The moderating effect of emotional intelligence on the relationship between job satisfaction and affective commitment, a study among employees in vietnam banking industry

46 Appendix 8: Scale reliability assessment with Cronbach’s alpha N=313 – Job satisfaction scale – Fringe benefits 47 Appendix 9: Scale reliability assessment with Cronbach’s alpha N=313

International School of Business

Phạm Ngọc Anh Thơ

THE MODERATING EFFECT OF EMOTIONAL INTELLIGENCE ON THE RELATIONSHIP BETWEEN JOB SATISFACTION AND AFFECTIVE COMMITMENT: A STUDY AMONG EMPLOYEES IN VIETNAM BANKING

International School of Business

Phạm Ngọc Anh Thơ

THE MODERATING EFFECT OF EMOTIONAL INTELLIGENCE ON THE RELATIONSHIP BETWEEN JOB SATISFACTION AND AFFECTIVE COMMITMENT: A STUDY AMONG EMPLOYEES IN VIETNAM BANKING

I would like to express my deep gratitude to my supervisor Tran Phuong Thao forher precious guidance, valuable suggestions and advice as well as her immenseknowledge to support me throughout the process.

In addition, I would like to thank my family, my friends and my colleagues fortheir constant encouragement and immediate helps without which the study would nothave been possible

The study investigates the influence of job satisfaction facets on affectivecommitment in the context of Vietnam banking industry It also examines the moderatingrole of emotional intelligence in the relationship between job satisfaction facets andaffective commitment The model was tested with a sample of 313 employees workingfor banks in Ho Chi Minh city The result reveals a good fit between collected data andmeasurement scales which were introduced and developed in the Western contexts Thefindings shows that job satisfaction facets have positive impacts on affectivecommitment, and that emotional intelligence significantly serves as a moderator for threeout of seven facets of job satisfaction and affective commitment relationship Thediscovery highlights the importance of documenting job satisfaction and emotionalintelligence in enhancing the employee commitment towards organization as well as callsthe management attention to their recruiting, training and long term developmentpolicies.

ABSTRACT

TABLE OF CONTENTS

1 Introduction 1

1.1 Background of the study 1

1.2 Research problem 3

1.3 Research objectives 5

1.4 Scope of the study 5

1.5 Significance of the study 6

2 Literature review 7

2.1 Foundational theories 7

2.2 Affective commitment 8

2.3 Emotional intelligence 9

2.4 Job satisfaction 12

3 Hypothesis development 13

3.1 Job satisfaction and affective commitment 13

3.2 Emotional intelligence and affective commitment 15

3.3 Emotional intelligence as a moderator 16

4 Research methodology 19

4.1 Procedure and sample 19

4.2 Questionnaire design 21

4.3 Measurement scale 21

4.4 Data analysis 22

5 Result 23

5.1 Sample profile 23

5.2 Measurement refinement 24

5.3 Measurement models: 26

5.3.1 Correlations analysis 26

5.3.2 Hierarchical regression analysis 27

5.4 Discussion 31

5.4.1 Theoretical implication 32

5.4.2 Managerial implication 33

5.5 Direction for further studies and limitations 34

5.6 Conclusion 35

References Error! Bookmark not defined.7

Appendix 2: Questionnaire survey – Vietnamese version 39

Appendix 3: Scale reliability assessment with Cronbach’s alpha (N=313) - Affective commitment scale 44

Appendix 4: Scale reliability assessment with Cronbach’s alpha (N=313) – Emotional intelligence scale 44

Appendix 5: Scale reliability assessment with Cronbach’s alpha (N=313) – Job satisfaction scale - Pay 45

Appendix 6: Scale reliability assessment with Cronbach’s alpha (N=313) – Job satisfaction scale - Promotion 46

Appendix 7: Scale reliability assessment with Cronbach’s alpha (N=313) – Job satisfaction scale - Supervisors 46 Appendix 8: Scale reliability assessment with Cronbach’s alpha (N=313) – Job satisfaction scale – Fringe benefits 47 Appendix 9: Scale reliability assessment with Cronbach’s alpha (N=313) – Job satisfaction scale – Co-workers 47 Appendix 10: Scale reliability assessment with Cronbach’s alpha (N=313) – Job satisfaction scale – Nature of work 48

Appendix 11: Scale reliability assessment with Cronbach’s alpha (N=313) – Job satisfaction scale -Communication 49

Appendix 12: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot study 49

Appendix 13: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot study 50

Appendix 14: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot studyJob satisfaction facets -Pay 51

Appendix 15: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot studyJob satisfaction facets -Promotion 51

Appendix 16: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot studyJob satisfaction facets -Supervisors 52

Appendix 17: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot study Job satisfaction facets -Fringe benefits 52

Appendix 18: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot study- Job satisfaction facets – Operating condition 53

Appendix 19: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot study Job satisfaction facets -Co-workers 53

Appendix 20: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot study Job satisfaction facets -Nature of work 54

Appendix 21: Scale reliability assessment with Cronbach’s alpha (N=158) – pilot study Job satisfaction facets -Communication 54

Appendix 22: Exploratory factor analysis (EFA) - Affective commitment 55

Appendix 23: Exploratory factor analysis (EFA) - Emotional intelligence 56

Appendix 24: Exploratory factor analysis (EFA) – Job satisfaction 57

Appendix 25: Normal distribution of residuals 60

Appendix 26: Homoscedasticity test 61

Appendix 27: Hierarchical regression analysis with emotional intelligence is a moderating value 62

Figure 1 The conceptual model 19

Figure 2 Moderating effect of emotional intelligence on the relationship between nature of works and affective commitment……….39

Table 1: Demographic statistics 22

Table 2: Cronbach’s Alpha analysis 26

Table 3: Correlations 27

Table 4: Hierarchical Regression Analysis with Emotional intelligence as a moderating variable 30

Table 5: Result Summary……… … 39

1 Introduction

The first part of the study is devoted to emphasizing the practical and theoreticalimportance of studying affective commitment, job satisfaction, emotional intelligenceand the relationships among them Besides, the study desires to present the interest in thetopic in the context of Vietnam banking The final section is the summary of objectiveswhich is followed by the scope of the study

1.1 Background of the study

Human resource has been considered as one of the most valuable asset of anyorganization According to Ferrary (2015), human capital has a direct relationship toorganizational performance in terms of company’s profit and targets Other studies foundthat maintaining a stable labour force is an optimal way to obtain effective costmanagement and enhance quality of service in a firm (Jago & Deery, 2002) as well asachieve sustainable economic development of a nation (Bousrih, 2013)

The concept of organizational commitment has been studied by many researchersover the last three decades Diversity of factors were found to have significantrelationships with organizational commitment namely: Job satisfaction, job performance,job stress, work attitude, organizational justice, tenure, gender educational level, turnoverintention, etc (Mathieu & Zajac, 1990; Meyer &Allen, 1999) Affective commitment -the strongest component among three commitments has received considerable attentionfrom researchers all over the word Being considered as a representative of organizationalcommitment, affective commitment has demonstrated its significant role in most of the

relationships related to employee’s engagement (Carmeli, 2000) Understanding theassociations of these factors with organizational commitment, especially affectivecommitment, is important for improving the employee performance and enhancingemployee productivity as well.

In order to expand the knowledge of this issue, numerous of new approaches havebeen undertaken to contribute to organizational commitment literature Lately, emotionalintelligence has captured the considerable attention from researchers of various fields Itwas explored to have a vital role in explaining variety of potential outcomes such asacademic achievement, human physical and mental health, social support, overall well-being (Meisler, 2013) Additionally, emotional intelligence has played its special function

as a mediator or a moderator which helps a wide array of studies obtain deeper insightsinto many relationships Clarke and Mahadi (2011) observed the linkage between leader-member exchange and a plethora of work-related outcomes, eventually they discoveredthat emotional intelligence – as a moderating effect, kept an important role in explainingthe variation of each specific pair of constructs Emotional intelligence was also found tohave significant influence on the stressor-mental health relation (Davis & Humphrey,2012), the psychological control-behaviour problem relation (Gugliandolo et.al, 2015)orthe stress-burnout relation (Görgens-Ekermans &Brand 2012)

Emotional intelligence was proved to have the moderating effect on relationshipsbetween affective reactions and specific behaviours (Jorden et al., 2002) Accordingly,previous studies on the job satisfaction – organizational commitment relation have

provided crucial contribution to the theoretical foundation but this is not sufficient toexplain the variation of employee commitment Based on the research of Jorden andothers, the recent study aims to investigate whether the interaction of emotionalintelligence and job satisfaction predicts affective commitment, in other words, whetherthe association between job satisfaction and affective commitment is stronger forindividuals who have high score of emotional intelligence and weaker for ones whopossesses a low level of emotional intelligence.

1.2 Research problem

Since joining the WTO (World Trade Organization) in 2007, Vietnam haswitnessed a significant economic growth in the latest decade due to the improvement ofbusiness environment and the high volume of foreign investments Along with theachievement in economy, Vietnam banking industry has also seen an outstandingenlargement and development in terms of scale and service quality (Leung, 2009),especially in Hanoi and Ho Chi Minh City

However, the increases in competition and financial innovation led to anextremely rapid expansion which resulted in banking system problems in the past fewyears Eventually, the project 254 called “Restructuring credit institutions system 2011-2015” was implemented by the Prime Minister in 2012 Accordingly, the State Bank ofVietnam has pushed banks to merge, with the goal of more than halving the number oflenders (Lane, 2016) In such an unfavourable condition, labour in this sector hasexperienced a considerable fluctuation with the highest average voluntary turnover rate

compared to other service sectors; with the latest figure in 2014 is about 15 % (Thanh,2015) Restructuring banking sector not only causes the workforce transition amongbanks but also between banking and non-banking sectors, accordingly, keeping theintellectual property for organizations is a crucial strategy for maintaining a sustainabledevelopment.

The literature has showed that a large number of empirical findings on employee’scommitment, emotional intelligence and job satisfaction originated from Westerncountries, this study attempts to widen the implication to a broader scope outside thetraditional samples by testing the relationships among mentioned constructs in thecontext of Vietnam The concept of affective commitment was studied in Vietnamservices industry (Choi, Tran, & Park, 2015), the finding showed a positive relationshipbetween inclusive leadership and employee work engagement with the mediating roles ofaffective commitment and employee creativity Besides, some empirical studies on jobsatisfaction were conducted in Vietnam in recent years: Job satisfaction was tested amongmarketers in Ho Chi Minh city by Nguyen and Nguyen (2011), along with jobattractiveness, job satisfaction was found to be positively influenced by firm-specificmarketing capital pool Job satisfaction was also examined in the association with work-family conflict in the context of state-owned and foreign invested enterprises (Nguyen,Jin, Kiong & Fah, 2016) Whilst organizational commitment and job satisfaction havebeen taken into consideration in Vietnam for several recent years, little is known aboutEmotional intelligence in this small Southeast Asia country Over all, gaining a deeper

understanding of factors associated with organizational commitment, specificallyaffective commitment, what directly and indirectly facilitate or hamper the degree of thecommitment, will help banks to set appropriate human resources policies to encourageemployee’s motivation as well as enhance bank’s performance.

1.3 Research objectives

The study aims to examine the relationship between affective commitment and jobsatisfaction’s facets with emotional intelligence in the role of a mixed moderator Threespecified objectives are given as follow:

(1) The relationship between job satisfaction’s facets and affective commitment

(2) The relationship between emotional intelligence and affective commitment

(3) The impact of emotional intelligence on the relationships between job

satisfaction’s facets and affective commitment as a mixed moderator

1.4 Scope of the study.

The study concentrates on the impacts of job satisfaction on affective commitmentand the mixed moderating effect of emotional intelligence on the relationships, the studyexcludes all other elements affecting the variables which potentially influence the results.Obviously, all the notions studied in this research relates to a wide range of sectors in theeconomy, the scope of the study is limited in banking industry

Moreover, data collection was only conducted in Ho Chi Minh city Located in aconvenient position, Ho Chi Minh city is known as the most important economic centre

which plays a vital driving impetus of financial, commercial and service hub of thecountry Particularly, the city is home to a large network of banking and financialinstitutions (Tuoi tre news, 2015).For these reasons, Ho Chi Minh city was selected forthe study.

1.5 Significance of the study

In the context of a transitional economy in Vietnam, the banking sector hassuffered from numerous of unfavourable situations came from stock markets, real estatemarkets and money market (Vuong, 2014); the turnover in banking segment caused bysuch inconvenient conditions has thus increased Thus, the employment shift preventscommercial banks from sustainably and stably operating as well as developing in the longterm Hence, the study is expected to offer a significant contribution to both theoreticaland empirical perspective:

From the academic viewpoint, the study provides insights into emotionalintelligence in the workplace of Vietnamese employees, unfolds the role ofemotional intelligence in moderating job satisfaction-affective commitmentrelation and discovers which facets of the job satisfaction have considerableinfluences on staffs’ commitment to the bank

From the practical viewpoint, the findings attract an interest of makers in emotional intelligence relating to recruiting, training andretaining activities For employees, once emotional intelligence is wellunderstood and concerned, the ability of managing and controlling

policy-emotions, moods, reactions is thus enhanced This advantage is expected tohelp bank staffs create stronger and more fulfilling relationships inworkplace.

2 Literature review

2.1 Foundational theories

Based on the idea of social exchange theory (Emerson, 1976) which wasconceptualized as a joint activity gained by exchanging the benefits, commitment wasstated to be directly tied to psychological contract in which the employees-organizationrelation is driven by a reciprocal obligation between two parties (Sarantinos, 2007).According to Daan and Sleebos(2006), job satisfaction and organizational commitmentare social exchange outcomes because the two constructs reflect a perception of theexchange quality of which the employee and the organization are required to fulfill theirobligations to each other and establish ongoing reciprocity

The Emotional Intelligence theory of Bar-On was first introduced in 1985 andfollowed by a series of other subsequent developed versions by Bar-On (2006), Mayerand Salovey (1990) and Goleman(1998) Even though there are a lot of such differentperspectives in doing research on emotional intelligence, since this construct wasconceptualized, IQ alone has no more dominated as a measure of success

Emotional intelligence was found to play roles in both building and maintainingsuccessful social relationships, predicting specific aspects of situations involving socialexchange (Reis et al., 2007) When an organization makes its employees satisfied with

their job, it will in turn obtain the commitment from employees In this situation,emotional intelligence will either foster or hamper this exchange process.

2.2 Affective commitment

Organizational commitment is defined the degree of attachment or loyaltyemployees feel towards the organisation (Allen & Meyer, 1990; Meyer et al., 2006;Mowday et al., 1979) It has received considerable attention as a result of its ability toproduce desirable outcomes for organizations Previous studies suggested thatorganizational commitment has the potential link to many outcomes such as jobperformance (Mathieu & Zajac, 1990), job satisfaction (Bailey, Albassami & Al-Meshal,2016; Vandenberg & Lance, 1992), sales (Boles et al 2007), turnover, absenteeism,tenure and organizational goals (Baek-Kyoo & Park, 2010; Meyer & Allen, 1999) Incontrast, there was another school of thoughts advocating that the variation oforganizational commitment can be explained by a plethora of antecedents such as jobsatisfaction, emotional intelligence, organizational justice, organizational culture,sociodemographic characteristics etc (Cohen-Charash & Spector, 2001; Jiang, 2015;Shalini, 2013; Sikorska-Simmons, 2005)

The study employs the three-component modelof Allen and Meyer (1990) toexamine the bank staffs’ commitment to the banks they are working for The three-component model which was re-conceptualized from numerous definitions has beenbroadly used in academic community thanks to its widespread application in variousdomains, accordingly, commitment was classified by three distinguishable mindsets

(Allen& Meyer, 1990): Affective commitment refers to the emotional attachment anemployee has towards the organization Employees with strong affective commitmentstay with the organization because they want to.Continuance commitment reflects anawareness of the costs of leaving the organization Finally, normative commitmentreflects a perceived obligation to stay with the organization.

Among three components of commitment, affective commitment was found to bemost important factor compared to continuance and normative commitment (Meyer et al.,2002) This can be explained by the argument that workers tend to make their owndecision on whether they want to remain with an organization (affective commitment)rather than they do so for obligation (normative commitment) or financial issues(continuance commitment) Once employee’s emotional bond to their workplace nolonger exists, they tend to leave for another job instead of staying for any other purposes(Rhoades et al., 2001) For this reason, the study concentrates on affective assessment tointerpret the job satisfaction-organizational commitment connection Moreover, sinceaffective commitment is considered as an emotional attachment between an individualand an organization (Meyer&Allen, 1999), it is expected to be directly or indirectlyexplained by another psychological construct: emotional intelligence in some certainways

2.3 Emotional intelligence

Emotional Intelligence is the ‘ability of an individual to monitor his own andothers’ feelings and emotions, to discriminate among them and to use this information to

guide other individual’s thinking and actions’ (Salovey & Mayer, 1990) Emotionalintelligence has been researched in a number of different ways to examine therelationships with other work-related factors including transformational leadership,conflict resolution, political skill, success in job interviews, and job performance (Day &Carroll, 2004; Joseph & Newman, 2010; Harms &Crede´, 2010; Mayer et al., 2008;Meisler, 2013) or explain work attitudes and behaviours (Carmeli, 2003; Wong & Law,2002) Cooper (1997) also pointed out that people having high emotional intelligence cancope with the organizational problems more effectively than those having low emotionalintelligence There has been numerous of debates around this topic in the academiccommunity, research on emotional intelligence thus keep increasing over twenty-fiveyears with different point of views although they overlap in some facets.

Ability model was initially introduced by Salovey and Mayer in early 1990’swhich measures a set of cognitive abilities: The ability to perceive emotions; the ability tointegrate emotion; the ability to facilitate thought; the ability to understand emotions; andthe ability to manage emotions using Emotional-Intelligence Test (MSCEIT: Mayer-Salovey-Caruso-Emotional-Intelligence Test) The model continued to be developed byWong and Law in 2002 with the scale named Wong and Law Emotional IntelligenceScale (WLEIS) The model supposes that emotional intelligence originates in childhoodand continue to develop subsequently The mixed model (Bar-On, 1997) regardsemotional intelligence as a combination of cognitive abilities and personality traits The

competency model (Goleman, 1998) considers emotional intelligence as being capable ofdevelopment in operating within working environment.

The current study opted the ability model to examine the influence of emotionalintelligence on the relationship between affective commitment and job satisfaction Based

on the theory of Salovey and Mayer (1990), Wong and Law (2002) conceptualizedemotional intelligence as composed of four distinct dimensions called WLEIS (Wong andLaw Emotional Intelligence Scale):Self emotional appraisal (SEA) refers to theindividual’s ability to understand their emotions, other’s emotional appraisal (OEA)refers to the individual’s ability to recognize and understand other people’s emotion,regulation of emotion (ROE) refers to ability of people who regulate their emotionswhich will enable a more rapid recovery from distresses and use of emotion (UOE) refers

to the ability of individuals to make use of their emotions by directing them towardsconstructive activities and enhancing performance

According to Meisler and Vigoda-Gadot (2014), compared to other tools, WLEIShas some predominant features: Firstly, it has been tested in different cultures, ethnic andgender groups by recent studies Additionally, measure of emotional intelligenceemploying WLEIS does not overlap with measure of personality traits Last but not least,WLEIS was designed to be used in organizations and it especially plays a vital role inpredicting job satisfaction For all these reasons, this approach of emotional intelligence

is the best choice to assess this construct in the context of the study

2.4 Job satisfaction.

Job satisfaction can be defined as the degree to which people like their jobs(Spector, 1997).Job satisfaction is associated with important work-related and generaloutcomes such as higher levels of job performance, organizational commitment;discretionary activities such as organizational citizenship behavior, and life satisfaction aswell as with lower levels of absenteeism, lateness, and turnover (Bailey, Albassami & Al-Meshal, 2016; Judge, Heller & Mount, 2002; Gyekye, & Haybatollahi, 2015;Wickramasinghe, 2010 ) Lack of job satisfaction results in a poorer performance, morefrequent absenteeism from work, lower work productivity, more serious depression(Meyer & Allen, 1999)

Job satisfaction is one of the most broadly investigated concept in academiccommunity in general and in organizational behaviour in particular Spector (1997)pointed out that job satisfaction consists pay, promotion, supervisor, marginal benefits,contingent rewards, areas of activity, colleagues, nature of work, communication.Examining job satisfaction with different facets, Tett and Mayer (1993) argued that thisconstruct should be assessed by three components: Satisfaction with career advancement,satisfaction with supervisors, and satisfaction with co-workers Likewise, Goetz et al.(2012) classified job satisfaction into two dimensions: Internal job satisfaction whichrefers to an individual’s motivation to willingly perform an assigned task for itself and forthe pleasure it brings, and, external job satisfaction which refers to external benefitsprovided by the organization

Even though global assessment of job satisfaction with first-order construct orone-item scale usage have been employed in many studies, it is not encouraged to be usedfor measuring the concept due to its inaccurate and overestimated assessment (Boles et.al., 2007) Specifically, every facet may not have the equal effect on an individual’ssatisfaction, one may be highly satisfied with job due to a high satisfaction with salary,promotion, supervisor rather than with the work itself The global or overall approach isunable to analyse job satisfaction in more details because the survey questions primarilyevaluate one’s general feelings or attitude towards the job, thus it is difficult for Humanresources policies to be set up appropriately As argued above, the Job SatisfactionSurvey of Spector (1997) including eight-facet assessment was adopted to collect therespondent’s detailed feelings about the job.

3 Hypothesis development 3.1 Job satisfaction and affective commitment

The relationship between job satisfaction and organizational commitment is not anew topic in academic community It has been reported in numerous articles by scholarsall over the world that these two concepts have a positive connection, however, theinvestigation results of this association reflected various scientific perspectives.According to some studies, job satisfaction was regarded as an antecedent toorganizational commitment, in other words, job satisfaction served as a predictor oforganizational commitment (Mowday et al., 1982; Porter et al., 1974)

Conflicting with the above argument, there was another school of thoughtsuggesting that organizational commitment is an outcome of job satisfaction, and that thesatisfaction can be explained by various degrees of organizational commitment (Bailey,Albassami, & Al-Meshal, 2015; Baker &Baker, 1999; Laschingeret al., 2001) In thecontext of this research, job satisfaction is identified as a determinant to organizationalcommitment In line with the former perspective, the current research supposes thatsatisfied employees are more committed to their job as well as to the bank whereasdissatisfied employees might want to leave the organization if their attitude to the job is

no longer positive

Although three components of organizational commitment share the similarimpact on an employee’s attachment to the organization, the nature of each commitmentmay differ and has the separate meanings In a study conducted in 2002, Meyer et al.concluded that affective commitment occupies a greater contribution to employee’sfavourable work behaviours compared with normative and continuance commitment As

a result, when it comes to organizational commitment, affective commitment alone wascommonly used in majority of studies instead of three types of commitments (Jiang,2014; Lövblad, Hyder, & Lönnstedt, 2012; Vandenberghe, & Bentein, 2009) This can beexplained by the fact that satisfied employees are committed to the organization for theiremotions rather than obligations or the cost they may pay if leaving In more detail, whenaffective commitment was separately tested in the research of Boles et al (2007), theresult revealed that different facets of job satisfaction have various degrees of effect on

the dependent variable Patrick and Sonia (2012) also discovered the significant role ofjob satisfaction in explaining 67% the variation of affective commitment.

Consistent with earlier research, this study consumes that eight dimensions of jobsatisfaction: Pay, promotion, supervisors, fringe benefits, operation condition, co-workers, nature of work, communication have the positive effects on affectivecommitment Therefore, the first hypotheses were formed as follow:

H1a: Pay is positively related to affective commitment

H1b: Promotion is positively related to affective commitment

H1c: Supervisors is positively related to affective commitment

H1d: Fringe benefits is positively related to affective commitment

H1e: Operation condition is positively related to affective commitment

H1f: Co-workers is positively related to affective commitment

H1g: Nature of work is positively related to affective commitment

H1h: Communication is positively related to affective commitment

3.2 Emotional intelligence and affective commitment

According to Wong and Law (2002), employees with high emotional intelligenceare more satisfied with their jobs, enjoy a higher level of job satisfaction, as a result, will

be less likely to leave their jobs On the contrary, low emotional intelligence employees,however, are more likely to leave their organizations due to less committed to their

organizations Drawing from a study of Abraham (2000), emotional intelligence wasfound to significantly associate with organizational commitment for most participants,this association was even stronger than that between job satisfaction and organizationalcommitment In line with Abraham’s finding, Gu¨leryu ¨z et al (2008) also pointed outthat all the direct path coefficients from dimensions of emotional intelligence toorganizational commitment were significant However, these findings were exploredalong with the discoveries in which either moderators or mediators were included Hence,

although the direct paths were found significant, R2 was greater with the presence ofmoderating and mediating effects

As previously discussed, among three types of commitment, affective commitmentbenefits the organization to a higher degree due to its strongest contribution to theemployee’s favourable work behaviours Therefore, based on the emotional intelligence-organizational commitment relationship which was concluded to be significant, the studydesired to examine whether an individual who is good at managing and regulatingemotions is more affectively attached to the organization than individual who hasproblems in controlling the feelings? The second hypothesis was thus stated as following:

H2: Emotional intelligence is positively related to affective commitment

3.3 Emotional intelligence as a moderator

Emotional intelligence has been reported as a moderator in numerousrelationships Clarke and Mahadi (2011) emphasized the moderating role in their researchbetween quality of leader-member exchange and a series of work-related outcomes

namely organizational citizenship behaviour, turnover intention, job satisfaction,organizational commitment, in-role performance andpsychological well-being, the resultafter using a hierarchical moderated regression model revealed that the higher relativescores on emotional intelligence fostered the magnitude of the effect between each pair ofconcepts Moreover, emotional intelligence was discovered to moderate the relationshipsbetween emotional labor and strain, emotional labor efforts and outcomes, emotionaldissonance and work-family conflict, organizational politics and organizationalcommitment, etc… (Cheung & Tang, 2012; Gao et al., 2013).

However, these above research heavily connected to the Western background andreceived little attention in Asia workplace environment Cultural differences between theEast and the West have been empirically proved to impact emotion experience anddisplay (Matsumoto & Hwang, 2012) To be specific, individual achievement andfeelings are valued in Western countries whereas the interdependence of the self andcommunity is more valued in Eastern countries (Scott-Halsell et al 2008) Thisdistinction will to some extent determine individual’s thinking and action of each culture.Hence, studying emotional intelligence in the context of an Asia nation like Vietnamshould be taken into account for adequate consideration

The study supposed that the impact of job satisfaction on affective commitment isstronger when an individual has a higher level of emotional intelligence and vice versa

Up to the present, the moderating role of emotional intelligence in the relation of jobsatisfaction and organizational commitment has not received much attention from

academic community In order to enrich the organizational behaviour literature, thecurrent study observed the potential mixed moderating effect of emotional intelligence toprovide a deeper understanding for job satisfaction – affective commitment relation Indoing so, the interactions of job satisfaction’s facets and emotional intelligence wereadded to test this potential effect The hypotheses were thus developed:

H3a: Emotional intelligence plays a mixed moderating role in the relationshipbetween pay and affective commitment

H3b: Emotional intelligence plays a mixed moderating role in the relationshipbetween promotion and affective commitment

H3c: Emotional intelligence plays a mixed moderating role in the relationshipbetween supervisors and affective commitment

H3d: Emotional intelligence plays a mixed moderating role in the relationshipbetween fringe benefits and affective commitment

H3e: Emotional intelligence plays a mixed moderating role in the relationshipbetween operation condition and affective commitment

H3f: Emotional intelligence plays a mixed moderating role in the relationshipbetween co-workers and affective commitment

H3g: Emotional intelligence plays a mixed moderating role in the relationshipbetween nature of work and affective commitment

H3h: Emotional intelligence plays a mixed moderating role in the relationshipbetween communication and affective commitment.

Figure 1 The conceptual model

4 Research methodology

4.1 Procedure and sample

The study was undertaken through two steps including pilot study (qualitative andquantitative) at the first stage and main survey (quantitative) at the second stage bycapturing responses of bank staffs of ten banks in Ho Chi Minh City

The qualitative pilot study involved an in-depth interview with 4 bank staffs Theinterview was implemented to assess the clarity of words, the content of the questionsand the ability to answer the questions of the interviewees Generally, all the items werewell understood The quantitative pilot study was then used to test the reliability and

validity of the scale with the sample of 158 employees selected by convenient method.The purpose of this stage was not only to assure that the instruments employed in thisstudy was appropriate for the context of Vietnam but also to well prepare for the finalmeasurement which was subsequently used in the main survey.

The main survey was conducted using convenience sampling As mentionedabove, the method implemented in the main survey was the quantitative approach,responses obtained from bank staffs were collected via their personal e-mails Surveyswere created by Google Form, the feedbacks then were used to assess the scale and testthe proposed hypotheses by means of Cronbach’s alpha and exploratory factor analysis(EFA)

Based on the rule of five observations per parameter estimated of Bollen (as cited

in Nguyen, 2013), the sample size of minimum 280 participants is needed for the surveythat consists 56 parameters Therefore, 350surveys were distributed to employees of tenbanks located in Ho Chi Minhcity The respondents included employees from variousdepartments were selected to assure that the variety and reliability of the survey wereachieved

There were 322 feedbacks sent back to the collector However, there were 9responses which were eliminated due to missing value or straight-lining (respondentsselect the same answer choice for all questions) Consequently, there were 313 validfeedbacks usable for further data analysis

4.2 Questionnaire design

A questionnaire of the survey included 56 main questions The questionnairesurvey used Likert scale seven levels (Totally disagree – totally agree) to measure bankstaffs’ attitude The remaining part contained questions related to gender, age, position,education…to collect biodata The questionnaire was originally composed in English,they were then translated into Vietnamese to assure that every question is wellunderstood by readers Finally, the responses were back translated to serve as materialsfor further data analysis All the questions were arranged randomly instead of orderly toprevent CMV (Common method variance) from taking place After respondents finishedfilling the questionnaire, the collector immediately checked for any missing fields andasked respondents to add in order to avoid insufficient information from any respondent.The questionnaire surveys in both languages are shown in Appendix 1 and Appendix 2

4.3 Measurement scale

Affective commitment was measured with Affective Commitment Scaledeveloped by Allen and Meyer (1990) included 8 items A seven-point Likertscale wasemployed from strongly disagree to strongly agree Interviewees showed their feelingsabout their banks by selecting one appropriate choice The higher the average score of aspecific dimension, the greater the level of that type of affective commitment

Emotional Intelligence was examined with self-report Wong Law EmotionalIntelligence Scale (WLEIS,Wong& Law, 2002) Applying the seven-point Likert, thescale consisted of 16 questions belong to four dimensions that were consistent with

Mayer and Salovey’s (1997) definition of Emotional intelligence However, these fourdimensions were combined into a single emotional intelligence measure rather than amultidimensional one because the current study was interested in the overall construct toexamine how the magnitude of the examined relationship changes with and without thepresence of emotional intelligence.

Job satisfaction was tested with the Job Satisfaction Survey (Spector, 1997)including 8 dimensions: Pay, promotion, supervisor, fringe benefits, operating conditions,co-workers, nature of work and communication The scale comprised 32 items, eachcategory consisted 4 statements which were scored on a seven-point Likert scale

4.4 Data analysis

Two first-order constructs (affective commitment and emotional intelligence) andone second-order construct (job satisfaction) were examined in the study The scales wererefined and validated using Cronbach’s Alpha and EFA from the pilot sample of 158participants to gain more appropriate measurements, the results after being refined werethen used for the main survey Based on the data collected from main survey, Cronbachalpha and EFA were once again employed with 313 respondents to assess how much themeasurement model fits the research data

In the next stage, three-step hierarchical regression was applied to analyse therelationships of job satisfaction, emotional intelligence and affective commitment At thefirst step, all facets of job satisfaction were input to test its relationship with affectivecommitment, the result demonstrated the first hypothesis Emotional intelligence was

added at the second step to investigate how much affective commitment variation isexplained by its presence, the second hypothesis was thus examined Finally, the entry ofthe interaction between job satisfaction facets and emotional intelligence revealedwhether emotional intelligence acted as a mixed moderator in the job satisfaction-

affective commitment association, by observing the change in R2 the third hypotheseswere proved because the result was significant

5.Result 5.1 Sample profile

350 online surveys were directly delivered to the employees (using Google Forms)via personal e-mails at identified banks in Ho Chi Minh city The employees were asked

to present their attitude towards the statements related to job satisfaction, affectivecommitment and emotional intelligence The seven point Likert scales were designed byselecting fixed choices that would enable interviewees to measure how much they agreewith a given issue There were 313 responses collected which were applicable for dataprocessing

Descriptive statistics provided detailed information of the collected data: Femalelabour force accounted for 59.4% of the sample, 18.8% higher than male participation.Besides, while more than a half of interviewees aged under 30 dominated the survey with56.2%, bank staffs aged from 30 to 40 and over 40 just occupied approximately 30% and15% respectively Most of the respondents working in banking sector had bachelor’degree or lower, about 235 out of 313, whereas this figure of master’s degree or higher

was 78 Finally, the bank type was classified into 3 categories namely State-owned banks,joint-venture and foreign banks and others with the percentage of participants for eachtype were 32.9, 26.2 and 40.9 respectively.

Table 1: Demographic statistics

for anomalies or missing values using “frequency” function Cronbach Alpha and EFAthen were used to measure the internal consistency, analyse the reliability and to refinethe scale The result revealed that the job satisfaction’s facet named “Operatingconditions” should be excluded due to low Cronbach Alpha coefficient 0.269 because thestudy of Nunnally and Bernstein suggested that Cronbach Alpha coefficient greater than0.6 is acceptable (as cited in Nguyen, 2013) The result table is illustrated in Appendix18.

The output gained from EFA analysis also showed satisfactory indexes: Affectivecommitment was extracted 1 factor, emotional intelligence was extracted 4 factors andjob satisfaction was extracted 7 factors after the “Operating conditions” was deleted Thenumber of factors in each specific construct were identified at a greater than 1Eigenvalue All factor loadings, extracted variances were higher than 0.5, KMO of threescales were greater than 0.5 and Bartlett’s tests were significant, convergent as well asdiscriminant validity was achieved In conclusion, the final scale which then would beused for the main survey has three constructs and 52 questions eventually The completedsurvey questionnaire can be seen in appendix 2

The process was repeated for the main survey of 313 responses Table 2 indicatesthe Cronbach’s Alpha coefficient which is correspondent with each factor The resultpointed out all the alphas were greater than 0.6 together with high item-total correlation(>0.5), a more detailed Cronbach Alpha analysis performance is presented in appendix ofthe study Consistent with the refinement of pilot study, EFA analysis (principle

component with varimax rotation) of main survey resulted in a similar assessment which

is depicted in Appendix 22, Appendix 23 and Appendix 24 The measurement modelachieved a good fit to the data

Table 2: Cronbach’s Alpha analysis

EI Emotional intelligence

regression For this reason, these variables should be converted into mean-deviatedvariables whose means = 0 and variances ≠ 1 before being input into Hierarchicalregression analysis in the next step.

Table 3: Correlations

1.Affective

1 commitment

** Correlation is significant at the 0.01 level (2-tailed).

5.3.2 Hierarchical regression analysis

In order to test the main effect and discover whether emotional intelligence moderates therelationship between job satisfaction’s facets and affective commitment, hierarchalregression analysis was conducted Linear regression was the next step after correlation,

variables.

Before regression conduction, assumptions were checked to assure that the results gainedfrom data analysis are valid This study assessed the assumptions related to residuals byusing histogram to test whether the errors of the regression are approximately normallydistributed; and; scatterplots to illustrate homoscedasticity of the data The outputspointed out that all these assumptions were met These can be found in appendix 25 and26.

As indicated in the first step, the value of R2 was 0.825 which revealed that seven facets

of job satisfaction explain 82.5% of Affective commitment All Beta coefficient of thefacets showed positive values at p<0.001 Except for the hypothesis H1e which was notsupported due to low Cronbach alpha coefficient of operation condition, the resultobtained from step 1 supported the remaining hypotheses of the study that pay,promotion, supervisors, fringe benefits, co-workers, nature of work and communicationpositively related to affective commitment respectively

At the second step, emotional intelligence was added, the coefficient of determinationincreased to 0.019 which was found to be significant The 1.9% higher in the variation ofdependent variable is explained by emotional intelligence This finding was consistent tothe statement of hypothesis 2 that emotional intelligence is positively related to affectivecommitment

By adding the interaction of emotional intelligence and various dimensions of jobsatisfaction in the third step, the mixed moderating role of emotional intelligence in thelast hypotheses including H3c, H3f, H3g were proved Withdrawing from the second and

the third step of regression analyses, emotional intelligence can be concluded to serve as

a mixed moderator (rather than a pure moderator) for job satisfaction’s facets andaffective commitment because it was also found to have direct influence on thedependent variable as mentioned

The output points out that the additional 2.7% in R2 was significant at 0.001 level whichmeans emotional intelligence has a considerable contribution to enhancing therelationship between job satisfaction’s facets and employees’ affective commitment Inmore detail, emotional intelligence exposed its mixed moderating impacts on the relationbetween affective commitment and employees’ satisfaction with supervisors (H3c), co-workers (H3f) and nature of work (H3g), however, this impact was found to be non-significant with pay (H3a), promotion (H3b), fringe benefit (H3d) and communication(H3h) H3e was excluded due to the elimination in Cronbach alpha analysis

As expected, the last step of table 4 shows that emotional intelligence moderated thesupervisors-affective commitment and co-workers-affective commitment relationpositively On the contrary, the moderating effect on nature of work-affectivecommitment relation was not in the predicted direction In general, emotional intelligence

is likely to boost the magnitude of the relationship between affective commitment andsatisfaction with human factors which are supervisors and co-workers rather thansatisfaction with the work itself which is nature of work Under high emotionalintelligence level, the effect of satisfaction with nature of work on affective commitmentwas negative and stronger, compared with it was under low emotional intelligence level

Table4: Hierarchical Regression Analysis with Emotional intelligence as a moderating variable

Supported H1c: Supervisors  Affective commitment

Supported H1d: Fringe benefits  Affective commitment

Supported H1e: Operation condition  Affective commitment

Supported H1f: Co-workers Affective commitment

Supported

H1g: Nature of work  Affective commitment

Supported H1h: Communication  Affective commitment

Supported H2: Emotional intelligence  Affective commitment

Supported H3a: Moderating effect of Emotional intelligence on Pay - Affective Not supported

commitment relation.

H3b: Moderating effect of Emotional intelligence on Promotion - Not supported

Affective commitment relation.

H3c: Moderating effect of Emotional intelligence on Supervisors - Supported

Affective commitment relation.

H3d: Moderating effect of Emotional intelligence on Fringe benefits - Not supported

Affective commitment relation

H3e: Moderating effect of Emotional intelligence on Operation Not supported

condition - Affective commitment relation

H3f: Moderating effect of Emotional intelligence on Co-workers - Supported

Affective commitment relation.

H3g: Moderating effect of Emotional intelligence on Nature of work - Supported

Affective commitment relation.

H3h: Moderating effect of Emotional intelligence on Communication - Not supported

Affective commitment relation.

5.4 Discussion

This study examines the relationship between affective commitment and job

satisfaction’s facets as well as between affective commitment and emotional intelligence

It also explores the mixed moderating effect of emotional intelligence on the mentioned

associations The results reveal that affective commitment is positively affected by

satisfaction with pay, promotion, supervisors, fringe benefits, co-workers, nature of work and communication Similarly, emotional intelligence is concluded to have positive influence on affective commitment Moreover, emotional intelligence is found to improvethe links between affective commitment and satisfaction with co-workers and with

supervisors yet reduce the impact of satisfaction with nature of work on affective

commitment The outcomes offer some considerable contribution in terms of theoretical and empirical perspective

5.4.1 Theoretical implication

The study confirms the positive relationship between domains of job satisfaction and employees’ affective commitment which is consistent with prior research (Boles et al., 2007; Shalini, 2013) In line with the findings of Abraham (2000) and Gu¨leryu ¨z et al (2008), this study once again suggests that emotional intelligence has a direct positive impact on affective commitment Hence, the current study provides a test of the theories

in an Asian culture background to support the generalizability Moreover, in the context

of Vietnam banking industry, this study is among the first research which examined the three constructs simultaneously

Additionally, this research contributions to the literature on emotional intelligence by

demonstrating its moderating role on the association between job satisfaction and affectivecommitment The results extend the literature in organizational settings by pointing out how the relation of affective commitment and satisfaction with co-workers

as well as the relation of affective commitment and satisfaction with supervisors are strengthened by the presence of emotional intelligence In contrast, the relation of

affective commitment and satisfaction of nature of work is weakened when emotional intelligence effect is added According to Goleman (as cited in Beigi & Shirmohammadi,1995), emotionally intelligent people are self-motivated which means everything they do

is driven by their inner ambition Therefore, they never stop looking for new challenges, thus, they are likely to leave for other challenging jobs once the nature of current work isclearly understood

5.4.2 Managerial implication

The findings of the research provide a closer understanding of job satisfaction’s crucial role in explaining the variance of employees’ affective commitment Once different aspects of the job are well managed, it will be easier for policy makers to devise a better behavioral forecast, to well manage the stability of labor force which can further magnifythe productivity of employees

Furthermore, the discovery of emotional intelligence’s influence on the relationship widens the administrator’s perception towards employee competency assessment To a large extent, once emotional intelligence is taken into consideration, it can assist policy makers to enhance the human resource management in terms of selecting and

appointing suitable bankers for every specific position

As mentioned above, emotional intelligence was shown to help in strengthening the linkbetween staff’s satisfaction with supervisors and co-workers and commitment because