WRITE 8-9 pages on Topic- EMOTIONAL INTELLIGENCE IMPACTS ON THE WORKFORCE

Miao_Ameta-analysisofemotionalintelligenceandworkattitudes.pdf

Home >Human Resource Management homework help >WRITE 8-9 pages on Topic- EMOTIONAL INTELLIGENCE IMPACTS ON THE WORKFORCE

Journal of Occupational and Organizational Psychology (2017), 90, 177–202

www.wileyonlinelibrary.com

A meta-analysis of emotional intelligence and work attitudes

Chao Miao 1 *, Ronald H. Humphrey

2 and Shanshan Qian

1 Finance, Accounting and Management Department, Jay S. Sidhu School of Business &

Leadership, Wilkes University, Wilkes-Barre, Pennsylvania, USA 2 Department of Leadership and Management, Lancaster University Management

School, Lancaster, UK 3 Department of Management, College of Business and Economics, Towson University, Towson, Maryland, USA

Our meta-analysis of emotional intelligence (EI) demonstrates that: First, all three types of

EI are significantly related to job satisfaction (ability EI: q̂ = .08; self-report EI: q̂ = .32; and mixed EI: q̂ = .39). Second, both self-report EI and mixed EI exhibit modest yet statistically significant incremental validity (DR2 = .03 for self-report EI and DR2 = .06 for mixed EI)and large relative importance (31.3% for self-report EIand 42.8% formixed EI) in

the presence of cognitive ability and personality when predicting job satisfaction. Third,

we found mixed support for the moderator effects (i.e., emotional labour demand of jobs)

for the relationship between EI and job satisfaction. Fourth, the relationships between all

three types of EI and job satisfaction are mediated by state affect and job performance.

Fifth, EI significantly relates to organizational commitment (self-report EI: q̂ = .43; mixed EI: q̂ = .43) and turnover intentions (self-report EI: q̂ = �.33). Sixth, after controls, both self-report EI and mixed EI demonstrate incremental validity and relative importance

(46.9% for self-report EI; 44.2% for mixed EI) in predicting organizational commitment.

Seventh, self-report EI demonstrates incremental validity and relative importance (60.9%)

in predicting turnover intentions.

Practitioner points

� Employees with higher emotional intelligence (EI) have higher job satisfaction, higher organizational commitment, and lower turnover intentions.

� Adding EI measures to the set of personality and cognitive measures currently being used can improve the ability to assess employee job satisfaction, organizational commitment, and turnover intentions.

� EI improves job satisfaction by helping employees reduce negative feelings, by increasing positive feelings, and/or by improving job performance.

� To produce productive and satisfied workers, organizations should incorporate EI in employee recruitment, training, and development programmes.

Emotional intelligence (EI; Mayer, Roberts, & Barsade, 2008; Mayer & Salovey, 1997) is

defined ‘as the set of abilities (verbal and non-verbal) that enable a person to generate,

recognize, express, understand, and evaluate their own, and others’, emotions in order

*Correspondence should be addressed to Chao Miao, Finance, Accounting and Management Department, Jay S. Sidhu School of Business & Leadership, Wilkes University, Wilkes-Barre, PA 18766, USA (email: [email protected]).

DOI:10.1111/joop.12167

177

to guide thinking and action that successfully cope with environmental demands and

pressures’ (Van Rooy & Viswesvaran, 2004, p. 72). There is a large volume of evidence

both confirming the predictive validity of EI and indicating that EI predicts outcomes

such as academic performance, emotional labour, job performance, organizational citizenship behaviour, workplace deviance, leadership, life satisfaction, stress, trust,

team process effectiveness, and work–family conflict (Ashkanasy & Daus, 2002; Bar-On, 2000; Gooty, Connelly, Griffith, & Gupta, 2010; Humphrey, 2002, 2013; Jordan,

Ashkanasy, Hartel, & Hooper, 2002; Kellett, Humphrey, & Sleeth, 2006; Kluemper,

DeGroot, & Choi, 2013; O’Boyle, Humphrey, Pollack, Hawver, & Story, 2011; Winkel,

Wyland, Shaffer, & Clason, 2011). However, the relationships between EI and work

attitudes remain indeterminate and unclear. Differences in effect sizes between primary

studies also suggest a need to meta-analyse the relationships between EI and work attitudes. It is also important to understand how EI influences work attitudes, and so we

unpack its relationship with job satisfaction by evaluating mediating and moderating

factors in this relationship. Finally, as organizations consider developing the EI of their

employees, it is important to understand how this variance affects worker attitudes.

In this study, we performed a meta-analysis on the relationships between EI and three

important work attitudes: job satisfaction, organizational commitment, and turnover

intentions. EI may influence work attitudes in a variety of ways. Job satisfaction reflects

appraisal-based reactions towards one’s job, meaning that favourable evaluations of work characteristics produce job satisfaction and unfavourable appraisals of work character-

istics engender job dissatisfaction (Breaux, Munyon, Hochwarter, & Ferris, 2009; Weiss,

2002). EI may have a considerable influence on how people appraise their jobs because EI

consists, in part, of the ability to reason effectively about events that produce positive or

negative emotions. Consequently, EI may have a strong influence on how employees

interpret and react to work events.

Job performance is a key work event, and to the extent that employees find that

performing well at work helps them meet their personal goals, then high job performance should increase job satisfaction (Locke, 1969) and other work attitudes.

Thus, it is not only the level of the performance, but the employees’ perceptual

processes and personal goals that determine whether job performance increases job

satisfaction (Munyon, Hochwarter, Perrew�e, & Ferris, 2010). EI may be a characteristic that predisposes employees to see job performance in a light that enhances job

satisfaction. Although many models of attitudes and behaviours assume that attitudes

cause behaviours, self-perception theory (Bem, 1967) suggests that people also observe

their behaviours in order to infer their own attitudes. There is considerable evidence that job performance can influence attitudes (Judge, Thoresen, Bono, & Patton, 2001; Locke

& Latham, 2002). Prior meta-analyses have established that EI is positively related to job

performance (O’Boyle et al., 2011). Thus, EI should have an indirect path through job

performance to job satisfaction.

These same processes may also occur when employees observe their other work

behaviours (i.e., besides their direct job performance) and the other aspects of their work

environment. In other words, EI may help cast a rosy glow over a wide variety of work

events and help employees interpret them in a positive light, one that promotes positive affect and diminishes negative affect. Employees who then observe their positive mood

and a positive workplace will naturally then infer that they have high job satisfaction.

Because positive affect and negative affect are important mediators according to Affective

Events Theory (AET; Weiss & Cropanzano, 1996), EI’s effects on work attitudes should be

at least partially mediated by positive and negative state affect.

178 Chao Miao et al.

Our research has several key purposes. No prior meta-analyses have examined EI

and work attitudes, so the following relationships have not been tested using meta-

analytic techniques for establishing the most accurate estimates of effects sizes,

incremental validity, moderators, and mediator effects. We have addressed this by first using meta-analysis to more accurately determine the overall size of the

relationships between EI and work attitudes (job satisfaction, organizational

commitment, and turnover intentions). Second, we use meta-analyses to control

for personality and cognitive intelligence and to test for EI’s incremental

predictability and relative importance when predicting work attitudes. Third, we

test for an important moderator of the EI–job satisfaction relationship. Fourth, we examine whether the EI–job satisfaction relationship is mediated by state affect and by job performance.

Theory and hypotheses

Employee EI and work attitudes

The classification of EI. The construct of EI has received substantial attention from

researchers in the fields of psychology and management (Joseph & Newman, 2010; Kellett et al., 2006; Kluemper et al., 2013; Mayer et al., 2008). EI is argued to be a predictor of job

performance (Goleman, 1995) and effective leadership (Walter & Bruch, 2009; Walter,

Cole, & Humphrey, 2011). To make sense of this considerable research, Ashkanasy and

Daus (2005, p. 441) categorized EI research into three streams. These have become

known as ability EI (stream 1), self-report EI (stream 2), and mixed EI (stream 3). To show

that EI measures can satisfy the traditional criteria for intelligence measures by having

objective right and wrong answers, Mayer, Salovey, Caruso, and Sitarenios (2003)

developed and refined their measure – MSCEIT V2.0. MSCEIT V2.0 is a representative ability EI measure. Researchers have found that subscales of the MSCEIT V2.0 have

predictive abilityfor important work-related variables even when controlling for cognitive

intelligence and the Big Five personality traits (Kluemper et al., 2013). Ability measures of

EI have also been shown to be related to emotion-focused coping, which in turn facilitates

performance (Gooty, Gavin, Ashkanasy, & Thomas, 2014). Other researchers believe that

self-reports are an excellent way to assess EI because intrapersonal processes, such as an

awareness of one’s emotions, are most easily measured by self-assessments of internal

states (Petrides & Furnham, 2003). These researchers often conceptualize EI as a trait rather than as an ability (Petrides, 2009a, 2009b; Petrides & Furnham, 2003). Represen-

tative measures in the stream 2 self-reports category include the Assessing Emotions Scales

(Schutte et al., 1998), the Workgroup Emotional Intelligence Profile (Jordan et al., 2002),

and the Wong & Law Emotional Intelligence Scale (Wong & Law, 2002). Representative

mixed EI measures include the Bar-On Emotional Quotient Inventory (Bar-On, 2000) and

the Emotional and Social Competency Inventory (Boyatzis, Brizz, & Godwin, 2011). Like

the stream 2 self-reports, mixed EI measures use self-report measures; however, they

include a broader set of variables and competencies as well as traits. It is noted that these three streams of EI are related yet still distinct from each

other in a number of ways. O’Boyle and his colleagues demonstrated that ‘Because

stream 3 measures overlap both in their measurement method and in the content of

their questions, while stream 2 measures only overlap with regard to the use of self-

reports, stream 3 measures should show higher relationships with personality factors

than stream 2 measures. . ...stream 3 measures, unlike stream 2, include measures of

Emotional intelligence and work attitudes 179

personality factors not directly related to EI, so it is likely that these measures will

overlap more with similar personality measures’ (2011, p. 793). It is worthwhile to

point out that some overlap between EI and other constructs is reasonable and is

indicative of construct validity because EI should be related to personality variables such as emotional stability (O’Boyle et al., 2011). Their meta-analytic results

confirmed this prediction and showed that corrected correlations between personality

and both stream 1 ability EI and stream 2 self-report EI range from weak to moderate,

whereas the corrected correlations between personality and stream 3 mixed EI range

from moderate to strong.

Meta-analytic findings of EI. There are multiple meta-analytic reviews that investigated the relationship between EI and job performance. Van Rooy and Viswesvaran (2004)

performed a meta-analysis and reported a .23 operational validity of EI in predicting

performance; they concluded that EI is indeed a valuable predictor of performance. They

also found that overall EI has small to moderate corrected correlations with personality

traits and a small corrected correlation with cognitive ability. Joseph and Newman (2010)

meta-analytically integrated the research on EI and job performance, and proposed a

cascading model of EI with emotional labour as a moderator. O’Boyle et al. (2011)

performed a meta-analysis of EI and contributed to the cumulative scientific knowledge by improving the two aforementioned meta-analyses. For instance, O’Boyle and his

colleagues included more studies and examined how each type of EI measure correlated

with Big Five personality measures and cognitive ability. They employed dominance

analysis to assess the relative importance of each EI stream in predicting job performance.

They found that all three streams of EI correlated with job performance and that self-report

EI and mixed EI exhibited incremental validities over and above cognitive intelligence and

the five factor model (FFM) of personality in predicting job performance. Dominance

analyses also demonstrated that all three streams of EI showed meaningful relative importance for the prediction of job performance in the presence of the FFM and

cognitive intelligence.

Job satisfaction. Job satisfaction is one of the most influential, important, and popular

constructs in the area of organizational psychology because it is a predictor of many

critical behavioural, attitudinal, and health-related outcomes, such as organizational

citizenship behaviour, counterproductive work behaviour, task performance, organiza- tional commitment, turnover intention, turnover, withdrawal cognitions and behaviours,

and physical and psychological health outcomes (Judge & Kammeyer-Mueller, 2012;

Schleicher, Hansen, & Fox, 2011). There exist multiple definitions of job satisfaction.

Weiss (2002) pointed out that the attitudinal approach to defining job satisfaction is the

most accepted one in the literature. This approach conceptualizes job satisfaction as

having both affective (emotional) and cognitive (belief) bases (Fisher, 2000; Weiss, 2002;

Weiss, Nicholas, & Daus, 1999). The affective base of job satisfaction refers to one’s

feelings about an attitude object, whereas the cognitive base of job satisfaction refers to one’s beliefs about an attitude object (Schleicher et al., 2011). This conceptualization of

job satisfaction dovetails with goal setting theory, which suggests that job satisfaction, at

its core, reflects goal achievement at work because both affective and cognitive bases of

job satisfaction are influenced by one’s progress towards goal achievement (Diener, Suh,

Lucas, & Smith, 1999; Locke, 1969, 1976; Locke, Cartledge, & Knerr, 1970). Goals serve as

180 Chao Miao et al.

the reference criteria for satisfaction versus dissatisfaction, meaning that for any given

trial, the achievement of goals produces satisfaction and failure to reach goals creates

dissatisfaction (Locke & Latham, 2002). Across trials, the more goals one reaches, the

higher one’s satisfaction is.

EI – job satisfaction, organizational commitment, and turnover intentions. Emo- tional intelligence should positively relate to job satisfaction and organizational commit-

ment, and be negatively related to turnover intentions. Emotionally intelligent individuals

are able to regulate their emotions, meaning they are less likely to leave an organization due

to emotional shocks and so may have reduced turnover intentions and greater

organizational embeddedness (Mitchell, Holtom, Lee, Sablynski, & Erez, 2001; Mitchell & Lee, 2001). Similarly, we would expect EI to positively predict organizational

commitment, as employees view work as instrumental in achieving their work-related

goals. As previously mentioned, job satisfaction consists of job appraisals, such that

satisfactory assessments of work characteristics produce job satisfaction and negative

judgements of work characteristics create job dissatisfaction (Breaux et al., 2009; Weiss,

2002). EI encompasses the ability to reason productively about positive and negative

workplaceeventsandthus shouldhave a stronginfluenceonhowemployeesinterpret and

respond to work events. When high job performance helps employees meet their personal goals, it should increase job satisfaction (Locke, 1969; Locke & Latham, 2002) and

organizational commitment and thereby reduce turnover intentions. Because EI improves

job performance (O’Boyle et al., 2011), it should indirectly influence job satisfaction.

Emotionally savvy individuals are inclined to interpret their jobs as more satisfying and

rewarding rather than threatening and hostile (Fox & Spector, 2000; Kong & Zhao, 2013;

Thoresen, Kaplan, Barsky, Warren, & de Chermont, 2003; Walter & Bruch, 2009). This is

because emotionally intelligent individuals are more resilient, are more likely to bounce

back from negative feelings, and are more adept at evaluating and regulating their own emotions (Sy, Tram, & O’Hara, 2006). Emotionally intelligent individuals have a greater

understanding of the causes of stress. Consequently, they know how to craft effective

plans to deal with negative outcomes in order to maintain positive feelings and high job

satisfaction. This may be one reason why people high on EI have better physical and

mental health, according to two meta-analyses (Martins, Ramalho, & Morin, 2010; Schutte,

Malouff, Thorsteinsson, Bhullar, & Rooke, 2007). In addition, people with high EI can

accurately read others’ emotions, and reading others’ emotions helps people understand

how to respond to others and how to act in appropriate ways in a variety of social situations (Byron, 2007). As a result, employees with high EI should have positive social

relationships with others in the workplace, and this should result in higher job satisfaction

and organizational commitment (Goleman, 1995; Kafetsios & Zampetakis, 2008).

Empirical findings support links between EI and overall job satisfaction (Kafetsios &

Zampetakis, 2008; Sy et al., 2006; Wong & Law, 2002) and between EI and organizational

commitment and turnover intentions (Jordan & Troth, 2011).

Hypothesis 1: EI should positively relate to job satisfaction and organizational commitment, and EI

should negatively relate to turnover intentions.

Incremental validity and relative importance of EI. Emotional intelligence denotes variation in the extent to which people can resolve a set of problems involving emotions,

thus differentiating EI from other intelligence factors that primarily centre on cognitive

Emotional intelligence and work attitudes 181

processes (Côt�e, 2014; Côt�e & Miners, 2006; Mayer et al., 2008). EI refers to a general intelligence in the realm of emotions, whereas cognitive intelligence refers to a general

intelligence in the realm of cognition (Côt�e, 2014; Côt�e & Miners, 2006). As such, EI differs from cognitive intelligence due to its unique representation of intelligence in the domain of emotion. EI differs from personality as well, because personality does not reflect one’s

ability/intelligence, whereas EI does (Joseph & Newman, 2010). Due to these reasons, EI

has unique content and has often displayed incremental validity in predicting outcomes

over other measures of intelligence, socio-emotional traits, and personality factors, which

has already been supported by meta-analytic findings (Côt�e, 2014; Mayer et al., 2008). Meta-analytic findings have demonstrated that the overlap between all three types of EI

and cognitive ability is weak to moderate and the overlap between all three types of EI and

FFM is weak to moderate in general (O’Boyle et al., 2011). It is worthwhile to point out that some overlap is reasonable and is indicative of construct validity, because EI should be

related to personality variables such as emotional stability (O’Boyle et al., 2011). EI should

also relate to cognitive ability because it is a form of intelligence (Côt�e, 2014). Taken altogether, despite the overlap between EI and cognitive ability and personality,

there is still much unique variance in EI that cannot be explained by personality and

cognitive ability, and this unique variance may predict job satisfaction, organizational

commitment, and turnover intentions above and beyond personality and cognitive ability

(Côt�e, 2014; O’Boyle et al., 2011; Sy et al., 2006). We accordingly offer the following hypotheses:

Hypothesis 2: EI should contribute incremental validity and relative importance in the presence of

the FFM and cognitive ability when predicting job satisfaction, organizational

commitment, and turnover intentions.

Differential validity of EI. Personality is a good predictor of job satisfaction because

personality traits reflect one’s affective disposition and influence one’s interpretation of

job characteristics and one’s mood at work (Judge, Heller, & Mount, 2002). On the other

hand, cognitive ability is a weak predictor of job satisfaction because cognitive ability is a

cognitive trait, whereas job satisfaction is primarily determined by affective dispositions,

such as personality traits like extraversion and neuroticism (Ganzach, 1998; Staw, Bell, &

Clausen, 1986). To support these arguments, meta-analytic findings have demonstrated

that personality is a good predictor of job satisfaction (Judge et al., 2002), whereas cognitive ability is a weak predictor of job satisfaction (Gonzalez-Mul�e, Carter, & Mount, 2014).

Mixed EI has the highest correlation with personality, self-report EI the next highest,

and ability EI the lowest correlation with personality (O’Boyle et al., 2011). In addition,

ability EI has the highest correlation with cognitive ability, while self-report EI and mixed

EI have small correlations with cognitive ability. Overall, this suggests that ability EI

measures may be similar to cognitive intelligence measures in their impact on job

satisfaction and thus have fairly small correlations with job satisfaction. In a similar vein, mixed EI measures have the highest associations with personality measures and thus

should have the largest correlation with job satisfaction. We thus propose the following

hypothesis:

Hypothesis 3: Mixed EI will show the strongest relationship with job satisfaction, self-report EI the

next strongest, and ability EI the weakest relationship with job satisfaction.

182 Chao Miao et al.

The mediating role of affect

Emotional intelligence may be a characteristic that inclines employees to view a wide

variety of organizational events in a manner that augments positive affect. Consistent with

self-perception theory (Bem, 1967), employees may observe their positive moods at work and infer that they have high job satisfaction. This is consistent with AET Weiss &

Cropanzano, 1996), which also provides explanations for the positive link between EI and

job satisfaction through state affect. State affect refers to ‘what one is feeling at any given

moment in time’ (Thoresen et al., 2003, p. 915). State positive affect (SPA) refers to

pleasant emotions such as feeling active, alert, and energetic at any given moment in time,

whereas state negative affect (SNA) refers to the momentary experience of anger, fear,

nervousness, and other negative emotions at any given moment in time (Watson, 2000;

Watson, Clark, & Tellegen, 1988). Meta-analytic findings have indicated that SPA is positively related to job satisfaction and personal accomplishment and negatively related

to emotional exhaustion and depersonalization, whereas SNA is negatively related to job

satisfaction and personal accomplishment and positively related to emotional exhaustion

and depersonalization (Thoresen et al., 2003).

Affective Events Theory suggests that each individual should have an average affective

mood level and that some people tend to be on the positive half, whereas others tend to be on

the negative half. Further, responding to discrete ‘affective events’ in the workplace will

influenceaffectiveresponses,thusleadingtoaffective,attitudinal,and behaviouraloutcomes; as such, this average mood level can be either diminished or raised by negative or positive

events at work. Hence, affective reactions generated by workplace events (i.e., SPA and SNA)

create ebb and flow in job satisfaction (Ashkanasy & Humphrey, 2011; Humphrey, 2013;

Johnson, 2009; Walter & Bruch, 2009; Weiss & Cropanzano, 1996; Weiss et al., 1999).

Building on AET, we argue that there are two prominent categories of reasons why EI is

associated with job satisfaction – enhancement of SPA and reduction in SNA. Job satisfaction has an affective (i.e., feeling) component (Weiss, 2002; Weiss et al., 1999).

We propose that EI may contribute to the affective base of job satisfaction by increasing SPA and decreasing SNA. Emotionally intelligent individuals are able to identify and

interpret cues that activate self-regulatory action in order to cultivate SPA and circumvent

SNA (Karim, 2009; Mayer & Salovey, 1997). High EI individuals are better at handling

affective processes because they can accurately perceive and monitor their own feelings

and precisely process emotional information in order to effectively respond to their

feelings. This allows them to develop appropriate strategies to regulate SNA and maintain

SPA (Dong, Seo, & Bartol, 2013). They are sensitive and reactive to positive emotion-

invoking experiences at work, thus making them feel more positive (more SPA) and less negative (less SNA) at work (De Clercq, Bouckenooghe, Raja, & Matsyborska, 2014). As

SPA is positively related to job satisfaction, whereas SNA is negatively related to job

satisfaction (Thoresen et al., 2003), high EI persons can increase their job satisfaction by

regulating their emotions to experience more intense SPA and less SNA.

A considerable number of studies have examined EI and job satisfaction, and the ample

number of studies has allowed us to test for mediators and moderators. However, fewer

studies have been conducted on the relationships between EI and organizational

commitment and turnover intentions, so wecould not examine mediators and moderators for these outcome variables. We advance the following hypotheses:

Hypothesis 4: SPA mediates the relationship between EI and job satisfaction.

Hypothesis 5: SNA mediates the relationship between EI and job satisfaction.

Emotional intelligence and work attitudes 183

Goal setting theory and the mediating role of job performance

Prior meta-analytic evidence has confirmed a positive relationship between EI and job

performance (Joseph & Newman, 2010; O’Boyle et al., 2011). EI may increase job

performance because emotionally intelligent individuals are able to regulate their emotions in order to experience positive emotions. Positive emotions widen employees’

behavioural repertoires, increase their behavioural flexibility, and boost their attentional

scope, thus resulting in increased job performance (Judge & Kammeyer-Muellar, 2008).

In line with goal setting theory, job satisfaction is an outcome of goal-directed

performance, because one’s progress towards goal accomplishment (i.e., goal-directed

performance) influences job satisfaction (Locke, 1969; Locke & Latham, 2002). Across

trials, the better one performs, the more goals one accomplishes and the higher job

satisfaction one has. Judge et al. (2001) pointed out that if effective job performance supports the accomplishment of major goals in work, individuals should have higher job

satisfaction as a result.

Taken altogether, we propose that job performance should mediate the relationship

between EI and job satisfaction because EI enables one to attain one’s performance goals,

and obtaining goals increases job satisfaction (Locke, 1976). Consistent with self-

perception theory, employees observe their level of performance and perceive a

corresponding level of job satisfaction (Bem, 1967). Emotionally savvy individuals have a

better understanding of themselves, and this increases both their ability to set self- motivating goals and their chances of achieving performance goals that lead to job

satisfaction (Kafetsios & Zampetakis, 2008; Mayer & Salovey, 1997; Spence, Oades, &

Caputi, 2004; Wong & Law, 2002). For example, emotionally intelligent individuals know

how to recognize their supervisors’ attitudes from emotional cues; moreover, they know

how to regulate their own emotion to act and communicate in ways that foster better

social relationships with their supervisors (Wong & Law, 2002), which in turn should lead

to higher performance appraisals and job satisfaction.

Emotionally intelligent individuals also regulate their emotions to deter the draining of resources that cause burnout, to quickly bounce back from negative feelings, and to

maintain positive feelings so that they can preserve and replenish cognitive and emotional

resources. According to the job demands–resources model, conserving these cognitive resources should enable employees to more effectively accomplish the performance goals

that lead to positive outcomes such as job satisfaction (Bakker & Demerouti, 2007;

Crawford, LePine, & Rich, 2010; Hobfoll, 2001). We suggest the following hypothesis:

Hypothesis 6: Job performance mediates the relationship between EI and job satisfaction.

Moderator

Emotional labour. Emotional labour refers to ‘the management of feeling to create a

publicly observable facial and bodily display’ (Hochschild, 1983, p. 7). Jobs that involve

emotional labour include face-to-face or voice-to-voice contact with the public, produce

an emotional state in another person, and allow employers to exercise a degree of control

over the emotional activities of employees (Hochschild, 1983). Thus, emotional labour

requires the act of both displaying the appropriate emotion (i.e., conforming to a display

rule) and regulating both feelings and expressions to forward organizational goals

(Ashforth & Humphrey, 1993; Grandey, 2000). Meta-analysis has related emotional labour to employee well-being, job satisfaction, and organizational attachment (Hulsheger &

Schewe, 2011).

184 Chao Miao et al.

It is likely that the association between EI and job satisfaction is conditioned by work

contexts (Côt�e, 2014). One such contextual variable is the emotional labour demand of jobs. Prior findings indicate that EI should predict criteria more strongly in jobs that

involve high emotional labour, because these jobs require employees to regulate their emotional expressions, and thus involve a high level of emotional regulation (Humphrey,

Ashforth, & Diefendorff, 2015; Johnson & Spector, 2007; Joseph & Newman, 2010; Wong

& Law, 2002). A meta-analysis found that people high on EI are more likely to use the most

effective form of emotional labour (Wang, Seibert, & Boles, 2011). The choice of

emotional labour demand as a contextual variable is consistent with trait activation

theory, which suggests that traits should more strongly predict outcomes when a context

has trait-relevant cues that activate the expression of traits (Tett & Burnett, 2003; Tett &

Guterman, 2000). We have predicted that the relationship between EI and job satisfaction will be

stronger whena job requires high levels of emotional labour.Whena job requires frequent

customer/interpersonal interaction (i.e., high emotional labour demand), the expression

of EI should be activated because employees need to rely more on their EI to regulate their

emotions in order to prevent emotional and cognitive resources from being drained, and

to effectively maintain and enhance job satisfaction. Where there is infrequent customer/

interpersonal interactions (i.e., low emotional labour demand), the expression of EI may

be suppressed because this job does not demand the use of EI to handle interpersonal interactions. Thus, we hypothesize:

Hypothesis 7: Emotional labour demand moderates the relationship between EI and job

satisfaction such that the relationship becomes stronger when emotional labour

demand is high.

Method

Literature search

We applied several search techniques to maximize the likelihood of capturing all relevant

studies. We set the range of dates starting from the earliest date of each database, journal,

and conference to year 2014. We used a list of keywords (and several variations in them)

for search, such as emotional intelligence, emotional competency, emotional ability, job satisfaction, work satisfaction, organizational commitment, and turnover intention.

First, we searched electronic databases, such as ABI/INFORM, EBSCO Host (e.g.,

Academic Search Complete and Business Source Complete), Google, Google Scholar,

JSTOR, ProQuest Dissertations and Theses, PsycNET (e.g., PsycInfo and PsycArticles), and

Social Science Citation Index. Second, major journals in psychology and management

were also searched, such as Academy of Management Journal, Administrative Science

Quarterly, Journal of Applied Psychology, Journal of Management, Journal of

Organizational Behavior, Journal of Occupational and Organizational Psychology,

Organization Science, and Personnel Psychology. Third, we searched major management

conferences, such as the Academy of Management, the Southern Management Association,

and the Society for Industrial and Organizational Psychology. We contacted scholars who

have published in the EI domain to ask for unpublished manuscripts, correlation matrices,

and raw data, and we completed our search in October 2014. We used the English language

to search for relevant studies. Our search returned a few articles written in foreign

languages that had English titles and abstracts. Two authors of this paper are bilingual and

were able to read some of these articles. Our search yielded 1,036 articles.

Emotional intelligence and work attitudes 185

Inclusion criteria

A study was deemed eligible for being included in the present meta-analysis if it met the

following criteria. First, primary studies had to be empirical and quantitative. All

qualitative studies were excluded. Second, primary studies had to report the correlation coefficients for the relationships between EI and job satisfaction, between EI and

organizational commitment, between EI and turnover intentions, and/or between EI and

state affect. When no such information existed in primary studies, sufficient statistics

needed to be provided in such studies to allow the conversion into effect sizes (we

employed Lipsey and Wilson’s (2001) as well as Peterson and Brown’s (2005) methods to

perform effect size conversions). Third, primary studies had to use real employee samples

in their research design. Studies based on non-employee samples (e.g., student samples)

were eliminated from our meta-analysis. Fourth, a study had to use scales designed to measure EI. Studies that used proxy measures of EI (e.g., self-monitoring scales) were not

eligible. When the above criteria were applied to screen the articles, it resulted in 119

studies. A list of tables describing the included studies and a list of references of the studies

included in the present meta-analytic review were uploaded as online supplemental

materials (see Tables S3–S10 in supplemental materials).

Coding procedures We coded different categories of EI (i.e., ability EI, self-report EI, and mixed EI) based on

Ashkanasy and Daus (2005) and O’Boyle et al. (2011). We coded emotional labour

demands according to the methods developed by Joseph and Newman (2010). The

occupations where there are frequent customer/interpersonal interactions that require

emotion regulation were coded as high emotional labour demand jobs. The occupations

where there are infrequent interpersonal/customer interactions that demand less

emotion regulation were coded as low emotional labour demand jobs. Joseph and

Newman (2010) categorized 191 jobs into high versus low emotional labour demand and we used their categorization to code the emotional labour demand of the studies that we

found. We adhered to the coding rules developed by Thoresen et al. (2003) to code state-

based affect (i.e., SPA and SNA). The studies where respondents were asked to rate their

experiences of positive affect and negative affect over the previous week (or less) were

coded as state affect. As argued by Thoresen et al. (2003), this one-week rule was in line

with Watson’s (2000) definition of state affect.

Two coders participated in coding and independently coded each sample. The initial

coding agreement was 95%. Coding discrepancies were resolved through discussion. Another author of this paper was invited to join the discussion to solve any remaining

coding disagreement when two coders could not reach consensus after discussion. All

codingdisagreementwas handled and resolved,and a 100%consensuswas finallyachieved.

Meta-analytic procedures

We performed psychometric meta-analysis by using the procedures developed by Hunter

and Schmidt (2004) to synthesize collected data. Statistical artefacts can have systematic downward bias effects on effect sizes, and one source of statistical artefacts is measurement

error (Hunter & Schmidt, 2004). We thereby corrected for measurement errors in both

independent and dependent variables for each individual correlation. We noted that some

primary studies did not report the reliability. Thus, we imputed the missing reliability for

both independent and dependent variables by using the mean of reliabilities of the studies

186 Chao Miao et al.

that reported reliability information (Hunter & Schmidt, 2004; also see supplements for

more details). We presented corrected sample-size-weighted mean correlation (q̂) as the estimate of population mean correlation. We calculated corrected 95% confidence intervals

to determine the statistical significance of effect sizes. Effect sizes are considered to be statistically significant when corrected 95% confidence intervals do not include zero. We

performed moderator analyses by using Hunter and Schmidt’s (1990) approach (i.e., z-test).

This test allows the examination of the statistical significance of between-group effect size

difference. We computed Varart%, 80% credibility intervals, and Q statistic to determine the

potential existence of moderators. Varart% denotes the percentage of the variance in q̂ explained by statistical artefacts. Hunter and Schmidt (2004) suggested that moderators

may exist if statistical artefacts explain less than 75% of the variance in the meta-analytic

correlations. We also reported corrected 80% credibility intervals to explore the potential existence of moderators because Whitener (1990) recommended that a wide 80%

credibility interval indicates the possible existence of moderators. In addition, a statistically

significant Q statistic suggests that heterogeneity exists in effect size distribution (i.e., the

potential existence of moderators).

We created meta-analytically derived corrected correlation matrices (see Tables S11(a)

to S12 in supplemental materials) and performed hierarchical multiple regression, relative

weight analyses, and meta-analytic structural equation modelling (Johnson, 2000;

Johnson & LeBreton, 2004; Viswesvaran & Ones, 1995). Along with all effect sizes derived from the present study, we also used corrected effect sizes from prior meta-

analytic reviews to complete the input correlation matrices for these three analyses. We

computed harmonic mean sample size (Viswesvaran & Ones, 1995) because sample sizes

differed across the cells in the correlation matrices. Harmonic mean sample size produces

more conservative estimates because less weight is given to large samples (Colquitt, Scott,

& LePine, 2007).

Results

Main and moderator effects

Because of limited sample sizes for ability EI measures, we were not able to examine ability

EI’s relationships with either turnover intentions or organizational commitment.

Likewise, there were not enough studies to allow us to perform meta-analysis on the

mixed EI–turnover intentions relationship. In the following sections, we will provide the results for EI measures only when the number of studies and sample sizes are large enough

to justify them. Table 1 contains the results of the relationships between EI and job

satisfaction, organizational commitment, and turnover intentions based on psychometric

meta-analysis. The relationship between ability EI and job satisfaction (k = 13, N = 1,927) was positive and statistically significant (q̂ = .08) because the corrected 95% confidence interval spanned from .01 to .15 and did not include zero. The effect sizes for the

relationships between the other two types of EI and job satisfaction (q̂ = .32 for self-report EI and q̂ = .39 for mixed EI) show similar patterns of results. We repeated the same procedures and found that self-report EI positively relates to organizational commitment

(q̂ = .43) and negatively relates to turnover intentions (q̂ = �.33). In addition, mixed EI positively relates to organizational commitment (q̂ = .43). As such, Hypothesis 1, which proposed that EI should positively relate to job satisfaction and organizational commit-

ment and negatively relate to turnover intentions, is supported.

We observed that there were substantial variations across effect sizes for three major

distributions for the relationship between EI and job satisfaction, because far less than 75%

Emotional intelligence and work attitudes 187

T a b le

1 . P sy c h o m e tr ic m e ta -a n a ly si s re su lt s

k N

� r o S D r

q̂ S D q

V a r a r t%

C o rr e c te d

9 5 % C I

C o rr e c te d

8 0 % C R

S ig n ifi c a n t

d if fe re n c e

x . A b il it y E I – jo b s a ti s fa c ti o n

1 3

1 ,9 2 7

.0 7

.1 2

.0 8

.1 0

4 9

.0 1 to

.1 5

�. 0 5 to

.2 0

2 7 .9 0 * *

y , z

E m o ti o n a l la b o u r

a . H ig h

5 6 0 9

.1 0

.1 6

.1 3

.1 5

3 2

.0 6 to

.2 0

�. 0 7 to

.3 3

1 7 .0 8 **

– b . L o w

4 4 9 6

.0 7

.1 2

.0 9

.1 0

5 3

.0 3 to

.1 4

�. 0 4 to

.2 1

7 .6 4 †

– A b il it y E I – st a te

p o si ti v e a ff e c t

2 3 7 3

�. 0 6

.1 1

�. 0 7

.0 9

4 5

�. 1 8 to

.0 5

�. 1 9 to

.0 5

4 .3 9 *

A b il it y E I – st a te

n e g a ti v e a ff e c t

2 3 7 3

�. 3 2

.0 6

�. 3 9

.0 0

1 0 0

�. 4 2 to

�. 3 5

�. 3 9 to

�. 3 9

1 .6 3

y . S e lf -r e p o r t E I – jo b s a ti s fa c ti o n

6 6

2 0 ,1 1 6

.2 8

.1 1

.3 2

.1 1

2 4

.2 9 to

.3 5

.1 8 to

.4 7

3 5 0 .6 2 * * *

x , z †

E m o ti o n a l la b o u r

a . H ig h

3 4

1 1 ,5 1 6

.3 1

.0 9

.3 5

.0 8

3 3

.3 2 to

.3 9

.2 5 to

.4 6

1 0 8 .8 1 ** *

b . L o w

1 0

2 ,4 9 7

.1 7

.1 3

.2 1

.1 4

2 4

.1 6 to

.2 6

.0 4 to

.3 8

4 8 .2 5 ** *

S e lf -r e p o r t E I – o r g a n iz a ti o n a l

c o m m it m e n t

3 0

7 ,6 7 5

.3 6

.1 7

.4 3

.1 7

1 3

.3 8 to

.4 8

.2 2 to

.6 4

2 5 9 .5 3 * * *

S e lf -r e p o r t E I – tu r n o v e r in te n ti o n s

1 7

5 ,0 0 4

�. 2 8

.2 1

�. 3 3

.2 5

6 �.

4 0 to

�. 2 5

�. 6 4 to

�. 0 1

3 2 3 .7 7 * * *

S e lf -r e p o rt E I – st a te

p o si ti v e a ff e c t

3 8 8 9

.4 2

.1 1

.4 7

.1 1

2 0

.4 0 to

.5 4

.3 3 to

.6 0

1 5 .7 9 ** *

S e lf -r e p o rt E I – st a te

n e g a ti v e a ff e c t

3 8 8 9

�. 3 6

.1 7

�. 4 2

.1 8

1 0

�. 5 1 to

�. 3 2

�. 6 4 to

�. 1 9

4 2 .5 8 ** *

z . M ix e d E I – Jo b S a ti s fa c ti o n

4 1

7 ,0 7 6

.3 3

.2 3

.3 9

.2 7

8 .3 1 to

.4 8

.0 5 to

.7 3

1 3 5 1 .2 5 * * *

x , y †

E m o ti o n a l la b o u r

a . H ig h

2 0

3 ,7 2 7

.3 3

.1 2

.3 9

.1 2

2 9

.3 4 to

.4 4

.2 3 to

.5 5

7 7 .6 6 ** *

– b . L o w

1 4

2 ,2 6 4

.3 9

.3 2

.4 6

.3 7

4 .3 3 to

.5 8

�. 0 2 to

.9 3

9 3 3 .9 6 ** *

– M ix e d E I – o r g a n iz a ti o n a l c o m m it m e n t

2 6

3 ,8 6 7

.3 6

.2 2

.4 3

.2 4

1 1

.3 5 to

.5 1

.1 2 to

.7 4

5 0 0 .4 1 * * *

M ix e d E I – st a te

p o si ti v e a ff e c t

1 4 7 5

.2 3

.0 0

.3 1

.0 0

N A

.3 1 to

.3 1

.3 1 to

.3 1

N A

M ix e d E I – st a te

n e g a ti v e a ff e c t

1 4 7 5

�. 1 8

.0 0

�. 2 4

.0 0

N A

�. 2 4 to

�. 2 4

�. 2 4 to

�. 2 4

N A

N o te s.

K , n u m b e r o f in d e p e n d e n t sa m p le s; N , sa m p le si ze ; � r o , u n c o rr e ct e d sa m p le -s iz e -w

e ig h te d m e an

c o rr e la ti o n ; S D r , sa m p le -s iz e -w

e ig h te d st an d ar d d e v ia ti o n o f

o b se rv e d m e an

c o rr e la ti o n s; q̂ , c o rr e c te d sa m p le -s iz e -w

e ig h te d m e an

c o rr e la ti o n ; S D q , sa m p le -s iz e -w

e ig h te d st an d ar d d e vi at io n o f c o rr e c te d m e an

c o rr e la ti o n s;

V ar

a r t% ,p e rc e n ta ge

o fv ar ia n c e in q̂ e x p la in e d b y st at is ti c al ar te fa c ts ;c o rr e ct e d 9 5 % C I, c o rr e c te d 9 5 % c o n fi d e n c e in te rv al ;c o rr e c te d 8 0 % C R ,c o rr e ct e d 8 0 % c re d ib ili ty

in te rv al ;Q

,a st at is ti c u se d to

as se ss th e h e te ro ge n e it y in e ff e ct si ze s ac ro ss st u d ie s; si gn ifi c an t d if fe re n c e = le tt e rs in th is c o lu m n c o rr e sp o n d to

th e le tt e rs in ro w s an d

in d ic at e th at e ff e c t si ze s si gn ifi c an tl y d if fe r fr o m o n e an o th e r at .0 5 le v e l. T h e le tt e rs w it h ‘† ’d e n o te th e st at is ti c al si gn ifi c an c e at .1 0 le v e l. T h e si gn

‘– ’s u gg e st s th e re

is n o

si gn ifi c an t b e tw

e e n -g ro u p d if fe re n c e .z -t e st is u se d to

e v al u at e th e st at is ti c al si gn ifi c an c e o fb e tw

e e n -g ro u p d if fe re n c e in e ff e c t si ze s. E I, e m o ti o n al in te lli ge n c e .F o r c la ri ty

o f re p o rt in g, th e m e ta -a n al y ti c d is tr ib u ti o n s th at

re fe r to

th e m ai n e ff e c t o f e ac h st re am

o f E I o n e ac h w o rk

at ti tu d e w e re

in d ic at e d in b o ld c h ar ac te rs .

† p < .1 0 ; * p < .0 5 ; * * p < .0 1 ; * * * p < .0 0 1 .

188 Chao Miao et al.

of the variance in q̂ (Varart%) was explained by statistical artefacts. This met Hunter and Schmidt’s (2004) 75% rule for indicating the potential existence of moderators. Q statistics

were significant as well, which further confirmed our conclusion that effect size

distributions were heterogeneous for all three types of EI. Therefore, performing further moderator analyses was justified.

The results of the effect size differences among different types of EI are shown in the

last column of Table 1. This column also displays the results of all other moderator

analyses as well. We performed z-tests to determine the statistical significance of the

between-group differences. Our results indicate that ability EI has the lowest relationship

with job satisfaction compared to the other two types of EI (self-report EI vs. ability EI,

Dq̂ = .24, p < .05; mixed EI vs. ability EI, Dq̂ = .31, p < .05). The relationship between mixed EI and job satisfaction is marginally significantly larger than the relationship between self-report EI and job satisfaction (Dq̂ = .07, p < .1). We therefore concluded that Hypothesis 3 is supported (see Table 2).

Emotional labour was a significant moderator only for the self-report EI–job satisfaction relationship. Thus, there was mixed support for Hypothesis 7.

Incremental validity, relative weight analyses, and meta-analytic structural

equation modelling

Incremental validity analysis

Table 3 displays the results of incremental validity analysis based on the hierarchical

multiple regression analysis. Whenthe dependent variable isjob satisfaction, the first model demonstrates that cognitive ability and the FFM in combination account for 15% (p < .001) of the variance in job satisfaction. The second, third, and fourth models illustrate the

incremental validity of each type of EI in the presence of cognitive ability and the FFM. The

second model shows that ability EI contributes no incremental validity (p = ns) in the

Table 2. Summary of results for all hypotheses

Hypotheses Results

Hypothesis 1: EI should positively relate to job satisfaction and

organizational commitment, and EI should negatively relate to

turnover intentions

Supported

Hypothesis 2: EI should contribute incremental validity and relative

importance in the presence of the FFM and cognitive ability

when predicting job satisfaction, organizational commitment,

and turnover intentions

Supported for self-report EI

and mixed EI, but not for

ability EI

Hypothesis 3: Mixed EI will show the strongest relationship with

job satisfaction, self-report EI the next strongest, and ability EI

the weakest relationship with job satisfaction

Supported

Hypothesis 4: SPA mediates the relationship between EI and

job satisfaction

Supported

Hypothesis 5: SNA mediates the relationship between EI and

job satisfaction

Supported

Hypothesis 6: Job performance mediates the relationship between

EI and job satisfaction

Supported

Hypothesis 7: Emotional labour demand moderates the relationship

between EI and job satisfaction such that the relationship becomes

stronger when emotional labour demand is high

Supported only

for self-report EI

Emotional intelligence and work attitudes 189

presence of cognitive ability and the FFM. On the other hand, the third and the fourth

models show that both self-report EI and mixed EI contribute an additional 3% (p < .001) and6%(p < .001)ofvariance,respectively,aboveandbeyondcognitiveabilityandtheFFM.

When the dependent variable is organizational commitment, self-report EI and mixed EI contribute an additional 9% (p < .001) and 8% (p < .001) of variance, respectively, above and beyond cognitive ability and the FFM. When the dependent variable is turnover

intentions, self-report EI contributes an additional 8% (p < .001) of variance above and beyond cognitive ability and the FFM.

Relative weight analysis

Because the predictors in our regression model are correlated, we performed relative weight analysis to determine the relative importance of each predictor in predicting

employee job satisfaction. Table 3 displays the results of relative weight analysis for all

three types of EI in the last two columns of each model. Ability EI only contributed 1.3% of

the explained variance, along with a R 2 contribution of .00 in Model 2. It failed to meet our

threshold for a small effect (see the section of supplemental notes in supplemental

materials for details about how we determined the criteria of small, medium, and large

effects). Further, ability EI demonstrated the least relative importance compared to all

other predictors in Model 2. Unlike ability EI, self-report EI and mixed EI all demonstrated relative importance in the

presence of the FFM and cognitive ability. Self-report EI is the most dominant predictor in

Model 3, capturing 31.3% of the explained variance along with an R 2 contribution of .06.

The second most dominant predictor in Model 3 was extraversion (RW% = 23.2; R 2 = .04), and the least dominant predictor was cognitive ability (RW% = 1.4; R2 = .00).

Mixed EI is the most dominant predictor relative to the FFM and cognitive ability in Model

4, contributing 42.8% of the explained variance as well as a R 2 contribution of .09. The

second most dominant predictor was extraversion (RW% = 17.2; R2 = .04), and the least dominant predictor was cognitive ability (RW% = 1.5; R2 = .00). Mixed EI had more than twice the relative importance of the second most dominant predictor (i.e., extraversion).

With regard to organizational commitment, self-report EI and mixed EI demonstrated

impressive relative importance of 46.9% (R 2 = .12) and 44.2% (R2 = .11), respectively.

When the dependent variable was turnover intentions, self-report EI showed large relative

importance of 60.9% (R 2 = .09). Because of the large effects for both self-report and mixed

EI, we hold that Hypothesis 2 is supported, but note that there is scale-based moderation

with regard to ability measures.

Meta-analytic structural equation modelling

We performed meta-analytic structural equation modelling to test the hypotheses related

to mediation. Mediation would exist if the test were to show a significant indirect path. We

separated mixed EI from both ability EI and self-report EI when performing mediation

testing, because mixed EI has moderate and high multicollinearity with ability EI and self-

report EI, respectively. The presence of multicollinearity would inflate standard errors, reduce statistical power, cause the issues of bouncing betas, and produce uninterpretable

results (Schwab, 2005). We still kept ability EI and self-report EI together when testing

mediation, because the correlation between ability EI and self-report EI was just .12,

which did not cause multicollinearity issues.

190 Chao Miao et al.

T a b le

3 . H ie ra rc h ic a l m u lt ip le re g re ss io n a n d re la ti v e w e ig h t a n a ly se s fo r a ll th re e st re a m s o f E I in p re d ic ti n g JS , O C , a n d T I

D V = JS

M o d e l 1

M o d e l 2

M o d e l 3

M o d e l 4

b R W

R W

% b

R W

% b

R W

% b

R W

C o g n it iv e a b il it y

.0 7 ** *

.0 0 3

2 .0

.0 6 ** *

.0 0 3

1 .7

.0 6 ** *

.0 0 3

1 .4

.0 7 ** *

.0 0 3

1 .5

N e u ro ti c is m

�. 1 8 ** *

.0 4 6

3 0 .3

�. 1 8 ** *

.0 4 6

3 0 .0

�. 1 4 ** *

.0 3 6

1 9 .7

�. 0 7 ** *

.0 3 1

1 4 .8

E x tr a v e rs io n

.2 5 ** *

.0 5 1

3 3 .5

.2 5 ** *

.0 5 1

3 3 .3

.2 2 ** *

.0 4 3

2 3 .2

.1 6 ** *

.0 3 6

1 7 .2

O p e n n e ss

�. 1 6 ** *

.0 0 7

4 .8

�. 1 6 ** *

.0 0 7

4 .9

�. 1 9 ** *

.0 1 1

5 .7

�. 2 1 ** *

.0 1 4

6 .6

A g re e a b le n e ss

.0 0

.0 0 9

5 .9

�. 0 0

.0 0 8

5 .6

�. 0 0

.0 0 8

4 .1

�. 0 3 *

.0 0 7

3 .4

C o n sc ie n ti o u sn e ss

.1 4 ** *

.0 3 6

2 3 .4

.1 4 ** *

.0 3 5

2 3 .2

.0 9 ** *

.0 2 7

1 4 .6

.1 1 ** *

.0 2 8

1 3 .7

A b il it y E I

.0 2

.0 0 2

1 .3

S e lf -r e p o rt E I

.2 1 ** *

.0 5 8

3 1 .3

M ix e d E I

.3 2 ** *

.0 8 9

4 2 .8

R 2

.1 5 ** *

.1 8 ** *

.2 1 ** *

D R 2

.0 0

.0 3 ** *

.0 6 ** *

H a rm

o n ic m e a n N

6 ,6 8 1

5 ,5 8 9

6 ,0 1 1

6 ,5 4 1

D V = O C

M o d e l 1

M o d e l 2

M o d e l 3

b R W

R W

% b

R W

% b

R W

C o g n it iv e a b il it y

�. 1 5 ** *

.0 2 0

1 1 .7

�. 1 6 ** *

.0 2 2

8 .5

�. 1 4 ** *

.0 2 0

8 .1

N e u ro ti c is m

.0 1

.0 1 1

6 .4

.0 8 ** *

.0 0 9

3 .3

.1 4 ** *

.0 1 1

4 .7

E x tr a v e rs io n

.1 8 ** *

.0 4 6

2 6 .8

.1 3 ** *

.0 3 5

1 3 .7

.0 7 ** *

.0 3 1

1 2 .6

O p e n n e ss

.1 0 ** *

.0 1 8

1 0 .7

.0 4 **

.0 1 3

5 .0

.0 3 *

.0 1 3

5 .3

A g re e a b le n e ss

.1 0 ** *

.0 2 5

1 4 .8

.0 9 ** *

.0 2 2

8 .5

.0 6 ** *

.0 2 0

8 .2

C o n sc ie n ti o u sn e ss

.2 1 ** *

.0 5 0

2 9 .6

.1 3 ** *

.0 3 6

1 4 .1

.1 7 ** *

.0 4 2

1 6 .9

S e lf -r e p o rt E I

.3 5 ** *

.1 2 1

4 6 .9

M ix e d E I

.3 8 ** *

.1 0 8

4 4 .2

R 2

.1 7 ** *

.2 6 ** *

.2 5 ** *

D R 2

.0 9 ** *

.0 8 ** *

H a rm

o n ic m e a n N

4 ,0 2 2

4 ,1 0 7

4 ,3 2 3

C o n ti n u e d

Emotional intelligence and work attitudes 191

T a b le

3 . (C o n ti n u e d )

D V = T I

M o d e l 1

M o d e l 2

b R W

R W

% b

R W

C o g n it iv e a b il it y

.0 6 ** *

.0 0 4

5 .9

.0 7 ** *

.0 0 5

3 .4

N e u ro ti c is m

.2 0 ** *

.0 3 7

5 6 .6

.1 4 ** *

.0 2 7

1 9 .4

E x tr av e rs io n

�. 0 6 **

.0 0 5

7 .3

�. 0 1

.0 0 4

2 .6

O p e n n e ss

.0 7 ** *

.0 0 2

3 .4

.1 2 ** *

.0 0 6

4 .5

A g re e a b le n e ss

�. 0 3

.0 0 7

1 0 .1

�. 0 2

.0 0 6

3 .9

C o n sc ie n ti o u sn e ss

�. 0 4 †

.0 1 1

1 6 .7

.0 4 †

.0 0 7

5 .2

S e lf -r e p o rt E I

�. 3 2 ** *

.0 8 6

6 0 .9

R 2

.0 7 ** *

.1 4 ** *

D R 2

.0 8 ** *

H a rm

o n ic m e a n N

3 ,6 3 3

3 ,7 6 1

N o te s.

b , st a n d a rd iz e d re g re ss io n w e ig h ts ; D V , d e p e n d e n t v a ri a b le ; E I, e m o ti o n a l in te ll ig e n c e ; JS , jo b sa ti sf a c ti o n ; O C , o rg a n iz a ti o n a l c o m m it m e n t; T I, tu rn o v e r

in te n ti o n s; h a rm

o n ic m e a n N ,h a rm

o n ic m e a n sa m p le si ze ;R

W ,r e la ti v e w e ig h t; R W

% ,p e rc e n ta g e o f re la ti v e w e ig h t (c o m p u te d b y d iv id in g in d iv id u a lr e la ti v e w e ig h t

b y th e su m o f in d iv id u a lr e la ti v e w e ig h t a n d m u lt ip ly in g b y 1 0 0 ); R 2 ,m

u lt ip le c o rr e la ti o n s; D R 2 ,i n c re m e n ta lc h a n g e in R 2 .I t is n o te d fr o m T a b le S 1 1 (a ) th a t a t b iv a ri a te

le v e l, o p e n n e ss is n o t si g n ifi c a n tl y re la te d to

jo b sa ti sf a c ti o n (q̂

= .0 2 ) a n d a g re e a b le n e ss is si g n ifi c a n tl y a n d p o si ti v e ly re la te d to

jo b sa ti sf a c ti o n (q̂

= .1 7 ). H o w e v e r, a s

d e m o n st ra te d in T a b le

3 ,M

o d e l1

sh o w s th a t o p e n n e ss is si g n ifi c a n tl y a n d n e g a ti v e ly re la te d to

jo b sa ti sf a c ti o n (b

= �. 1 6 ), w h e re a s a g re e a b le n e ss is n o t si g n ifi c a n tl y

re la te d to

jo b sa ti sf a c ti o n (b

= .0 0 ). O u r re su lt s a re

c o n si st e n t w it h Ju d g e e t a l.’ s (2 0 0 2 ) m e ta -a n a ly ti c fi n d in g s, w h ic h a ls o sh o w e d so m e in c o n si st e n c ie s b e tw

e e n

b iv a ri a te

a n d m u lt iv a ri a te

re su lt s o f o p e n n e ss a n d a g re e a b le n e ss .A

c c o rd in g to

Ju d g e e t a l. (2 0 0 2 ), o p e n n e ss is d e sc ri b e d a s a ‘d o u b le -e d g e d sw

o rd ’t h a t m a y p ro m p t

in d iv id u a ls to

se n se

b o th

th e g o o d a n d th e b a d m o re

d e e p ly . A lt h o u g h a g re e a b le in d iv id u a ls sh o u ld

e x p e ri e n c e h ig h e r jo b sa ti sf a c ti o n d u e to

th e ir m o ti v a ti o n to

a c h ie v e in te rp e rs o n a l in ti m a c y , it m a y o n ly b e a g re a t p re d ic to r o f jo b sa ti sf a c ti o n in so c ia l o c c u p a ti o n s w h e re

tr a it a g re e a b le n e ss

is m o re

re le v a n t.

† p < .1 0 ; * p < .0 5 ; * * p < .0 1 ; * * * p < .0 0 1 .

192 Chao Miao et al.

We used meta-analytic structural equation modelling to compare a list of alternative

models (see Table S14 in supplemental materials). For Test 1 in Table S14, we assessed

how state affect and job performance mediate the relationships between ability EI, self-

report EI, and job satisfaction. We compared all the other models with Model 1 – a partial mediation model with direct paths from both ability EI and self-report EI to job satisfaction.

v2 difference test showed that the differences between all three alternative models and Model 1 are consistently not statistically significant, meaning that making the model more

parsimonious does not worsen model fit. We chose Model 4 (full mediation model)

because it is the most parsimonious one among all four models and it also fits the data very

well, v2(2) = 3.53 (p = .17), CFI = 1.00, NFI = 1.00, GFI = 1.00, SRMR = .01. We applied the same method for Test 2 in Table S14, where we assessed how SPA,

SNA, and job performance mediate the relationship between mixed EI and job satisfaction. Although v2 difference test showed that partial mediation model (Model 1) demonstrates better model fit than full mediation model (Model 2; Dv2[1] = 88.26), we still decided to choose full mediation model (Model 2) due to three reasons. First,

sample size greatly influences the v2 difference and our meta-analytic sample size was large (Kline, 2011). Therefore, even a negligible difference between models may still

have produced a statistically significant v2 statistic in the present study (Berry, Lelchook, & Clark, 2012). Second, partial mediation model (Model 1) is a saturated

model and we cannot derive any conclusion from this model. As such, Model 2, a non- saturated model, is more preferable relative to Model 1. Third, full mediation (Model 2)

not only displays acceptable model fit (CFI = .92, NFI = .92, GFI = .98, SRMR = .05), but is also more parsimonious than Model 1. Hence, we opted to choose Model 2 in

Test 2 due to the aforementioned reasons. Both chosen models based on the results of

model comparison were indicated with bold characters in Table S14.

Figure 1 presents the results of the examination of mediation, along with all

standardized path coefficients for all chosen models. Figure 1a corresponds to Model 4

under Test 1 in Table S14. Figure 1b corresponds to Model 2 under Test 2 in Table S14. With regard to Figure 1a, we assessed how SPA, SNA, and job performance mediated

the relationship between ability EI and self-report EI and job satisfaction. We performed

three sets of mediation tests – Sobel test, Aroian test, and Goodman test. For instance, the indirect paths from self-report EI to job satisfaction through SPA (b = .15) and SNA (b = .08) were statistically significant. Similarly, the indirect effect from self-report EI to job satisfaction through job performance (b = .04) was statistically significant as well. We repeated the same procedures for all the other models in Figure 1. We found that all

indirect paths were statistically significant. As such, all mediation hypotheses (Hypothe- ses 4–6; see Table 2 for specific hypotheses) are supported. The results of mediation examination are shown directly below each figure.

Publication bias analyses

We performed three different types of publication bias analyses and found no evidence of

publication bias inflating reported effect sizes (see supplemental materials for details).

Discussion

Emotion is an integral part of organizational life and is often functional for the

organization, and the proper management of emotions can lead to increased job

Emotional intelligence and work attitudes 193

satisfaction (Ashforth & Humphrey, 1995). We presented the first meta-analytic review of

the relationship between employee EI and employee job satisfaction and found a positive

and significant relationship between all three types of EI and job satisfaction. In addition,

EI is also positively related to organizational commitment and negatively related to

turnover intentions. Thus, emotionally savvy individuals are not only high-performing

(O’Boyle et al., 2011) but are also more satisfied with their jobs.

Theoretical implications

Although the relationship between EI and job satisfaction is positive and statistically

significant, the variation in effect sizes across studies is substantial (according to Hunter

and Schmidt’s 75% rule and Q statistic) for the relationships between all three types of EI

and job satisfaction. We found that the relationship between self-report EI and job

satisfaction is higher when emotional labour demand is high. This coincides with Joseph

and Newman’s (2010) findings, suggesting that when a job involves frequent customer/ interpersonal interaction (i.e., high emotional labour demand) it requires employees to

use their EI to regulate their emotions. However, emotional labour was not a moderator

for either ability EI or mixed EI. This may be because recent research suggests that

emotional labour is used in a wide variety of jobs (Humphrey et al., 2015). These mixed

findings warrant more research on the EI–emotional labour relationship. The pattern of results upholds the categorization of EI measures into three streams/

types (Ashkanasy & Daus, 2005; O’Boyle et al., 2011). Due to differential relationships

with cognitive ability and personality, we found that mixed EI has the highest association with employee job satisfaction (q̂ = .39), self-report EI the next highest (q̂ = .32), and ability EI the lowest relationship with employee job satisfaction (q̂ = .08). These results are consistent with our expectation because ability EI is more cognitively loaded and thus

should have the lowest relationship with job satisfaction, because cognitive ability is a

AEI: Mediation effect of SPA –.128[.026] × .311[.029] = –.04*** AEI: Sobel Test: –4.47; Aroian test = –4.46; Goodman test = –4.49 AEI: Mediation effect of SNA –.345[.025] × –.222[.028] = .08 *** AEI: Sobel Test: 6.87; Aroian test = 6.86; Goodman test = 6.89 AEI: Mediation effect of JP .207[.028] × .155[.027] = .03*** AEI: Sobel Test: 4.53; Aroian test = 4.51; Goodman test = 4.56 SEI: Mediation effect of SPA .485[.026] × .311[.029] = .15*** SEI: Sobel Test: 9.30; Aroian test = 9.29; Goodman test = 9.31 SEI: Mediation effect of SNA –.379[.025] × –.222 [.028] = .08 *** SEI: Sobel Test: 7.03; Aroian test = 7.01; Goodman test = 7.04 SEI: Mediation effect of JP .275[.028] × .155[.027] = .04*** SEI: Sobel Test: 4.96; Aroian test = 4.94; Goodman test = 4.98

Mediation effect of SPA .310[.026] × .311[.026] = .10*** Sobel Test: 8.44; Aroian test = 8.43; Goodman test = 8.46 Mediation effect of SNA –.240[.026] × –.222[.025] = .05 *** Sobel Test: 6.40; Aroian test = 6.38; Goodman test = 6.42 Mediation effect of JP .280[.026] × .155[.025] = .04*** Sobel Test: 5.37; Aroian test = 5.36; Goodman test = 5.39

AEI

SEI

SPA

SNA JS MEI

SPA

SNA JS

–.13***

JP JP

–.34***

.21***

.49***

–.38***

.28***

.31***

–.22***

.16***

.31***

–.24***

.28***

.31***

–.22***

.16***

(a) (b)

Figure 1. Path models of the mediating roles of state affect and job performance in the relationship

between emotional intelligence (EI) and job satisfaction. Note. Standardized path coefficients are

reported. Standard errors are reported in brackets. AEI, ability EI; SEI, self-report EI; MEI, mixed EI; JS, job

satisfaction; SPA, state positive affect; SNA, state negative affect; JP, job performance. (a) Model 4 under

Test 1 in Table S14. (b) Model 2 under Test 2 in Table S14. Fit indices for each model are reported in

Table S14. We omitted covariance for clarity of reporting. ***p < .001.

194 Chao Miao et al.

weak predictor of job satisfaction (Gonzalez-Mul�e et al., 2014). Mixed EI has the largest relationship with other personality traits and should thus have the strongest relationship

with job satisfaction because personality is a much better predictor of job satisfaction than

cognitive ability (Judge et al., 2002). Our results indicate that both self-report EI and mixed EI not only display incremental

validity above and beyond cognitive ability and the FFM, but that they also show large

relative importance (31.3% relative importance for self-report EI and 42.8% relative

importance for mixed EI) in the explained variance in job satisfaction. In particular, mixed

EI alone impressively accounts for nearly half of the explained variance in job satisfaction

compared to cognitive ability (1.5% relative importance) and the FFM (55.7% relative

importance for five personality traits as a whole set). We found similar effects for the

incremental validity and relative importance of EI with regard to organizational commitment and turnover intentions. These findings are consistent with – and add to prior meta-analytic findings on – how EI contributes relative importance with regard to job performance (O’Boyle et al., 2011).

Our study also explored the theoretical mechanisms through which EI influences job

satisfaction. Building on goal setting theory and self-perception theory (Bem, 1967; Locke,

1976), we found that the relationship between EI and job satisfaction is mediated by both

state affect and job performance. EI may be a characteristic that causes employees to see

both their work performance and their job in a rosy light, one which promotes positive affect. Employees high on EI may then observe their positive affect at work and deduce

that they have high job satisfaction. Building on goal setting theory and self-perception

theory (Bem, 1967; Locke, 1976), we weaved prior meta-analytic findings on EI–job performance relationships into our mediation model and found that job performance

mediates the relationship between EI and job satisfaction. This shows EI’s relevance to the

goal setting literature and indicates that EI helps employees to reach their performance

goals. Employees may then deduce their own level of job satisfaction from their levelof job

performance. These findings open multiple avenues for future research on EI, goals, and work criteria. Locke and Latham (2002) suggested a set of moderators (e.g., goal

importance, goal commitment, and task complexity), and future researchers may

consider developing models that include these additional moderators in order to derive a

more thorough picture of the interrelationships among EI, goals, and work criteria.

Limitations and future directions

First, there were a small number of samples for some of our meta-analytic distributions, which makes the results subject to second-order sampling error. For the same reason, we

were not able to analyse some moderators for some types of EI. Therefore, we encourage

readers to exercise caution when interpreting our results based on a small number of

samples, and we acknowledge that the results based on a small number of samples are

preliminary. This partly explains why the results of our moderator analyses are

inconsistent across three EI types. Moderator testing in meta-analysis is a low power

test (Steel & Kammeyer-Mueller, 2002). Therefore, if the number of samples across

different levels of moderators is small (ability EI distributions in particular), then the results of moderation can hardly be significant, which is why we identified some

inconsistencies in our results across three types of EI. We thus encourage readers to

interpret the results of moderator analyses based on a small number of samples with

caution.

Emotional intelligence and work attitudes 195

Second, the present meta-analytic review was dominated by the studies using cross-

sectional design. Future studies should use longitudinal designs and conduct advanced

analyses, such as latent growth modelling (Bliese & Ployhart, 2002), to draw robust causal

inferences. Third, at bivariate level, we found a significant moderator effect of the emotional

labour demand of jobs on the relationship between self-report EI and job satisfaction. We

suspect that this moderator may also function in our mediation model in such a way that

people under high emotional labour demands may have high job satisfaction with high EI

through affect or job performance. This moderated mediation model may help us better

integrate our variables. However, we cannot use meta-analysis to test this model because

moderated mediation models have to be tested based on raw data, whereas ours – like all other meta-analyses – is also based on correlation matrices without raw data. For this reason, we encourage future studies to collect primary data to assess the moderated

mediation model described here.

Practical implications

Job satisfaction, organizational commitment, and turnover intentions are important

attitudes related to many critical workplace outcomes, such as job performance, turnover,

profits, and psychological well-being. Our investigations provide insights and evidence regarding the importance of employees’ EI in determining employees’ work attitudes. To

produce satisfied and productive workers, organizations can incorporate EI in employee

education, training, and development (Walter et al., 2011).

Job satisfaction is a very important form of employee job attitude in organizations,

because job satisfaction is knownto improve physical and psychological health outcomes,

to be positively related to organizational commitment, organizational citizenship

behaviour, and task performance, and to be negatively related to turnover intention,

turnover, and withdrawal cognitions and behaviours (Schleicher et al., 2011). Impor- tantly, our research findings suggest a low-cost, yet effective, way to staff an organization

with satisfied employees, which is to hire emotionally intelligent people. Incorporating a

measure of EI during the selection process would help an organization to find satisfied

employees because emotionally intelligent employees are more satisfied, according to our

research findings. Nonetheless, hiring people high in EI does not mean that organizations

are free of their obligations to reduce workplace stress and strain and to improve overall

working conditions. Organizations with good values can increase employees’ organiza-

tional commitment and reduce turnover intentions (Abbott, White, & Charles, 2005). Equally importantly, organizations that are perceived to support their employees have

employees who are more committed (Loi, Hang-Yue, & Foley, 2006).

Although ability EI tests did not show incremental validity, they may still have

considerable practical importance. Their objective nature means that they are not

susceptible to test takers’ self-serving biases, so they may be useful when hiring new

employees, and also for giving feedback to current employees who are resistant to advice

from their peers (O’Boyle et al., 2011; Walter et al., 2011). For practitioners who care

little about the overlap between self-report and mixed EI and other psychological constructs,our results suggest that one should consider utilizing self-report/peer-report EI

measures, because the validity of both self-report EI and mixed EI in predicting job

satisfaction is much larger than that of ability EI. We also recommend the use of mixed EI as

a shorthand alternative to a lengthy battery of a few traditional personnel tests, because

mixed EI captures a compound of different constructs and demonstrates reasonable

196 Chao Miao et al.

criterion-related validity. Because self-report measures and mixed measures show

incremental validity over cognitive ability and personality measures, organizations that

have lengthy batteries of such measures can still increase their ability to predict job

satisfaction, organizational commitment, and turnover intentions by incorporating self- report and/or mixed EI measures.

Acknowledgements

We would like to thank Associate Editor Tim Munyon and the anonymous reviewers for their

knowledgeable advice and guidance. They helped with theory development, statistical

analysis, and presentation and writing. Tim’s advice was invaluable, and the article’s focus and

clarity was much improved with his help. In addition, Editor Sharon Clarke also made useful

suggestions that made the paper more focused and concise. Thanks to all for their help.

References

Abbott, G. N., White,F. A.,&Charles, M. A. (2005). Linking values and organizational commitment: A

correlational and experimental investigation in two organizations. Journal of Occupational and

Organizational Psychology, 78, 531–551. doi:10.1348/096317905X26174 Ashforth, B. E., & Humphrey, R. H. (1993). Emotional labor in service roles: The influence of identity.

Academy of Management Review, 18, 88–115. doi:10.5465/AMR.1993.3997508 Ashforth, B. E., & Humphrey, R. H. (1995). Emotion in the workplace: A reappraisal. Human

Relations, 48, 97–125. doi:10.1177/001872679504800201 Ashkanasy, N. M., & Daus, C. S. (2002). Emotion in the workplace: The new challenge for managers.

Academy of Management Executive, 16, 76–86. doi:10.5465/AME.2002.6640191 Ashkanasy,N. M., & Daus,C. S. (2005). Rumors ofthe death ofemotional intelligence in organizational

behavior are vastly exaggerated. Journal of Organizational Behavior, 26, 441–452. doi:10.1002/ job.320

Ashkanasy, N. M., & Humphrey, R. H. (2011). Current emotion research in organizational behavior.

Emotion Review, 3, 214–224. doi:10.1177/1754073910391684 Bakker, A. B., & Demerouti, E. (2007). The job demands–resources model: State of the art. Journal

of Managerial Psychology, 22, 309–328. doi:10.1108/02683940710733115 Bar-On, R. (2000). Emotions and social intelligence: Insights from the emotional quotient inventory.

In R. Bar-On & J. D. A. Parker (Eds.), The handbook of emotional intelligence: Theory,

development, assessment, and application at home, school, and in the workplace (pp. 363– 388). San Francisco, CA: Jossey-Bass.

Bem, D. J. (1967). Self-perception: An alternative interpretation of cognitive dissonance

phenomena. Psychological Review, 74, 183–200. doi:10.1037/h0024835 Berry, C. M., Lelchook, A. M., & Clark, M. A. (2012). A meta-analysis ofthe interrelationships between

employee lateness, absenteeism, and turnover: Implications for models of withdrawal behavior.

Journal of Organizational Behavior, 33, 678–699. doi:10.1002/job.778 Bliese, P. D., & Ployhart, R. E. (2002). Growth modeling using random coefficient models: Model

building, testing, and illustrations. Organizational Research Methods, 5,362–387. doi:10.1177/ 109442802237116

Boyatzis, R., Brizz, T., & Godwin, L. (2011). The effect of religious leaders’ emotional and social

competencies on improving parish vibrancy. Journal of Leadership & Organizational Studies,

18, 192–206. doi:10.1177/1548051810369676 Breaux, D. M., Munyon, T. P., Hochwarter, W. A., & Ferris, G. R. (2009). Politics as a moderator of the

accountability – Job satisfaction relationship: Evidence across three studies. Journal of Management, 35, 307–326. doi:10.1177/0149206308318621

Emotional intelligence and work attitudes 197

http://dx.doi.org/10.1348/096317905X26174

http://dx.doi.org/10.5465/AMR.1993.3997508

http://dx.doi.org/10.1177/001872679504800201

http://dx.doi.org/10.5465/AME.2002.6640191

http://dx.doi.org/10.1002/job.320

http://dx.doi.org/10.1177/1754073910391684

http://dx.doi.org/10.1108/02683940710733115

http://dx.doi.org/10.1037/h0024835

http://dx.doi.org/10.1002/job.778

http://dx.doi.org/10.1177/109442802237116

http://dx.doi.org/10.1177/1548051810369676

http://dx.doi.org/10.1177/0149206308318621

Byron, K. (2007). Male and female managers’ ability to read emotions: Relationships with

supervisors’ performance ratings and subordinates’ satisfaction ratings. Journal of Occupa-

tional and Organizational Psychology, 80, 713–733. doi:10.1348/096317907X174349 Colquitt, J. A., Scott, B. A., & LePine, J. A. (2007). Trust, trustworthiness, and trust propensity: A

meta-analytic test of their unique relationships with risk taking and job performance. Journal

of Applied Psychology, 92, 909–927. doi:10.1037/0021-9010.92.4.909 Côt�e, S. (2014). Emotional intelligence in organizations. Annual Review of Organizational

Psychology and Organizational Behavior, 1, 459–488. doi:10.1146/annurev-orgpsych- 031413-091233

Côt�e, S., & Miners, C. T. (2006). Emotional intelligence, cognitive intelligence, and job performance. Administrative Science Quarterly, 51, 1–28. doi:10.2189/asqu.51.1.1

Crawford, E. R., LePine, J. A., & Rich, B. L. (2010). Linking job demands and resources to employee

engagement and burnout: A theoretical extension and meta-analytic test. Journal of Applied

Psychology, 95, 834–848. doi:10.1037/a0019364 De Clercq, D., Bouckenooghe, D., Raja, U., & Matsyborska, G. (2014). Unpacking the goal

congruence–organizational deviance relationship: The roles of work engagement and emotional intelligence. Journal of Business Ethics, 124, 695–711. doi:10.1007/s10551-013-1902-0

Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. (1999). Subjective well-being: Three decades of

progress. Psychological Bulletin, 125, 276–302. doi:10.1037/0033-2909.125.2.276 Dong, Y., Seo, M. G., & Bartol, K. (2013). No pain, no gain: An affect-based model of developmental

job experience and the buffering effects of emotional intelligence. Academy of Management

Journal, 57, 1056–1077. doi:10.5465/amj.2011.0687 Fisher, C. D. (2000). Mood and emotions while working: Missing pieces of job satisfaction? Journal

of Organizational Behavior, 21, 185–202. doi:10.1002/(SICI)1099-1379(200003)21:2<185:: AID-JOB34>3.0.CO;2-M

Fox, S., & Spector, P. E. (2000). Relations of emotional intelligence, practical intelligence, general

intelligence, and trait affectivity with interview outcomes: It’s not all just ‘G’. Journal of

Organizational Behavior, 21, 203–220. doi:10.1002/(SICI)1099-1379(200003)21:2<203::AID- JOB38>3.0.CO;2-Z

Ganzach, Y. (1998). Intelligence and job satisfaction. Academy of Management Journal, 41, 526– 539. doi:10.2307/256940

Goleman, D. (1995). Emotional intelligence: Why it can matter more than IQ. New York, NY:

Bantam Books.

Gonzalez-Mul�e, E., Carter, K., & Mount, M. K. (2014). Is ignorance really bliss? A meta-analysis of the relationship between general mental ability and attitudes. Working Paper.

Gooty, J., Connelly, S., Griffith, J., & Gupta, A. (2010). Leadership, affect and emotions: A state of the

science review. The Leadership Quarterly, 21, 979–1004. doi:10.1016/j.leaqua.2010.10.005 Gooty, J., Gavin, M. B., Ashkanasy, N. M., & Thomas, J. S. (2014). The wisdom of letting go and

performance: The moderating role of emotional intelligence and discrete emotions. Journal of

Occupational and Organizational Psychology, 87, 392–413. doi:10.1111/joop.12053 Grandey, A. A. (2000). Emotion regulation in the workplace: A new way to conceptualize emotional

labor. Journal of Occupational Health Psychology, 5, 95–110. doi:10.1037/1076-8998.5.1.95 Hobfoll, S. E. (2001). The influence of culture, community, and the nested-self in the stress process:

Advancing conservation of resources theory. Applied Psychology: An International Review, 50,

337–370. doi:10.1111/1464-0597.00062 Hochschild, A. R. (1983). The managed heart: Commercialization of human feelings. Berkeley,

CA: University of California Press.

Hulsheger, U. R., & Schewe, A. F. (2011). On the costs and benefits of emotional labor: A meta-

analysis of three decades of research. Journal of Occupational Health Psychology, 16, 361–389. doi:10.1037/a0022876

Humphrey, R. H. (2002). The many faces of emotional leadership. The Leadership Quarterly, 13,

493–504. doi:10.1016/S1048-9843(02)00140-6

198 Chao Miao et al.

http://dx.doi.org/10.1348/096317907X174349

http://dx.doi.org/10.1037/0021-9010.92.4.909

http://dx.doi.org/10.1146/annurev-orgpsych-031413-091233

http://dx.doi.org/10.2189/asqu.51.1.1

http://dx.doi.org/10.1037/a0019364

http://dx.doi.org/10.1007/s10551-013-1902-0

http://dx.doi.org/10.1037/0033-2909.125.2.276

http://dx.doi.org/10.5465/amj.2011.0687

http://dx.doi.org/10.1002/(SICI)1099-1379(200003)21:2<185::AID-JOB34>3.0.CO;2-M

http://dx.doi.org/10.1002/(SICI)1099-1379(200003)21:2<203::AID-JOB38>3.0.CO;2-Z

http://dx.doi.org/10.2307/256940

http://dx.doi.org/10.1016/j.leaqua.2010.10.005

http://dx.doi.org/10.1111/joop.12053

http://dx.doi.org/10.1037/1076-8998.5.1.95

http://dx.doi.org/10.1111/1464-0597.00062

http://dx.doi.org/10.1037/a0022876

http://dx.doi.org/10.1016/S1048-9843(02)00140-6

Humphrey, R. H. (2013). Effective leadership: Theory, cases, and applications. Los Angeles, CA:

Sage.

Humphrey, R. H., Ashforth, B. E., & Diefendorff, J. M. (2015). The bright side of emotional labor.

Journal of Organizational Behavior, 36, 749–769. doi:10.1002/job.2019 Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting error and bias in

research findings. Beverly Hills, CA: Sage.

Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analysis: Correcting error and bias in

research findings. Newbury Park, CA: Sage.

Johnson, J. W. (2000). A heuristic method for estimating the relative weight of predictor variables

in multiple regression. Multivariate Behavioral Research, 35, 1–19. doi:10.1207/S15327906 MBR3501_1

Johnson, S. K. (2009). Do you feel what I feel? Mood contagion and leadership outcomes. The

Leadership Quarterly, 20, 814–827. doi:10.1016/j.leaqua.2009.06.012 Johnson, J. W., & LeBreton, J. M. (2004). History and use of relative importance indices in

organizational research. Organizational Research Methods, 7, 238–257. doi:10.1177/1094428 104266510

Johnson, H. A. M., & Spector, P. E. (2007). Service with a smile: Do emotional intelligence, gender,

and autonomy moderate the emotional labor process? Journal of Occupational Health

Psychology, 12, 319. doi:10.1037/1076-8998.12.4.319

Jordan, P. J., Ashkanasy, N. M., Hartel, C. E. J., & Hooper, G. S. (2002). Workgroup emotional

intelligence: Scale development and relationship to team process effectiveness and goal focus.

Human Resource Management Review, 12, 195–214. doi:10.1016/S1053-4822(02)00046-3 Jordan, P. J., & Troth, A. (2011). Emotional intelligence and leader member exchange: The

relationship with employee turnover intentions and job satisfaction. Leadership &

Organization Development Journal, 32, 260–280. doi:10.1108/01437731111123915 Joseph, D. L., & Newman, D. A. (2010). Emotional intelligence: An integrative meta-analysis and

cascading model. Journal of Applied Psychology, 95, 54–78. doi:10.1037/a0017286 Judge, T. A., Heller, D., & Mount, M. K. (2002). Five-factor model ofpersonality and job satisfaction: A

meta-analysis. Journal of Applied Psychology, 87, 530–541. doi:10.1037/0021-9010.87.3.530 Judge, T. A., & Kammeyer-Muellar, J. D. (2008). Affect, satisfaction, and performance. In N. M.

Ashkanasy & C. L. Cooper (Eds.), Research companion to emotion in organizations (pp. 136– 151). Cheltenham, UK: Edward Elgar.

Judge, T. A., & Kammeyer-Mueller, J. D. (2012). Job attitudes. Annual Review of Psychology, 63,

341–367. doi:10.1146/annurev-psych-120710-100511 Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K. (2001). The job satisfaction–job performance

relationship: A qualitative and quantitative review. Psychological Bulletin, 127, 376–407. doi:10.1037/0033-2909.127.3.376

Kafetsios, K., & Zampetakis, L. A. (2008). Emotional intelligence and job satisfaction: Testing the

mediatory role of positive and negative affect at work. Personality and Individual Differences,

44, 712–722. doi:10.1016/j.paid.2007.10.004 Karim, J. (2009). Emotional intelligence and psychological distress: Testing the mediatory role of

affectivity. Europe’s Journal of Psychology, 5, 20–39. doi:10.5964/ejop.v5i4.238 Kellett, J. B., Humphrey, R. H., & Sleeth, R. G. (2006). Empathy and the emergence of task and

relations leaders. The Leadership Quarterly, 17, 146–162. doi:10.1016/j.leaqua.2005.12.003 Kline, R. B. (2011). Principles and practice of structural equation modeling. New York, NY:

Guilford.

Kluemper, D. H., DeGroot, T., & Choi, S. (2013). Emotion management ability: Predicting task

performance, citizenship, and deviance. Journal of Management, 39, 878–905. doi:10.1177/ 0149206311407326

Kong, F., & Zhao, J. (2013). Affective mediators of the relationship between trait emotional

intelligence and life satisfaction in young adults. Personality and Individual Differences, 54,

197–201. doi:10.1016/j.paid.2012.08.028 Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.

Emotional intelligence and work attitudes 199

http://dx.doi.org/10.1002/job.2019

http://dx.doi.org/10.1207/S15327906MBR3501_1

http://dx.doi.org/10.1016/j.leaqua.2009.06.012

http://dx.doi.org/10.1177/1094428104266510

http://dx.doi.org/10.1037/1076-8998.12.4.319

http://dx.doi.org/10.1016/S1053-4822(02)00046-3

http://dx.doi.org/10.1108/01437731111123915

http://dx.doi.org/10.1037/a0017286

http://dx.doi.org/10.1037/0021-9010.87.3.530

http://dx.doi.org/10.1146/annurev-psych-120710-100511

http://dx.doi.org/10.1037/0033-2909.127.3.376

http://dx.doi.org/10.1016/j.paid.2007.10.004

http://dx.doi.org/10.5964/ejop.v5i4.238

http://dx.doi.org/10.1016/j.leaqua.2005.12.003

http://dx.doi.org/10.1177/0149206311407326

http://dx.doi.org/10.1016/j.paid.2012.08.028

Locke, E. A. (1969). What is job satisfaction? Organizational Behavior and Human Performance,

4, 309–336. doi:10.1016/0030-5073(69)90013-0 Locke, E. A. (1976). The nature and causes of job satisfaction. In M. D. Dunnette (Ed.), Handbook of

industrial and organizational psychology (pp. 1297–1349). Chicago, IL: Rand-McNally. Locke, E. A., Cartledge, N., & Knerr, C. S. (1970). Studies of the relationship between satisfaction,

goal-setting, and performance. Organizational Behavior and Human Performance, 5, 135– 158. doi:10.1016/0030-5073(70)90011-5

Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task

motivation: A 35-year odyssey. American Psychologist, 57, 705–717. doi:10.1037/0003-066X. 57.9.705

Loi, R., Hang-Yue, N., & Foley, S. (2006). Linking employees’ justice perceptions to organizational

commitment and intention to leave: The mediating role of perceived organizational support.

Journal of Occupational and Organizational Psychology, 79, 101–120. doi:10.1348/ 096317905X39657

Martins, A., Ramalho, N., & Morin, E. (2010). A comprehensive meta-analysis of the relationship

between emotional intelligence and health. Personality and Individual Differences, 49, 554– 564. doi:10.1016/j.paid.2010.05.029

Mayer, J. D., Roberts, R. D., & Barsade, S. G. (2008). Human abilities: Emotional intelligence. Annual

Review of Psychology, 59, 507–536. doi:10.1146/annurev.psych.59.103006.093646 Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & D. J. Sluyter (Eds.),

Emotional development and emotional intelligence: Educational implications (pp. 3–25). New York, NY: Basic Books.

Mayer, J. D., Salovey, P., Caruso, D., & Sitarenios, G. (2003). Measuring emotional intelligence with

the MSCEIT V2.0. Emotion, 3, 97–105. doi:10.1037/1528-3542.3.1.97 Mitchell, T. R., Holtom, B. C., Lee, T. W., Sablynski, C. J., & Erez, M. (2001). Why people stay: Using

job embeddedness to predict voluntary turnover. Academy of Management Journal, 44, 1102– 1121. doi:10.2307/3069391

Mitchell, T. R., & Lee, T. W. (2001). 5. The unfolding model of voluntary turnover and job

embeddedness: Foundations for a comprehensive theory of attachment. Research in

Organizational Behavior, 23, 189–246. doi:10.1016/S0191-3085(01)23006-8 Munyon, T. P., Hochwarter, W. A., Perrew�e, P. L., & Ferris, G. R. (2010). Optimism and the nonlinear

citizenship behavior – Jobsatisfaction relationship in three studies. Journal of Management, 36, 1505–1528. doi:10.1177/0149206309350085

O’Boyle, E. H., Humphrey, R. H., Pollack, J. M., Hawver, T. H., & Story, P. A. (2011). The relation

between emotional intelligence and job performance: A meta-analysis. Journal of

Organizational Behavior, 32, 788–818. doi:10.1002/job.714 Peterson, R. A., & Brown, S. P. (2005). On the use of beta coefficients in meta-analysis. Journal of

Applied Psychology, 90, 175–181. doi:10.1037/0021-9010.90.1.175 Petrides, K. V. (2009a). Psychometric properties of the Trait Emotional Intelligence Questionnaire

(TEIQue). In C. Stough, D. H. Saklofske & J. D. A. Parker (Eds.), Assessing emotional intelligence:

Theory, research, and applications (pp. 85–101). New York, NY: Springer Science. Petrides, K. V. (2009b). Technical manual for the Trait Emotional Intelligence Questionnaire

(TEIQue). London, UK: London Psychometric Laboratory.

Petrides, K. V., & Furnham, A. (2003). Trait emotional intelligence: Behavioral validation in two

studies of emotion recognition and reactivity to mood induction. European Journal of

Personality, 17, 39–57. doi:10.1002/per.466 Schleicher, D. J., Hansen, S. D., & Fox, K. E. (2011). Job attitudes and work values. In S. Zedeck (Ed.),

APA handbook of industrial and organizational psychology, vol. 3: Maintaining, expanding,

and contracting the organization. APA handbooks in psychology (pp. 137–189). Washington, DC: American Psychological Association.

Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., Cooper, J. T., Golden, C. J., & Dornheim, L.

(1998). Development and validation of a measure of emotional intelligence. Personality and

Individual Differences, 25, 167–177. doi:10.1016/S0191-8869(98)00001-4

200 Chao Miao et al.

http://dx.doi.org/10.1016/0030-5073(69)90013-0

http://dx.doi.org/10.1016/0030-5073(70)90011-5

http://dx.doi.org/10.1037/0003-066X.57.9.705

http://dx.doi.org/10.1348/096317905X39657

http://dx.doi.org/10.1016/j.paid.2010.05.029

http://dx.doi.org/10.1146/annurev.psych.59.103006.093646

http://dx.doi.org/10.1037/1528-3542.3.1.97

http://dx.doi.org/10.2307/3069391

http://dx.doi.org/10.1016/S0191-3085(01)23006-8

http://dx.doi.org/10.1177/0149206309350085

http://dx.doi.org/10.1002/job.714

http://dx.doi.org/10.1037/0021-9010.90.1.175

http://dx.doi.org/10.1002/per.466

http://dx.doi.org/10.1016/S0191-8869(98)00001-4

Schutte, N. S., Malouff, J. M., Thorsteinsson, E. B., Bhullar, N., & Rooke, S. E. (2007). A meta-analytic

investigation of the relationship between emotional intelligence and health. Personality and

Individual Differences, 42, 921–933. doi:10.1016/j.paid.2006.09.003 Schwab, D. P. (2005). Research methods for organizational studies. New York, NY: Psychology

Press.

Spence, G., Oades, L. G., & Caputi, P. (2004). Trait emotional intelligence and goal self-integration:

Important predictors of emotional well-being? Personality and Individual Differences, 37,

449–461. doi:10.1016/j.paid.2003.09.001 Staw, B. M., Bell, N. E., & Clausen, J. A. (1986). The dispositional approach to job attitudes: A lifetime

longitudinal test. Administrative Science Quarterly, 31, 56–77. doi:10.2307/2392766 Steel, P. D., & Kammeyer-Mueller, J. D. (2002). Comparing meta-analytic moderator estimation

techniques under realistic conditions. Journal of Applied Psychology, 87, 96–111. doi:10.1037/ 0021-9010.87.1.96

Sy, T., Tram, S., & O’Hara, L. A. (2006). Relation of employee and manager emotional intelligence to

job satisfaction and performance. Journal of Vocational Behavior, 68, 461–473. doi:10.1016/ j.jvb.2005.10.003

Tett, R. P., & Burnett, D. D. (2003). A personality trait-based interactionist model of job performance.

Journal of Applied Psychology, 88, 500–517. doi:10.1037/0021-9010.88.3.500 Tett, R. P., & Guterman, H. A. (2000). Situation trait relevance, trait expression, and cross-situational

consistency: Testing a principle of trait activation. Journal of Research in Personality, 34, 397– 423. doi:10.1006/jrpe.2000.2292

Thoresen, C. J., Kaplan, S. A., Barsky, A. P., Warren, C. R., & de Chermont, K. (2003). The affective

underpinnings of job perceptions and attitudes: A meta-analytic review and integration.

Psychological Bulletin, 129, 914–945. doi:10.1037/0033-2909.129.6.914 Van Rooy, D. L., & Viswesvaran, C. (2004). Emotional intelligence: A meta-analytic investiga-

tion of predictive validity and nomological net. Journal of Vocational Behavior, 65, 71–95. doi:10.1016/S0001-8791(03)00076-9

Viswesvaran, C., & Ones, D. S. (1995). Theory testing: Combining psychometric meta-analysis and

structural equations modeling. Personnel Psychology, 48, 865–885. doi:10.1111/j.1744-6570. 1995.tb01784.x

Walter, F., & Bruch, H. (2009). An affective events model of charismatic leadership behavior: A

review, theoretical integration, and research agenda. Journal of Management, 35, 1428– 1452. doi:10.1177/0149206309342468

Walter, F., Cole, M. S., & Humphrey, R. H. (2011). Emotional intelligence: Sine qua non of leadership

or folderol? The Academy of Management Perspectives, 25, 45–59. doi:10.5465/AMP.2011. 59198449

Wang, G., Seibert, S. E., & Boles, T. L. (2011). Synthesizing what we know and looking ahead: A meta-

analytical review of 30 years of emotional labor research. Research on Emotion in

Organizations, 7, 15–43. doi:10.1108/S1746-9791(2011)0000007006 Watson, D. (2000). Mood and temperament. New York, NY: Guilford Press.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of

positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology,

54, 1063–1070. doi:10.1037/0022-3514.54.6.1063 Weiss, H. M. (2002). Deconstructing job satisfaction: Separating evaluations, beliefs and affective

experiences. Human Resource Management Review, 12, 173–194. doi:10.1016/S1053-4822 (02)00045-1

Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion of the

structure, causes, and consequences of affective experiences at work. In B. M. Staw & L. L.

Cummings (Eds.), Research in organizational behavior (pp. 1–74). Greenwich, CT: JAI Press. Weiss, H. M., Nicholas, J. P., & Daus, C. S. (1999). An examination of the joint effects of affective

experiences and job beliefs on job satisfaction and variations in affective experiences over time.

Organizational Behavior and Human Decision Processes, 78, 1–24. doi:10.1006/obhd.1999. 2824

Emotional intelligence and work attitudes 201

http://dx.doi.org/10.1016/j.paid.2006.09.003

http://dx.doi.org/10.1016/j.paid.2003.09.001

http://dx.doi.org/10.2307/2392766

http://dx.doi.org/10.1037/0021-9010.87.1.96

http://dx.doi.org/10.1016/j.jvb.2005.10.003

http://dx.doi.org/10.1037/0021-9010.88.3.500

http://dx.doi.org/10.1006/jrpe.2000.2292

http://dx.doi.org/10.1037/0033-2909.129.6.914

http://dx.doi.org/10.1016/S0001-8791(03)00076-9

http://dx.doi.org/10.1111/j.1744-6570.1995.tb01784.x

http://dx.doi.org/10.1177/0149206309342468

http://dx.doi.org/10.5465/AMP.2011.59198449

http://dx.doi.org/10.1108/S1746-9791(2011)0000007006

http://dx.doi.org/10.1037/0022-3514.54.6.1063

http://dx.doi.org/10.1016/S1053-4822(02)00045-1

http://dx.doi.org/10.1006/obhd.1999.2824

Whitener, E. M. (1990). Confusion of confidence intervals and credibility intervals in meta-analysis.

Journal of Applied Psychology, 75, 315–321. doi:10.1037/0021-9010.75.3.315 Winkel, D. E., Wyland, R. L., Shaffer, M. A., & Clason, P. (2011). A new perspective on psychological

resources: Unanticipated consequences of impulsivity and emotional intelligence. Journal of

Occupational and Organizational Psychology, 84, 78–94. doi:10.1348/2044-8325.002001 Wong, C.-S., & Law, K. S. (2002). The effects of leader and follower emotional intelligence on

performance and attitude: An exploratory study. The Leadership Quarterly, 13, 243–274. doi:10.1016/S1048-9843(02)00099-1

Received 13 October 2015; revised version received 26 November 2016

Supporting Information

The following supporting information may be found in the online edition of the article:

Table S1. Dimensions/facets represented by three types of EI. Table S2. Job satisfaction measures. Table S3. Main codes for the studies included in the EI – JS meta-analysis. Table S4. Main codes for the studies included in the EI – OC meta-analysis. Table S5. Main codes for the studies included in the EI – TI meta-analysis. Table S6. Main codes for the studies included in the MSCEIT dimensions - JS meta- analysis.

Table S7. Main codes for the studies included in the WLEIS dimensions – JS meta- analysis.

Table S8. Main codes for the studies included in the EQ-i Dimensions – JS meta- analysis.

Table S9. Main codes for the studies included in the EI - job satisfaction dimensions meta-analysis.

Table S10. Main codes for the studies included in the EI – state affect meta-analysis. Table S11. (a) Meta-analytically derived corrected intercorrelation matrix for hierarchical multiple regression and relative weight analyses for predicting job

satisfaction. (b) Meta-analytically derived corrected intercorrelation matrix for

hierarchical multiple regression and relative weight analyses for predicting organiza- tional commitment. (c) Meta-analytically derived corrected intercorrelation matrix for

hierarchical multiple regression and relative weight analyses for predicting turnover

intention.

Table S12. Meta-analytically derived corrected intercorrelation matrix for Figure 1 (a) and (b).

Table S13. Supplemental psychometric meta-analysis results. Table S14. Comparison of the fit of the alternative models. Appendix S1. Supplemental notes.

202 Chao Miao et al.

http://dx.doi.org/10.1037/0021-9010.75.3.315

http://dx.doi.org/10.1348/2044-8325.002001

http://dx.doi.org/10.1016/S1048-9843(02)00099-1

Copyright of Journal of Occupational & Organizational Psychology is the property of Wiley- Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.