Does Emotional Intelligence make the Difference?
The Relationship Between Leaders’ Emotional Intelligence, Subordinates’
Empowerment and Organisational Citizenship Behaviour, and the Moderating
role of Subordinates’ Work Locus of Control.
Merel Nieuwkerk | 10998462 | 03-07-2017 | Executive programme in management studies Leadership & management | University of Amsterdam
First supervisor: dr. A.H.B. de Hoogh Second supervisor: prof. dr. D.N. den Hartog
2
Statement of originality
This document is written by Merel Nieuwkerk who declares to take full responsibility for the contents of this document.
“I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.”
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
Emotional intelligence is the ability to perceive, facilitate, understand and manage emotions. In this study, a mediated moderation model proposing the moderating role of internal locus of control and the mediating role of empowerment in the relation between leader’s emotional intelligence and subordinates’ organizational citizenship behaviour is examined. A valuable contribution of this research is the measurement of emotional intelligence using three different methods: subordinate-rated, leaders’ self-assesment and situational judgement test. Employees of a large financial institution in the Netherlands filled out the survey (leader N = 63, subordinate N = 178). Leaders filled out the Wong and Law Emotional Intelligence Scale (WLEIS) to self-asses their emotional intelligence, and they conducted the Situational Test of Emotional Management (STEM). Furthermore, they rated the organisational citizenship behaviour of their subordinates. The employees rated their leader’s emotional intelligence, and their own psychological empowerment and work locus of control. As expected, the model whereby subordinates rated their leader’s emotional intelligence, displays a positive relationship between emotional intelligence and empowerment, and empowerment and organisational citizenship behaviour. A full mediation has been confirmed. Moreover, the model including the leader self-rated emotional intelligence shows a relationship between emotional intelligence and organisational citizenship behaviour. Following, theoretical and practical implications of the findings are discussed, and areas for future research are identified.
Keywords: Emotional intelligence (EI); psychological empowerment; organizational
4
Table of contents
1. Introduction ________________________________________________________________________________________ 6 2.1 Emotional intelligence __________________________________________________________________________ 9 2.2 Organisational citizenship behaviour ________________________________________________________ 15 2.3 Emotional intelligence and organisational citizenship behaviour _________________________ 18 2.4 Emotional intelligence and empowerment __________________________________________________ 20 2.5 Empowerment and organisational citizenship behaviour __________________________________ 22 2.6 Emotional intelligence, empowerment and organisational citizenship behaviour _______ 24 2.7 Work locus of control __________________________________________________________________________ 25 3. Research design ___________________________________________________________________________________ 28 3.1 Method __________________________________________________________________________________________ 28 3.2 Procedure ______________________________________________________________________________________ 28 3.3 Sample __________________________________________________________________________________________ 29 4. Measures ___________________________________________________________________________________________ 31 4.1 Emotional intelligence _________________________________________________________________________ 31 4.1.1 Leader self-‐reported (WLEIS) _____________________________________________________________ 31 4.1.2 Leader ability situational judgement scored (STEM) ____________________________________ 32 4.1.3 Subordinates’ score of leader (WLEIS) ____________________________________________________ 34 4.2 Empowerment _________________________________________________________________________________ 35 4.3 Organisational citizenship behaviour ________________________________________________________ 35 4.4 Work locus of control __________________________________________________________________________ 36 4.5 Control variables _______________________________________________________________________________ 36
5
5. Data analysis and results ________________________________________________________________________ 38 5.1 Procedure ______________________________________________________________________________________ 38 5.2 Team aggregation reliability __________________________________________________________________ 39 5.3 Preliminary analysis ___________________________________________________________________________ 40 5.4 Hypotheses testing ____________________________________________________________________________ 42 5.4.1 Testing Model A (individual-‐level) ________________________________________________________ 42 5.4.2 Testing Model B (team-‐level/ self-‐rated) _________________________________________________ 45 5.4.3 Testing Model C (team-‐level/ SJT) _________________________________________________________ 47 6. Discussion _________________________________________________________________________________________ 52
6.1 Theoretical implications ______________________________________________________________________ 52 6.2 Managerial implications _______________________________________________________________________ 57 6.3 Strengths, limitations and future research ___________________________________________________ 58 7. Conclusion _________________________________________________________________________________________ 61 References ____________________________________________________________________________________________ 62 Appendices ___________________________________________________________________________________________ 70 Appendix I. Cover letter for leader _______________________________________________________________ 70 Appendix II. Survey leader ________________________________________________________________________ 71 Appendix III. Survey subordinate _________________________________________________________________ 80 Appendix IV. Principal component analysis _____________________________________________________ 84 Appendix V. Hypotheses testing (incl. all control variables) ___________________________________ 87 Testing model A (individual-‐level) ______________________________________________________________ 87 Testing model B (team-‐level/ self-‐rated) _______________________________________________________ 89 Testing model C (team-‐level/ SJT) _______________________________________________________________ 90
6
1. Introduction
Today, we are on the cusp of a mayor change. Some would even call it a revolution. The rapid evolution of technology is affecting the world, as we know it. It will continue to develop and fundamentally change the way we live, and impact the way we work. Developments in artificial intelligence might enable computers to do a great variety of jobs previously reserved for humans. Certain forms of labour may very well disappear for good. While this transformation may be inevitable, technological progress will not deprive us from our emotions. It is our ability to empathise and our emotional intelligence sets us apart from automated processes. These skills are needed to ensure that individuals and societies are resilient and adaptable in the context of change. Indeed, Brooks, and Naffukho (2006), argue that it is in these qualities that human labour shall reside in the future, and our leaders should be equipped to lead within this emotional dimension.
In recent years, emotional intelligence (EI) has gained a prominent character in literature and has surfaced among social and organizational psychologists (see, e.g., Fineman, 1993; Salovey & Mayer, 1997; Schutte, Malouff, Bobik, Coston, Greeson, Jedlicka, & Wendorf, 2001). Despite its appearance in managerial publications, empirically based knowledge about EI is limited. Even though, some practitioners and researchers are confident about the importance of EI in organisations, there has been much debate about the concept and theory of EI, and how it should be assessed (Landy & Conte, 2004; Zeidner, Roberts, & Matthews, 2002). This discussion is duet o the fact that previously used measures of EI proved to score low on reliability and especially on discriminant validity (Conte, 2005). Only recently, researchers are starting to identify valid EI measures (Ciarrochi, Chan, & Caputo, 2000; Mayer, Caruso, & Salovey 1999; Schutte et al., 2001) using non-self reported measures of EI, and knowledge on the effects of leader EI is accumulating.
7
EI of leaders is known to lead to positive interactions with subordinates. Antonakis, Ashkanasy, and Dasborough (2009), and Day and Carroll (2004) note that leaders who can perceive their subordinates’ emotions and understand their impact on others, and can manage their own emotions and have empathy for others, will have a greater probability of providing effective leadership. Subordinates are more likely to feel understood, encouraged and supported by their EI leader. Subsequently, subordinates will feel more worthwhile, influential and effective (Kark, Shamir, & Chen, 2003), and ultimately more empowered. Yen, Lin and Tai (2004) note that psychological empowerment may boost one’s personal motivation and stimulate organisational citizenship behaviour (OCB). Thus, when employees feel empowered they believe they are capable and able to make a difference. Therefore, they might be inclined to put in extra effort, which can be considered as OCB. However, there is little empirical evidence in the literature about the relationship between the EI of the leaders, follower empowerment and OCB.
Scholars indicate that subordinates’ responses to leadership might differ based on subordinates’ personality (Ehrhart & Klein, 2001). The personality trait ‘internal locus of control’ forms a source of psychological empowerment, resulting in individuals who believe they can exert greater effort on work tasks (Koberg, Boss, Senjem, & Goodman, 1999). De Hoogh and Den Hartog (2009) suggest that: “individuals with a high internal locus are more likely to see themselves, rather than the environment or others such as their leaders, as the cause of things that happen to them and are less likely to feel the need for leaders’ guidance” (p. 10). Therefore, the present study extends existing research by examining the moderating role of work locus of control (WLOC) in the relationship between leader’s EI, subordinates’ empowerment and OCB. Expected is that he positive relationship between leader’s EI and subordinates’ outcomesis stronger when subordinates’ internal locus of control is lower. In
8 Internal locus
of control
(subordinate)
addition, in respons to suggestions of Wong and Law (2002), compared is the validity of three different measures of emotional intelligence; namely self-report and subordinates’ ratings of EI as well as a measure of emotional intelligence-ability (emotion management).
Knowledge of leaders’ EI in relation to behaviour of employees is of relevance for human capital optimization to cope with the challenges of the future. Consequently, the managerial implications of this study are for human resource development purposes and outcomes might be applicable to the selection, promotion and training of leaders to cultivate EI and thereby stimulate positive subordinate outcomes, such as empowerment and OCB. This research aims to answer the following question:
What is the relationship between a leader’s emotional intelligence, empowerment and organisational citizenship behaviour, and the moderating role of subordinates’ work locus of control?
The conceptual model to support this research question can be found in Figure 1.
Figure 1 Conceptual model Emotional intelligence (leader) Empowerment (subordinate) Organisational citizenship behaviour (subordinate)
9
2. Literature review
This research is based on the expectation that there is a relationship between a leader’s EI and subordinates’ empowerment, which in turn influences subordinates’ OCB. The personality of subordinates is likely to influence the effect of leaders, and therefore WLOC is included as a moderating variable. In the following section, EI is defined and an introduction to OCB is given. Subsequently, hypotheses on the relationships will be provided.
2.1 Emotional intelligence
The theory of EI is rooted in the concept of social intelligence. Thorndike (1920), an American psychologist, defined social intelligence as “the ability to understand and manage men and women, boys and girls—to act wisely in human relations” (p. 228). Thorndike (1920) suggested that social intelligence, combined with intellectual intelligence, is a predictive factor for individual success. Even though the theory of social intelligence resembles features of EI, the construct-name emotional intelligence, referring to ones ability to deal with his or her emotions, has only been introduced in 1990. Salovey and Mayer (1990) specified EI as “the subset of social intelligence that involves the ability to monitor one’s own and others’ feelings and emotions, to discriminate among them and to use this information to guide one’s thinking and actions” (p. 189).
Later, EI has been carried forward by the definition of Davies, Stankov and Roberts (1998). They qualitatively summarized the EI literature, and consecutively developed a four-dimensional definition, which will be used in this study. The dimensions consist of: (1) appraisal and expression of emotion in oneself, (2) appraisal and recognition of emotion in others, (3) regulation of emotion in oneself, (4) use of emotion to facilitate performance.
‘Appraisal and expression of emotion in oneself’ refers to the understanding of deep emotions
10
acknowledge their emotions better than most people (Law, Wong, & Song 2004). Watson and Tellegen (1985) were the first to describe a two-dimensional construction of sensing and understanding emotions by (1) feelings and (2) thoughts.Later, others (Booth- Butterfield & Booth-Butterfield, 1990; Mayer & Stevens, 1994; Swinkels & Guiliano, 1995) noted that this aspect of EI not only exists of understanding of feelings, but also about the thoughts concerning that feeling, and thus the reflective factor. All are considered as relevant for this dimension. The second dimension, ‘appraisal and recognition of emotion in others’, relates to the ability to perceive and understand emotions of others. It can be characterised as sensitivity to emotions of others and being more capable of predicting others’ emotional responses, also known as empathy (Salovey & Mayer, 1990). Thirdly, the element ‘regulation of emotion’ is paired with being able to have better control over emotions and to recover more rapidly from psychological distress (e.g. return to normal state after revel or disordered). It also includes the ability to “alter the affective reactions of others” (Davies et al., 1998, p. 991); e.g., the ability to calm distressing emotions in other individuals (Salovey & Mayer, 1990). Lastly, the ‘use of emotion to facilitate performance’ is considered as the ability of one to “make use of his or her emotions by directing them toward constructive activities and personal performance” (Law et al., 2004, p. 484). Overall, the four dimensions focus on two elements: the emotion within ’self’, and emotion of ‘others’.
In a review of the EI literature, Ciarrochi et al. (2000) mentioned: “while the definitions of EI are often varied for different researchers, they nevertheless tend to be complementary rather than contradictory” (p. 540). Regardless of the existence of one definition of EI, discussion about the construct and its dimensions is on-going. Scholars have lacked consensus upon the construct of EI and literature has been countered; is EI an ability or a trait?
11
traditional standards for intelligence measures, given its roots in social intelligence. According to conceptual criterion of intelligence, EI can be distinguished as a set of abilities of a person, instead of a personality trait or preferred way of behaving. Intelligence has been defined by Gardner (1993) as “the ability to solve problems, or to fashion products, that are valued in one or more cultural or community settings” (p. 7). The concept of EI as ability focuses on the set of abilities that people use to understand, regulate and exploit their emotion. Doing so reflects the ability to manage ones emotions. This means that some have the cognitive capability to do so, while others do not have these skills. Both Mayer and Salovey (1997, 2000), and Wong and Law (2002) have proven in multiple studies that EI can be characterized as ability. They show that EI is different from personality traits, such as the Big Five personality dimensions. Furthermore, they have build empirical foundation that EI relates to other abilities, but is significantly distinct from general mental abilities. It is thus independent from other types of intelligence. Lastly, they established that EI is indeed an intelligence factor, as in nature one is able to develop this intelligence over time, and thus increases over time by age and experience. Thereby proving “EI qualifies as ability rather than a personality trait” (Law et al., 2004, p. 485).
On the other hand, others (Bar-on, 1997; Goleman, 1995; Petrides & Furnham, 2001) label EI as “an umbrella term that encompasses a constellation of personality traits, affect, and self-perceived abilities, rather than actual aptitude” (Joseph, Jin, Newman, & O’ Boyle, 2015, p. 299). This type of EI has been defined as a trait or mixed EI. Within this ‘umbrella’ the definition of mixed EI ranges between being identical to ability EI and overlapping Big Five components. For example, Bar-on’s definition encompasses similarities to the definition of ‘ability EI’ and include five key components: (a) the ability to recognise, understand, and express emotions and feelings; (b) the ability to understand how others feel and relate with
12
them; (c) the ability to manage and control emotions; (d) the ability to manage change, adapt, and solve problems of a personal and interpersonal nature; and (e) the ability to generate positive affect and be self motivated. Goleman (1998, p. 21) describes mixed EI as basic abilities, which are the foundation for “specific capabilities of leadership self-awareness, managing emotions, motivating others, showing empathy, and staying connected”. Also Goleman (1998) describes an overlap between ability and trait EI, by mentioning that, although being a trait, EI can be learned and developed. Later, Mayer, Roberts and Barsade (2008) have made an effort to summarize the construct of ‘mixed EI’ and formulated four aspects. Two of these aspects are (a) achievement motivation, and (b) control- related qualities, such as impulse control and flexibility. Joseph et al. (2015) mention the overlap of these aspects with the consciousness dimension of the Big Five (Roberts, Chernyshenko, Stark, & Goldberg, 2005). The two other dimensions are (c) assertiveness gregariousness (known as two facets of Extraversion; Costa and McCrae, 1992), and (d) self-related qualities. In this manner, Mayer et al. (2008) suggest, “mixed EI overlaps with conscientiousness and extra- version of the Big Five, and other self-related qualities such as general self-efficacy” (Joseph et al., 2015, p. 299). Overall, one can say that the disparate terminology even within the so-called ‘mixed EI’ does not contribute to definitional clarity.
In line with the lack of consensus amongst researchers about the definition of EI are the conflicting results about the elusive construct ‘emotional intelligence’. Even though numerous scholars have proposed that EI is classified as intelligence, in the traditional sense, contemporary research and theorizing lack any conceptual model of intelligence within which the construct might be placed (Davies et al., 1998). For example, researchers who tried to establish, with considerable research, that social intelligence is a part of intelligence, but distinctive from other types of intelligence, have found unsatisfactory results (Ford & Tisak,
13 1983; Brown & Anthony, 1990; Carroll, 1993).
The lacking of a common construct is one of the reasons validity in EI research is flawed (Conte, 2005). The unclear definition is accountable for the difficulty to examine construct validity, because researchers are not precise about what is being measured. Similarly, content validity is deficient due to vague theoretical development for a variety of measures and thus missing validity evidence. Reliability, however, has been demonstrated amongst several EI measures. Both self-report EI measures, as scales for ability-based measures do have acceptable internal consistency, but do not provide evidence whether it is measuring other more established constructs, such as general intelligence or personality dimensions (Day & Carroll, 2004; Conte, 2005).
In search of establishing EI as a validated construct several researchers have reviewed different measurement instruments of EI. Mayer and Salovey (1999) noted that self-report measures appear to assess existing personality characteristics or possibly emotional competencies, but it does not assess intelligence. They also mention that EI has the discriminant validity lacking in measures of social intelligence (Mayer & Salovey, 1993, 1997). Davies et al. (1998) review several self-report instruments that measure different factors of the four dimensions of EI as proposed by Salovey and Mayer (1990). They examine these measurement tools on two levels (personality and intelligence) and provide a display of their search for a more objective measurement method. They conclude by saying that measures are too similar to entrenched personality traits and objective measures are to be considered unreliable, in line with the conclusions about self-reports of Mayer et al. (1999). Conte (2005) also reviewed four EI measures: Emotional Competence Inventory (ECI), Bar-On Emotional Quotient Inventory (EQ-i), and Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT). Conte (2005) agrees with Davies et al. (1998) that some measures
14
demonstrate adequate internal consistency reliability, but do lack content validity evidence. Harms and Credé (2010, p. 7) describe that “although there have been efforts to create psychometrically valid measures of EI, there is no single universally accepted measure of EI”. Accordingly, this research aims to contribute by comparing the validity of three different and recently established measures of EI, namely self-report of the leader using the Wong and Law Emotional Intelligence Scale (WLEIS), and subordinates’ ratings of EI using the WLEIS, as well as the Situational Test of Emotional Management (STEM) filled out by the leader measuring the dimension ’emotion management’ of ability EI. This way, the validity of the different measures can be established, and outcomes can be matched to ensure higher reliability.
Conforming to the difficulties of authenticating the construct and validity of EI, the link between EI and its outcomes, and link to leadership has been an area of major controversy (Cherniss, 2010; Murphy, 2006; Joseph et al., 2015). There has been no scarcity in research investigating the relationship between EI and positive outcomes since Goleman (1995) made the concept of EI widely popular. Researchers have indeed found positive affiliation for EI with school and work performance (Van Rooy & Viswesvaran, 2004; Joseph & Newman, 2010; O’Boyle, Humphrey, Pollack, & Hawyer, 2008). For example, Zeidner, Matthews and Roberts (2004), mention that individuals with greater EI are likely to be better at communicating and motivating (Goleman, 1998), better at teamwork (Mayer & Salovey, 1997). Moreover, more emotional intelligent leaders are able to positively affect relationships in work settings, which consecutively impacts organisational commitment of subordinates (Cherniss, 2010). Thus, leaders’ EI has been noted as making a difference in effectiveness of leadership results (Harms & Credé, 2010).
15
Both Antonakis et al. (2009), and Day and Carroll (2004) claim that leaders with a high level of EI will have a greater probability of providing effective leadership. Effective leadership is defined as “a leader’s performance in influencing and guiding the activities toward achievement of goals” (Judge, Bono, Ilies, & Gerhardt, 2002, p. 767). “In order to provide this effective leadership, a leader has to be able to manage complex social and personal dynamics, all centred in the concept of emotional intelligence” (e.g., Cann, 2004; Mayer, DiPaolo, & Salovey, 1990; Weisinger, 1998). George (2000) asserts that by “accurately identifying how subordinates feel, leaders better appraise and influence subordinates’ emotions so they are supportive of leader’s goals and objectives, thus insuring a shared vision” (Zeidner et al., 2004, p. 387). He also mentions that leaders can use emotions to generate enthusiasm, and create an environment of trust and cooperation. Some, like Goleman (1995) and Kemper (1999), have claimed that EI is that what distinguishes outstanding from average leaders by more than 90%. However, Harms and Credé (2010) note that “there has been widespread scepticism of the link between EI and leadership outcomes” (Antonakis et al., 2009; Landy, 2005; Locke, 2005) and many studies have failed to find significant relationships between EI and leadership (e.g., Moss, Ritossa, & Ngu, 2006; Weinberger, 2004). Therefore, this study adds to literature by examining the positive outcomes of leader’s EI on subordinates’ behaviour.
2.2 Organisational citizenship behaviour
The OCB construct has received increased research attention over the past years, as organizations expect employees to go beyond their formal job descriptions to cope with the challenges of the future (Parker, Bindl, & Strauss, 2010). Researchers have distinguished various forms of OCB (Podsakoff, MacKenzie, Paine, & Bachrach, 2000) and have build the foundation for further research by suggesting that OCB has a positive impact on
16
organizational effectiveness and individual performance evaluation (Avila, Fern, & Mann, 1988; MacKenzie, Podsakoff & Fetter, 1993; Podsakoff & MacKenzie, 1994; Van Scotter & Motowidlo, 1996; Walz & Niehoff, 1996). In a meta-analysis by Podsakoff, Whiting, Podsakoff, and Blume (2009), the positive consequences at individual level are described: better performance evaluations, lower turnover intentions, and absenteeism. On organisational and unit-level, they have found OCB to be positively related to a variety of effectiveness measure, such as productivity, efficiency, costs and profitability. While researchers are finding accumulated evidence in support of the OCB construct, discussions and questions about the concept and the content have arisen in the last years.
Originally, Organ (1988) defined OCB, or extra- role behaviour, as “individual behaviour that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization” (p. 4). It concerns ones voluntary altruistic or helpful acts to enhance the organization. This definition of OCB is build upon earlier work of Barnard (1938) who refers to “willingness to cooperate” and Katz (1964) who made a distinction between dependable role performance and “innovative and spontaneous behaviours”. Later, Organ (1997) adapted his definition to “performance that supports the social and psychological environment in which task performance takes place” (p. 95). Hereby, he does not mention the formal reward system, but emphasizes the importance of the contextual performance.
OCB is a distinct construct from in-role behaviours as theorized by a number of researchers (Barnard, 1938; Katz, 1964; Katz & Kahn, 1978) and later demonstrated by others (MacKenzie, Podsakoff, & Fetter, 1991; Williams & Anderson, 1991). Even though some question this differentiation, leaders and employees differentiate between in- or extra role behaviours in day-to-day business. Van Dyne and LePine (1998) mention that absence of
in-17
role behaviour leads to negative consequences, or reprimands, while extra-role behaviour is positive and seen as open to choice. They note that: “supervisors value extra-role behaviour because dynamic environments do not allow anticipation or specification of all desired employee behaviours” (p. 108). Thus, from a managerial perspective, OCB is valued and seen as important for business. In line with this, OCB is often considered as important for cultivating organisational success (Barnard, 1938; Katz & Kahn, 1966).
Furthermore, there are disparate approaches to the dimensions of OCB. Organ (1988) categorized OCB in five dimensions: Altruism (help others to resolve problems), conscientiousness (working hard to exceed standards), sportsmanship (obeying rules and endure imperfections), courtesy (prevent problems and considerate of others), and civic virtue (attentive and proactive). Van Dyne, Cummings and Parks (1995) suggested that OCB could be characterised into two types: affiliative and challenging. Displays of challenging
behaviours are change oriented, such as actively trying to improve conditions, and
maintaining high standards. Affiliative behaviours are other-oriented, such as helping others and defending organizational objectives (Van Dyne et al., 1995). These are spontaneous, voluntary contributions, demonstrating the “willingness” of employees to contribute to the organizational goals, beyond the role expectations and contractual agreements. The two dimensions are often noted as ‘voice’ and ‘helping’. Employee initiative and innovativeness, covered by OCB, can provide the organizations with adaption and innovation that are necessary for long-term survival and growth (Katz, 1964; Van Dyne, Graham & Dienesch, 1994; Yadav & Punia, 2016). As OCB can contribute to sustainable business development, it is important to understand the antecedents of OCB. In the following chapters the influence of leaders, motivational aspects and effect of personality of the subordinate on OCB is discussed.
18
2.3 Emotional intelligence and organisational citizenship behaviour
“Given the importance of OCB, organisations should make every effort to motivate their employees to perform these behaviors”(Koning & Van Kleef, 2015, p. 490). As OCB is described as extra-role behavior and is not included in formal job requirements, it is essential for organizations to find other ways to motivate employees to engage in OCB. Earlier, researchers have addressed how OCB can be stimulated in numerous ways. This research concentrated on the employee, and its characteristics and perceptions that are conditions for OCB (for example, Bateman & Organ, 1983; Smith, Organ, & Near, 1983). According to Organ and Ryan (1995) ‘morale’ is the underlying characteristic in factors such as perceived fairness and perceived leader support, contributing to OCB. Podsakoff et al. (2000) suggest that other variables, such as trust, may comprise employee morale.
Research that has been conducted later, focused on the influence of external factors in promoting employee OCB, such as leaders and leadership. Specific types of leadership, such as transformational and charismatic leadership have been found to impact OCB. The leadership behaviors employed by these leadership styles, articulating vision and setting goals, have been found to promote OCB behavior of employees (e.g., Chen, Tsui, & Farh, 2002; Podsakoff, MacKenzie, Moorman, & Fetter, 1990). Hereby illustrating that leaders take part in stimulating OCB. Podsakoff et al. (2000) mention that supportive leadership behaviour is most strongly related to OCB and “may underlie the effects of perceived organisational support on OCB”. Supportive behaviour of the leader can be displayed through recognizing and understanding emotions of subordinates and thereby leaders’ EI should be considered. Although literature claims EI of the leader generates subordinates’ enthusiasm, trust and cooperation (George, 2000), research on the part of EI of the leader on OCB is still scarce.
19
Research that has been conducted on the relationship between EI and OCB shows promising results. For example, Yadav and Punia (2016) are more specific on leaders’ EI and state that: “emotional intelligence as such has an exceptional potential to enhance human emotional understanding and cognitive aspects and the ambiance in which an individual’s work is more likely to affect the OCB of employees” (p. 48). According to Organ (1988) the willingness of subordinates to engage in OCB is determined by employees’ feelings or moods during job-related activities.Respectively, accumulated empirical evidence has provided that leaders’ EI can elevate subordinates’ OCB (Modassir & Singh, 2008; Sarlak, Danaee, & Sahafi, 2011; Yunus, Othman, Noormala, & Che Norlia, 2010).
Caza, Zhang, Wang, and Bai (2015) found that the display of leaders’ (sincerity of) emotions affect subordinates responses to the leader. The perceived emotional sincerity is proven to enhance trust in the leader. If leaders thus are able to understand and manage their emotions, one dimension of EI, it affects the subordinates’ trust. According to Fulmer and Gelfand (2012), subordinates’ who trust their leader are more satisfied with the leadership, and as a result they are willing to devote time and attention toward the leaders’ ends (Colquitt, Scott, & LePine, 2007). As well, they are able to focus on work tasks, as they do not need to worry about vulnerability towards the leader (Mayer & Gavin, 2005). Subordinates that trust their leader are more effective in their work, “also in discretionary or extra-role efforts” and thus OCB (Caza et al., 2015, p. 522).
However, research on the direct effect of EI on OCB has not been published in top tier journals. Hence, researchers (e.g. Joseph & Newman, 2010; MacCann & Roberts, 2008) advocate the need for collecting multi-source data, and combining different measures to challenge or endorse these findings. In order to continue with and elaborate on previous research, the following is predicted:
20
Hypothesis 1: Emotional intelligence of the leader is positively related to subordinates’
organisational citizenship behaviour.
2.4 Emotional intelligence and empowerment
Empowerment is a process by which people, organizations, and communities gain influence over issues important to them (Rappaport, 1987). The root constructs of empowerment are ‘power’ and ‘control’. Power and control are motivational and expectancy-belief-states that are internal to individuals (Conger & Kanungo, 1988). Power, motivational-wise, refers to an intrinsic need for determination (Deci, 1975), or confidence in self-efficacy (Bandura, 1986). To empower means ‘to enable’, and any (leadership) strategy or technique that enhances self-determination or belief of self-efficacy of subordinates will make individuals feel more powerful. Thus, Conger and Kanungo (1988) define empowerment as “the process of enhancing feelings of self-efficacy among organizational members through the identification of conditions that foster powerlessness and through their removal by both formal organizational practices and informal techniques of providing efficacy information” (p. 474). In other words, they note that is about motivating self-efficacy.
While Conger and Kanungo (1988) describe empowerment as a single concept, Thomas and Velthouse (1990) propose a multifaceted concept. They describe four cognitions, task assessments, which are seen as having extra motivational effect to an individual’s work role: meaningfulness, competence (similar to efficacy of Conger & Kanungo, 1988), self-determination and impact. Impact refers to the degree an individual feels it is "making a difference" in terms of accomplishing the goal of the assignment. Self-determination, or choice, is the initiation and continuation of work behaviours and processes (Bell & Staw, 1989; Spector, 1986). Deci and Ryan (1985) note that the central aspect of self-determination is the experience of choice, and mention that this results in more flexibility, creativity,
21
initiative, resiliency, and self-regulation. Meaningfulness entails the “value of the task goal or purpose, judged in relation to the individual's own ideals or standards” (Thomas & Velthouse, 1990, p. 672). The last of the four assessments is competence: the belief of an individual that he or she has the capability to perform an activity. Competence is similar to the before mentioned description of Conger and Kanungo (1988) on self-efficacy. Bandura (1977) has studied self-efficacy, and found that it contributes to initiating behaviours, high effort, and persistence when facing obstacles. When individuals are empowered, they establish a sense of “personal mastery or “can do” attitude” (Conger & Kanungo, 1988, p. 475).
Studies on leadership note that the practice of empowering subordinates is a core component of organisational and managerial effectiveness (Bennis & Nanus, 1985; Kanter, 1979, 1983; McClelland, 1975). Conger and Kanungo (1988) conceptualise empowerment as ‘psychological enabling’. They treat empowerment as motivational construct, which concentrates on subordinates’ need for self-determination and personal efficacy. Leaders enable empowerment by creating a climate where employees feel inspired and self-confident (Bass & Avolio, 1993; Bennis & Nanus, 1985; Edwards & Collinson, 2002). Grant (2007) mentions that empowering leaders demonstrate faith in the quality and importance of employee contributions, and assure that they can make a difference, they can promote a sense of ‘meaning’. Overall, leaders are seen as important factor to establish a subordinates’ positive self-concept.
Leaders may empower subordinates directly through interactions with subordinates (e.g. assuring subordinates of their competency) (De Hoogh & Den Hartog, 2009). Recognizing, properly responding to, and being able to influence subordinates’ emotions is crucial for leaders to create high quality interactions (Salovey & Mayer, 1989-90). George even characterizes leadership as an “emotion-laden process” (George, 2000, p. 1046). Indeed, “The
22
ability to manage subordinates’ feelings is seen as a critical leadership function (Humphrey, 2002, 2008) In agreement; Law et al. (2004) mention that the core factors of these interactions are the ‘understanding and regulation of one’s emotion’ as well as ‘understanding of emotion of others’. Because emotional intelligent leaders are able to identify and understand subordinates’ emotions and formulate emotional responses, they may be capable of expressing support, sympathy and thereby generate subordinate confidence (resulting in the empowering factors ‘self-determination’, ‘impact’ and ‘competence’). Thereby, the provide inspiration and a sense of meaning, which might refer to the ‘meaningfulness’ cognition of empowerment (Prati, Douglas, Ferris, Ammeter, & Buckley, 2003). Accordingly, hypothesised is:
Hypothesis 2: Emotional intelligence is positively related to subordinates’ empowerment.
2.5 Empowerment and organisational citizenship behaviour
Due to the positive established work-related outcomes of empowerment, such as increased job-satisfaction, organisational commitment, lower turnover intentions and strain (Seibert, Wang, Courtright, & Kozlowski, 2011), empowerment has been gaining attention in managerial literature. Seibert et al. (2011) argue that empowerment is not only related to work attitudes, but also to positive forms of work performance, due to the more active demeanor empowered employees have toward their work. This is in line with Thomas and Velthouse (1990) who consider empowerment as a ‘motivational process’ in employees. Because people with empowerment are more motivated they take on an active role, and therefore Seibert et al. (2011) propose both in-role task performance as OCB as behavioural outcomes of empowerment.
23
Indeed, scholars have identified the motivational model as one of the major explanatory frameworks to identify the causes of citizenship behaviours (Huang, Iun, Liu, & Gong, 2010; Bateman & Organ, 1983; Organ, 1977). Organ, Podsakoff, and MacKenzie (2006, p. 110) explain that when employees engage in intrinsically motivating tasks, they may “find the performance of job-related activities to be more rewarding, and as a result, they are motivated to expand greater effort to achieve their task objectives”. Similar to earlier research, Raub and Robert (2010) also mention motivational processes as antecedent of OCB. Bandura (1997) specifically mentions ‘competence’ and ‘impact’ as likely internal motivating factors that further encourage OCB. The factors competence and impact wil make the subrodinate feel capable of achieving positive outcomes. These two factors are also part of the empowerment-construct and therefore, employees who are intrinsically motivated or empowered will display more OCB.
Furthermore, Morrison (2011, p. 382) argues “the primary driving motive behind the challenging behaviour, or voice, is the desire to benefit the organisation or work unit”. However, in order to display voice one considers the consequences (Detert & Burris, 2007; Morrison & Milliken, 2000). Two consequences of speaking up are noted in particular: whether it is effective (perceived efficacy) and judgement about the negative consequence (related to trust). As noted earlier, subordinates who are empowered take an active role towards work, show self-determination, believe their actions have impact and thus exhibit self-efficacy. As such, challenging OCB needs a ‘step over the line’ of the management authority and requires individuals to “adopt an active orientation toward their work roles, and such an active orientation is a characteristic of employees who feel psychologically empowered” (Raub & Robert, 2010, p. 1750). This means that subordinates with high
24
empowerment are inclined to believe in their own “can do” or self-efficacy, and more probably display challenging behaviour.
Moreover, Bandura (2001) also suggests that empowerment increases resilience, creativity, and above all, initiative. Comparably, Choi (2007) and Thomas and Velthouse (1990) argue that psychological empowerment is associated with flexibility and a sense that one is less constrained by supervisory instructions and organizational rules. Because empowered individuals feel less constrained by their jobs, they have more discretion to act as moral agents to help others (Bandura, 2001). Thus, employees who are empowered experience freedom to take initiative and cooperate with others, which can be considered as affiliative OCB.
Although affiliative and challenging behaviours might be seen as opposites (Choi, 2007), the following can be considered: if affiliative behaviour is only customer oriented it can be perceived as stretching organisational rules and thereby ‘challenging’ from an organisational standpoint. Therefore, in this research OCB is taken together as one construct.
Research by Yen, Lin and Tai (2004) confirms that psychological empowerment may boost one’s personal motivation and stimulate OCB in general. In agreement, Seibert et al. (2011) confirm and prove that psychological empowerment is positively related to OCB. Therefore, expected is:
Hypothesis 3: Empowerment is positively related to subordinates’ organisational
citizenship behaviour.
2.6 Emotional intelligence, empowerment and organisational citizenship behaviour The EI of a leader is acknowledged to result in positive interactions with subordinates. It is known that when leaders can perceive, manage and understand their own emotions, and
25
understand and have empathy with others, this will be likely to commence in effective leadership (Antonakis et al., 2009; Day & Carroll, 2004). Poskadoff et al. (2000) note that leadership might also play a role in determining OCBs. However, they also mention that the mechanisms through which these behaviours influence the employee are unclear. Therefore, Poskadoff et al. (2000) suggest focusing on indirect relationships through which OCB exists for future research.
The mediation of empowerment in the relation of EI and OCB unfolds, because subordinates are more likely to feel understood, encouraged and supported by their (emotional intelligent) leader. Kark et al. (2003) note that this may lead to feeling more worthwhile (or more self-esteem according to Brockner, 1988), influential and effective, which is related to empowerment. According to Bandura (1977), this “positive self-evaluation extends to the workplace, resulting in a more positive and desirable work behaviour“ (Wang, Zhang, & Jackson, 2013). When employees feel empowered, they believe that they are capable and are more likely to take a pro-active and initiating attitude towards work. Thus, the employee is willing to ‘go the extra mile’, put in extra effort, and thus show more OCB. Accordingly, predicted is:
Hypothesis 4: Empowerment of the subordinate mediates the positive relation between
leader’s emotional intelligence and subordinates’ organizational citizenship behaviour.
2.7 Work locus of control
Personality of a subordinate may be a foundation for differentiations in the response to leadership (Ehrhart & Klein, 2001). Howell and Shamir (2005) even state that subordinates’ characteristics are powerful determinants of reactions to leaders. The personality trait ‘locus of control’ (LOC) is defined by Rotter (1966) as the degree to which individuals’ belief
26
outcomes in life are controlled by one’s own actions or by environmental factors. The scale upon which LOC is determined is called ‘internal’ or ‘external’. In other words; the degree to which one is responsible for one’s own outcomes in life (internality), rather than uncontrollable external forces (externality) (Wang et al., 2013). According to Phares (1968), people with a high internal LOC attempt greater efforts to control their environment, display more desirable learning, seek and use new information more eagerly. Furthermore, they exhibit less conformity and are less troubled by social demands of a situation than people with external LOC. Later, Spector (1988) presented a more specific work-LOC (WLOC), which is more applicable in organisational settings than general LOC.
Individuals with high internal WLOC are implied to be better at creating and shaping their work experiences in a way that is favourable to them (Phares, 1976). De Hoogh and Den Hartog (2009) note that individuals with low (internal) WLOC are less motivated and energized, and more likely to turn to their leader for direction. Confirmed is that these individuals indeed feel a need for leader guidance and support (Ng, Sorensen, & Eby, 2006). Ng. et al. (2006) advocate that individuals with a high internal WLOC are less likely to feel the need for leaders’ guidance, because they recognize their own responsibility. Thus, individuals with a high internal WLOC are expected to be less likely receptive to the influence of the leaders’ EI. WLOC has not previously been examined as a moderator of the
relationship between EI and empowerment, and EI and OCB. In order to extend existing
research, the following is suggested:
Hypothesis 5: Subordinates’ internal locus of control moderates the relationship between
leader’s emotional intelligence and subordinates’ empowerment, such that the positive relationship between emotional intelligence and empowerment and is stronger when subordinates’ internal locus of control is lower.
27
Hypothesis 6: Subordinates’ internal locus of control moderates the relationship between
leader’s emotional intelligence and subordinates’ organisational citizenship behaviour, such that the positive relationship between emotional intelligence and organisational citizenship behaviour is stronger when subordinates’ internal locus of control is lower.
Subsequently, taken the before stated hypotheses and the overall model into consideration, the proposed hypothesis is:
Hypothesis 7: Leader emotional intelligence is related to subordinates’ organisational
citizenship behaviour via conditional indirect effects, such that the interaction between leader emotional intelligence and subordinates’ internal locus of control is related to subordinates’ empowerment, which in turn is related to subordinates’ organisational citizenship behaviour.
28
3. Research design
3.1 Method
The hypotheses are tested in a deductive explanatory study by conducting quantitative cross-sectional research. The intent is to explain the relationship between a leader’s EI and the subordinates’ empowerment and OCB, moderated by WLOC. Data was collected from employees in one of the largest financial institutions in the Netherlands. The survey included previously validated measures and questions were translated into Dutch to match the working language of the participants. Paper-based questionnaires were distributed as a packet to leaders with a minimum of three employees in their team. This way a greater probability of receiving complete packets (one leader, three team members) was expected. The survey itself was introduced as measuring ‘leadership in organisations’ to prevent participant bias. The leaders were asked to rate themselves on EI, complete a situational judgement test (SJT) on emotion management, and rate OCB of three subordinates. Employees were asked to score their immediate supervisor/ leader on EI and were provided with questions to rate their own empowerment and WLOC. Questions were entered into SPSS manually and missing values were given the code ‘99’. Afterwards, the data was tested for significant correlations, mediation and moderation effects, and analysed.
3.2 Procedure
Survey packets were distributed to leaders, or to one point of contact (e.g. the business manager or secretary) within a department, and in turn distributed it to the leaders of the specific department. The survey packets included an instruction letter and four paper-based surveys, one for the leader and three for the subordinates. If the location of the work location of the leader was not similar to that of the researcher, return-envelopes were provided to allow
29
for sending the completed surveys via mail directly to the researcher. The cover letter explained the study purpose, the voluntary nature of participation, and assured confidential treatment of data. The complete content of the packet (cover letter, surveys) can be found in Appendix I, II and III. Contact details of the researcher were enclosed to provide the ability for asking questions or express the interest in receiving outcomes of the study. After reading the instructions, the leader randomly selected three direct subordinates to fill out the survey. Questionnaires were returned anonymously to the point of contact of the department, or put in an envelope and collected from the leader by the researcher within two weeks. Reference codes enabled matching of respondents with distributed packets, and matching leaders with participants. In case the packets were not returned after two weeks, the contact persons received a reminder via e-mail in an attempt to increase the response rate.
3.3 Sample
Within this financial institution, participants were chosen from different disciplines to ensure a proper representation of the company. Departments included: Human Resources, Audit, IT, Risk Management, Treasury, Digital Banking, Global Wholesale Product Clients (e.g. Markets, Acquisition Finance, Trade Commodity Finance, Asset Based Finance), and Private & Business Banking. Data sampling was done via non-probability convenience sampling. Part of the sample was contacted through one point of contact in a department, and others were contacted directly (leaders). In cases whereby one person was contacted as representative, the contact person was asked to distribute the surveys among leaders within this specific department. In total, 125 leaders and 325 employees were asked to contribute to this study (total N= 450). At the time of analysis, a total of 63 leaders, with a minimum of three subordinates, completed the survey (50,4% response). Of the 325 employees, 178 returned the survey (54,8% response).
30
Additional data was received later, but not integrated in this study due to time- limitations. For two teams, leader surveys were not received and data is excluded. In four cases only one employee survey was returned and as a result, both the employee and leader data is excluded for analysis. Consequently, analysis is executed on teams with data from one leader and at least two employees. Furthermore, incomplete datasets on the explanatory variables were deleted. Overall, the final sample size with items excluded resulted in leader
N= 59 and subordinate N= 173.
The average age of the total 232 respondents is ± 42, ranging between 25 and 63 years (SDage = 8,99). The population consists of 150 males (64,7%) and 82 females (35,3%). The distribution of gender between leader and subordinates is similar to the total dataset. Most respondents completed university as highest education (132 persons, 57,1%), others have noted they completed high school (4; 1,7%), MBO (10; 4,3%), HBO (80; 34,6%), or ‘other’ and filled out they completed post-doctoral education (5; 2,2%). Leaders are educated on HBO- (64/80), university- (91/132) and post-doctoral-level (3/5). On average, the population worked for 12,71 years at the company (= 12 years and 8 months), ranging between 0,5 to 40 years (SDtenure= 8,74). The employees worked with their leader for 25 months on average,
31
4. Measures
Employees completed surveys rating their leader’s EI, their own feelings of empowerment, and WLOC. The leaders rated the OCB of subordinates, and were asked to self-assess EI and fill in the ability EI measure (STEM). Almost all items were rated on a 5-point Likert Scale (e.g. response choices are: “strongly disagree”, “disagree”, “neutral”, “agree”, “strongly agree”). Exception to this rule is the measurement filled out by the leader to rate ability EI, whereby four possible answers were provided as options. Questions were retrieved from validated surveys and formerly translated. A translation and back-translation procedure (Brislin, 1980) was used to ensure the reliability and validity of the instruments.
4.1 Emotional intelligence
4.1.1 Leader self-reported (WLEIS)
The measure used for the leader’s self-reported EI is the WLEIS, a 16-item cross-validated “Emotional Intelligence Questionnaire” developed by Wong and Law (2002). The outcomes of the validation show that EI and personality are distinct constructs (Law et al., 2004) and consequently EI can be measured as a separate construct. The four different dimensions measured are: (1) self-emotion appraisal, (2) others-motion appraisal, (3) use of emotion and (4) regulation of emotion (Davies et al., 1998). The sixteen items are divided into four items per dimension. A sample item from the Self-Emotion Appraisal (SEA) is “I have a good understanding of my own emotions.” A sample item from the Others’ Emotion Appraisal (OEA) is “I have good understanding of the emotions of people around me.” The Use of Emotion (UOE) to facilitate performance dimension is measured by, for example: “I always tell myself I am a competent person.” An item from the dimension Regulation of
32
Emotion (ROE) dimension is “I am quite capable of controlling my own emotions.” In the initial development study of Wong and Law (2002), Cronbach’s alpha provided for the four dimensions (self-rated) are between .79 and .86. The alphas in the present study are moderate: .68 for Self-Emotion Appraisal, .75 for Other’s Emotion Appraisal, .68 for Use of Emotion, and .83 for the Regulation of Emotion. An increase in alpha for the UOE scale could be realized by eliminating item 9 “I always tell myself I am a competent person”, from .68 to .73. However, removing this item would decrease the overall Cronbach’s alpha (from .83 to .82). Thus, item 9 is retained to measure the complete EI construct.
4.1.2 Leader ability situational judgement scored (STEM)
Secondly, EI is measured as a set of abilities, by using the short version of the STEM by MacCann and Roberts (2008). The 30 most reliable questions out of the 44 original questions were used. The MSCEIT (as founded by Mayer, Salovey, Caruso, & Sitarenios, 2003) is the precursor of the STEM and assesses four branches of emotion-related abilities:” (a) the perception and expression of emotions; (b) the integration of emotions into thought processes; (c) understanding the relations between, and transitions among, emotions and between emotions and circumstances; and (d) the management of emotions to moderate negative, and enhance positive, emotions” (Mayer et al., 2000). MacCann and Roberts (2008) have taken ‘understanding’ and ‘management’ and created an alternative situational test of emotional understanding/ management (STEU and STEM) to measure EI theoretically keyed.
The preliminary validity of the STEM as an EI ability-test is based on four important criteria: it has a strong degree of association with existing measures of intelligence, has meaningful relationships with other EI tests, relates to outcomes indicative of emotions, and correlates to personality in the similar range of other intelligence tests. The 30-item STEM is
33
developed based on three steps: item generation, response-option generation and expert scoring. First, 50 individuals (psychology students and volunteers) were asked to describe encountered emotional situations, which resulted in 180 items of a one to two sentence description of emotional aspects of a situation. Accordingly, these items were divided into three samples and three groups of undergraduate students were created to describe “the best thing to do in this situation” and “what they would do if”. Lastly, expert groups scored the multiple answers. In this research, the STEM is given in a multiple-choice format, and scored and analysed according to the given expert proportions. The expert-scoring technique means that the correct answer is determined by combining the judgment of experts. The STEM-measurement has been accumulated in two different ways: (1) the ratings of the leader have been converted to the expert ratings and treated as continuous variables (compared to that used in cognitive ability tests), (2) the ratings of the leader have been converted to 1 for the best possible answer and 0 for all three others based on psychologist’ ratings, and thus treated as ‘right’ or ‘wrong’ (binary variable) (similar to research of Hartman, 2011). Thus, the ability-EI data is analysed in two different ways: as continuous and as binary variable.
In order to fill out the measurement, leaders are presented with brief details about an emotional situation and asked to choose from the four responses the most effective course of action to manage both the emotions the person is feeling and the problems they face in the situation. The respondent’s choice is assigned a numerical value based on the expert weights previously calculated. An example item is: “Alan helps Trudy, a peer he works with occasionally, with a difficult task. Trudy complains that Alan’s work isn’t very good, and Alan responds that Trudy should be grateful he is doing her a favour. They argue. What action
would be the most effective for Alan? (a) Stop helping Trudy and don’t help her again, (b) try
34
advice”. Reported Cronbach’s α by MacCann and Roberts (2008) is 0.92, thus excellent internal consistency. The current research provided for an estimated internal consistency of .67. This is remarkably lower than the .92 of MacCann and Roberts (2008), but similar to the internal consistency reported by Austin (2010). After eliminating the negatively correlated items >.1 (item 16, 19 and 29), an increase in alpha has been realised, resulting in an acceptable α = .72 for the STEM as continuous variable. The STEM treated as binary variable shows a Cronbach’s alpha of .23, which suggests the items may not be measuring the same underlying construct. Removing items 16, 19 and 29 results in α = .36.
4.1.3 Subordinates’ score of leader (WLEIS)
The subordinates have undertaken the third measurement-method of EI. The subordinates scored their leader on the ‘Regulations of emotion’ (ROE) and ‘Others’ emotion appraisal’ (OEA) of the WLEIS. A similar set-up has been used as the leader’s self-assessment, and questions have been amended to fit the scoring of the leader. An example from the original WLEIS OEA is “I have a good understanding of the emotions of people around me”. The amendment for the subordinate scoring is: “My leader has a good understanding of the emotions of people around him/her”. An example of the ROE questions is: ”My leader is quite capable of controlling his/her own emotions”. In previous research, the reliability estimates for the subordinate-rated were ROE and OEA were: ROE: .79; OEA: .82 (Wong & Law, 2002). The present study reports a Cronbach’s alpha of ROE = .86, and OEA = .89. This is considered as highly reliable and slightly higher than earlier research has reported by Wong & Law (2002). For this study, the factors are taken together, and continued is with EI subordinate-rated as one factor (α = .85).
35 4.2 Empowerment
To measure subordinates’ psychological empowerment, a 12-item scale developed by Spreitzer (1995) is used. The scale has four dimensions, which include: meaningfulness, competence, self-determination and impact. In order to measure each dimension, the subordinates rate three questions per dimension. Examples of these questions are: “The work I do is very important to me” (meaning), “I am confident about my ability to do my job” (competence), “I have significant autonomy in determining how I do my job (self-determination), and “My impact on what happens in my department is large” (impact). Spreitzer (1995) reported a Cronbach’s α of 0.72 (test 1) and 0.62 (test 2), and noted that this is acceptable due to multiple aspects loaded on the empowerment scale. In the present research, the distinct factors were taken as one to measure empowerment as one construct; Cronbach’s alpha is .83.
4.3 Organisational citizenship behaviour
Van Dyne et al. (1995) were the first to suggest that extra-role behaviours, or OCB, could be categorized. The typology used is affiliative behaviours (e.g., helping, conscientiousness, and courtesy) and challenging behaviours (e.g, speaking up, and voice). In the current research, these two components of OCB are measured using two different scales. Challenging OCB included five items based on Van Dyne and LePine (1998). Questions reflect the voice-behaviour of employee (e.g. “This employee develops and makes recommendations concerning issues that affect this work group”). Research by Van Dyne and LePine (1998) shows the Cronbach’s α of the supervisor-scored OCB is .94. The seven questions of affiliative OCB are based on research of MacKenzie et al. (1991), Morrison (2011) and Podsakoff et al. (2000). Corresponding items with affiliative OCB include: “This employee is always ready to lend a helping hand to those around him/ her”. In the present research,
36
Cronbach's alphas for the 5 voice and 7 helping items were .80 and .78, respectively. However, the data fits the one-factor OCB (α = .85) better, and therefore one factor of OCB is considered for the current research.
4.4 Work locus of control
LOC in work settings is measured with a summated rating WLOC Scale. In total eight items from the summated 16-item scale of Spector (1988) are used. The eight questions presented to the respondents were items that focused on internal locus of control. To illustrate, items include “A job is what you make of it” and “People who perform their jobs well generally get rewarded for it”. Higher scores imply a higher internal locus of control. De Hoogh and Den Hartog (2009) reported a coefficient alpha of two samples: .78 and .70, respectively. In the present study, Cronbach’s alpha of WLOC is .67, and no substantial increase in alpha could have been achieved by eliminating items. Other studies advocate refinement of the WLOC scale (e.g., Oliver, Jose, & Brough, 2006).
4.5 Control variables
Several potentially relevant control variables have been included in this research, including: age, gender, level of education, years of work experience at employer, and length of working relationship between leader and subordinate. Previous research by MacCann and Roberts (2008) shows a relationship between the abilities of EI and the age and level of education of the assessor. Previous research on the MSCEIT has found that older participants score more highly on both Situational test of Emotional Understanding (STEU) and Management (STEM) and that this difference is larger for Emotion Management (Bastian, 2005). Furthermore, several researches have suggested differences across all facets of the MSCEIT influenced by gender, whereby women perform better than men (Ciarrochi, et al.,
37
2000; Day & Carroll, 2004; Mayer et al., 2000; Petrides & Furnham, 2000). As the STEM is derived from the MSCEIT, Hartman (2011) researched the effect of gender on the outcomes, but found no significant relationship on group-level. To parse out the variance between our independent variable and dependent variables, the variables age (in years), education (0 = high school, 1 = MBO, 2 = HBO, 3 = university, 4 = PhD) and gender (0 = male, 1 = female) are controlled for in the analysis.
Furthermore, in previous empirical research, Becker (1964) presents that “certain individual characteristics such as tenure, and work experience positively affect attitudes and behaviours”. OCB is one of these behaviours as mentioned by Becker (1964). Therefore, the controls ‘work experience’ (in years), and ‘length of working relationship’ (in months) have been included in the survey, to be able to eliminate alternative explanations and to demonstrate the unique relationship between leader’s EI and subordinates’ OCB.
