analysis of the relationship between emotional intelligence and health.
[Link] Personality and Individual Differences, 49, 554 –564. [Link]
Lievens, F., & Sackett, P. R. (2006). Video-based versus written situational 10.1016/[Link].2010.05.029
ⴱ
judgment tests: A comparison in terms of predictive validity. Journal of Matešić, K. (2015). The relationship between cognitive and emotional
Applied Psychology, 91, 1181–1188. [Link] intelligence and high school academic achievement. Collegium Antropo-
9010.91.5.1181 logicum, 39, 371–375. [Link]
ⴱ
Lochner, J. L. (2016). The relationship of emotional intelligence, self- Matsumoto, D., LeRoux, J., Wilson-Cohn, C., Raroque, J., Kooken, K.,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

efficacy and gender in predicting grade point average for visual arts Ekman, P., . . . Goh, A. (2000). A new test to measure emotion
This document is copyrighted by the American Psychological Association or one of its allied publishers.

students (Unpublished doctoral dissertation). Capella University, Min- recognition ability: Matsumoto and Ekman’s Japanese and Caucasian
neapolis, MN. Brief Affect Recognition Test (JACBART). Journal of Nonverbal Be-
ⴱ
Loera, D. L. (2013). Emotional intelligence and student success among havior, 24, 179 –209. [Link]
White and first- and second-generation Latina/o college students (Un- Matthews, G., Zeidner, M., & Roberts, R. D. (2002). Emotional intelli-
published doctoral dissertation). University of La Verne, La Verne, CA. gence: Science and myth. Cambridge, MA: MIT Press. [Link]
Lopes, P. N., Brackett, M. A., Nezlek, J. B., Schütz, A., Sellin, I., & 10.7551/mitpress/2704.001.0001
Salovey, P. (2004). Emotional intelligence and social interaction. Per- Mattingly, V., & Kraiger, K. (2019). Can emotional intelligence be
sonality and Social Psychology Bulletin, 30, 1018 –1034. [Link] trained? A meta-analytical investigation. Human Resource Management
.org/10.1177/0146167204264762 Review, 29, 140 –155. [Link]
Lopes, P. N., Salovey, P., Coté, S., Beers, M., & Petty, R. E. (2005). Mavroveli, S., Petrides, K. V., Rieffe, C., & Bakker, F. (2007). Trait
Emotion regulation abilities and the quality of social interaction. Emo- emotional intelligence, psychological well-being and peer-rated social
tion, 5, 113–118. [Link] competence in adolescence. British Journal of Developmental Psychol-
ⴱ
Lu, S.-C. (2010). Validation of a measure of emotional intelligence for ogy, 25, 263–275. [Link]
ⴱ
first year college students in a Hispanic serving institution (Unpublished Mavroveli, S., Petrides, K. V., Sangareau, Y., & Furnham, A. (2009).
doctoral dissertation). Texas A&M University – Corpus Christi, Corpus Exploring the relationships between trait emotional intelligence and
Christi, TX. objective socio-emotional outcomes in childhood. British Journal of
ⴱ
Lui, M. M. (2009). Can I succeed as an adolescent mother? Examining Educational Psychology, 79, 259 –272. [Link]
the role of emotional intelligence in predicting self-efficacy, academic 000709908X368848
ⴱ
achievement, and school attendance (Unpublished doctoral dissertation). Mavroveli, S., Petrides, K. V., Shove, C., & Whitehead, A. (2008).
Temple University, Philadelphia, PA. Investigation of the construct of trait emotional intelligence in children.
Lyons, L. C., & Morris, W. A. (2017). The meta-analysis calculator. European Child & Adolescent Psychiatry, 17, 516 –526. [Link]
Retrieved from [Link] .org/10.1007/s00787-008-0696-6
ⴱ
MacCann, C. (2010). Further examination of emotional intelligence as a Mavroveli, S., & Sánchez-Ruiz, M. J. (2011). Trait emotional intelligence
standard intelligence: A latent variable analysis of fluid intelligence, influences on academic achievement and school behaviour. British Jour-
crystallized intelligence, and emotional intelligence. Personality and nal of Educational Psychology, 81, 112–134. [Link]
Individual Differences, 49, 490 – 496. [Link] 2044-8279.002009
.2010.05.010 Mayer, J. D., Caruso, D. R., & Salovey, P. (1999). Emotional intelligence
MacCann, C., Fogarty, G. J., & Roberts, R. D. (2012). Strategies for meets traditional standards for an intelligence. Intelligence, 27, 267–298.
success in education: Time management is more important for part-time [Link]
than full-time community college students. Learning and Individual Mayer, J. D., Caruso, D. R., & Salovey, P. (2000). Selecting a measure of
Differences, 22, 618 – 623. [Link] emotional intelligence: The case for ability scales. In R. Bar-On &
.015 J. D. A. Parker (Eds.), Handbook of emotional intelligence (pp. 320 –
ⴱ
MacCann, C., Fogarty, G. J., Zeidner, M., & Roberts, R. D. (2011). 342). San Fransisco, CA: Jossey-Bass.
Coping mediates the relationship between emotional intelligence (EI) Mayer, J. D., Caruso, D. R., & Salovey, P. (2016). The ability model of
and academic achievement. Contemporary Educational Psychology, 36, emotional intelligence: Principles and updates. Emotion Review, 8, 290 –
60 –70. [Link] 300. [Link]
MacCann, C., Joseph, D. L., Newman, D. A., & Roberts, R. D. (2014). Mayer, J. D., Roberts, R. D., & Barsade, S. G. (2008). Human abilities:
Emotional intelligence is a second-stratum factor of intelligence: Evi- Emotional intelligence. Annual Review of Psychology, 59, 507–536.
dence from hierarchical and bifactor models. Emotion, 14, 358 –374. [Link]
[Link] Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P.
ⴱ
MacCann, C., & Roberts, R. D. (2008). New paradigms for assessing Salovey & D. Sluyter (Eds.), Emotional development and emotional
emotional intelligence: Theory and data. Emotion, 8, 540 –551. http:// intelligence: Educational implications (pp. 3–31). New York, NY: Basic
[Link]/10.1037/a0012746 Books.
ⴱ
Malik, F., & Shujja, S. (2013). Emotional intelligence and academic Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. (2001). Emo-
achievement: Implications for children’s performance in schools. Jour- tional intelligence as a standard intelligence. Emotion, 1, 232–242.
nal of the Indian Academy of Applied Psychology, 39, 51–59. http:// [Link]
[Link]/ Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. (2003). Mea-
ⴱ
Margavio, T. M., Margavio, G. W., Hignite, M. A., & Moses, D. R. suring emotional intelligence with the MSCEIT V2.0. Emotion, 3, 97–
(2012). A comparative analysis of the emotional intelligence levels of 105. [Link]
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 31

ⴱ
McClain, E. K. (2009). The relationship of emotional intelligence to verbal Accuracy Scale. Journal of Nonverbal Behavior, 18, 9 –35. http://
academic performance and perceived stress in first year medical stu- [Link]/10.1007/BF02169077
ⴱ
dents (Unpublished doctoral dissertation). University of Kansas, Law- Nwabuebo, E. (2013). Emotional intelligence and academic achievement
rence, KS. among senior Black students enrolled in Baccalaureate nursing pro-
ⴱ
Menzie, T. A. (2005). Emotional intelligence and social and academic grams: Implications for nursing education (Unpublished doctoral dis-
competence in middle school youth (Unpublished doctoral dissertation). sertation). Capella University, Minneapolis, MN.
Rutgers The State University of New Jersey, New Brunswick, NJ. O’Boyle, E. H., Jr., Humphrey, R. H., Pollack, J. M., Hawver, T. H., &
ⴱ
Mestre, J. M., Guil, R., Lopes, P. N., Salovey, P., & Gil-Olarte, P. (2006). Story, P. A. (2011). The relation between emotional intelligence and job
Emotional intelligence and social and academic adaptation to school. performance: A meta-analysis. Journal of Organizational Behavior, 32,
Psicothema, 18, 112–117. 788 – 818. [Link]
Mestre, J. M., MacCann, C., Guil, R., & Roberts, R. D. (2016). Models of O’Connor, M. C., & Paunonen, S. V. (2007). Big five personality predic-
cognitive ability and emotion can better inform contemporary emotional tors of post-secondary academic performance. Personality and Individ-
intelligence frameworks. Emotion Review, 8, 322–330. [Link] ual Differences, 43, 971–990. [Link]
10.1177/1754073916650497 .017
ⴱ
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Miao, C., Humphrey, R. H., & Qian, S. (2017). Are the emotionally O’Connor, R. M., Jr., & Little, I. S. (2003). Revisiting the predictive
This document is copyrighted by the American Psychological Association or one of its allied publishers.

intelligent good citizens or counterproductive? A meta-analysis of emo- validity of emotional intelligence: Self-report versus ability-based mea-
tional intelligence and its relationships with organizational citizenship sures. Personality and Individual Differences, 35, 1893–1902. [Link]
behavior and counterproductive work behavior. Personality and Indi- .[Link]/10.1016/S0191-8869(03)00038-2
ⴱ
vidual Differences, 116, 144 –156. [Link] Ogundokun, M. O., & Adeyemo, D. A. (2010). Emotional intelligence and
.04.015 academic achievement: The moderating influence of age, intrinsic and
ⴱ
Mitrofan, N., & Cioricaru, M.-F. (2014). Emotional intelligence and extrinsic motivation. The African Symposium: An online journal of the
school performance-correlational Study. Procedia: Social and Behav- African Educational Research Network, 10, 127–141. Retrieved from
ioral Sciences, 127, 769 –775. [Link] [Link]
ⴱ
.03.352 Olatoye, A. R., Akintunde, O. S., & Yakasai, I. M. (2010). Emotional
Moeyaert, M., Ugille, M., Natasha Beretvas, S., Ferron, J., Bunuan, R., & intelligence, creativity and academic achievement of business adminis-
Van den Noortgate, W. (2017). Methods for dealing with multiple tration students. Electronic Journal of Research in Educational Psychol-
outcomes in meta-analysis: A comparison between averaging effect ogy, 8, 763–786.
sizes, robust variance estimation and multilevel meta-analysis. Interna- Olderbak, S., Semmler, M., & Doebler, P. (2019). Four-branch model of
tional Journal of Social Research Methodology, 20, 559 –572. [Link] emotional intelligence with fluid and crystallized intelligence: A meta-
.[Link]/10.1080/13645579.2016.1252189 analysis of relations. Emotion Review, 11, 166 –183. [Link]
ⴱ
Naeem, N., van der Vleuten, C., Muijtjens, A. M. M., Violato, C., Ali, 10.1177/1754073918776776
ⴱ
S. M., Al-Faris, E. A., . . . Naeem, N. (2014). Correlates of emotional Olson, L. G. (2008). An investigation of factors that influence academic
intelligence: Results from a multi-institutional study among undergrad- achievement in Christian higher education: Emotional intelligence, self-
uate medical students. Medical Teacher, 36, S30 –S35. [Link] esteem and spiritual well-being (Unpublished doctoral dissertation).
10.3109/0142159X.2014.886008 Capella University, Minneapolis, MN.
ⴱ
Nata, G., Pereira, M. J., & Neves, T. (2014). Unfairness in access to higher Opateye, J. A. (2014). The relationship between emotional intelligence,
education: An 11 year comparison of grade inflation by private and test anxiety, stress, academic success and attitudes of high school stu-
public secondary schools in Portugal. Higher Education, 68, 851– 874. dents towards electrochemistry. Ife Psychologia, 22, 239 –249.
[Link] Open Science Collaboration. (2015). PSYCHOLOGY. Estimating the re-
National Council of the Teachers of English. (1996). Standard for the producibility of psychological science. Science, 349, aac4716. [Link]
English language arts. Retrieved from [Link] .[Link]/10.1126/science.aac4716
NCTEFiles/Resources/Books/Sample/[Link]?_ga⫽2 Palmer, B. R., Gignac, G., Manocha, R., & Stough, C. (2005). A psycho-
.146967543.198597852.1548738430-683684206.1548738430 metric evaluation of the Mayer-Salovey-Caruso emotional intelligence
Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., Ceci, test version 2.0. Intelligence, 33, 285–305. [Link]
S. J., . . . Urbina, S. (1996). Intelligence: Knowns and unknowns. .intell.2004.11.003
ⴱ
American Psychologist, 51, 77–101. [Link] Parker, J. D., Creque, R. E., Sr., Barnhart, D. L., Harris, J. I., Majeski,
066X.51.2.77 S. A., Wood, L. M., . . . Hogan, M. J. (2004). Academic achievement in
ⴱ
Nelson, P. D. (2010). Emotional intelligence and academic achievement high school: Does emotional intelligence matter? Personality and Indi-
in 11th grade at-risk students (Unpublished doctoral dissertation). Wal- vidual Differences, 37, 1321–1330. [Link]
den University, Minneapolis, MN. .2004.01.002
ⴱ ⴱ
Nesari, A. J., Karimi, L., & Filinezhad, N. (2011). On the relationship Parker, J. D., Duffy, J. M., Wood, L. M., Bond, B. J., & Hogan, M. J.
between emotional intelligence and vocabulary learning of Iranian EFL (2005). Academic achievement and emotional intelligence: Predicting
learners at the intermediate level. Procedia: Social and Behavioral the successful transition from high school to university. Journal of the
Sciences, 28, 900 –903. [Link] First-Year Experience & Students in Transition, 17, 67–78.
ⴱ ⴱ
Newsome, S., Day, A. L., & Catano, V. M. (2000). Assessing the Parker, J. D., Summerfeldt, L. J., Hogan, M. J., & Majeski, S. A. (2004).
predictive validity of emotional intelligence. Personality and Individual Emotional intelligence and academic success: Examining the transition
Differences, 29, 1005–1016. [Link] from high school to university. Personality and Individual Differences,
8869(99)00250-0 36, 163–172. [Link]
Nisbett, R. E., & Wilson, T. D. (1977). The halo effect: Evidence for Pekrun, R., Elliot, A. J., & Maier, M. A. (2009). Achievement goals and
unconscious alteration of judgments. Journal of Personality and Social achievement emotions: Testing a model of their joint relations with
Psychology, 35, 250 –256. [Link] academic performance. Journal of Educational Psychology, 101, 115–
.250 135. [Link]
Nowicki, S., Jr., & Duke, M. P. (1994). Individual differences in the Pekrun, R., Goetz, T., Daniels, L. M., Stupnisky, R. H., & Perry, R. P.
nonverbal communication of affect - the Diagnostic-Analysis of Non- (2010). Boredom in achievement settings: Exploring control-value an-
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

32 MACCANN ET AL.

ⴱ
tecedents and performance outcomes of a neglected emotion. Journal of Rankin, B. (2013). Emotional intelligence: Enhancing values-based prac-
Educational Psychology, 102, 531–549. [Link] tice and compassionate care in nursing. Journal of Advanced Nursing,
a0019243 69, 2717–2725. [Link]
ⴱ
Pekrun, R., Goetz, T., Titz, W., & Perry, R. P. (2002). Academic emotions Rastegar, M., & Karami, M. (2013). On the relationships among emo-
in students’ self-regulated learning and achievement: A program of tional intelligence, affective and social strategy use, and academic
qualitative and quantitative research. Educational Psychologist, 37, 91– achievement of Iranian EFL learners. Theory and Practice in Language
105. [Link] Studies, 3, 389 –396. [Link]
Peña-Sarrionandia, A., Mikolajczak, M., & Gross, J. J. (2015). Integrating R Core Team. (2017). R: A language and environment for statistical
emotion regulation and emotional intelligence traditions: A meta- computing. Vienna, Austria: R Foundation for Statistical Computing.
analysis. Frontiers in Psychology, 6, 160. [Link]
Perera, H. N., & DiGiacomo, M. (2013). The relationship of trait emotional Ree, M. J., & Earles, J. A. (1992). Intelligence is the best predictor of job
intelligence with academic performance: A meta-analytic review. Learn- performance. Current Directions in Psychological Science, 1, 86 – 89.
ing and Individual Differences, 28, 20 –33. [Link] [Link]
ⴱ
.lindif.2013.08.002 Rice, D. M. (2007). An examination of emotional intelligence: Its rela-
ⴱ
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Petersen, V. C. (2010). The relationship between emotional intelligence tionship to academic achievement in Army JROTC and the implications
This document is copyrighted by the American Psychological Association or one of its allied publishers.

and middle school students with learning disabilities (Unpublished for education (Unpublished doctoral dissertation). Capella University,
doctoral dissertation). Fairleigh Dickinson University, Madison, NJ. Minneapolis, MN.
Petrides, K. V. (2009). Psychometric properties of the Trait Emotional Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates
Intelligence Questionnaire (TEIQue). In C. Stough, D. H. Saklofske, & of university students’ academic performance: A systematic review and
J. D. A. Parker (Eds.), Assessing emotional intelligence: Theory, re- meta-analysis. Psychological Bulletin, 138, 353–387. [Link]
search, and applications (pp. 85–101). New York, NY: Springer Sci- 10.1037/a0026838
ⴱ
ence ⫹ Business Media. Richardson, T. L. (2000). Emotional intelligence as a mediator of tran-
Petrides, K. V., & Furnham, A. (2000). On the dimensional structure of sition trauma in students progressing from elementary to middle school
emotional intelligence. Personality and Individual Differences, 29, 313– (Unpublished doctoral dissertation). The Florida State University, Tal-
320. [Link] lahassee, FL.
ⴱ
Petrides, K. V., & Furnham, A. (2003). Trait emotional intelligence: Rivers, S. E., Brackett, M. A., Reyes, M. R., Mayer, J. D., Caruso, D. R.,
Behavioural validation in two studies of emotion recognition and reac- & Salovey, P. (2012). Measuring emotional intelligence in early ado-
tivity to mood induction. European Journal of Personality, 17, 39 –57. lescence with the MSCEIT-YV: Psychometric properties and relation-
[Link] ship with academic performance and psychosocial functioning. Journal
Petrides, K. V., Perez-Gonzalez, J. C., & Furnham, A. (2007). On the of Psychoeducational Assessment, 30, 344 –366. [Link]
criterion and incremental validity of trait emotional intelligence. Cog- .1177/0734282912449443
nition and Emotion, 21, 26 –55. [Link] Roberts, B. W., Lejuez, C., Krueger, R. F., Richards, J. M., & Hill, P. L.
30601038912 (2014). What is conscientiousness and how can it be assessed? Devel-
Petrides, K. V., Pita, R., & Kokkinaki, F. (2007). The location of trait opmental Psychology, 50, 1315–1330. [Link]
emotional intelligence in personality factor space. British Journal of a0031109
Psychology, 98, 273–289. [Link] Roberts, R. D., Schulze, R., & MacCann, C. (2008). The measurement of
X120618 emotional intelligence: A decade of progress? In G. Boyle, G. Matthews,
Petrides, K. V., Sangareau, Y., Furnham, A., & Frederickson, N. (2006). & D. Saklofske (Eds.), The Sage handbook of personality theory and
Trait emotional intelligence and children’s peer relations at school. assessment (pp. 461– 482). New York, NY: Sage. [Link]
Social Development, 15, 537–547. [Link] .4135/9781849200479.n22
ⴱ
9507.2006.00355.x Rode, J. C., Mooney, C. H., Arthaud-Day, M. L., Near, J. P., Baldwin,
ⴱ
Pishghadam, R. (2009). A quantitative analysis of the relationship be- T. T., Rubin, R. S., & Bommer, W. H. (2007). Emotional intelligence
tween emotional intelligence and foreign language learning. Electronic and individual performance: Evidence of direct and moderated effects.
Journal of Foreign Language Teaching, 6, 31– 41. Retrieved from Journal of Organizational Behavior, 28, 399 – 421. [Link]
[Link]/ .1002/job.429
ⴱ
Pool, L. D., & Qualter, P. (2012). Improving emotional intelligence and Rodeiro, C. L. V., Emery, J. L., & Bell, J. F. (2012). Emotional intelli-
emotional self-efficacy through a teaching intervention for university gence and academic attainment of British secondary school children: A
students. Learning and Individual Differences, 22, 306 –312. [Link] cross-sectional survey. Educational Studies, 38, 521–539. [Link]
.[Link]/10.1016/[Link].2012.01.010 .org/10.1080/03055698.2011.643115
ⴱ
Por, J., Barriball, L., Fitzpatrick, J., & Roberts, J. (2011). Emotional Rojstaczer, S., & Healy, C. (2012). Where A is ordinary: The evolution of
intelligence: Its relationship to stress, coping, well-being and profes- American college and university grading. Teachers College Record, 114,
sional performance in nursing students. Nurse Education Today, 31, 1940 –2009. Retrieved from [Link]
855– 860. [Link] Rossen, E., Kranzler, J. H., & Algina, J. (2008). Confirmatory factor
Poropat, A. E. (2009). A meta-analysis of the five-factor model of person- analysis of the Mayer-Salovey-Caruso Emotional Intelligence Test V 2.0
ality and academic performance. Psychological Bulletin, 135, 322–338. (MSCEIT). Personality and Individual Differences, 44, 1258 –1269.
[Link] [Link]
ⴱ
Qualter, P., Gardner, K. J., Pope, D. J., Hutchinson, J. M., & Whiteley, Roth, B., Becker, N., Romeyke, S., Schafer, S., Domnick, F., & Spinath,
H. E. (2012). Ability emotional intelligence, trait emotional intelligence, F. M. (2015). Intelligence and school grades: A meta-analysis. Intelli-
and academic success in British secondary schools: A 5-year longitudi- gence, 53, 118 –137. [Link]
ⴱ
nal study. Learning and Individual Differences, 22, 83–91. [Link] Saklofske, D. H., Austin, E. J., Mastoras, S. M., Beaton, L., & Osborne,
.org/10.1016/[Link].2011.11.007 S. E. (2012). Relationships of personality, affect, emotional intelligence
ⴱ
Radford, M. (2011). Emotional intelligence as a predictor of nursing and coping with student stress and academic success: Different patterns
student success (Unpublished doctoral dissertation). Union University, of association for stress and success. Learning and Individual Differ-
Jackson, TN. ences, 22, 251–257. [Link]
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 33

ⴱ
Salgado, J. F. (1998). Big Five personality dimensions and job perfor- Singh, B., Singh, M., & Singh, K. (2009, March). The influence of
mance in army and civil occupations: A European perspective. Human emotional intelligence and learning style on student’s academic
Performance, 11, 271–288. [Link] achievement. Paper presented at the Conference on Scientific and
hup1102&3_8 Social Research. Retrieved from [Link]
Salgado, J. F., Anderson, N., Moscoso, S., Bertua, C., de Fruyt, F., & The-Influence-of-Emotional-Intelligence-and-Learning-Style-on-Student-s-
Rolland, J. P. (2003). A meta-analytic study of general mental ability Academic-Achievement
validity for different occupations in the European community. Journal of Skaalvik, E. M., & Skaalvik, S. (2007). Dimensions of teacher self-efficacy
Applied Psychology, 88, 1068 –1081. [Link] and relations with strain factors, perceived collective teacher efficacy,
9010.88.6.1068 and teacher burnout. Journal of Educational Psychology, 99, 611– 625.
Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, [Link]
ⴱ
Cognition and Personality, 9, 185–211. [Link] Skipper, C. O., & Brandenburg, S. (2013). Emotional intelligence and
DUGG-P24E-52WK-6CDG academic performance of engineering students. Engineering Project
ⴱ
Samples, G. M. (2010). Emotional intelligence and academic success Organization Journal, 3, 13–21. [Link]
among Bible college students (Unpublished doctoral dissertation). Re- .2012.738669
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ⴱ
gent University, Virginia Beach, VA. Song, L. J., Huang, G.-H., Peng, K. Z., Law, K. S., Wong, C.-S., & Chen,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Sánchez-Álvarez, N., Extremera, N., & Fernandez-Berrocal, P. (2016). The Z. (2010). The differential effects of general mental ability and emo-
relation between emotional intelligence and subjective well-being: A tional intelligence on academic performance and social interactions.
meta-analytic investigation. The Journal of Positive Psychology, 11, Intelligence, 38, 137–143. [Link]
276 –285. [Link] Steinkamp, M. W., & Maehr, M. L. (1983). Affect, ability, and science
ⴱ
Sanchez-Ruiz, M.-J., Mavroveli, S., & Poullis, J. (2013). Trait emotional achievement: A quantitative synthesis of correlational research. Review
intelligence and its links to university performance: An examination. of Educational Research, 53, 369 –396. [Link]
Personality and Individual Differences, 54, 658 – 662. [Link] 00346543053003369
ⴱ
10.1016/[Link].2012.11.013 Stewart, M. F., & Chisholm, C. U. (2012). Assessment of the impact of
Schlegel, K., Grandjean, D., & Scherer, K. R. (2014). Introducing the emotional competencies on academic achievement within engineering
Geneva emotion recognition test: An example of Rasch-based test de- and other student cohorts. World Transactions on Engineering and
velopment. Psychological Assessment, 26, 666 – 672. [Link] Technology Education, 10, 110 –115. Retrieved from [Link]
10.1037/a0035246 .[Link]/
ⴱ
Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection Stottlemyer, B. G. (2002). An examination of emotional intelligence: Its
methods in personnel psychology: Practical and theoretical implications relationship to achievement and the implications for education (Unpub-
of 85 years of research findings. Psychological Bulletin, 124, 262–274. lished doctoral dissertation). Texas A&M University - Corpus Christi,
[Link] Corpus Christi, TX.
ⴱ
Schneider, F. W., & Coutts, L. M. (1982). The high school environment: A Suliman, W. A. (2010). The relationship between learning styles, emo-
comparison of coeducational and single-sex schools. Journal of Educa- tional social intelligence, and academic success of undergraduate nurs-
tional Psychology, 74, 898 –906. [Link] ing students. The Journal of Nursing Research, 18, 136 –143. [Link]
.74.6.898 .[Link]/10.1097/JNR.0b013e3181dda797
ⴱ
Schulte, M. J., Ree, M. J., & Carretta, T. R. (2004). Emotional intelligence: Sunbul, A. M. (2007, September). The relationship between emotional
Not much more than g and personality. Personality and Individual intelligence and achievement among 1st and 4th grade faculty students.
Differences, 37, 1059 –1068. [Link] Paper presented at the First European SEBCD Conference, Malta.
ⴱ
.014 Szuberla, A. L. (2005). Emotional intelligence and school success (Un-
ⴱ
Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., Cooper, J. T., published doctoral dissertation). Walden University, Minneapolis, MN.
Golden, C. J., & Liane, D. (1998). Development and validation of a Taylor, R. D., Oberle, E., Durlak, J. A., & Weissberg, R. P. (2017).
measure of emotional intelligence. Personality and Individual Differ- Promoting positive youth development through school-based social and
ences, 25, 167–177. [Link] emotional learning interventions: A meta-analysis of follow-up effects.
Schutte, N. S., Malouff, J. M., & Thorsteinsson, E. B. (2013). Increasing Child Development, 88, 1156 –1171. [Link]
emotional intelligence through training: Current status and future direc- .12864
tions. International Journal of Emotional Education, 5, 56 –72. Tett, R. P., Freund, K. A., Christiansen, N. D., Fox, K. E., & Coaster, J.
Schutte, N. S., Malouff, J. M., Thorsteinsson, E. B., Bhullar, N., & Rooke, (2012). Faking on self-report emotional intelligence and personality
S. E. (2007). A meta-analytic investigation of the relationship between tests: Effects of faking opportunity, cognitive ability, and job type.
emotional intelligence and health. Personality and Individual Differ- Personality and Individual Differences, 52, 195–201. [Link]
ences, 42, 921–933. [Link] 10.1016/[Link].2011.10.017
ⴱ
Shao, K., Yu, W., & Ji, Z. (2013). The relationship between EFL students’ Tok, S., & Morali, S. L. (2009). Trait emotional intelligence, the Big Five
emotional intelligence and writing achievement. Innovation in Language personality dimensions and academic success in physical education
Learning and Teaching, 7, 107–124. [Link] teacher candidates. Social Behavior and Personality, 37, 921–931.
17501229.2012.725730 [Link]
ⴱ
Shipley, N. L., Jackson, M. J., & Segrest, S. L. (2010). The effects of Tonidandel, S., & LeBreton, J. M. (2011). Relative importance analysis: A
emotional intelligence, age, work experience, and academic perfor- useful supplement to regression analysis. Journal of Business and Psy-
mance. Research in Higher Education, 9, 1– 8. Retrieved from https:// chology, 26, 1–9. [Link]
ⴱ
[Link]/journal/11162 Trapp, C. S. (2011). The association among emotional intelligence, resil-
ⴱ
Sierra, M. D. V., Borges del Rosal, M. A., Romero, N. R., Villegas, K., ience, and academic performance of preservice teachers (Unpublished
& Lorenzo, M. (2013). Emotional intelligence and its relationship with doctoral dissertation). University of Phoenix, Phoenix, AZ.
gender, academic performance and intellectual abilities of undergradu- Tupes, E. C., & Christal, R. E. (1992). Recurrent personality factors based
ates. Electronic Journal of Research in Educational Psychology, 11, on trait ratings. Journal of Personality, 60, 225–251. [Link]
395– 412. 10.1111/j.1467-6494.1992.tb00973.x
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

34 MACCANN ET AL.

van der Linden, D., Pekaar, K. A., Bakker, A. B., Schermer, J. A., Vernon, approach to etiology and treatment (pp. 13–37). New York, NY: Guil-
P. A., Dunkel, C. S., & Petrides, K. V. (2017). Overlap between the ford Press.
general factor of personality and emotional intelligence: A meta- Westrick, P. A. (2017). Reliability estimates for undergraduate grade point
analysis. Psychological Bulletin, 143, 36 –52. [Link] average. Educational Assessment, 22, 231–252. [Link]
bul0000078 .1080/10627197.2017.1381554
van der Linden, D., te Nijenhuis, J., & Bakker, A. B. (2010). The general ⴱ
Willis, M. L. (2015). Emotional intelligence as a predictor of college
factor of personality: A meta-analysis of Big Five intercorrelations and student retention and success (Unpublished doctoral dissertation). Uni-
a criterion-related validity study. Journal of Research in Personality, 44, versity of Nebraska, Lincoln, NE.
315–327. [Link] ⴱ
Woitaszewski, S. A., & Aalsma, M. C. (2004). The contribution of
Van Rooy, D. L., & Viswesvaran, C. (2004). Emotional intelligence: A
emotional intelligence to the social and academic success of gifted
meta-analytic investigation of predictive validity and nomological net.
adolescents as measured by the Multifactor Emotional Intelligence Scale -
Journal of Vocational Behavior, 65, 71–95. [Link]
Adolescent Version. Roeper Review: A Journal on Gifted Education, 27,
S0001-8791(03)00076-9
ⴱ 25–30. [Link]
Vargas, G. A. (2014). The role of emotional intelligence skills in the ⴱ
Wraight, M. E. (2008). The impact of trait emotional intelligence and
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

academic achievement of students pursuing associate degrees at a South

Texas college (Unpublished doctoral dissertation). Texas A&M Univer- cognitive style on the academic achievement and life satisfaction of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

sity – Kingsville, Kingsville, TX. college students (Unpublished doctoral dissertation). The State Univer-
ⴱ
Veitch, B. L. (2011). Developmental education and emotional intelligence in sity of New York at Buffalo, Buffalo, NY.
ⴱ
three rural East Texas community colleges (Unpublished doctoral disserta- Wurf, G., & Croft-Piggin, L. (2015). Predicting the academic achievement
tion). Texas A&M University – Corpus Christi, Corpus Christi, TX. of first-year, pre-service teachers: The role of engagement, motivation,
ⴱ ATAR, and emotional intelligence. Asia-Pacific Journal of Teacher
Vela, R. H., Jr. (2004). The role of emotional intelligence in the academic
achievement of first year college students (Unpublished doctoral disser- Education, 43, 75–91. [Link]
tation). Texas A&M University – Kingsville, Kingsville, TX. .932328
ⴱ ⴱ
Victoroff, K. Z., & Boyatzis, R. E. (2013). What is the relationship Xu, J. (2018). Emotion regulation in mathematics homework: An empir-
between emotional intelligence and dental student clinical performance? ical study. The Journal of Educational Research, 111, 1–11. [Link]
Journal of Dental Education, 77, 416 – 426. .[Link]/10.1080/00220671.2016.1175409
Viswesvaran, C., & Ones, D. S. (1999). Meta-analyses of fakability esti- ⴱ
Yahaya, A., Ee, N. S., Bachock, J., Yahaya, N., Boon, Y., Hashim, S., &
mates: Implications for personality measurement. Educational and Psy- Lee, G. M. (2012). The impact of emotional intelligence element on
chological Measurement, 59, 197–210. [Link] academic achievement. Archives des Sciences, 65, 1–17. Retrieved from
00131649921969802 [Link]
ⴱ
Walker, M. E. B. (2006). Emotional intelligence and academic success in ⴱ
Zandi, M. (2012). The role of emotional intelligence in French language
college (Unpublished doctoral dissertation). The University of Southern learning and academic success of the female students. Procedia: Social
Mississippi, Hattiesburg, MS.
ⴱ and Behavioral Sciences, 46, 4986 – 4989. [Link]
Walsh-Portillo, J. (2011). The role of emotional intelligence in college
.sbspro.2012.06.372
students’ success (Unpublished doctoral dissertation). Florida Interna- ⴱ
Zarezadeh, T. (2013). The effect of emotional intelligence in English
tional University, Greater Miami, FL. [Link]
language learning. Procedia: Social and Behavioral Sciences, 84, 1286 –
.FI11120510
Walter, F., Cole, M. S., & Humphrey, R. H. (2011). Emotional intelligence: 1289. [Link]
ⴱ
Sine qua non of leadership or folderol? The Academy of Management Zirak, M., & Ahmadian, E. (2015). Relationship between emotional
Perspectives, 25, 45–59. [Link] intelligence & academic achievement emphasizing on creative thinking.
Wechsler, D. (1943). Non-intellective factors in general intelligence. The Mediterranean Journal of Social Sciences, 6, 561–570. [Link]
Journal of Abnormal and Social Psychology, 38, 101–103. [Link] 10.5901/mjss.2015.v6n5s2p561
ⴱ
.org/10.1037/h0060613 Zysberg, L., Levy, A., & Zisberg, A. (2011). Emotional intelligence in
Werner, K., & Gross, J. J. (2009). Emotion regulation and psychopa- applicant selection for care-related academic programs. Journal of Psy-
thology: A conceptual framework. In A. M. Kring & D. M. Sloan choeducational Assessment, 29, 27–38. [Link]
(Eds.), Emotion regulation and psychopathology: A transdiagnostic 0734282910365059

(Appendix follows)
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 35

Appendix
Brief Descriptions of All Data Sources Meeting Inclusion Criteria

Education
Authors Year Na kb Country level Publication type Streamc rd

Abdo 2012 266 1 USA secondary dissertation 1 .12

Abdullah et al. 2004 205 1 Kuala Lumpur & secondary article 3 .19
Malaysia
Abel 2013 78 1 USA secondary dissertation 1 .15
Adeyemo 2007 300 1 Nigeria tertiary article 2 .33
Afolabi et al. 2009 110 1 Nigeria tertiary article 1 .20
Agnoli et al. 2012 352 2 Italy primary article 1 .16
Ahammed et al. 2011 204 7 UAE tertiary article 3 ⫺.08
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Ahmad 2011 291 6 Pakistan tertiary article 1 .17

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Ahmad & Rana 2012 538 1 Pakistan tertiary article 1 .08

Alhashemi 2014 136 1 Bahrain tertiary article 2 .10
Alumran & Punamaki 2008 312 5 Bahrain mixed article 1 .08
Amdurer et al. 2014 266 1 USA tertiary article 1 ⫺.02
Anand et al. 2016 390 1 India tertiary article 1 .74
Arbabisarjou et al. 2013 250 12 Iran tertiary conference paper 1 .62
Aremu et al. 2006 300 1 Nigeria secondary article 2 .31
Barchard 2003 150 13 USA tertiary article 2, 3 .05
Barisonek 2005 44 15 USA primary dissertation 1 .02
Bastian et al. 2005 185 10 Australia tertiary article 2, 3 .09
Berenson et al. 2008 82 1 USA tertiary article 2 .33
Biggart 2019 89 1 UK tertiary unpublished 3 .19
Billings et al. 2014 407 8 Australia primary article 2 .05
Bowman 2007 48 27 USA tertiary dissertation 3 .15
Boyce 2002 57 5 USA tertiary dissertation 3 .16
Brackett & Mayer 2003 207 9 USA tertiary article 1, 2, 3 .10
Brackett et al. 2004 330 6 USA tertiary article 3 .23
Bradshaw 2008 60 7 USA tertiary dissertation 1, 3 .17
Brouzos et al. 2014 103 20 Egypt primary article 1 .17
Carrothers et al. 2000 147 2 USA secondary article 1 .11
Castro-Johnson & Wang 2003 520 2 USA tertiary article 3 .15
Catalina et al. 2012 92 1 Romania tertiary article 2 ⫺.08
Cavins 2006 73 1 USA tertiary dissertation 1 .33
Chapman & Hayslip 2005 292 1 USA tertiary article 2 .08
Cheshire et al. 2015 85 10 USA tertiary article 3 .11
Chew et al. 2013 163 14 Malaysia tertiary article 3 .17
Clark 2004 161 2 USA secondary dissertation 1 .20
Codier & Odell 2014 72 2 USA tertiary article 3 .25
Collins 2013 65 5 USA tertiary dissertation 3 ⫺.01
Colston 2008 115 1 USA tertiary dissertation 1 .17
Costa & Faria 2015 380 54 Portugal secondary article 2, 3 .21
Cyr 2007 237 10 USA tertiary dissertation 3 .03
Di Fabio & Palazzeschi 2009 124 10 Italy secondary article 1, 3 .22
Di Fabio & Palazzeschi 2015 133 1 Italy secondary article 1 .53
Doring 2006 72 4 USA primary dissertation 2, 3 ⫺.02
Downey et al. 2014 242 4 Australia secondary article 2 .08
Downey et al. 2008 158 45 Australia secondary article 2 .10
Drago 2005 32 5 USA tertiary dissertation 3 .31
Drati 2010 165 5 USA secondary dissertation 1 .11
Edison 2003 61 15 USA tertiary dissertation 3 .08
Evenson 2008 100 6 USA tertiary dissertation 1 .18
ETS 2019 590 2 USA Tertiary Unpublished 3 .15
Fahim & Pishghadam 2007 508 6 Iran tertiary conference paper 1 .18
Fallahzadeh 2011 223 6 Iran tertiary conference paper 1 .09
Fannin 2002 115 8 USA secondary dissertation 3 .37
Fatum 2008 75 3 USA primary dissertation 1 .10
Fayombo 2012 151 7 Barbados tertiary article 1 .25
Feldman 2004 79 10 USA tertiary dissertation 3 ⫺.12
Garg et al. 2016 299 12 Canada tertiary article 1 .08
Gibson 2005 40 6 USA tertiary dissertation 2, 3 ⫺.01

(Appendix continues)
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

36 MACCANN ET AL.

Appendix (continued)

Education
Authors Year Na kb Country level Publication type Streamc rd

Gilbert 2003 170 12 UK tertiary dissertation 2 .04

Glickman-Rogers 2010 205 8 USA secondary dissertation 3 .32
Goodwin 2016 198 4 USA tertiary dissertation 1 .32
Hall & West 2011 74 3 USA tertiary article 3 .05
Hogan 2003 673 10 Canada tertiary dissertation 1 .05
Hogan et al. 2010 44 1 Canada secondary dissertation 1 .35
Holt 2007 124 10 USA tertiary dissertation 3 .09
Humphrey-Murto et al. 2014 113 12 Canada tertiary article 3 .09
Izaguirre 2008 199 6 USA tertiary dissertation 1 .06
Jaeger 2003 75 24 USA tertiary article 1 .20
Jang et al. 2019 278 27 USA tertiary unpublished 1, 2, 3 ⫺.02
Johnson 2008 111 4 USA tertiary dissertation 3 .02
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Jones 2014 38 1 USA secondary dissertation 3 .28

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Jordan et al. 2010 86 20 Ireland secondary article 1 .08

Kaliska 2015 169 10 Slovak Republic secondary article 1 .19
Khajehpour 2011 300 1 Iran secondary conference paper 2 .32
Killen 2016 47 2 USA secondary dissertation 1 .60
Kracher 2009 109 1 USA tertiary dissertation 1 .01
Kumar et al. 2016 200 1 India tertiary article 1 .24
Kumar et al. 2013 450 1 India secondary article 1 .08
Kvapil 2007 237 25 USA secondary dissertation 3 .30
Lanciano & Curci 2014 89 5 Italy tertiary article 3 .44
Lasser 1997 64 1 USA tertiary dissertation 1 .09
Lawrence & Deepa 2013 400 1 India secondary article 1 .17
Leddy et al 2011 330 2 USA tertiary article 3 ⫺.09
Lewis 2004 93 2 USA tertiary dissertation 2, 3 ⫺.01
Libbrecht et al. 2014 367 6 Europe tertiary article 3 .13
Lochner 2016 119 1 USA tertiary dissertation 2 .05
Loera 2013 105 10 USA tertiary dissertation 3 .17
Lu 2010 2195 13 USA tertiary dissertation 1 .05
Lui 2009 108 48 USA & Canada secondary dissertation 1 .08
MacCann 2019 39 14 Australia tertiary unpublished 3 ⫺.03
MacCann & Roberts 2008 175 6 Australia tertiary article 3 .30
MacCann & Burrows 2013 120 2 Australia tertiary article 2 ⫺.14
MacCann et al. 2011 186 5 USA secondary article 3 .34
Malik & Shujja 2013 204 5 Pakistan primary article 1 .12
Margavio et al. 2012 409 1 China/USA tertiary article 2 .62
Marquez et al. 2006 77 1 Spain secondary article 3 .46
Martin 2011 170 12 USA primary dissertation 1 .15
Matešić 2015 369 3 Croatia secondary article 1 .00
Mavroveli et al. 2009 140 2 UK primary article 1 .25
Mavroveli et al. 2008 70 10 UK secondary article 1 .21
McClain 2009 85 1 USA tertiary dissertation 2 .23
Menzie 2005 55 1 USA secondary dissertation 1 .21
MHS 2019 1771 1 USA tertiary unpublished 3 .13
Mikolajczak 2019 74 1 Belgium tertiary unpublished 1 .23
Mitrofan & Cioricaru 2014 136 1 Romania secondary conference paper 1 ⫺.06
Naeem et al. 2014 467 1 Saudi Arabia tertiary article 2 .14
Nelson 2010 141 1 USA secondary dissertation 3 .38
Nesari et al. 2011 120 1 Iran mixed article 1 .14
Newsome et al. 2000 180 6 Canada tertiary article 1 ⫺.02
Nwabuebo 2013 60 1 USA tertiary dissertation 1 .19
O’Connor, Little 2003 90 22 USA tertiary article 1, 3 .15
Ogundokun et al. 2010 1563 1 Nigeria secondary article 1 .74
Olatoye et al. 2010 235 1 Nigeria tertiary article 2 ⫺.18
Olson 2008 74 6 USA tertiary dissertation 1 .12
Opateye 2014 600 1 Nigeria tertiary article 1 .10
Parker et al. 2004 667 5 USA secondary article 1 .25
Parker et al. 2005 1426 5 USA tertiary article 1 .12
Parker et al. 2004 372 10 Canada tertiary article 1 .10
Petersen 2010 51 6 USA secondary dissertation 1 .11
Pishghadam 2009 508 30 Iran tertiary article 1 .12

(Appendix continues)
 APA NLM
tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z xppws S⫽1 11/21/19 4:31 Art: 2017-1125

EMOTIONAL INTELLIGENCE AND ACADEMIC PERFORMANCE 37

Appendix (continued)

Education
Authors Year Na kb Country level Publication type Streamc rd

Por et al. 2011 130 1 UK tertiary article 2 ⫺.11

Qualter et al. 2012 207 64 UK secondary article 1, 3 .20
Radford 2011 115 1 USA tertiary dissertation 1 .10
Rankin 2013 178 5 UK tertiary article 2 .13
Rastegar & Karami 2013 106 1 Iran tertiary article 2 .23
Rice 2007 486 13 USA secondary dissertation 1 .14
Richardson 2000 98 6 USA primary dissertation 1 .20
Rivers et al. 2012 66 3 USA primary article 3 .51
Rodeiro, Emery & Bell 2012 874 5 UK secondary article 1 .26
Rodrigo-Ruiz 2017 232 19 Spain secondary dissertation 1, 3 .24
Saklofske et al. 2012 163 5 UK tertiary article 1 .05
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Saklofske et al. 2019 277 1 Canada tertiary unpublished 1 .01

Samples 2010 111 5 USA tertiary dissertation 3 .15
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Sanchez-Ruiz et al. 2013 323 1 Cyprus tertiary article 1 .35

Schutte et al. 1998 64 1 USA tertiary article 2 .32
Shipley et al. 2010 193 4 USA tertiary article 1 .08
Sierra et al. 2013 129 15 Spain tertiary article 3 .06
Singh et al. 2009 389 4 Malaysia tertiary conference paper 2 .18
Skipper & Brandenburg 2013 142 1 USA tertiary article 1 .01
Song et al. 2010 173 2 China tertiary article 2 .27
Stewart & Chrisholm 2012 154 90 UK or Canada tertiary article 1 .11
Stottlemyer 2002 185 26 USA secondary dissertation 1 .01
Suliman 2010 98 1 Saudi Arabia tertiary article 1 ⫺.15
Sunbul 2007 311 5 Turkey tertiary conference paper 1 .16
Szuberla 2005 61 12 USA primary dissertation 3 .30
Tok & Morali 2009 295 3 Turkey tertiary article 2 .05
Trapp 2011 118 5 USA tertiary dissertation 1 .07
Vargas 2014 402 14 USA tertiary dissertation 1 .04
Veitch 2011 251 14 USA tertiary dissertation 1 .04
Vela 2004 760 14 USA tertiary dissertation 1 .07
Victoroff & Boyatzis 2013 100 12 USA tertiary article 1 .06
Walker 2006 1205 30 USA tertiary dissertation 1 .03
Walsh-Portillo 2011 45 1 USA tertiary dissertation 1 .18
Willis 2015 591 1 USA tertiary dissertation 1 .06
Woitaszewski & Aalsma 2004 39 1 USA secondary article 3 .05
Wraight 2008 243 1 USA tertiary dissertation 1 .17
Wurf & Croft-Piggin 2015 81 10 Australia tertiary article 2 .14
Xu 2016 1799 1 China secondary article 2 .08
Zandi 2012 239 6 Iran tertiary conference paper 1 .09
Zarezadeh 2013 330 6 Iran tertiary conference paper 1 .11
Zirak & Ahmadian 2015 337 5 Iran primary article 1 .10
Zysberg et al. 2011 102 2 Israel tertiary article 3 .28
a
N ⫽ number of participants (if multiple samples in the paper/source, N is the average). b k ⫽ number of correlation coefficients (e.g., if correlations of
the four MSCEIT branches with GPA reported, k ⫽ 4). c Stream ⫽ 1 (ability EI), 2 (self-rated EI), or 3 (mixed EI). d r ⫽ unweighted, uncorrected
average correlation across all observations in the citation.

Received December 19, 2017

Revision received August 30, 2019
Accepted October 22, 2019 䡲
JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 2 OUTPUT: Thu Nov 21 04:31:41 2019
/tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z

AUTHOR QUERIES

AUTHOR PLEASE ANSWER ALL QUERIES 1

AQau—Please confirm the given-names and surnames are identified properly by the colors.
⫽ Given-Name, ⫽ Surname
The colors are for proofing purposes only. The colors will not appear online or in print.
AQ1—Author: This article has been lightly edited for grammar, style, and usage. Please compare
against your original document and make changes on these paged proofs. If no change is
required in response to a question, please write “OK as set” in the margin.
AQ2—Author: Please be sure to provide the name of the department(s) with which you and your
coauthors are affiliated at your respective institutes if you have not already done so. If you are
affiliated with a governmental department, business, hospital, clinic, VA center, or other
nonuniversity-based institute, please provide the city and U.S. state (or the city, province, and
country) in which the institute is based. Departments should be listed in the author footnote
only, not the byline. If you or your coauthors have changed affiliations since the article was
written, please include a separate note indicating the new department/affiliation: [author’s
name] is now at [affiliation].
AQ3—Author: Your article contains supplemental material. If you have queries for references that
are not cited in the text these can be ignored if the references are found in the supplemental
material.
AQ4 —Author: Confirm/correct as necessary all heading levels as set throughout the article.
AQ5—Author: Please review the typeset table(s) carefully against your original table(s) to verify
accuracy of editing and typesetting.
AQ6 —Author: Cohen (1962) is not in your references. Please add to the reference list or delete
from the text.
AQ7—Author: Cohen (1988) is not in your references. Please add to the reference list or delete
from the text.
AQ8 —Author: Confirm/correct as necessary “k ⫽ xx.”
AQ9 —Author: Gibbs (1995) is not in your references. Please add to the reference list or delete
from the text.
AQ10 —Author: Hodzic et al. (2018) is not in your references. Please add to the reference list or
JOBNAME: AUTHOR QUERIES PAGE: 2 SESS: 2 OUTPUT: Thu Nov 21 04:31:41 2019
/tapraid5/z2r-psybul/z2r-psybul/z2r99918/z2r2709d18z

AUTHOR QUERIES

AUTHOR PLEASE ANSWER ALL QUERIES 2

delete from the text (cited twice).

AQ11—Author: Mattingly & Kraiger (2018) is not in your references. Please add to the reference
list or delete from the text (cited twice).
AQ12—Author: Schutte et al. (2013) is not in your references. Please add to the reference list or
delete from the text (cited twice).
AQ13—Author: Azizollah et al. (2013), Bar-On (1997), Borenstein et al. (2009), Bramley & Dhawan
(2011), Bryant (2007), Cochran (1954), Cronbach (1960), Fayombo (2012a), Grandey (2003),
Higgins & Thompson (2002), Hunter & Schmidt (1990), Hunter & Schmidt (2004), Joibari &
Mohammadtaheri (2011), Kuncel et al. (2001), Mayer et al. (2002), McGrew (2009), Myers &
Pyles (1992), Sackett (2007), Schneider & Newman (2015), Spearman (1904), Spearman (1931),
and Steel & Kammeyer-Mueller (2002) are not cited in your text. Please cite in the text or
delete from the references.
AQ14 —Author: Provide page numbers for reference Gross & Thompson (2007).
AQ15—Author: Please confirm that all authors’ institutional affiliations (including
city/state/country locations) are correct as shown in the affiliations footnote and provide
complete correspondence information.
AQ16 —Author: Please confirm figure(s) and legend(s) as set, and cite Figures 3– 6 in the article
text.
AQ17—Author: Olderbak, Semmler, & Doebler, 2017 is not in your references. Please add to the
reference list or delete from the text.