Emotional intelligence cognitive intelligence and job performance pdf

The primary objective of this study was to explore the EI among nurses who had direct contact with inpatients diagnosed with novel COVID The secondary objective was to explore the effects of EI on JP when dealing with such a crisis. The results of this study suggest that a high level of EI was typical among critical care unit nurses, nurses in intensive care units, neonatal intensive care units, then general nurses, and finally, nurses working in respiratory therapy with the lowest levels.

The study showed moderate relations between EI and JP, and the latter could be affected by how some nurses perceive their emotions. While Birks et al found similar results concerning levels of EI among various healthcare providers, the present study reported the highest average level of EI among critical care unit nurses. This suggests that high EI may be a prerequisite for finding work in nursing, and that it may exist in many candidates prior to recruitment.

In previous major studies, this impression was reinforced during interviews with participants. Gradually, this reliance on EI may increase their awareness of their own EI-related behaviors, allowing them to actively take part in ensuring the delivery of high-quality interdisciplinary care.

Hospital staff, particularly nurses, reflect innately skilled communication of empathy. But if the importance of EI to the practice of nursing is obvious, other aspects of the situation are still likely to be daunting. The present results attempt to induce basic conclusions regarding the sources and attainment of empathy or other components of EI, or at least how EI is affected by intrinsic and extrinsic factors. The current results are, however, consistent with those of Nightingale et al, who reported that high EI in nurses was associated with both corporeal and emotional caring; administrators and doctors may share such essence of EI, even if they have lower levels of EI.

The interpretation of a minor decrease in performance while having a slightly high EI is attributed to the nature of the catastrophe of COVID, where many patients are almost at the edge of death.

Perhaps coping with clinical situations and enforcing tasks in helpful manners may increase EI, but still, there is a point where nurses can no longer adhere to their emotions. Again, it is well documented that EI is a predictor of JP, as cited in many research studies, but with some variations. The present results are consistent with those reported by Trivellas et al and Mo et al, who found a strong prediction of JP during crisis.

However, in some fields, EI scores were linked negatively with measures such as stress despite their positive link to JP. For instance, Akhtar et al found that EI displayed a strong positive relationship with JP, which was amplified when organizational support for EI training increased considerably.

The significance of this is that the present results are consistent with other investigations, many of which employed large sample sizes and demonstrated positive relations between EI and JP among their participants. This study has some limitations, including the small sample size as well as the ratio of males to females which represented almost one-third.

To facilitate the effective implications of EI in health organizations, it may be helpful in the future to investigate the internal factors that predict the levels of EI during the COVID crisis, including other healthcare providers and patients in different stages. One way to do so would be to develop reliable and consistent means of key performance indicators that are related to various EI levels among nurses and other healthcare professionals, including communication skills and burnout domains.

These assessments could then be administered routinely, in the form of a checklist for instance, to the benefit of both the nursing and the health institution. Moreover, it would be advantageous for researchers to examine the impact of hospital leadership on EI and JP, as well as explore the relationships between EI, burnout, and employee turnover among nurses during such crisis.

In general, nurses have high EI, especially among critical care medically oriented professionals. Results should be interpreted with caution. These findings may suggest that constructing an empathic environment among healthcare providers may yield significant benefits in the form of increased employee performance and an increase in operational best practices in patient-centered institutions.

It is recommended that EI may be treated as a valuable asset among nurses, and that it could be incorporated into healthcare practice guidelines and performance evaluations. There was no written consent, as replying to the assigned online survey implied agreement to participate, and the committee approved this procedure. The author certifies that there are no conflicts of interest with any financial organization or anything else regarding the material discussed in the manuscript.

National Center for Biotechnology Information , U. Psychol Res Behav Manag. Published online Sep Wadi B Alonazi 1. Author information Article notes Copyright and License information Disclaimer. Received May 22; Accepted Sep 8. This work is published and licensed by Dove Medical Press Limited. By accessing the work you hereby accept the Terms.

Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed. For permission for commercial use of this work, please see paragraphs 4. This article has been cited by other articles in PMC.

Abstract Background During a pandemic, healthcare professionals encounter various health hazards that affect their personal life and workplace. Open in a separate window. Figure 1. Methods In a quantitative study, data were collected from three large tertiary hospitals in Saudi Arabia. Participants Each one of the three settings has employed an average of skillful nurses. Measure The online survey itself was designed with the goal of investigating whether any relationships existed between EI levels independent variable and levels of JP dependent variables.

Results Out of health providers who received the survey by email, only provided valid responses that were used in the present study, yielding an Instrument Adaptation Factor analysis is a data reduction technique used to determine whether multiple items are measuring the same underlying cause. Notes: r, 0. Abbreviation: CI, confidence interval for B.

Discussion The primary objective of this study was to explore the EI among nurses who had direct contact with inpatients diagnosed with novel COVID Limitations and Implications This study has some limitations, including the small sample size as well as the ratio of males to females which represented almost one-third.

Disclosure The author certifies that there are no conflicts of interest with any financial organization or anything else regarding the material discussed in the manuscript. References 1. Keeping positive and building strength: the role of affect and team leadership in developing resilience during an organizational crisis. Group Organ Manag. The impact of emotional intelligence in health care professionals on caring behaviour towards patients in clinical and long-term care settings: findings from an integrative review.

Int J Nurs Stud. Iran J Manag Stud. Elsevier; Career indecision versus indecisiveness: associations with personality traits and emotional intelligence. J Career Assess. Coetzee M, Harry N.

J Vocat Behav. Emotional intelligence of nursing applicants and factors related to it: a cross-sectional study. Nurse Educ Today. Emotional intelligence and perceived stress in healthcare students: a multi-institutional, multi-professional survey.

BMC Med Educ. Professional tears: developing emotional intelligence around death and dying in emergency work. J Clin Nurs. Convergent and divergent validity of integrative versus mixed model measures of emotional intelligence. Benefits of mindfulness at work: the role of mindfulness in emotion regulation, emotional exhaustion, and job satisfaction. J Appl Psychol. Do we develop public health leaders? The impact of emotional intelligence on nursing: an overview.

Emotional intelligence assessment in a graduate entry medical school curriculum. Aftab HA. Ram P. Relationship between job satisfaction and job performance in the public sector-a case study from India. Factors affecting performance of hospital nurses in Riyadh Region, Saudi Arabia. A multilevel moderated mediational model of the daily relationships between hassles, exhaustion, ego-resiliency and resulting emotional inertia.

J Res Pers. J Aging Health. The effects of leader and follower emotional intelligence on performance and attitude: an exploratory study. Leadersh Q. The impact of emotional intelligent leadership on staff nurse empowerment: the moderating effect of span of control. J Nurs Manag. Freshman B, Rubino L. Emotional intelligence: a core competency for health care administrators.

Health Care Manag. Emotional intelligence and authentic leadership among Saudi nursing leaders in the Kingdom of Saudi Arabia. J Prof Nurs. Effects of job conditions, occupational stress, and emotional intelligence on chronic fatigue among Chinese nurses: a cross-sectional study. Coleman D. Emotional Intelligence. The impact of emotional intelligence on job outcomes and turnover intention in health care.

Measured emotional intelligence in baccalaureate nursing education: a longitudinal study. Nurs Educ Perspect. J Travel Med. We disagree with Schmidt and Hunter because their definition of intelligence can be applied to specific content domains such as the domain of emotion and, therefore, intelligence does not always correspond to general intelligence. Hence, general intel- ligence is the general ability to reason correctly with abstrac- tions concepts and solve problems.

Emotional intelligence can be conceptualized as the ability to grasp and reason cor- rectly with emotional abstractions emotional concepts and solve emotional problems. Second, the construct of emotional intelligence meets the conceptual, correlational, and developmental criteria of an intelligence proposed by Mayer, Caruso, and Salovey , based on their review of the research on intelligence.

This criterion is based on an important distinction between abilities and personality traits. As such, abilities represent what a person can do in specific situations—situations in which conditions are favorable.

By contrast, personality traits represent what a person typically does across situations and over time McCrae and John, Mixed models of emotional intelligence include abilities, per- sonality traits, and various other concepts Bar-On, ; Tett, Fox, and Wang, Mixed models include concepts that are outside the defini- tion of abilities and hence do not meet the conceptual criteri- on of an intelligence.

To meet the correlational criterion of an intelligence Carroll, ; Neisser et al. Correlations between emotional intelligence and cognitive and verbal intel- ligence have been found to range from. These findings suggest that as much as 20 percent of emo- tional intelligence overlaps with other intelligences and, therefore, over 80 percent of emotional intelligence is sepa- rate from other intelligences.

These findings reveal that emo- tional intelligence meets the correlational criterion of an intel- ligence.

To meet the developmental criterion of an intelligence, a con- struct must have the potential to improve over time. Empiri- cal evidence accords an important role to experience and learning in the development of emotional intelligence.

There is also evidence that individuals can be trained to rec- ognize emotions in photographs of facial expressions Elfen- bein, and to use better strategies to manage their emotions Totterdell and Parkinson, Normative data showing that older individuals obtain higher emotional intelli- gence test scores than younger individuals Mayer et al.

Emotional Intelligence and Cognitive Intelligence Emotional intelligence and cognitive intelligence are related yet distinct constructs in our compensatory model. In the dominant model of mental abilities, the different abilities are structured hierarchically Carroll, General intelligence sub- sumes several sets of abilities that represent specializations of general intelligence into broad content or process areas in ways that reflect experience and learning Carroll, We conceptualize emotional intelligence and cognitive intelli- gence as separate broad sets of abilities that are subsumed under general intelligence in the hierarchical model.

Emotion- al intelligence represents the specialization of general intelli- gence in the area of emotions in ways that reflect experience and learning about emotions. Cognitive intelli- gence represents the specialization of general intelligence in the domain of cognition in ways that reflect experience and learning about cognitive processes such as memory Schaie, ; Brody, Emotional intelligence and cognitive intelligence should be positively associated because they are both subsumed under general intelligence.

The research reviewed above reveals that people with high cognitive intelligence tend to have high emotional intelligence and that people with low cognitive intelligence tend to have low emotional intelligence.

Emotion- al intelligence and cognitive intelligence are separate con- structs, however, because they represent the specialization of general intelligence in separate content domains. The influ- ence of the familial environment explains, in part, why emo- tional intelligence and cognitive intelligence do not corre- spond perfectly. Individuals raised in familial environments that are favorable to the develop- ment of emotional abilities may thus have high emotional intelligence despite having low cognitive intelligence.

Con- versely, individuals raised in emotionally impoverished familial environments may have low emotional intelligence despite having high cognitive intelligence.

Emotional Intelligence, Cognitive Intelligence, and Job Performance Cognitive intelligence is positively related to the dimensions of job performance—task performance and organizational citi- zenship behavior OCB —in most, if not all jobs Motowidlo and Van Scotter, ; Schmidt and Hunter, ; Chan and Schmitt, Task performance concerns the core substan- tive duties that are formally recognized as part of a job, and OCB concerns activities that contribute to the achievement of the objectives of an organization but that are not necessar- ily formally recognized as part of a job Organ, ; Borman and Motowidlo, ; Rotundo and Sackett, Cognitive intelligence enhances OCB through the knowledge of facts, procedures, and rules relevant to effec- tive helping, cooperating, and endorsing the organization Motowidlo, Borman, and Schmit, Individuals with low cognitive intelligence may reap relatively large returns from high emotional intelligence because they tend to exhibit low job performance in most, if not all jobs.

When job performance is low, the room for correction and improvement is large. For example, a salesperson who fails to hold the interest of potential customers has high potential for reducing mistakes in the future.

Job performance that is not attained through cognitive intelligence may be attained through emotional intelligence via multiple complementary mechanisms. The first mechanism concerns expertise at identifying and understanding the emotions of other individu- als. In most, if not all jobs, organization members interact with supervisors, coworkers, support staff, and outsiders such as customers, clients, or patients. These individuals publicly display their emotions through facial, vocal, and bodi- ly signals that provide important information about their goals, attitudes, and intentions Rafaeli and Sutton, ; Sutton, This information may, in turn, be converted into high task performance by individuals with high emotional intelligence and low cognitive intelligence.

A second mechanism by which emotional intelligence may enhance the job performance of individuals with low cogni- tive intelligence concerns how regulating emotion influences the quality of social relationships. Employees who generate and display genuine emotions elicit more favorable reactions than employees who choose to display fake emotions Grandey, ; Grandey et al.

Individuals with high emotional intelli- gence and low cognitive intelligence may employ their abili- ties to manage emotions to develop good social relationships Wong and Law, that may in turn enhance task perfor- mance via advice and social support Sparrowe et al. Good social relationships may also compel employees to engage in OCB frequently to benefit close colleagues. Emotionally intelligent individuals with low cognitive intelligence may achieve high levels of task performance and OCB in most, if not all jobs by managing their emotions in ways that enhance their moti- vation and the quality of their decisions Law, Wong, and Song, A manager who understands that anger tends to lead people to underestimate the degree of risk in situa- tions Lerner and Keltner, may suppress anger before making an important financial decision and, in turn, exhibit good task performance.

In addition, an organization member who understands that motivation is often enhanced by posi- tive emotions Erez and Isen, and successfully boosts positive emotions may exert more effort to engage in OCB. The preceding discussion suggests that emotional intelli- gence may positively relate to the job performance of organi- zation members with low cognitive intelligence and, as such, compensate for low cognitive intelligence.

Emotional intelli- gence, however, should become less positively associated with job performance as cognitive intelligence increases. Indi- viduals with high cognitive intelligence are expected to exhib- it high job performance and hence leave little room for cor- rection and improvement. For example, a doctor who makes few diagnostic errors has little potential for reducing errors in the future.

The compensatory model thus suggests the following hypotheses: Hypothesis 1: The association between emotional intelligence and task performance becomes more positive as cognitive intelligence decreases. Hypothesis 2: The association between emotional intelligence and organizational citizenship behavior OCB becomes more positive as cognitive intelligence decreases.

Their mean age was 41 s. Participants had an aver- age of 19 full years of work experience s. We did not code the job of one per- son who did not answer the questions concerning occupation and work tasks. We recruited participants via an e-mail message sent to approximately managerial, administrative, and profession- al staff.

The message invited full-time employees of the uni- versity who had been working in their job under the same supervisor for at least three months to participate in a study of important work outcomes.

Before taking part in the study, participants obtained a verbal agreement from their supervi- sors that they would provide ratings of job performance, to maximize the response rate.

To maximize the rate of response to our e-mail invitation, we stressed the confiden- tiality of the information provided Fink, We empha- sized the importance of research to the status of the organi- zation, because individuals are motivated to enhance the status of the groups to which they belong Ellemers, Wilke, and van Knippenberg, We provided a monetary incen- tive Fink, Finally, we indicated that participants would receive a report on the results.

This technique is effective because people like to learn about their own behavior and that of their colleagues Alreck and Settle, We obtained a response rate of 23 percent, which is a con- servative estimate calculated with the assumption that each person received, opened, and read the e-mail invitation. This response rate is similar to the typical response rate for mailed organizational surveys Fink, ; Alreck and Settle, To ascertain the degree of similarity between our sam- ple and the entire managerial, administrative, and profession- al workforce of the organization, we compared their demo- graphic characteristics.

There were two stages of data collection. In the first stage, we scheduled interested potential participants for a minute session held in a laboratory room.

A total of indi- viduals completed the first stage. No more than five partici- pants were scheduled for each session. We administered the test of cognitive intelligence first because it requires a standard procedure. The order in which the other tests and questionnaires were administered was randomly assigned and counterbalanced to alleviate fatigue and carryover effects Bickart, Participants completed all tests and question- naires on a computer except for the test of cognitive intelli- gence, which they completed using paper and pencil.

Supervisors who preferred to complete the question- naire on paper received a paper copy along with a self- addressed stamped envelope. We obtained assess- ments of job performance for , or 85 percent, of the employees. Some of the employees were rated by the same supervisor. The assessments were provided by supervisors. The participants for whom measures of job performance were available are the focus of the study. Measures Emotional intelligence. The MSCEIT contains tasks that ask respondents to identify emotions in photographs of faces and in images and landscapes, compare different emotions to different sensa- tions such as colors, indicate how emotions influence think- ing and reasoning, assemble emotions into complex feelings, identify how emotions transition from one to another, and rate the effectiveness of different emotion regulation strate- gies in both intrapersonal and interpersonal contexts.

Respondents receive credit to the extent that their answers match the answers provided by experts who are members of the International Society for Research on Emotion. Total scores are converted to interpretable nor- malized standard scores with a mean of and a standard deviation of For the sake of brevity, we report only the results with the expert scoring system.

The MSCEIT is an ability test in which respondents are presented with emo- tional problems and asked to choose the best answer among a set of options Mayer, Caruso, and Salovey, The abili- ty test approach to measuring emotional intelligence address- es some of the serious limitations of the competing self- report and peer-report approaches. In addi- tion, there is compelling evidence for its validity. A confirma- tory factor analysis of 2, responses showed that the fac- tor structure of the responses corresponds to the conceptual model Mayer et al.

Studies have also shown appro- priate discriminant validity with personality traits Brackett and Mayer, and criterion validity with outcomes such as the quality of social interactions Lopes et al. Cognitive intelligence. Fifty items are divided into four timed sub- tests that involve different perceptual tasks to avoid reliance on a single skill. The first item subtest lasts three minutes and presents the respondent with an incomplete series of figures.

The respondent chooses, from among the choices provided, the answer that best continues the series. The sec- ond item subtest lasts four minutes and presents the respondent with five figures and asks which two figures are in some way different from the three others.

The third item subtest lasts three minutes and asks the respondent to choose one of six figures that completes a matrix of figures. The final item subtest lasts two and a half minutes and requires the respondent to select, from five provided choic- es, the one that duplicates the conditions in a comparison fig- ure.

Correct answers are counted, and raw totals are convert- ed to interpretable normalized standard scores with a mean of and a standard deviation of 15 using a conversion table in the manual. We chose the Culture Fair Intelligence Test for several rea- sons. The test exhibits adequate internal reliability. The validity of the test is supported by consider- able evidence. The test predicts job outcomes e. Finally, at Given the long duration of our study, a longer test may have resulted in a decline in the quality of the data due to fatigue.

Job performance. We administered a five-item scale adapted from McCarthy and Goffin to assess task perfor- mance. Consistent with past research Williams and Anderson, ; Organ, , we analyzed the dimensions of OCB separately because different results are often obtained for them. We chose supervisory ratings of task performance and OCB because of the several advantages that they provide Rynes, Gerhart, and Parks, Supervisors provide the majority of the performance ratings in organizations Bretz, Milkovich, and Read, Supervisory ratings can be used for virtually any type of job, including jobs in which objective perfor- mance is difficult or impossible to measure Landy and Farr, A different approach would have required us to exclude participants in jobs such as manager from the study and, in turn, would have reduced our ability to generalize from the results.

For these reasons, researchers consider supervi- sory ratings to be most likely valid reflections of actual perfor- mance Arvey and Murphy, ; Guion, Although performance ratings collected for research purposes exhibit a. Accord- ingly, we ensured supervisors that performance ratings were obtained solely for research purposes.

Control variables psychological. We controlled for leader- member exchange to rule out an alternative explanation of any results. Evidence linking emo- tional intelligence to the quality of social relationships Lopes et al. Moreover, leader-mem- ber exchange is related to job performance Gerstner and Day, , and supervisors may provide lenient ratings to subordinates with whom they have good relationships. Thus, individuals with high emotional intelligence and low cognitive intelligence could have received high ratings because they developed good relationships with their supervisors.

To rule out this possibility, supervisors completed the LMX7 mea- sure for leaders Graen and Uhl-Bien, , as recommend- ed by Gerstner and Day following their literature review and meta-analysis. The LMX7 measure includes seven items rated on 5-point scales with varying anchors. The interac- tion between leader-member exchange and cognitive intelli- gence may carry a joint association of emotional intelligence and cognitive intelligence with job performance, even in the presence of leader-member exchange by itself Hull, Tedlie, and Lehn, That is, there could be an interaction between leader-member exchange and cognitive intelligence that looks essentially the same as the hypothesized interac- tion between emotional intelligence and cognitive intelli- gence, so that the association between leader-member exchange and job performance becomes more positive as cognitive intelligence decreases.

We verified that this was not the case by including both leader-member exchange and the interaction term for leader-member exchange and cogni- tive intelligence as control variables Hull, Tedlie, and Lehn, We also controlled for the Big Five traits of personality that reflect typical ways of acting at a broad level of analysis McCrae and Costa, ; McCrae and John, because they may act as confounding third variables. The Big Five traits are conceptually related to the emotion system Izard, , and a quantitative review revealed that the Big Five traits exhibit correlations ranging from.

Moreover, every Big Five trait is related to job performance, either universally or in certain jobs Barrick, Mitchell, and Stewart, Control variables demographic. To conduct conservative tests of the hypotheses, we controlled for education level, the number of hours worked per week, and occupation.

Education level was positively correlated with OCBO. The number of hours worked per week was positively correlated with both task performance and OCBO. Participants in man- agement occupations exhibited higher levels of task perfor- mance and both dimensions of OCB than participants who were not in these occupations.

Participants in office and administrative occupations exhibited lower levels of task per- formance and OCBO than participants who were not in these occupations. Finally, participants in computer and mathemati- cal occupations exhibited lower levels of OCBO than partici- pants who were not in these occupations. We used other occupations as the comparison category.

Nei- ther emotional intelligence nor job performance was related to age, gender, tenure in the organization, or total years of work experience. Subsidiary analyses showed that the sub- stantive conclusions were virtually identical when we also controlled for these demographic variables. Analysis Construct adequacy and discriminant validity of the test of emotional intelligence. We used confirmatory factor analysis to test whether the observed measures were associ- ated with their respective constructs.

We examined the fit of the model and verified that each indicator loaded significantly with its intended construct. We then formally assessed the discriminant validity of the test of emotional intelligence by testing additional models that constrain the association 1 between two latent constructs to 1 and using a chi-square There are at least three explanations of test of the difference in fit between the model with the the differences in job performance across occupations.

First, managers may have unconstrained association and the model with the con- exhibited higher performance than non- strained association Bagozzi and Phillips, ; Anderson managers because strong performers were promoted to managerial jobs.

Sec- and Gerbing, Discriminant validity is inferred if the ond, differences in job performance model with the constrained association provides a significant- across occupations may have emerged as ly worse fit to the data than the model with the uncon- a result of the characteristics of group leaders, given evidence that the produc- strained association.

Additional evidence of discriminant valid- tivity of business units depends on the ity is provided if the percent confidence interval for the transformational qualities of leaders Howell and Avolio, Third, the uni- latent correlation between two constructs in the uncon- versity may have had more success strained model does not include 1.

For these reasons, the differences gence, cognitive intelligence, agreeableness, conscientious- in job performance across occupations most likely reflect true variance. To reduce the num- ber of estimated parameters, we used the four subtests of the emotional intelligence test that correspond to the abilities to perceive, use, understand, and regulate emotions and the four subtests of the cognitive intelligence test as indicators.

We aggregated the items from the measures of each of the Big Five traits, leader-member exchange, and the OCB dimensions by randomly assigning them to three aggregate indicators per construct Bagozzi and Edwards, We did not aggregate the task performance items because there were only five of them.

Each item loaded significantly with its intended construct, as evidenced by the t-values greater than 6. The results of the tests of discriminant validity, displayed in table 1, support the discriminant validity of the test of emo- tional intelligence. Analytical strategy. We used hierarchical multiple regression to test the hypotheses. We centered the continuous predic- tors to facilitate the interpretation of the findings Cohen et al. We entered education level, the number of hours worked per week, the dummy codes for occupation, the Big Five traits, leader-member exchange, the leader-member exchange by cognitive intelligence interaction, emotional intelligence, and cognitive intelligence in step 1 and the emo- tional intelligence by cognitive intelligence interaction in step 2.

If the change in R2 in step 2 is significant, we can con- clude that the interaction between emotional intelligence and cognitive intelligence is a unique predictor of job performance Aiken and West, ; Cohen et al.

Five participants did not reveal their educational level, the number of hours they worked per week, or both. We conducted the analyses using listwise deletion, resulting in a sample of partici- pants. Analyses repeated using the mean replacement proce- dure revealed the same substantive conclusions. Graphical and statistical analyses Fox, ; Roth and Switzer, ; Cohen et al. The means and standard deviations for emotional intelligence and cognitive intelligence were close to the normative means and standard deviations of and 15, respectively.

Both emo- tional intelligence and cognitive intelligence were positively correlated with all three dimensions of job performance. Agreeableness and extraversion were positively correlated with task performance and OCBO. Openness to experience was positively correlated with OCBO. None of the personality 2 traits was correlated with OCBI. Hypothesis 1 predicted that the association significant correlation is consistent with between emotional intelligence and task performance will other research.

In their meta-analysis, become more positive as cognitive intelligence decreases. Tett, Jackson, and Rothstein found a percent confidence interval for the The results shown in table 3 reveal that the interaction uncorrected correlation between consci- between emotional intelligence and cognitive intelligence entiousness and job performance ranging from —.

The interaction, graphed in correlation of —. Past research also figure 1, is consistent with hypothesis 1. Education 3. Number of hours worked per week Computer and mathematical occupations Management occupations Office and admin. Other occupations Agreeableness 4. Conscientiousness 4. Emotional stability 3. Extraversion 3. Openness to experience 3.

Leader-member exchange 4. Emotional intelligence Cognitive intelligence Task performance 5. OCBO 3. OCBI 3. Emotional stability. Openness to experience. Leader-member exchange. Emotional intelligence —.

Cognitive intelligence —. Task performance —. OCBI —. Thus, hypothesis 1 is supported. Emotional intelligence, cognitive intelligence, and organi- zational citizenship behavior OCB.

Hypothesis 2 predicted that the association between emotional intelligence and OCB will become more positive as cognitive intelligence decreas- es. The results shown in table 3 reveal that the interaction between emotional intelligence and cognitive intelligence predicts OCBO.

The interaction, graphed in figure 2, is consistent with hypothesis 2. Moderating effect of cognitive intelligence on the association between emotional intelligence and task performance. Inspection of table 3 also reveals that the interaction of cognitive intelligence and emotional intelligence is not related to OCBI. Thus, hypothesis 2 is partially sup- ported. Mediational Analyses We measured leader-member exchange to rule out its poten- tial mediational effects.

The analyses reported in table 3, however, do not represent complete tests of mediation. Accordingly, we conducted additional analyses to test for mediation. Both partial and full mediation can be ruled out if the predictor is not related to the potential mediator James and Brett, ; Kenny, Kashy, and Bolger, There was no evidence that leader-member exchange mediated the associations among emotional intelligence, cognitive intelligence, and job performance.

Moderating effect of cognitive intelligence on the association between emotional intelligence and organizational citizenship behavior directed at the organization. We also examined whether the Big Five traits acted as medi- ators. There was no evidence that the Big Five traits mediated the associations among emotional intelligence, cog- nitive intelligence, and job performance.

Subsidiary Analyses Involving the Emotional Demands of Jobs We examined whether the emotional demands of the job affected how emotional intelligence and cognitive intelligence were associated with job performance. In the compensatory model, emotional intelligence compensates for low levels of cognitive intelligence in most, if not all jobs, because effec- tive social interaction, good decisions, and high motivation contribute to job performance in most, if not all jobs.

Our the- orizing differs from that of Wong and Law , who pro- posed that emotional intelligence is more strongly related to job performance in emotionally demanding jobs than in other jobs.

The interaction between supervisor-rated emotional demands and emotional intelligence, however, was not a significant predictor of job performance by conventional standards. The internal reliability coefficient was. The standardized scores were —. We regressed job performance on all of the control variables, emotional intelligence, cognitive intelligence, emotional demands, the three two-way interactions, and the three-way interaction of emotional intelligence, cognitive intelligence, and emotional demands Aiken and West, The pre- sent study reveals, instead, that emotional intelligence and cognitive intelligence are compensatory with respect to task performance and organizational citizenship behavior directed at the organization OCBO.

Emotional intelligence becomes a stronger predictor of task performance and OCBO as cogni- tive intelligence decreases. Our results reveal that employees with low cognitive intelligence perform tasks correctly and engage in OCBO frequently if they are emotionally intelligent.

Our findings provide a potential way to reconcile the diver- gent findings in past research. It is difficult to ascertain whether this is the case, however, because only some of the past studies mea- sured cognitive intelligence, and different measures of cogni- tive intelligence with scores calculated on different scales were administered.

Our findings address the controversy about the usefulness of emotional intelligence for organizational research and man- agerial practice. Landy and Zeidner, Matthews, and Roberts argued against using emotional intelligence to predict job performance unless it makes an incremental linear contribution to prediction. Our results suggest that this argu- ment is overly simplistic. Predictors of job performance such as emotional intelligence may be important in ways other than their incremental linear effects Murphy, ; Hough, Our results show that emotional intelligence is an important predictor of task performance and OCBO because of its interactive effect with cognitive intelligence.

Our results also reveal that using cognitive intelligence tests alone to pre- dict performance entails risk, because employees with low cognitive intelligence can perform effectively if they have high emotional intelligence.

Our hypothesis concerning OCBI was not supported. One possible explanation concerns how abilities and personality traits may predict job performance criteria differently Hough, For example, one study found that cognitive intelli- gence predicts the technical proficiency of soldiers better than their personal discipline, and the opposite is true for per- sonality traits McHenry et al.

Task performance con- cerns core substantive duties, and OCBO includes taking action to protect the organization from potential problems and offering ideas to improve the functioning of the organiza- tion. Future Directions Our performance measures consisted of ratings by the supervisors. These measures may be limited by the relatively low interrater reliability of supervisory ratings, which implies that any one supervisory rating includes measurement error Viswesvaran, Ones, and Schmidt, Even so, future research with performance measures other than supervisory ratings, if available, would help to address this limitation of our study.

Our measures of performance invite some alternative expla- nations of the results. One alternative explanation is that the emotional intelligence of individuals with low cognitive intelli- gence enhances their likeability and, in turn, their perfor- mance ratings. Our research design sheds light on the viabili- ty of this alternative explanation.

The mediation analyses also revealed no evidence of intervening roles of agreeableness or extraversion that reflect, in part, likeability Costa and McCrae, ; Lopes et al. It is difficult to explain why likeable employees may have received inflated ratings on only two of the three criteria. Finally, past research showing that supervi- sors like subordinates who perform well Robbins and DeNisi, suggests that strong performance by the sub- ordinate leads to both high ratings and liking from the super- visor and casts further doubt on the possibility that individu- als with high emotional intelligence and low cognitive intelligence received high ratings because they were liked.

Another alternative explanation is that emotional intelligence assisted individuals with low cognitive intelligence to manage their impressions well and, in turn, receive high performance ratings. There is conceptual and empirical evidence that sub- ordinates who manage their impressions successfully devel- op good relationships with their supervisors Liden and Mitchell, ; Wayne and Green, Therefore, the con- trol for leader-member exchange casts some doubt on a role for impression management.

It is also difficult to explain why employees who managed impressions well may have received inflated ratings of task performance and OCBO but not OCBI. From a different perspective, impression manage- ment could be viewed as a job skill that supervisors should take into account when rating performance because it may help employees to perform tasks correctly and engage in OCBO frequently.

In future research, it would be useful to examine whether impression management plays a substan- tive role in the associations between emotional intelligence, cognitive intelligence, and job performance.

It would also be interesting for future research to examine whether emotional intelligence is associated with job perfor- mance in individuals with low cognitive intelligence because it helps them reach advantageous positions in social net- works. Emotional intelligence may help individuals develop links to friends and coworkers that provide assistance that, in turn, contributes to high performance Mehra, Kilduff, and Brass, ; Sparrowe et al.

In addition, individuals who connect disparate groups achieve high performance by brokering between these groups Burt, Establishing these connections may often be emotionally challenging because different groups often have different norms and cul- tures that may be incongruent with one another.

Emotional intelligence may help individuals with low cognitive intelli- gence manage the emotional discomfort of connecting dis- parate groups and, in turn, achieve high performance. Emotionally intelligent individuals with low cognitive intelli- gence may also achieve high job performance because they manage conflict successfully. In addition, negative emotions exacerbate the detrimental impact of relationship conflict on performance Jehn, Evidence that emotionally intelligent individuals tend to develop close social relationships e.

Exploring the potential role of emotional contagion in the associations between emotional intelligence, cognitive intelli- gence, and job performance represents another interesting avenue for future research. It is important to examine further whether the associations between emotional intelligence, cognitive intelligence, and job performance depend on the nature of the job.

Thus, although we can be confident that our analyses would have detected medium-sized effects, we can- not be confident that small-sized effects do not exist. Studies with more complete measures of emotional demands and larger samples are needed.

Finally, future research should examine how to improve the measurement of emotional intelligence. We used an ability test of emotional intelligence to circumvent the limitations of commonly used self-report and peer-report measures. Despite its advantages, this test may not reflect exactly how people process emotional information in real-life social inter- actions Matthews, Zeidner, and Roberts, In addition, the test may require some cognitive abilities such as abilities in the domain of language to complete successfully.