Importance of followers' emotions in effective leadership more

The importance of followers’ emotions in effective leadership Lars Glasø1 2, Guy Notelaers3 4 2 & Anders Skogstad2 Abstract Within the framework of Affective Events Theory (AET) this study examines emotional experiences as a potential mediator between followers’ perceptions of employee-centered leadership and their experiences of job engagement and intention to leave the organisation, respectively. The results showed that the relationships between employee-centered leadership and job engagement, as well as turnover intentions, were fully mediated by the followers’ positive emotional experiences. Negative emotional experiences yielded insignificant mediation effects, a finding that may be explained by characteristics of the leadership style studied. The present study substantiates that followers’ emotions bridge the ‘gap’ between leader behaviour and follower attitudinal outcomes and, hence, supports the notion that followers’ emotions are essential in the study of effective leadership. Key Words: Leadership; Emotions; Affective Events Theory; job engagement; Intention to leave. Received: 06.07.11, Accepted: 01.12.11, Published 20.12.11 Introduction Leadership research has shown that leaders’ behaviour is related to several important outcomes such as job satisfaction, performance and well being among followers (see e.g., Judge & Piccolo, 2004); and it is has also been substantiated that various situational factors (Robbins & Judge, 2008) and follower characteristics, e.g., the Big 5 personality factors (see Judge, Bono, Ilies & Gerhardt, 2002; Kirkpatrick & Locke, 1991) influence such outcomes. Further, some studies have shown that followers’ experienced emotions during interactions with their immediate superior are associated with job satisfaction and well-being (e.g., Glasø & Einarsen, 2006). However, studies examining whether and how emotions may influence the relationship between leader behaviour and attitudinal and behavioural outcomes are still limited. Hence, the aim of the present study is to test empirically whether followers’ positive and negative emotions mediate the relationships between their immediate superiors’ employee-centered behaviour, and their job engagement and intention to leave the organisation, respectively. Leadership has been defined in terms of mobilizing the workforce towards attaining organisational goals through a process in which the leader’s attempt to influence the behaviour of the followers is imperative (Yukl, 2010). In this respect, the behavioural style of the leader is considered to be particularly important. Leadership styles such as relational (Blake & Mouton, 1964), inspirational (Conger & Kanungo, 1998) and transformational leadership (Bass, 1990; Bass & Riggio, 2006) have been related to central organisational outcomes such as employees’ well-being (Van Dierendonc, Haynes, Borrill & Stride, 2004), leadership trust (Podsakoff, MacKenzie, & Bommer,1996), self-efficacy beliefs (Kirkpatrick & Locke, 1996), leadership satisfaction (Yammarino & Bass, 1990), worker absenteeism (George & Jones, 1996), job satisfaction (Judge & Piccolo, 2004;Weiss & Cropanzano, 1996) and salesmen’s performance Lars Glasø, PhD. Professor. Dep. of leadership and organizational behaviour BI Norwegian Business School 0442 Oslo, Norway E–mail: Lars.Glaso@bi.no 1. Department of leadership and organizational behavior. BI Norwegian Business School 2. University of Bergen, Norway. 3. School of Business and Economics, Maastricht University. The Netherlands. 4. Work, Organizational and Personnel Psychology, KU-Leuven, Belgium. Scandinavian Journal of Organizational Psychology 2011, 3:2 (McColl-Kennedy & Andersen, 2002). The positive relationship between transformational leadership and task performance is supported by several meta-studies. For example, in Piccolo and Colqiutt’s (2006) metaanalysis of 247 articles and 34 doctoral dissertations, the estimated correlation between these two variables was 0.44 (see also Judge & Piccolo, 2004: Lowe, Kroeck & Sivasubramavian, 1996). However, many empirical studies regarding the influence of leadership behaviour on organisational outcomes have been one-sided, focusing mainly on leader behaviour without taking into sufficient consideration features of the followers that might influence the relationship between leadership behaviour and subsequent outcomes (Alimo-Metcalfe & Alban-Metcalfe, 2002). Reasonably, the leader’s behaviour per se cannot account for all the variation in different follower outcomes because the followers’ appraisal and subsequent reactions to leader behaviour probably will systematically influence those outcomes (see e.g., Lazarus and Folkman, 1984; Lazarus, 1999). Hence, it is quite likely that various characteristics of the situation and the follower will influence attitudinal outcomes of leadership as well. In this respect, researchers have proposed several mediators (e.g., leader-member exchange, core job characteristics, intrinsic motivation, and goal commitment) to explain the relationship between leadership behaviour and employee outcomes (see e.g., Piccolo & Colquitt 2006; Wang, Law, Hackett, Wang & Chen 2005). According to Muchinsky (2000), feelings are fundamental and must be taken into account if one wants to understand people's reactions at work. We believe, in accordance with Muchinsky, that this is a fruitful and applicable point of departure when studying followers’ exposure to different events at work, such as leadership behavior. In the present study we employ Weiss and Cropanzano’s (1996) affective Events Theory (AET) as a theoretical frame of reference. AET offers a “macrostructure” that incorporates affective elements to explain work behaviour (Weiss & Beal, 2005) and “enables the consideration of interactive relationships” (Walter & Bruch, 2009, p. 1435). More specifically, within an AET framework Weiss and Beal (2005) argue that an individual’s positive and negative affective responses to workplace events mediate the relationship between such events and his or her cognition and behaviour. Weiss and Cropanzano (1996) state that workplace events trigger affective responses which, after being accumulated over time, will influence work related attitudes such as job satisfaction. These attitudes will in turn impact upon workplace behaviour such as absenteeism, lateness, turnover, and productivity. 18 Accordingly, the theory is based on the assumption that emotions are not equal to, for example, the attitude of job satisfaction. However, although AET is regarded as an important contribution in explaining the causes and consequences of emotions at work, empirical examinations of the basic assumptions put forward in the theory are limited (Weiss & Beal, 2005). AET as a theoretical framework does not specify the kinds of work environment factors or work events that may be associated with positive or negative affective reactions, except from labelling positive events as ‘uplifts’ and negative events as ‘hassles’. Few studies have explored the specific events that might arouse affect at work (Bach and Fisher, 2000). In the present study we will focus on examples of presumed positive events within the theory, namely the immediate superior’s employeecentered leadership behaviors at work. Regarding relationships between experienced emotions and organisational outcomes, research has shown that individuals’ affective states can influence a variety of performance-relevant outcomes including judgments (Robbins & DeNisi, 1994; 1998), creative problem solving (Isen, 1999), attitudinal responses (George & Bettenhausen, 1990), helping behaviour, and risk taking (Isen, 2000). Following Cherulnick, Donley, Wiewel, and Miller (2001), who proposed that leadership behaviours elicit emotional arousal in their subordinates, Lewis (2000) found, in a laboratory study, that followers observing a leader expressing anger felt more nervous and less relaxed than followers observing a leader expressing sadness or no emotion at all. She also found that followers observing a leader expressing sadness felt less enthusiasm and more fatigue than followers observing a leader expressing anger or no emotion at all. Such findings have made several scholars to make convincing arguments that leaders actually perform emotional regulation and express different kinds of emotional displays at work to influence followers and team members (see e.g., Bono, Foldes, Vinson & Muros, 2007; Glasø & Einarsen, 2008; Humphrey, 2008). In line with AET, McColl- Kennedy and Anderson (2002) have shown that followers’ emotions systematically influence the relationship between leader behaviour and the follower’s effectiveness. In a study of 121 sales representatives in Australia they found that both negative (frustration) and positive (optimism) emotions fully mediated the relationship between transformational leadership and the followers’ performance which was provided from company records. More recently in Germany, Rowold and Rohmann (2009) examined musicians' perceptions of their orchestra conductors' leadership behaviors and 19 related those behaviors to performance. Their positive emotions were associated with both transactional and transformational leadership. Negative emotions partially mediated the influence of transformational leadership on performance. These studies substantiate that followers’ intrapsychic reactions, including emotional states, have a systematic influence on performance. In support of this notion, Glasø and Einarsen (2006) demonstrated that leaders having a high-quality relationship with their subordinates (i.e., trust, confidence and support ) tended to evoke positive emotions in their subordinates, while poor relationships encompassing mistrust and lack of support evoked their negative emotions. The study showed that the subordinates’ affective experiences were significantly related to their levels of job satisfaction and well-being. Furthermore, Tsai, Chen and Cheng (2009) showed, in a longitudinal study of 282 employees and their immediate superiors in 10 insurance companies in Taiwan, that transformational leadership both directly influenced task performance and helping co-worker behaviour and had an indirect effect through employee positive moods. Also, in a Swiss study, meditational analyses showed that employees’ positive and negative emotions accounted for the relations between perceived supervisor support and cynicism and psychological hardiness and cynicism (Cole, Bruch & Vogel, 2006). In sum, research indicates that emotions may mediate the relationship between leader behaviour and different outcomes. Even though these studies yield very interesting findings, it still seems that the leadership literature is populated with many more ideas about the leader’s role in the generation of followers’ emotions and outcomes than it is populated with confirmatory empirical findings (Brief & Weiss, 2002). In fact, there is still a general lack of organisational studies focusing on the role followers’ emotions might play in explaining a broad range of essential organisational outcomes. Two such important attitudinal outcomes are employees’ job engagement and their intention to leave the organisation. Hence, the aim of the present study is to test empirically whether followers’ positive and negative emotions mediate the relationships between followers’ perceived employeecentered leadership behaviour and their job engagement and intention to leave the organisation, respectively. Job engagement, referring to a positive affectivemotivational state of fulfilment characterized by vigour, dedication, and absorption (Schaufeli, Salanova, Gonzalez-Roma, & Bakker, 2002), has been pointed to Glasø, Notelaers & Skogstad as a crucial factor in sustaining employees’ well-being and productivity in organisations, and has been linked to performance and creativity as well as to health (Bakker& Schaufeli, 2008). However, these studies, which are highly relevant and important, do not explicitly address leadership behaviour as a potential job resource or an essential event at work. In line with the AET framework, we need to disentangle whether different leadership behaviour and followers’ subsequent affective experiences affect job engagement. Likewise, employees’ intentions to stay or leave the organisation are vital for organisational functioning. Indeed, employees actually quitting their jobs may result in loss of valuable, organisation-specific knowledge that is costly and time-consuming to replace (Beehr, Glazer, Nielson, & Farmer, 2000). George (1989) has shown that both positive and negative moods predicted turnover intentions. Furthermore, George and Jones (1996) examined the interactive effects of values, positive moods, and satisfaction in predicting turnover intentions and found that positive mood predicted turnover intentions both in isolation and through their interaction with satisfaction and value fulfilment. In a longitudinal field study Pelled and Xin (1999) showed that both positive and negative mood at work predicted subsequent absenteeism, positive mood being the more influential of the two. In contrast, only negative mood predicted subsequent turnover. These studies demonstrate the importance of including affective experiences when studying withdrawal behaviours among employees, and, further scrutinise the potential influence of positive and negative emotions on attitudinal outcomes. The present study examines to what extent followers’ experiences of positive and negative emotions mediate the relationships between their exposure to supportive leadership and their job engagement and turnover intentions, respectively. Based on previous research and relationships postulated within an AET framework, we hypothesize the following: Hypothesis 1: Followers’ positive emotions mediate the relationship between followers’ experience of employee-centered leadership and their job engagement. Followers’ experience of employeecentered leadership will be related to an increase in positive emotions, which further will be related to an increase in job engagement. Scandinavian Journal of Organizational Psychology 2011, 3:2 Hypothesis 2: Followers’ positive emotions mediate the relationship between followers’ experience of employee-centered leadership and their turnover intentions. Followers’ experience of employee-centered leadership will be related to an increase in positive emotions, which further will be related to a decreased intention to leave the organisation. Hypothesis 3: Followers’ negative emotions mediate the relationship between followers’ experience of employee-centered leadership and their job engagement. Followers’ experience of employeecentered leadership will be associated with a decrease in negative emotions, which further will be related to an increase in job engagement. Hypothesis 4: Followers’ negative emotions mediate the relationship between followers’ experience of employee-centered leadership and their turnover intentions. Followers’ experience of employee-centered leadership will be associated with a decrease in negative emotions, which further will be related to a decreased intention to leave the organisation. 20 fit in a confirmatory factor model, a LISREL confirmatory factor analysis showed that 3 of the items obtained a very good fit. Hence these three items, reported in the following, will represent employeecentered leadership in the present study: “Is considerate”, “Is just in treating subordinates” and, “shows regard for the subordinates as individuals”. The scale has 5 response categories ranging from ”Absolutely disagree” to ”Absolutely agree” (Cronbach’s alpha = 0.73). Emotional experiences were measured by the international short Positive and Negative Affect Schedule (Thompson, 2007) which is an international valid 10 item PANAS (Watson, Clarke & Tellegen, 1988). PANAS consists of two scales referring to 5 negative emotions (Cronbach’s alpha = 0.73), such as “upset”, “hostile” and “ashamed” and 5 positive emotions, such as “alert”, “inspired” and “determined” (Cronbach’s alpha = 0.80). The respondents are asked to what extent they generally have felt those emotions during the last two weeks. Response categories are: “Never”, “a little”, “moderately”, “quite a bit”, and “always”. Job engagement was measured with the nine-item version (Cronbach’s alpha = 0.95) of the Utrecht Work Engagement Scale (UWES; Schaufeli, Bakker & Salanova, 2006). The UWES reflects three underlying though highly correlated dimensions, which are measured with three items each: Vigor (e.g., ‘‘At my work, I feel bursting with energy”), Dedication (e.g., ‘‘My job inspires me”), and Absorption (e.g., ‘‘I get carried away when I am working”). The high correlations between the three dimensions have led to a research practice where all items are conceived to measure the overall concept of engagement (De Lange, De Witte & Notelaers, 2008; Le Pine and Rich, 2010). High scores on the items indicate high work engagement. The respondents rated each item on a seven-point Likert scale (from "Never the last year" to "Daily"). Since the first item of the UWES is responsible for a very large deterioration of fit (1 degree of freedom was associated with an 188.12 point increase in χ2), it was left out of the analysis. Turnover intention was measured with a three-item scale (Cronbach’s alpha = 0.87) devised to measure an overall turnover propensity (Sjøberg & Sverke, 2000). The respondents are asked about whether they "Absolutely disagree", "disagree", "Neither agree nor disagree", "Agree" or "Absolutely agree" with the following statements: "I'm actively looking for other jobs", "I feel that I could leave this job”, and "If I was completely free to choose I would leave this job". Method Sample Data were collected by means of an anonymous self report questionnaire and distributed to 837 employees in a maritime transportation company in Norway. A total of 462 respondents completed the survey (response rate of 55. 2 %). The response rate is slightly above the mean found in surveys of this kind (see Baruch & Holtom, 2008). Only 13, 9% were women. The age of the sample ranged from 17 to 66 years with a mean of 45 years (SD = 11. 77). A total of 57.6% of the sample were leaders (captains, officers) and 42.4% were employees without personnel responsibilities. Most of them were working on the ferries (82%), while 14.1% were working on express steamers or in the catering (4%). The average tenure was 11.5 years (SD = 10. 53) and average job tenure in the company is 4.5 years (SD = 5. 15). Instruments Employee-centered leadership was measured by a short version of Ekvall and Arvonen's (1991; 1994) leadership questionnaire which includes employeecentered as well as production and change-centered leadership. While the items of the original employeecentered leadership subscale were associated with a bad 21 Construct dimensions in AET. A central tenet of the AET is that experienced emotional states and attitudinal variables, such as job satisfaction, job engagement and intention to leave are related, but clearly distinguishable constructs (see Weiss & Cropanzano, 1996). A necessary condition to proceed in applying AET was therefore to assess the dimensionality of the present study’s variables. The results of the analyses lend support to the assumptions in AET since both job engagement and positive and negative emotions (r = .58; p <.001 and r = -.24 p<.001, respectively) as well as turnover intentions and positive and negative emotions (r = -.38; p<.001 and r = .25; p<.001, respectively) were moderately correlated. However, we wanted to explore this issue further by comparing different factor models. The analyses showed that the 5 factor model distinguishing between the 5 factors in the current study (employee-centered leadership, positive emotions, negative emotions, job engagement and turnover intentions) obtained a better fit ; χ2 = 687.31; df = 242) than other factor models where for instance emotions and job engagement were conceived as one factor ( χ2 = 1528.26; df = 249) or a factor model where emotions and intention to leave were combined into one latent variable (χ2 = 1343.09; df = 249), and finally, where next to employee-centered leadership only one factor was differentiated (χ2 = 1601.87; df = 251). Thus, our findings support the idea of AET, that job engagement, turnover intentions and emotions are distinct empirical and theoretical constructs, and should be explicated and studied as such (see also Ashkanazy, Zerbe & Härtel, 2002). Glasø, Notelaers & Skogstad Analyses and statistics As shown in figure I, the conceptual model includes two mediators (positive and negative emotions) which we intend to test simultaneously with respect to the two outcomes job engagement and turnover intention. In order to test a mediation a model with more then three variables the traditional Baron and Kenny (1986) approach is not appropriate (MacKinnon, Lockwood, Hoffman, West & Sheets, 2002; McQueen, Getz, & Bray, 2003). Structural equation modeling, however, allows analyzing several mediators and several outcome variables simultaneously. In the structural equation modeling approach, the full mediation model was the starting point for testing mediation (James, Mulaik & Brett, 2006). Maximum likelihood estimates were used to obtain the parameters in the models and where adjusted for non normality by the use of an asymptotic covariance matrix in LISREL 8.88. Scandinavian Journal of Organizational Psychology 2011, 3:2 22 Negative emotions Employeecentered leadership Turnover intentions Positive emotions Job engagement Figure 1. Hypothesised model 23 In the present study we build upon an analytical strategy employed by Geurts, Kompier, Roxburgh and Houtman (2003) and distinguish between three models in order to assess mediation. In the first model, we assume that emotions fully mediate the relation between employee-centered leadership and the outcome variables. In this model, only job engagement and turnover intentions are directly explained by emotions. In the second model, we assume that the relation between leadership behavior and the two outcome variables is partially mediated. In this model, engagement, and turnover intention are explained by emotions and follower perception of employee-centered leadership. If emotions still are contributing to the explained variance of the criteria, we can conclude that emotions partially mediate the relationship between leadership behavior and the outcomes. To evaluate the extent of mediation, a third model was estimated. This model differs from the second model because the path coefficients between emotions and two outcome variables are fixed. These parameters were set equal to the estimated parameters of the first model, in which complete mediation was estimated. In addition, the covariance between positive and negative emotions and the covariance between engagement and turnover intention was fixed to those in the full mediation model as they may influence the meditational paths. The difference in χ² between the second and third model indicates whether emotions mediate the relation between leadership behavior and the two outcome variables. If the difference in χ² between the second and third model is not significant, the model assuming complete mediation is the most appropriate one. If the Glasø, Notelaers & Skogstad difference in χ² between the second and third model is significant, the model in which partial mediation was assumed is the most suitable one. And finally, when the difference in χ² between the second and third model is significant, and the relation between emotions and outcome variables is not significant in the second model, the model assuming no mediation is the most suitable one. These three models are further described on the basis of various commonly reported statistical criteria. We used goodness-of-fit indices; the root-mean square error of approximation (RMSEA), the non normed fit index (NNFI), the Comparative Fit Index (CFI), and the Standardized root mean residual (SRMR). A RMSEA and SRMR that are smaller than .08 are indicative of a satisfactory approximate fit of the theoretical model (Hu & Bentler, 1999). For the other indices like the NNFI and the CFI, values larger than .90 (and preferably greater than .95) are considered to indicate a good fit. Results Table 1 shows the means, standard deviations and Pearson’s correlation coefficients for the variables in the study. All correlation coefficients between the variables were significantly different from zero (p < .001). Table 2 contains the fit measures of the LISRELanalysis (Jöreskog, & Sörbom, 1993) for the first model (which assumed complete mediation by negative emotions), as well as the fit measures for models where emotions partially mediated the relation between employee-centered leadership and the outcome variables of job engagement, and turnover intention. Table 1. Correlations between - , and reliabilities of latent variables Variables 1. Empl.-centered leadership 2. Positive emotions 3. Negative emotions 4. Job engagement 5. Turnover intentions 1. .73 .35 -.33 .32 -.30 .80 -.25 .58 -.38 .73 -.24 .25 .95 -.64 .87 2. 3. 4. 5. Off diagonal: correlations are all significant at .001 level Diagonal: reliability (Chronbach’s alpha) Scandinavian Journal of Organizational Psychology 2011, 3:2 The first model, or basic model, in which a complete mediation was assumed, fitted the data reasonably well. However, with 244 degrees of freedom the χ² of 695.74 is too high to obtain a perfect fit with 323 observations. Still, RMSEA as well as SRMR were below .08 and other descriptive statistics (Table 2) showed that this model had a satisfactory fit. In this model, the path coefficients between positive emotions and job engagement and between positive emotions and turnover intention were significant (t > 1.96). However, negative emotions were not significantly (t < 1.96) related to the two outcomes. Hence, negative emotions cannot be conceived as a possible mediator for the relationship between employee-centered leadership on the one hand and job engagement and turnover intention on the other hand. After deleting the non-significant paths from negative emotions to both outcome variables we obtained a new base model (M1) that fitted the data reasonably well. In this new base model, only positive emotions are identified as a possible mediator. Model 2 was the model in which positive emotions partially mediated the relationship between employee-centered leadership, and job engagement and turnover intention. This partial mediation model did lead to a significant improvement of fit (Δχ² (2) = 13.1), and fitted the data also reasonably well. To test the extent of the mediation we conceptualised model 3 where both the path coefficients between positive emotions and the two outcome variables and the covariance between emotions and between outcomes variables were fixed to the estimates obtained in the full mediation model. As shown in Table 2, this model does 24 not lead to a significant deterioration of fit (Δχ² (2) = 4.82). Hence, positive emotions do fully mediate the relationships between employee-centered leadership, and job engagement and turnover intentions, respectively. The path coefficients of the full mediation model are shown in Figure 2. The total percentage of explained variance was 33% for job engagement and 16 % for turnover intention. Thus, our first hypothesis, stating a mediation effect of positive emotions on the relationship between perceived employee-centered leadership and the two outcomes was fully supported, while hypothesis 2, stating a corresponding mediation effect of negative emotions, was not supported. Table 2. Goodness-of-fit indices and model comparisons for evaluating mediation of emotions χ 2 (d.f.) Δχ2 (Δdf) p RMSEA NNFI CFI SRMR Model 0 Model 1 Model 2 695.74 (244) 703.37 (246) M1 vs M2 690.29 (244) 13.1*** (2) M2 vs M3 0.00 0.00 0.00 .076 .076 .075 .94 .95 .95 .96 .96 .96 .067 .074 .068 .075 0.00 .96 .96 .069 Model 3 695.11 (269) 4.82 (ns) (2) 25 NA2 .68 Glasø, Notelaers & Skogstad NA4 .63 NA6 .79 NA7 .51 NA10 .49 Negative emotions Turnover intentions Pl 3 .57 -.33 -.27 -.40 .35 -.64 .73 .87 .88 Ti 1 Ti 2 Ti 3 Pl 8 Pl 13 .77 Employeecentered leadership V4 .84 .90 .71 .35 V8 D5 D7 D10 A9 A11 A14 .85 Positive emotions .60 Job engagement .88 .84 .90 .80 .72 .77 .57 .68 .63 .75 PA1 PA3 PA5 PA8 PA9 Figure 2. Full mediation model. Gray dotted lines are the direct relationships explained away by the full mediation model. Other lines are specified in table 1. Scandinavian Journal of Organizational Psychology 2011, 3:2 26 outcomes of work events. Hence, our study gives empirically evidence to the assumption that supportive leadership behaviour by influencing the followers’ positive emotions may prevent large organisational problems such as turnover intentions and loss of job engagement. As such, our results confirm the relevance of these variables for organisations in motivating and retaining their personnel, where failure may involve huge financial losses. In line with this, Hoel, Einarsen and Cooper (2003) have estimated that costs related to absence and replacement due to destructive leadership and harassment in Great Britain accounted for close to £2 billion annually. Moreover, these results give truly needed empirical evidence to the vast ‘emotional intelligence industry’, which emphasises the importance of leaders possessing high emotional intelligence to understand and manage their followers’ emotions in order to achieve organisational goals (e.g., Goleman, Boyatzis & McKee, 2003). The influence of follower emotions will, certainly, vary depending on the context and what is on stake. Regarding emotions which are evoked during leaderfollower interactions, however, there is reason to believe that they are important intervening variables, which is in line with the results of the present study. One reason why emotions may play a major role is due to their information value (Hochschild, 1983) in that emotions often contain valuable information about the quality of the leader-follower relationship. E.g., a smile as compared to a frown from the superior may have strong and opposite effects upon the follower (see also Glasø & Einarsen, 2006). In view of some earlier studies finding no mediation effects for negative emotions in the present study was surprising. However, a study by George (1989), showing that negative moods predict turnover intentions, and Pelled and Xin’s (1999) study showing that negative emotional states predicted subsequent turnover intentions reflect main effects only, and are, as such, not directly comparable with the present study. The lack of mediating effects of the negative emotions may, at least partly, be explained by a process where negative emotions are simply not systematically activated when followers’ experience employee-centered leadership, which for most followers will represent a positive event. Hence, the influence of negative events, such as destructive leadership behaviour (see e.g., Aasland, Skogstad, Notelaers, Nielsen, & Einarsen, 2009; Hoel, Glasø, Hetland, Cooper & Einarsen, 2010; Tepper, 2007) on follower emotions and attitudinal outcomes should also be investigated in future studies. Discussion The current study shows that the relationships between followers’ experienced employee-centered leadership and job engagement and intentions to leave the organisation, respectively, are fully mediated by their positive emotional experiences, while no such mediation effects were found for negative emotions. Hence, our results confirmed the first and second hypotheses while the third and fourth hypotheses were disconfirmed. The full mediating effect of positive emotions may be explained by a central facet of employee-centered leadership, namely its potential for strengthening the follower’s sense of being a valuable and competent person (see e.g., Ashkanasy & Tse, 2000). Consistent with such a view, Dasborough and Ashkanasy (2002) asserted that leadership is intrinsically an emotional process, where leaders display emotions and in this regard attempt to evoke appropriate emotions in their followers. Thus, experiencing employee-centered leadership behaviour, and followers’ subsequent emotional reactions, may interact in a constructive circle of events. Such a notion supports a reciprocal model where various types of positive leader behaviours and follower emotions evolve into a cycle that determines followers’ successful adaptation to their work environments. A similar positive spiral is described by Schaufeli and colleagues (2009) where initial job engagement predicts an increase in job resources, which, in its turn, further increases job engagement. In line with this, positive emotions may have important behavioural implications as regards the followers’ coping efforts. For instance, a study has shown that positive emotions play a crucial role in enhancing coping resources in the face of negative events (Tugade, Fredrickson & Barrett, 2004). Experiences of employee-centered leadership may, thus, both explain and enhance the followers’ positive emotional state as well as their positive attitude to others. However, because our cross-sectional study cannot draw conclusions about causal relationships, reverse relationships are also conceivable. Hence, increased job engagement as well as low turnover intention (i.e., indicating an intention to stay) may contribute to increased positive emotions, which again may affect the superior’s leadership behaviour to be even more participative. The fact that the mediating effect of positive emotions was strong lends support to AET conceptions that affective experiences are crucial for attitudinal 27 Glasø, Notelaers & Skogstad changes may, also, have influenced the followers’ emotional states. The cross-sectional design may be seen as a barrier for investigating the proposed model. For instance, one may also conceive an alternative hypothesis, namely that positive emotion is a consequence rather than a cause of job engagement, since it seems reasonable to expect that high levels of job engagement may bring about positive emotions as well as causing it. Moreover, when engagement and positive emotions fuel each other a spiraling effect may occur. Yet, reversed, reciprocal and normal causation cannot be determined with the current study. We will therefore recommend that such an alternative hypothesis should be studied within a longitudinal framework in future research. Even though PANAS has shown satisfactory psychometric qualities, and the fact that this instrument is extensively used in organisational studies (see e.g., Brief & Weiss, 2002; Watson, Clark & Tellegen, 1988; Weiss & Cropanzano, 1996), it should be noted that the measurement of emotions is a complex and difficult task, emotional experiences being both variable and transient. Accordingly, studies have shown that people retrospectively may overestimate the frequency and intensity with which they have experienced positive and negative affects as compared to real-time reports (Barrett, 1997). Hence, future studies should aim at examining these relationships longitudinally on an hourly or daily basis. Methodological issues Some important limitations should be considered when interpreting the results of the present study. Although the use of self-reports on all study variables are highly relevant in the present study, common method variance may have enhanced the overall strength of the associations (see e.g., Podsakoff, MacKenzie, Lee & Podsakoff, 2003). Furthermore, the sample representing a marine division of a major transport company in Norway puts limitations to generalizations. On the other hand, studying one specific type of organizations has its strengths. In line with this, an increasing number of scholars point out important limitations associated with a general research approach (De Croon, Blonk, De Zwart, Frings- Dresen & Broersen, 2002; Glasø, Bele, Nielsen & Einarsen, 2011; McClenahan, Giles & Mallett, 2007), and argue that the mainstream studies needs to be complemented by studies that incorporate factors and concerns specific to a particular working environment. Future studies should, therefore, try to replicate the present results in other samples representing various organisations and occupations. Asking respondents to which extent they experience positive and negative emotions with a time frame of two weeks seems quite suitable considering the fact that AET emphasises the point of measuring accumulated emotions in explaining the outcome variables examined in the present study. Furthermore, when employeecentered leadership behaviour becomes more or less enduring, it may both act as a daily uplift as well as constituting a more permanent feature of the working environment. According to AET, work environment features influence attitudes directly, through a cognitive route, as well as indirectly through an affective route, the latter by determining the occurrence of positive or negative affective work events (Weiss & Cropanzano, 1996) According to AET, an employee-centered leadership style should, then, have both direct and indirect effects on outcomes, which is in line with the findings of the present study. The work environment features defined in the AET model also point to the fact that situational and contextual issues, e.g. organization culture, climate and organizational changes, may be systematically related to outcome variables. The participants in the present study have experienced a period of more organizational change during last year in the form of new owners, new CEO and restructuring of more departments. Accordingly, these organizational Implications and conclusion The present study has shown that followers’ positive emotions fully mediate the relationships between followers’ experience of employee-centered leadership and their job engagement and intentions to leave the organisation, respectively. Hence, it is substantiated that followers’ positive emotions may play a central role in explaining the relationship between leadership behaviour and followers’ attitudinal outcomes. In accordance with this, leadership research would profit from employing an AET framework, as contrasted to the dominant tradition of studying leader behaviours’ direct effects on followers’ attitude outcomes (see e.g., Judge & Piccolo, 2004). The present results also give direction for future studies to include constructive as well as destructive forms of leadership, and the measurement of behavioural in addition to attitudinal outcomes, The results also have important practical implications in that Scandinavian Journal of Organizational Psychology 2011, 3:2 leaders should try to enhance their employee-centered leadership style during interaction with their followers. Such recommendations are in line with the notion stating that leaders’ emotional states and emotional intelligence are crucial in attaining organisational goals (see e.g., Ashkanasy & Tse, 2000; George, 2000, Goleman, Boyatzis & McKee, 2003). However, it should be noted that some scholars recently have taken a more critical view, suggesting a lack of relevance of emotional intelligence to leadership (e.g., Antonakis, Ashkanasy & Dasborough, 2009). We can learn from this empirical study that followers’ positive emotions fully mediate the relationships between employee-centered leader behaviour and two essential outcomes, namely job engagement and turnover intentions. As such, the study indicates that followers’ positive emotions seem to bridge the ‘gap’ between leader behaviour and follower attitudinal outcomes and, hence, supports the notion that followers’ emotions are essential in the study of effective leadership. Therefore, leader and organisation development programs should emphasise the important role of emotions in organisations, both from the standpoint of leaders and followers as well as the interdependency of their emotions. Finally, the study has provided findings that support the basic assumptions of the AET framework, thus, suggesting further application of this model in the working place. 28 Ashkanasy, N. M., Zerbe, W.J., & Härtel, C.E.J. (2002). Managing emotions in a changing workplace. In N. M. Ashkanasy, W. J. Zerbe, & C. E. J. Härtel (Eds.). Managing emotions in the workplace (pp. 3 – 22). New York: M. E. Sharpe. Baron, M. R., & Kenny, D. A. (1986). The ModeratorMediator Variable Distinction in Social Psychological Research: Conceptual, Strategic and Statistical Considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182. Baruch, Y., & Holtom, D. A. (2008). Survey response rate levels and trends on organizational research. Human Relations, 61(8), 1139-1160. Bakker, A. B., & Schaufeli, W. B. (2008). Positive organizational behavior: Engaged employees in flourishing organizations. Journal of Organizational Behavior, 29, 147–154. Barrett, L. F. (1997). The relationships among momentary emotion experiences, personality descriptions, and retrospective ratings of emotion. Personality and Social Psychology Bulletin, 23, 1100-1110. Bass, B. M. (1990). From transactional to transformational leadership: Learning to share a vision. Organizational dynamics, 18(3), 19–32. Bass, B. M., & Riggio, R. (2006). Transformational leadership. Mahwah, New Jersey: Lawrence Erlbaum Associates. Beehr, T.A., Glazer, S. N., Nielson, L., & Farmer, S. J. (2000).Work and non-work predictors of employees' retirement ages. Journal of Vocational Behavior, 57, 206-225. Blake, R. R., & Mouton, J. S. (1964). The managerial grid. Gulf Publishing, Houston (Tx). Bono, J. E., Foldes, H. J., Vinson, G., & Muros, J. P. (2007). Workplace Emotions: The Role of Supervision and Leadership. Journal of Applied Psychology, 92(5), 1357-1367. Brief, A. P., & Weiss, H. M. (2002). Organizational Behaviour: Affect in the Workplace. Annual Review of Psychology, 53, 279–307. Conger, J. A., & Kanungo, R. N. (1998). Charismatic leadership in organizations. Thousand Oaks, CA: Sage. Dasborough, M. T., & Ashkanasy, N. M. (2002). Emotion and attribution of intentionality in leadermember relationships. Leadership Quarterly, 13, 615-634. References Aasland, M.S., Skogstad, A., Notelaers, G., Nielsen, M.B., & Einarsen, S. (2009). The prevalence of destructive leadership behaviour. British Journal of management, 21, 438-452. Alimo-Metcalfe, B., & Alban-Metcalfe, R. J. (2002). Leadership. In P. Warr (Ed.). Psychology at work (pp. 300–325). London: Penguin. Antonakis, J., Ashkanasy, N.M., &Dasborough, M. T. (2009). Does Leadership need Emotional Intelligence? Leadership Quarterly, 20, 247-261. Ashkanasy, N., M., & Tse, B. (2000). Transformational Leadership as Management of Emotion: A Conceptual Review. In N. M. Ashkanasy, C. E. J. Härtel and W. J. Zerbe (Eds.). Emotions in the work place. Research, Theory, and Practice (pp.221-235). Westport, CT: Quorum Books. 29 De Croon, E. M., Blonk, R. W. B., De Zwart, B. C. H., Frings-Dresen, M. H. W. & Broersen, J. P. J. (2002). Job stress, fatigue, and job dissatisfaction in Dutch lorry drivers: Towards an occupation specific model of job demands and control. Occupational and Environmental Medicine, 59(6), 356–361. Ekvall, G., & Arvonen, J. (1991). Change-centred leaders: Empirical evidence of a third dimension of leadership. Leadership and Organizational Development Journal, 12, 18-23. George, J. M. (2000). Emotions and leadership: the role of emotional intelligence. Human Relations, 53(8), 1027–1055. George, J. M. (1989). Mood and absence. Journal of Applied Psychology, 74, 317–24. George, J. M., & Bettenhausen, K. (1990). Understanding pro-social behaviour, sales performance, and turnover: A group-level analysis in a service context. Journal of Applied Psychology, 75, 698–709. George J. M., & Jones, G. R. (1996). The experience of work and turnover intentions: Interactive effects of value attainment, job satisfaction, and positive mood. Journal of Applied Psychology, 81,318–25. Geurts, S. A. E., Kompier, M. A., J., Roxburgh, S., & Houtman, I. L. D. (2003). Does work-home interference mediate the relationship between workload and well-being? Evaluating model fit. In R. H. Hoyle (Ed.). Structural equation modelling: Concepts, issues and applications. (pp. 76-99). Thousand Oaks, CA: Sage. Glasø, L, Bele, E., Nielsen, M. B. & Einarsen, S. (2011). Bus drivers’ exposure to bullying at work: An occupation-specific approach. Scandinavian Journal of Psychology 52, 484–493. Glasø, L., & Einarsen, S. (2006). Experienced affects in leader-subordinate relationships. Scandinavian Journal of Management, 22, 49-73. Glasø, L., & Einarsen, S. (2008). Emotion regulation in leader-follower relationships. European Journal of Work and Organizational Psychology, 17(4), 482500. Goleman, D., Boyatzis, R., & McKee, A. (2003). Primal leadership: realizing the power of emotional intelligence. Boston: Harvard Business School Press. Hochschild, A. (1983). The managed heart. Berkeley: University of California Press. Glasø, Notelaers & Skogstad Hochschild, A. R. (1983). The managed heart. Berkeley: University of California Press Hoel, H., Einarsen, S. & Cooper, C. L. (2003). Organisational effects of bullying. In S. Einarsen, H. Hoel, D. Zapf and C. L. Cooper (Eds.). Bullying and Emotional Abuse in the Workplace. International Perspectives in Research and Practice (pp. 145-162). London: Taylor & Francis. Hoel, H., Glasø, L., Hetland, J., Cooper, C. L., & Einarsen, S. (2010). Leadership Styles as Predictors of Self-Reported and Observed Workplace Bullying. British Journal of Management, 21(2), 453-468. Humphrey, R. H. (2002). The many faces of emotional leadership. Leadership Quarterly, 13, 493–504. Humphrey, R. H. (2008). Affect and Emotion. New Directions in Management Theory and Research. IAP: NC. Isen A. M. (1999). Positive affect and creativity. In S. Russ (ed.). Affect, Creative Experience, and Psychological Adjustment. (pp. 3–17). Philadelphia: Bruner/Mazel. Isen A. M. (2000). Positive affect and decision making. In Lewis, M. and J. M. Haviland-Jones (eds.). Handbook of emotions (pp. 417–35). New York: Guilford Press. James, L. R., Mulaik, S. A., & Brett, J. M. (2006). A tale of two methods. Organizational Research Methods, 9(2), 233-244. Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and Leadership: A Qualitative and Quantitative Review Journal of Applied Psychology 87, No. 4, 765–780 Judge, T. A., & Piccolo, R. F. (2004). Transformational and Transactional Leadership: A Meta-Analytic Test of Their Relative Validity. Journal of Applied Psychology, 89, 755-768. Kirkpatrick, S. A., & Locke, E. A. (1991). Leadership: Do traits matter? Academy of Management Executive, 5, 48–60. Lazarus, R. S. (1999). Stress and emotion: A new synthesis. New York: Springer. Lazarus R. S., & Folkman, S. (1984). Stress, Appraisal and Coping. New York: Springer. Lewis, K. M. (2000). When leaders display emotion: How followers respond to negative emotional expression of male and female leaders. Journal of Scandinavian Journal of Organizational Psychology 2011, 3:2 30 Organizational Behaviour, 21 (special issue), 221Robbins, T. L., & DeNisi, A. S. (1998). Mood vs. 234. interpersonal affect: identifying process and rating distortions in performance appraisal. Journal of Lowe, K.B., Kroeck, K. G., & Sivasubramaniam, N. Business Psychology, 12,313–25. (1996). Effectiveness Correlates of Robbins, S., & Judge, T. A. (2008). Organizational Transformational and Transactional Leadership, A Behavior. 13th ed. Upper Saddle River, NJ: Prentice Meta-analytic Review. Leadership Quarterly, 3, Hall, 2003. 385–425. Rowold, J., & & Rohmann, A. (2009). Transformational McClenahan, C. A., Giles, M. L. & Mallett, J. (2007). and transactional leadership styles, followers' The importance of context specificity in work stress positive and negative emotions and performance in research: A test of the Demand- Control-Support German nonprofit orchestras. Nonprofit model in academics. Work & Stress, 21(1), 85–95. Management and Leadership, 20, 41-59. McColl-Kennedy, J. R, & Andersen, R. D. (2002). Salovey, P., B., Bedell, T., Detweiler, J. B., & Mayer, Impact of leadership style and emotions on J. D. (2000). Current directions in emotional subordinate performance. Leadership Quarterly, 13, intelligence research. In M. Lewis and J. M. 545–559. Haviland-Jones (Eds.). Handbook of emotions MacKinnon, D. P., Lockwood, C. M, Hoffman, J. M., (pp.504–519). New York: Guilford Press. West, S. G., & Sheets, V. (2002). A comparison of Schaufeli,W. B., Bakker, A. B., & Salanova, M. (2006). methods to test mediation and other intervening The Measurement of Work Engagement with a Short variable effects. Psychological Methods, 7, 83-104. Questionnaire. A Cross-National Study. Educational McQueen, A., G., Getz, J., & Bray, J. H. (2003). and Psychological Measurement, 66 (4), 701-716. Acculturation, substance use, and deviant behavior: Schaufeli,W. B., Bakker, A. B., & Van Rhenen, W. Examining separation and family conflict as (2009). How changes in job demands and resources mediators. Child Development, 74, 1737-1750. predict burnout, work engagement, and sickness Muchinsky, P. M. (2000). Psychology applied to work: absenteeism. Journal of Organizational Behavior, 30, An introduction to industrial and organizational 893–917. psychology (6th ed.). Belmont, CA: Schaufeli,W. B., Salanova, M., Gonzalez-Romá, V., & Wadsworth/Thomas Learning. Bakker, A.B. (2002). The measurement of Pelled L. H., & Xin, K. R. (1999). Down and out: an engagement and burnout: A confirmative analytic investigation of the relationship between mood and approach. Journal of Happiness Studies, 3, 71-92. employee withdrawal behaviour. Journal of Sjöberg, A., & Sverke, M. (2000). The interactive effect Management 25, 875–95. of job involvement and organizational commitment Piccolo, R.F., & Colquitt, J. A. (2006). on job turnover revisited: A note on the mediating Transformational Leadership and Job Behaviors: The role of turnover intention. Scandinavian Journal of Mediating Role of Core Job Characteristics. Psychology, 41(3), 247-252. Academy of Management Journal, 49, 327–340. Tepper, B. J. (2007). Abusive Supervision in Work Podfsakoff, P., MacKenzie, S., & Bommer, W. (1996). Organization: Review, Synthesis, and Research Transformational Leader Behaviors and Substitutes Agenda. Journal of Management, 33(3), 261-281. for Leadership as Determinants of Employee Thompson, E. R. (2007). Development and Validation Satisfaction, Commitment, Trust, and Organizational of an Internationally Reliable Short-Form of the Citizenship Behaviors. Journal of Management, Positive and Negative Affect Schedule (PANAS). 22(2), 259 - 298. Journal of Cross-Cultural Psychology, 38(2), 227Podsakoff, P. M., MacKensey, S. B., Lee, J. Y., & 242. Podsakoff, N. P. (2003). Common Method Biases in Tugade, M. M., Fredrickson, B. L., & Barrett, L. F. Behavioral Research: A critical Review of the (2004). Psychological Literature and Recommended Remedies. Journal of Applied Psychology, 88, 879-903. resilience and emotional granularity: Examining the benefits of positive emotions on emotion regulation Robbins, T. L., & DeNisi, A. S. (1994). A closer look at and health. Journal of Personality, 72, 1161 – 1190. interpersonal affect as a distinct influence on cognitive processing in performance evaluations. Yukl, G. A. (2010). Leadership in organizations. Upper Journal of .Applied Psychology 79, 341–53. Saddle River, NJ: Pearson Prentice Hall. 31 Yammarino, F.J., & Bass, B. M. (1990). Transformational Leadership and Multiple Levels of Analysis. Human Relations, 43(10), 975-995. Wang, H., Law, K. S., Hackett, B., Wang, R. D., & Chen, Z. X. (2005). Leader-member Exchange as a Mediator of the Relationship between Transformational Leadership and Followers Performance and Organizational Citizenship Behavior. Academy of Management Journal, 48, 420-432. Walter, W., & Bruch, H. (2009). An Affective Events Model of Charismatic Leadership Behavior: A Review, Theoretical Integration; and Research Agenda. Journal of Management, 35(6), 1428-1452. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validitation of brief measures of positive and negative affect: The PANAS Scale. Journal of Personality and Social Psychology, 54, 1063-1070. Weiss, H. M., & Beal, D. J. (2005). Reflections on Affective Events Theory. I: N. M. Ashkanasy, W. J. Zerbe and E. J. Hartel (Eds.). The effect of affect in organizational settings (pp. 1-22). Oxford: Elsvier Ltd. Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: a theoretical discussion of the structure, causes, and consequences of affective experiences at work. Research in Organizational Behaviour, 18, 1–74. Glasø, Notelaers & Skogstad Acknowledgement We would like to thank Professor Ståle Einarsen, Professor Stig Berge Matthiesen, Associate professor Jørn Hetland and Researcher Lars Johan Hauge for their valuable contribution in the collection of the data employed in the present study.