Promoting employee flourishing and performance: the roles of perceived leader emotional intelligence, positive team emotional climate, and employee emotional intelligence

A'yuninnisa, Rizqi N.; Carminati, Lara; Wilderom, Celeste P. M.

doi:10.3389/forgp.2024.1283067

ORIGINAL RESEARCH article

Front. Organ. Psychol, 26 February 2024

Sec. Employee Well-being and Health

Volume 2 - 2024 | https://doi.org/10.3389/forgp.2024.1283067

Promoting employee flourishing and performance: the roles of perceived leader emotional intelligence, positive team emotional climate, and employee emotional intelligence

$\r\nRizqi N. A&#x;yuninnisa,$ Rizqi N. A'yuninnisa^1,2^*

Lara Carminati¹

Celeste P. M. Wilderom¹

¹Faculty of Behavioral Management and Social Sciences, University of Twente, Enschede, Netherlands
²Faculty of Psychology, Universitas Gadjah Mada, Yogyakarta, Indonesia

Introduction: Through the lens of the of Emotion as Social Information (EASI) model, we examined the influence of perceived leader emotional intelligence on employee job flourishing and performance, produced via employee emotional intelligence, under the condition of positive team emotional climate.

Methods: We collected data with a two-wave survey design involving 1,210 primary school teachers and their 169 leaders. The hypothesized model was then tested using a multilevel structural equation model in Mplus.

Results: As hypothesized, the results suggest that perceived leader emotional intelligence positively affects the emotional intelligence level of followers, which enhances their job flourishing and performance. Reflecting the emotional display rules within the team, positive team emotional climate moderates this effect from leader to followers, in such a way that the relationship between perceived leader emotional intelligence and employee emotional intelligence is stronger when this positive team emotional climate is high rather than low.

Discussion: Our findings shed light on the affective impact of leaders and their teams as relevant sources of socio-emotional information for employees at work. Limitations of the present study and directions for future studies are included in the paper.

1 Introduction

In today's workplaces, the constantly changing nature of work and its increasingly digitalized features have imposed more emotional energy and stress on workers' relationships (Shoss, 2021; Gallup, 2022). Hence, affective job resources play a crucial role in obtaining optimal performance outcomes (Marín-López et al., 2019; Gamero et al., 2021). Among various affective factors at work, emotional intelligence has captivated organizational behavior scholars' interest due to its positive influence on numerous outcomes such as job performance, wellbeing, motivation and attitudes (Dasborough et al., 2022; Dogru, 2022). Emotional intelligence can be considered a personal resource able to aid motivational processes that enhance job performance (Bakker and Demerouti, 2017; Vermooten et al., 2021) and to favor the development of interpersonal relationships that are instrumental to achieve goals at work (Ashkanasy, 2003; Miao et al., 2017). Such function becomes prominent in leadership contexts where more and more often leaders are required to co-manage their followers' tasks and (some of) their emotions at work (Walter et al., 2012; Ashkanasy and Dorris, 2017). Therefore, leaders with a high level of emotional intelligence are likely to boost not only their own performance, but also benefit their followers' outcomes (Miao et al., 2016, 2018).

The positive effects of leader emotional intelligence on followers' outcomes have been mostly explained through the mechanisms of leader behaviors (e.g., Edelman and van Knippenberg, 2018; Roux and Gorgens-Eskermans, 2021) and/or group affective processes (e.g., Majeed and Jamshed, 2021; Zhang and Hao, 2022). However, exactly how leader behaviors and group processes relate to follower behavior at the individual level has been less explored, as such inquiry would need multilevel research designs (Tse et al., 2021; Ashkanasy and Kay, 2023). Despite indications of a positive relationship between leader and follower emotional intelligence, explanations on this relationship are still elusive (e.g., Wong et al., 2010; Chang et al., 2012). Hence, a more comprehensive understanding of the affective underpinning of this relationship is needed to enhance clarity on the mechanisms through which leader emotional intelligence influences followers' job outcomes, especially in light of the apparent joint impact of follower emotional intelligence on these outcomes.

The Emotion as Social Information (EASI) theory, offering a framework for comprehending the interpersonal impacts of emotions in the workplace, has the potential to unveil these mechanisms (Van Kleef, 2008). The theory posits that perceived emotional expressions can evoke emotional responses and inferences, thereby instigating behavioral changes among the observers. Human emotions are communicated through various channels, including facial expressions, vocal cues, postural signals, or symbolic cues, all of which influence the cognition, affect, and behaviors of those perceiving these emotions (Van Kleef and Côté, 2022). Leader emotional intelligence, which is expressed through this wide spectrum of emotional cues, is regularly observed by followers. These impressions exert both direct and indirect effects on followers' behavioral and emotional outcomes. In direct processes, perceived emotional expressions prompt affective and inference processes, leading followers to experience similar and/or other related emotions and work motivations, ultimately influencing job flourishing and performance. In indirect processes, the perceived emotional intelligence of the leader motivates followers/employees to emulate these expressions and cues, catalyzing adjustments in their own emotional intelligence, which, in turn, affect job flourishing and performance.

Despite EASI's plausible theoretical lens on interpersonal affective mechanisms, to this date it has fueled limited discourse regarding multilevel, group-individual contexts. The inquiry into how the social sharedness or climate of emotions within groups can influence interpersonal affective processes remains unanswered, which calls for much more empirical exploration (Barsade and Knight, 2015; Van Kleef et al., 2017; Van Kleef and Côté, 2022). According to EASI's perspective, we may expect that the context of groups likely moderates emotionally loaded interpersonal expressions of emotions. More specifically, the emotional climate, reflecting social norms and shared emotional expressions within groups, may facilitate and strengthen the interpersonal effects of emotions (Van Kleef et al., 2017; Van Kleef and Côté, 2022).

Taken together, our present study focusses on explanations for the multilevel interpersonal affective influences, at both the individual follower and group/work team context, between perceived leader emotional intelligence and desirable followers' outcomes. In line with positive occupational health psychology, we examine job flourishing and performance as the primary indicators of effective individual outcomes at work (Bakker and Derks, 2010). Specifically, we test how perceived leader emotional intelligence can promote employees' job flourishing and performance, either directly or indirectly via employee emotional intelligence, under the boundary conditions of positive team emotional climate. By doing so, we strive to bridge the gap in the literature and offer comprehensive insights and compelling arguments for managers regarding the acquisition of leader emotional intelligence and its contribution to employee motivation and performance at work (Ashkanasy, 2003; Bakker and Demerouti, 2017).

Thus, our empirical study contributes to the emotion-at-work literature and enriches the EASI theory in two main ways. First, we bridge the gap in the literature by providing multilevel evidence on specific affective mechanisms between leader and follower emotional intelligence. This is done under the boundary condition of positive team emotional climate, with the aim of better understanding their combined influence on follower job flourishing and performance. Second, shedding light on the beneficial moderating job-performance and -flourishing effects of sharing positive emotions in groups at work in coherence with high levels of emotional intelligence of leaders and their followers addresses the potentially key role of interpersonal influences of emotions at work. Hence, our results suggest clear practical implications for organizational experts or professionals (like trainers and teachers): to prioritize emotional competency of leaders and positive affective climates at work, which seem pivotal to promote desirable follower outcomes. In the remaining part of the paper, after explicating the theoretical frame, we report on the evidence and our interpretations, as well as potential implications for future research in this area.

2 Theory and hypotheses development

2.1 The importance of employee emotional intelligence on job flourishing and performance

Flourishing is a mental health state in which individuals experience a high frequency of emotional, psychological, and social wellbeing (Keyes, 2002, 2016). In this sense, flourishing goes beyond the absence of mental illness to experience high levels of wellbeing. At work contexts, flourishing individuals tend to feel positive emotions (i.e., happy and satisfied with their job), thrive, have good social relationships, are self-determined, competent, and purposeful (Bono et al., 2012; Rothmann, 2013). These positive features at work constitute both eudemonic and hedonic wellbeing, which make flourishing the gold-standard of overall individual wellbeing (Seligman, 2011; Huppert and So, 2013; Weziak-Bialowolska et al., 2019). In other words, job flourishing entails some emotional aspects as well as some behavioral predispositions, reflecting the state of an individual's job motivation (Bakker and Demerouti, 2017; A'yuninnisa, 2023).

Existing evidence has demonstrated its favorable effects on various attitudes (e.g., organizational commitment, knowledge sharing attitude; Diedericks and Rothmann, 2014; Khari and Sinha, 2018) and behavioral aspects in the workplace (e.g., creativity, in-role and extra-role performance; Redelinghuys et al., 2019; Singh et al., 2019, 2021), as well as health-related variables (e.g., physical health and vitality; Hori et al., 2019; Wissing et al., 2021). Therefore, job flourishing is considered as an ideal indicator of motivation at work and organizational effectiveness, in addition to job performance. In other words, employee job flourishing and job performance can be perceived as the ultimate goals of positive psychology in organizational contexts (Luthans, 2002; Cameron and Dutton, 2003; Bakker and Derks, 2010).

One crucial resource for reaching such goal at work is emotional intelligence (Mayer et al., 2004; Bakker and Demerouti, 2017). Emotional intelligence depicts the abilities to perceive emotions, access and generate emotions in assisting thought, understand emotions and emotional knowledge, and regulate emotions in promoting emotional and intellectual growth (Mayer et al., 2004). Emotional intelligence is thus an individual difference variable to better understand affective cues within and between people and organize them to successfully interact with one's environment. These abilities are pivotal to manage internal motivations at work which is instrumental during goal attainment processes (Mayer et al., 2004; Zeidner et al., 2004). Furthermore, the utilization of emotional intelligence is essential in social interaction and leadership processes, emphasizing its integral role in navigating interpersonal dynamics (Ashkanasy, 2003; Miao et al., 2017).

As a personal resource, employees' emotional intelligence could influence their wellbeing and job performance through different mechanisms. First, emotional intelligence is a functional resource that facilitate social interactions during the goal attainment process (Bakker and De Vries, 2020; Ruble et al., 2022). Second, emotional intelligence can optimize other resources at work to affect work outcomes (e.g., psychological capital, social support; Gong et al., 2019; Tesi, 2021). Third, it can help to manage job demands and buffer their negative impacts on the outcomes (Wang and Shi, 2020; Mérida-López et al., 2023). Evidence has demonstrated that employees' emotional intelligence could affect various wellbeing related states such as happiness (Callea et al., 2019), psychological wellbeing (Prajapati et al., 2021), work engagement (Molero Jurado et al., 2020; Wang and Shi, 2020), and flourishing (Callea et al., 2019; Nel, 2019). Furthermore, meta-analytic findings have shown that an employee's emotional intelligence is positively related to job performance (Dogru, 2022), even after controlling for the effects of cognitive ability and personality (O'Boyle et al., 2011). Hence, we expect:

Hypothesis 1a. Employee emotional intelligence is positively associated with job flourishing.

Hypothesis 1b. Employee emotional intelligence is positively associated with job performance.

2.2 Direct effects of perceived leader emotional intelligence on employee outcomes

Individuals' emotions and behaviors at work are largely influenced by leadership (Ashkanasy and Humphrey, 2011). At its core, effective leadership can be considered as an emotional process requiring the leaders to develop an optimistic vision, convey it in an emotionally captivating manner, and assist followers to stay motivated in achieving their goals (Walter et al., 2012). Leaders should be able to not only manage their own affective experience, but also manage the emotions of their followers during their goal attainment process. For these purposes, leaders need emotional intelligence as a personal resource and a potential job resource for their employees (Walter et al., 2012; Miao et al., 2016; Ashkanasy and Dorris, 2017). Consequently, leader emotional intelligence has demonstrated incremental value for followers' wellbeing beyond the effect of their own level of emotional intelligence, as well as their job performance more than the effects of leaders' cognitive ability and personality (Miao et al., 2016, 2018).

Indeed, according to the EASI theory, which accentuates the role of perceptions in understanding others' emotions, emotions of a leader affect followers' affective, cognitive and behavioral functioning. An individual's emotions provide information (i.e., thoughts, feelings, intentions) that are communicated to other people (i.e., observers) through affective reaction processes encompassing reciprocal emotional reactions and sentiments (Van Kleef, 2008). Reciprocal emotional reactions occur when an observer mirrors the emotions displayed by an expresser; for example, an observer might share in the happiness expressed by the individual. Additionally, the individual's emotional expressions have the potential to evoke sentiments in observers, such as liking, and can also significantly influence their behaviors (Van Kleef and Côté, 2022). In leadership, for instance, it is evident that the emotional expressions of the leaders is associated with followers outcomes, such that positive emotional expression is related to higher job performance (Liu et al., 2017).

Within the scope of our study, leaders function as emotional expressers and are closely observed by their employees. A leader's emotional intelligence, as conveyed through various channels including facial expressions, vocal cues, posture, textual communication, and symbolic gestures, plays a pivotal role in shaping employee outcomes by influencing their emotional responses and sentiments. When followers perceive that their leader is effectively managing one's own emotions, thereby showing positive emotional expressions, there is a higher likelihood that they will feel positive emotions as well (Clarkson et al., 2020). Furthermore, given that emotional intelligence encompasses a motivational component related to goal achievement, the observation of one's leaders with high levels of emotional intelligence fosters positive sentiments and job motivation among followers (Priest, 2021). As a result, this positive influence on employee emotions and motivation contributes to the overall enhancement of both job flourishing and performance. Thus, we hypothesize:

Hypothesis 2a. Perceived leader emotional intelligence is positively associated with employee job flourishing.

Hypothesis 2b. Perceived leader emotional intelligence is positively associated with job performance of employees.

2.3 Indirect effects of perceived leader emotional intelligence on employee outcomes via employees' emotional intelligence

Beyond its direct impact on followers' outcomes, leader emotional intelligence has been identified as influential in shaping the conditions of the followers. At the group level, leader emotional intelligence can be assumed as positively related to followers' emotional intelligence (Koman and Wolff, 2008; Chang et al., 2012). It can also affect the emotional climate of a workgroup/team (Ashkanasy and Dorris, 2017; Buskila and Chen-Levi, 2018; Majeed and Jamshed, 2021). Also at the individual level, leader emotional intelligence has a positive association with follower emotional intelligence (Wong and Law, 2002; Wong et al., 2010; Lim et al., 2018). Yet, little is known about the potential explanations underpinning this cluster of relationships (Miao et al., 2016).

Aligned with the principles of EASI theory, we posit that leader emotional intelligence can have a positive effects on follower emotional intelligence, operating through either affective or inferential processes (Van Kleef, 2008). Affective reactions processes are similar to emotional contagion processes, where employees, as the observers, mimic the emotions of their leader, as the emotions expresser. Leader emotional intelligence, evident in emotional expressions, regulations, cues, and behaviors vis-à-vis followers, are consistently observed and emulated by them. A high level of leader emotional intelligence can induce a similar level of emotional expressions among followers. The perceived positive leader emotional intelligence inclines employees toward positive sentiments, leading them to model the same positive emotional cues and regulations strategies. These imitative effects often occur automatically, with minimal involvement of cognition processes.

When operating through inferential processes, leader emotional intelligence may affect followers' emotional intelligence by engaging their cognitive functioning which stimulates more effectively behavioral changes (Van Kleef and Côté, 2022). In these processes, observers seek information about expresser's dispositions, goals, and intentions in expressing emotions and emotional behaviors. Consequently, observers draw inferences about that emotional expresser's motivation, the context of their emotions, and even their evaluations of the observers. Thus, perceived leader emotional intelligence informs employees about their leader's expectations regarding emotional display rules and work practices. A high level of perceived leader emotional intelligence provides valuable feedback for employees on how to manage their emotions, mirroring the leader's approach. These follower inferences contribute to an increased sense of affiliation, a fostering of goal-congruent emotions and ultimately to the displaying of higher levels of follower/employee emotional intelligence. Thus, we expect:

Hypothesis 3a. The positive relationship between perceived leader emotional intelligence and job flourishing of employees is mediated by employee emotional intelligence.

Hypothesis 3b. The positive relationship between perceived leader emotional intelligence and job performance of employees is mediated by employee emotional intelligence.

2.4 Positive team emotional climate as boundary condition

In addition to perceived leader emotional intelligence, another factor that can affect employee emotional intelligence is the affective tone or emotional climate of employees' workgroup/team. Formed by the day-to-day (emotions-loaded) interactions and shared experiences within an organizational unit, team affective climate entails employee perceptions of emotions and the expressions of a group of employees (De Rivera, 1992; De Rivera and Páez, 2007; Härtel et al., 2008). Thus, the emotional climate of a team reflects the meso-level processes of affective states and the emotional contagion occurring within the team (Dasborough et al., 2009).

Inviting further empirical investigation, EASI theory posits that the emotional climate, reflecting emotional sharedness within a group, is a boundary condition which can influence how emotions are communicated at work (Van Kleef et al., 2017; Van Kleef and Côté, 2022). Team emotional climate serves as a context which can influence (emotional) behaviors and social beliefs of and within the group (Páez et al., 2012; Sabucedo et al., 2017; Yzerbyt and Mahjoub, 2017). Furthermore, it can also affect each group member's emotional responses by imposing social norms able to either restrict or create emotional display rules during social interaction (Barsade and Gibson, 1998; Liu and Liu, 2013). As for leaders' emotional intelligence, such functions of team emotional climate induce the inference of appropriateness to adopt certain emotional expressions and behaviors (Van Kleef and Côté, 2022). When a positive team's emotional climate is highly positive, employees, particularly those with low emotional intelligence, would be inclined to adjust their affective expressions and behaviors to become more positive to comply with the group norms. Spurred by the direct and constant contact with their team members, employees would then be more likely to mimic any positive emotional cues and behaviors they perceive at work.

In other words, one may assume that positive team emotional climate could strengthen the adoption of emotional expressions and behaviors observed from the leaders, which is then manifested through the followers' own emotional cues and expressions, such that they enhance their own emotional intelligence level. Indeed, recent empirical research has demonstrated how a positive affective climate intervention given to work units could change the units' members' emotional intelligence (Mira-Galvañ and Gilar-Cobi, 2021). Through such influence, positive team emotional climate, boosting the influence of leaders' emotional intelligence, can induce positive change at the individual level, which in turn can have a positive impact on individual-employee outcomes. Hence, we hypothesize:

Hypothesis 4. Positive team emotional climate moderates the indirect effect of perceived leader emotional intelligence on job flourishing and job performance of employees via their enhanced level of emotional intelligence, such that this effect is stronger when positive team emotional climate is higher.

3 Methods

3.1 Sample and procedure

Given that the teaching profession often faces emotionally demanding situations which require an abundant level of affective job resources (Hakanen et al., 2006), the participants of our study were school teachers and their leaders (i.e., the school principals). After receiving approval from the university's ethics commission, we invited primary schools in the province of Yogyakarta, Indonesia. To ensure sample representativeness in relation to the targeted population and its geographical area, we employed a random sampling by inviting 50% of the schools in the city of Yogyakarta and its surrounding area (260 schools). Given the COVID-19 pandemic, the participants could choose whether they preferred to participate through an online or paper-pencil survey. After receiving their informed consent, we administered the questionnaire after a back-translation procedure from English to the Indonesian language (Brislin, 1970). To reduce common-method variance, the survey was administered twice with a 4–6-week gap, within the period of November 2020-April 2021. Such temporal separation was chosen considering not only the possibility of participants forgetting the questions but also the state-like components of the variables examined which may fluctuate over time (Podsakoff et al., 2003; Warr, 2013). To match the responses of the participants across the survey rounds, we generated a unique code for each participating individual.

In total, 179 schools/teams (response rate 68.84%) participated in the first survey. Six teams quit after the first administration of the survey, leaving 173 teams (96.65%) with 1,246 teachers completing the second survey. After removing invalid responses due to observed leniency and teams with < 3 participating teachers, the final sample consisted of 169 teams/school leaders (97.69%) with 1,210 individual teachers (97.11%). On average, each team had seven participating teachers (ranging from 3 to 17; SD = 2.57).

Sixty five percent of the participating school principals were female, and their ages ranged from 30 to 60 (M = 51; SD = 6.77). They were slightly more inclined to participate in the survey using pencil-paper mode, both at time 1 (51.2%) and time 2 (52.3%). In terms of tenure, 12.9% of the school principals had been working for < 1 year, 23.1% for 1 to 3 years, 10.4% for 3 to 5 years, and 53.6% for more than 5 years. Most of them had pursued university degrees: master's (29.2%), bachelor's (67%), diploma (2.6%), while very few had a non-university degree (1.2%).

In the teacher sample, females were dominant at 76.4%. Their age range was from 21 to 60 (M = 37; SD = 10.63). Regarding the mode of participation, 44% of them used the paper-pencil mode at time 1, while at time 2 the proportion rose to 50.5%. Among them, 65.9% had been working as teachers for more than 5 years, 9.8% for 3 to 5 years, 15.5% for 1 to 3 years, and 8.8% for < 1 year. Moreover, 31.2% of the teachers had been working with their current school principal for < 1 year, 35.4% for 1 to 3 years, 16.3% for 3 to 5 years, and 16.9% for more than 5 years. Regarding employment status, they were civil-worker teachers employed by the nation (41%), permanent teachers (19.8%), and non-permanent teachers employed by the schools (39.2%). Most of them had attained university degrees: master's (4.4%), bachelor's (92.7%), diploma (1.7%), while very few had earned a non-university degree (1.2%).

3.2 Measures

Unless otherwise noted, all measures used a 7-point Likert-type scale, ranging from 1 = strongly disagree/not at all to 7 = strongly agree/completely (see Supplementary material). The internal consistencies of the scales (Cronbach's α) are shown in Table 1.

Table 1

Table 1. Descriptive statistics, zero-order correlations, and Cronbach's alphas.

Emotional intelligence. Perceived leader and employee emotional intelligence were measured at time 1. The teachers were asked to rate their own and their leader's emotional intelligence on a 16-item questionnaire (Wong and Law, 2002), comprising four dimensions: self-emotional appraisal (e.g., “I/My school principal really understand(s) what I/he/she feel(s)”); others' emotional appraisal (e.g., “I am/My school principal is a good observer of others' emotions”); regulation of emotion (e.g., “I have/My school principal has good control of my/her/his own emotions”); and use of emotion (e.g., “I am/My school principal is a self-motivated person”). This measure has exhibited satisfactory construct validity across its four dimensions suitable for various age groups and genders, while clearly diverging from traditional personality constructs (Law et al., 2004; Kong, 2017).

Positive team emotional climate. The teachers were asked to rate positive team emotional climate using an 8-item measure (Liu et al., 2014) comprising: ego-focused positive emotion (e.g., “The teachers in this school feel energetic”); and other-focused positive emotion (e.g., “The teachers in this school get on well with each other”). To reduce common-method variance and given that team emotional climate is considered to be relatively stable within the 4–6 weeks of measurement gap, this moderator variable was measured at time 2 (Podsakoff et al., 2003). During the analysis, one item was removed since it did not fit with the sociocultural working context in Indonesia (please see Supplementary material). Further, we calculated the r_WG(J), and intraclass correlation coefficients (i.e., ICC(1) and ICC(2); LeBreton and Senter, 2008) to support aggregation to the team level. Regarding the r_WG(J), we accounted for potential biases in raters' judgement by applying three different distributions (i.e., uniform, slightly skewed and moderately skewed). The uniform and slightly-skewed distributions showed no out-of-range values, while the moderately-skewed distributions showed three out-of-range values. Given that the presence of out-of-range values is common in the skewed distributions, we conclude that no observed variance exceeded the theoretical null variance. Hence, we used the uniform distribution as the main reference point [r_WG(J) = 0.94] which indicated very strong agreement (James et al., 1984). Furthermore, the ICC(1) was 0.13, indicating a medium effect of group membership. The ICC(2) was 0.57, which was below the commonly used reliability cut-off point of 0.70, due to small group sizes of our study (ranging from 3 to 17). Thus, we justify the aggregation of emotional climate to the team level (LeBreton and Senter, 2008; Biemann et al., 2012).

Job flourishing. Job flourishing was assessed with the 6-point scale (1 = Never, 6 = Every day) flourishing-at-work (Rautenbach, 2015) at time 2. This 17-item scale includes three dimensions of wellbeing: emotional (e.g., “During the past month at work, how often did you feel happy?”); psychological (e.g., “During the past month at work, how often did you feel your work is meaningful?”); and social wellbeing (e.g., “During the past month at work, how often did you feel that you really belong to your school?”). Given the Indonesian background of this study's participants which does not fit with the sociocultural background in which the measure was developed, some items seemed to change in meaning and did not correspond to the designated factor structure (please see Supplementary material). Thus, we proceeded with the analysis with 13 items of the scale (three for emotional wellbeing, five for psychological wellbeing, and five for social wellbeing; Keyes, 2007; Rautenbach and Rothmann, 2017).

Job performance. Job performance was measured at time 2 with a 4-item measure (Gibson et al., 2009). This variable was measured using employees' leader rating as well as self-rating to reduce common-method bias (Podsakoff et al., 2003). Moreover, in the school context, school principals' ratings of teacher performance reflect more objective evaluation toward teachers' performance (Orphanos, 2014). We asked the school principals to rate all individual teachers and we asked the teachers to rate themselves. An example item from the leader's rating is “She/he performs high quality work,” while an example from self-rated job performance is “I perform high quality work.” One item was omitted since it did not fit the teachers' performance context in Indonesia.

Control variables. We controlled for participants' gender, age and employment status which have been found to be associated with flourishing (Hone et al., 2015; Hori et al., 2019) and job performance (Jimenez et al., 2022). We also controlled for the employee-leader (relationship) tenure since it could confound the hypothesized model (Sin et al., 2009; Mumtaz and Rowley, 2020).

3.3 Analytic strategy

To test the hypothesized model, we conducted multilevel structural equation modeling (MSEM) using Mplus 8.5 (Muthén and Muthén, 2017). This procedure allows us to decompose individual level variables into within- and between-level parts in testing mediation and moderation relationships (Preacher et al., 2010). We group-mean centered all the within-level variables to simplify interpretations of our findings (Aiken and West, 1991).

We compared a series of model fit indices to observe the size effects whenever new parameters were added. We also ran a series of chi-square tests based on the deviance index (−2^*log-likelihood; Bryk and Raudenbush, 1992). Table 3 presents the model-testing steps and lists the changes in deviances and variances. The direct effects of perceived leader emotional intelligence and employee emotional intelligence were calculated by modeling the direct paths from the predictors to job flourishing and the two job performance scores. When testing the mediation effects, we specified the fixed effects of perceived leader emotional intelligence on follower emotional intelligence at the within level. The indirect effects were estimated by adopting the Monte Carlo simulation (Selig and Preacher, 2008) to estimate bias corrected confidence intervals (CIs) via the software R. The cross-level moderation effect was estimated by including the random effects of positive team emotional climate on individual-level employee emotional intelligence. Finally, the full model was tested by examining the direct and indirect hypothesized relationships while controlling for the effects of all the stipulated control variables.

4 Results

4.1 Preliminary analysis

We conducted multilevel confirmatory factor analyses (CFA) prior to the hypotheses testing to examine the construct validity of all the variables. Following the procedure recommended by Hox et al. (2018), CFA was initially run at the individual level involving perceived leader emotional intelligence, employee emotional intelligence, job flourishing, self-rated and leader-rated job performance. We compared three models: (M1) a 5-factor model, in which the items of each construct were loaded on their respective general factor; (M2) an oblique 13-first-order-factor model, in which the items of each construct were loaded on their respective first order factor (4 factors for each of the subdimensions of perceived leader emotional intelligence and employee emotional intelligence subdimensions, 3 factors for job flourishing subdimensions, and 2 factors for both self-rated and leader-rated job performance); and (M3) a second-order-factor model —with 13 first-order factors and 3 second-order factors— where the items of each construct were loaded on their respective first-order factors, and only three constructs (perceived leader and employee emotional intelligence, job flourishing) were loaded on a higher order factor (Table 2).

Table 2

Table 2. Goodness of fit indices and chi-square difference test of multilevel CFA.

The models were evaluated based on goodness of fit indices using chi-square (χ²), comparative fit index (CFI > 0.90), root mean square error of approximation (RMSEA < 0.06), and standardized root mean squared residual (SRMR < 0.08) (Hu and Bentler, 1999; Nye and Drasgow, 2011). Among the three, the M2 showed the best fit in terms of a significant χ² difference. However, given that the significance of the χ² could be affected by our sample size and the model not being the best representation of original constructs' structures, we proceeded to the next step of analysis with M3 (Credé and Harms, 2015).

In addition to the preferred within-level model (M3), we compared four models at the between level by involving positive team emotional climate, employee emotional intelligence and leader-rated job performance. These models were: (M4) an independence model in which we listed all the variables at the between level with no relationships, (M5) a 3-factor model in which we loaded each construct's items on their respective general factor, (M6) a second-order-factor model (used only for employee emotional intelligence), and (M7) a saturated model which allowed all the observed variables to correlate with each other. The results showed that M6 fit both the data and original factor-structure the best, compared to the other models, with which we proceeded to the MSEM.

4.2 Hypotheses testing

We extended our measurement model to a structural model to test the hypotheses. However, due to its complexity the model did not converge. Thus, we run the MSEM analysis with observed variables. Tables 3, 4 depict the hypotheses testing results.

Table 3

Table 3. Model comparison and variance terms for MSEM.

Table 4

Table 4. Unstandardized coefficients of the multilevel analyses of the final model.

Hypotheses 1a and 1b assume that employee emotional intelligence is positively related to both job flourishing and performance. To test this assumption, we entered emotional intelligence as the predictor of job flourishing and both job performance scores (Table 3, Model 3 and 4). Given the improvement of the models, we found that emotional intelligence was positively related to job flourishing (b = 0.12, p < 0.001) and self-rated job performance (b = 0.36, p < 0.001), but not to leader-rated job performance (b = 0.02, p = 0.52). Thus, hypothesis 1a was supported and hypothesis 1b was partially supported.

Regarding hypotheses 2a and 2b, which concern the positive relationship between perceived leader emotional intelligence and both job flourishing and performance, the deviance decreased when a path was added from perceived leader emotional intelligence to job flourishing and both job performance scores (Model 5 and 6). This result implied that perceived leader emotional intelligence was positively related to job flourishing (b = 0.15, p < 0.001) and job performance. However, for the latter, the relationship was only significant for self-rated job performance (b = 0.18, p < 0.001), not for leader-rated job performance (b = 0.04, p = 0.07). Hence, hypothesis 2a was supported and hypothesis 2b was partially supported.

Hypotheses 3a and 3b assume a mediation role of employee emotional intelligence between perceived leader emotional intelligence and follower outcomes. The model testing for indirect effects of perceived leader emotional intelligence on the outcomes via emotional intelligence (Model 7) demonstrated significant results for job flourishing (b = 0.04, 95% CI [0.015, 0.052]) and self-rated job performance (b = 0.11, 95% CI [0.073, 0.129]), but not for leader-rated job performance (b = 0.01, 95% CI [-0.012, 0.024]). Finally, the total effect of perceived leader emotional intelligence produced via employee emotional intelligence on the outcomes was significant for job flourishing (b = 0.19, 95% CI [0.141, 0.238]), self-rated job performance (b = 0.29, 95% CI [0.227, 0.354]), and leader-rated job performance (b = 0.05, 95% CI [0.006, 0.092]). These results supported hypothesis 3a and partially supported hypothesis 3b.

Hypothesis 4 predicted that positive team emotional climate moderates the indirect effect of perceived leader emotional intelligence on job flourishing and job performance via emotional intelligence. We found that the variation in slopes was reduced, improving the model significantly (Δdeviance = 5.99^* on 1 df, p = 0.014), when the effect of positive team emotional climate was added to the slope between perceived leader emotional intelligence and emotional intelligence. Positive team emotional climate explained 50% [Table 2, Model 10 variance = (0.012–0.006)/0.012] of the between unit variation in the effect of perceived leader emotional intelligence on employee emotional intelligence. The cross-level effect of positive team emotional climate on the relationship between perceived leader emotional intelligence on employee emotional intelligence was significant (b = 0.16, S.E. = 0.064, p = 0.013). We estimated the slopes of the regression lines further for low (-1 SD) and high (+1 SD) levels of positive team emotional climate (Aiken and West, 1991). As illustrated in Figure 1, when positive team emotional climate was high, the slope was positive and significant. Consequently, when positive team emotional climate was high, the conditional indirect effect of perceived leader emotional intelligence produced via employee emotional intelligence on the outcomes was significant for job flourishing (b = 0.06, 95% CI [0.023, 0.029]) and self-rated job performance (b = 0.17, 95% CI [0.104, 0.0234]), but not for leader-rated performance. When positive team emotional climate was low, the conditional indirect effect of perceived leader emotional intelligence via employee emotional intelligence was only significant for self-rated job performance (b = 0.05, 95% CI [0.005, 0.096]), but not for job flourishing and leader-rated job performance. Hence, hypothesis 4 was supported. Table 5 shows the conditional indirect and total effects of perceived leader emotional intelligence on the three outcomes. In Figure 2 the tested research model and its regression coefficients are reflected.

Figure 1

Figure 1. Moderating effect of positive TEC on the relationship between perceived leader EI and employee EI.

Table 5

Table 5. Conditional indirect and total effects of perceived leader's EI on employee job flourishing, self-rated job performance, and leader-rated job performance via employee EI, conditional on positive TEC.

Figure 2

Figure 2. Hypothesized and tested path model with unstandardized coefficients.

5 Discussion

The present study examines multilevel affective determinants of employee job flourishing and performance. More specifically, by utilizing the theoretical perspective of the EASI model, we argue that a high level of perceived leader emotional intelligence affects high job flourishing and performance of followers through an enhanced level of follower emotional intelligence, under the boundary condition of a positive team emotional climate. The results support our hypotheses emphasizing the influential roles of perceived leader and employee emotional intelligence and positive team climate on followers' job flourishing and performance.

5.1 Theoretical implications

Our study contributes to the literature on emotion-at-work literature and extend the EASI model in two important ways. First, we unravel multilevel interpersonal affective mechanisms of employee job flourishing and performance by highlighting the roles of employee emotional intelligence as promoted by multilevel affective job resources from leader and team. At the intrapersonal individual level, emotional intelligence is instrumental at work because it can help to recognize and manage emotions well during the goal-attainment process, thereby maintaining or enhancing motivation (Mayer et al., 2004; Wang and Shi, 2020). It can also help to manage job demands and optimize job resources (Bakker and De Vries, 2020; Ruble et al., 2022). Our findings supported such functions of emotional intelligence as a personal resource (Bakker and Demerouti, 2007; Vermooten et al., 2021) by showing that employee emotional intelligence is able to predict both job flourishing and self-rated job performance.

At the higher level, leaders who are perceived to have high levels of emotional intelligence could bring positive interpersonal effects on employee outcomes by enhancing the employees' emotional intelligence levels. In line with the limited research on this topic (e.g., Wong et al., 2010; Chang et al., 2012), this study offers a further explanation on the effect that perceived leader emotional intelligence may have on the employee emotional intelligence promotion. Indeed, corroborating recent evidence suggesting that other-rated emotional intelligence can be more accurate than self-rated measures (Sweis et al., 2022), we opted to evaluate leader emotional intelligence based on employees' perceptions. This was deemed more appropriate in our study given our research goal. Furthermore, supported by the EASI model, emotions can be understood as a form of communication which is conveyed by an expresser to observers. The way emotional expressions are perceived by the observers plays a pivotal role in shaping not only emotional responses but also inferences triggering behavioral changes (Van Kleef, 2008, 2016). A leader's emotional intelligence level, manifested through his/her emotional expressions, cues, and regulatory behaviors, is constantly observed and internalized. Those constantly perceived emotions can thus elicit similar emotional reactions, as well as inferences about the meaning of communicated leader's emotions (e.g., emotional display rules), which can ultimately trigger employees to model and reproduce the same emotional expressions, cues, and regulatory behaviors (Van Kleef and Côté, 2022). Through such affective and cognitive processes, high perceived leader emotional intelligence could indirectly affect the level of employees' emotional intelligence, which consequently promotes job flourishing and self-rated performance.

Additionally, the perceived leader emotional intelligence could directly impact job flourishing and self-rated performance by eliciting positive emotional reactions and sentiments, which promote individuals' positive emotions and motivation. Interestingly, our results suggested that perceived leader emotional intelligence imposes a relatively equal direct effect on job flourishing as employee emotional intelligence. This implies that perceived leader emotional intelligence holds significant value and is able to impact job flourishing not solely through elevated emotional intelligence levels but also beyond. Yet, it is worth noting that the effects of perceived leader emotional intelligence on job performance were only observed for the employee self-rated job performance. Leader-rated job performance was not found to be significant as predicted directly and indirectly by perceived leader emotional intelligence via employee emotional intelligence. This might be the case since the application of MSEM controlled for the effects of team-level variance and potential confounding variables (e.g., employee-leader tenure) that might have influenced the result (Sin et al., 2009; Preacher et al., 2010). Even so, we still observe a significant total effect of perceived leader emotional intelligence on both scores of job performance. Therefore, overall, we provide support for the incremental effect of perceived leader emotional intelligence on followers' job flourishing and performance, beyond the effects of followers' own level of emotional intelligence (Miao et al., 2016, 2018).

To complete the multilevel perspective, serving as our second contribution, this study enriches the EASI model about the potential function of an emotionally positive team context as the moderator of the interpersonal dynamics at work. To complement perceived leader emotional intelligence's influence, we thus co-examined the influence of positive team emotional climate. As a higher-level factor, workgroup climate has been mostly investigated at the team/organizational level with a lack of explanation regarding its effect at the individual level (e.g., Knight et al., 2018). Here, we found it to be positively associated with perceived leader emotional intelligence (Majeed and Jamshed, 2021; Zhang et al., 2023). Hence, a team's positive emotional climate, as a contextual factor, are shown here to provide insight to top-down affective processes; With our study we answered calls for (more and stronger) evidence for the role of positive team emotional climate's influence on the interpersonal effects of emotions (Barsade and Knight, 2015; Van Kleef et al., 2017; Van Kleef and Côté, 2022). We found that positive team emotional climate acts as a moderator in the indirect effect of perceived leader emotional intelligence on employees' job outcomes via employee emotional intelligence. Thus, positive team emotional climate, serving as a conditional factor, can facilitate the emotional reactions and inferences of the employees toward perceived leader emotional intelligence (Van Kleef and Côté, 2022). When a positive team emotional climate is high, team members perceive positivity as the main emotional display rule/norm within the team, which affects them positively in the way they display their emotions (Barsade and Gibson, 1998; Liu and Liu, 2013). Therefore, a highly positive team emotional climate can persuade its members to improve their emotional regulation and expressions (i.e., their own emotional intelligence level), thereby reinforcing the type of behaviors produced by high levels of perceived leader emotional intelligence. In other words, our study sheds light on how follower perceptions of high leader emotional intelligence, together with the sharing of mainly positive emotional expressions within their group, may not only facilitate a high level of follower emotional intelligence but also elevated levels of follower job wellbeing and performance.

Lastly, it is worth noting that, even though our analysis did not include an effect of team emotional climate on leader-rated job performance, it demonstrated a positive cross-level effect on employee emotional intelligence and their two examined job outcomes. This divergence might stem from the differing perspectives people tend to have on a teacher's job performance. The emotional climate, as perceived by teachers, might not align with the aspects that school principals prioritize while assessing teacher performance. Aligned with previous research results, our findings suggest that the emotional climate within the school environment can influence teacher-rated job performance and flourishing (Rivai et al., 2019; Mailool et al., 2020). Yet in practice, such influence may still not be considered essential by school principals when evaluating their teachers' performance. This aligns with existing literature that emphasizes other factors such as class management, student improvement, and teaching abilities as primary considerations in principals' school performance assessments (Harris and Sass, 2014; Orphanos, 2014; Kraft et al., 2020).

Overall, our findings offer multilevel evidence on specific interpersonal affective mechanisms affecting the promotion of both job flourishing and performance. This empirical evidence supports the explanations for the importance of (perceived) leader emotional intelligence and a positive emotional team context at work as well as their effects on the promotion of follower emotional intelligence, which subsequently contribute to high employee job flourishing and performance.

5.2 Practical implications

As implications for practice, we found that perceived leader emotional intelligence is important factors which can contribute to facilitating workers' job flourishing and performance, particularly if the emotional climate within the team is highly positive. Thus, organizations should pay more attention to the emotional intelligence level of their leaders, given that leaders' emotional expressions are highly important for their followers who do pay (often implicit) attention to these expressions. For instance, organizations must place high value on high levels of emotional intelligence during the selection and promotion of employees for leadership positions (Lievens and Chan, 2017). Furthermore, organizations could assign coaches to already appointed leaders to enable them to refine their emotional competencies to optimize their followers' job flourishing and performance (Dippenaar and Schaap, 2017). Given the function of emotional intelligence also as a personal resource, the attention on emotional intelligence can go beyond workers at leadership positions. Once organizations value emotional intelligence for all their personnel, they could attract and retain more emotionally intelligent workers. For this purpose, organizations' identity-type statements should emphasize the importance of perceiving, understanding, and managing employee emotions (Daus et al., 2012). Moreover, at a more individual level, training and development opportunities pertaining to emotional competencies should be made available for all employees to enhance their emotional intelligence (Hodzic et al., 2018; Mattingly and Kraiger, 2019).

With regard to the effects of positive team emotional climate, organizations should consider inducing positive work climates at the team level in several ways (Parke and Seo, 2017). First, organizations should present positive expectations for workers to display authenticity by instructing, for instance, the leaders at any level to be the role model of acting authentically as well as in a functionally constructive manner (Medler-Liraz and Seger-Guttmann, 2018). Second, organizations should facilitate the utilization of positive emotions which can be done also through the initiation of programs that spark positive employee experiences, such as greeting coworkers and customers with smiles and related positive vibes but also promoting the giving of constructive feedback on job performance and the behaviors involved (Diener et al., 2020; Lane, 2021). Finally, more organizations should enable employees to engage voluntarily in training to improve their affective regulation strategies (Parke and Seo, 2017) so they could contribute better to the team climate and their own level of emotional intelligence and job flourishing.

5.3 Limitations and future research

As with all research, our study is not without limitations. First, although we have tried to minimize common-method variance by utilizing a multilevel, multi-raters, time-lagged survey in which we counterbalanced the order of the predictor, criterion, mediator, and moderator measurement (Podsakoff et al., 2003), most of the variables were measured using self-report questionnaires. Yet, we are confident that common-method bias in our data, if any, was very minimal due to the significant finding of the cross-level moderating effect (Lai et al., 2013). Future studies should minimize the risk of such bias by employing objective means (i.e., objective job performance scores; Podsakoff et al., 2003; Aguinis, 2013). Alternatively, measuring the variables over time by employing a full longitudinal design is recommendable, not only to reduce the bias but also to examine the relationship dynamics among the variables (Menard, 2011; Kline, 2014). Moreover, our study design does not infer causation. Alternative explanations may underlie the relationships between variables, such how followers' emotional intelligence or emotional expression could also affect leader emotional expression (Ashkanasy and Dorris, 2017). While beyond the scope of our study, investigating this is worth future exploration.

Second, despite the multilevel perspective that we adopted, most of the variables were measured and analyzed as individual-level variables, including perceived leader emotional intelligence. Even though the findings shed light on the potential cross-level mechanisms, it would be also beneficial to examine the mechanisms or process at the higher level simultaneously as an alternative explanation (Yammarino and Gooty, 2019). For instance, future research could examine how the collective perception of leader emotional intelligence could generate various higher-level processes and outcomes (e.g., positive team emotional climate, emotional competency, and outcomes (Maamari and Majdalani, 2017; Majeed and Jamshed, 2021; Zhang et al., 2023). Moreover, future research could explore more specific mechanisms by including some potential boundary conditions such as susceptibility of emotional contagion (Johnson, 2008; Jia and Cheng, 2021) and affect intensity (Härtel and Page, 2009; van Mierlo and Bakker, 2018).

Third, the generalizability of this study might be limited. We selected teachers as the core participants of our study to match its aim. The characteristics of teaching entails various emotional demanding conditions fitting with the goal of this study (Hakanen et al., 2006). Nevertheless, due to these characteristics and the specific sociocultural background of Indonesia, the results should be interpreted cautiously. Future replication studies must consider other professions and different cross-cultural contexts.

Lastly, some measures used in our study did not include all of the original items. Despite the back-translation procedure from English to the Indonesian language (Brislin, 1970), some items from the positive team emotional climate, job flourishing, and job performance scales were excluded from the analysis. These items seem to have been misunderstood by our Indonesian participants (Heggestad et al., 2019) who tend to demonstrate high power distance and collectivism as well as medium long-term orientation, masculinity, and indulgence (Hofstede, 2011; Hofstede Insights, 2018). For instance, a positive team emotional climate item (“enjoy a relaxed, easy-going working climate”) might have been interpreted as being lazy and not working hard. Furthermore, a job flourishing item (“feeling confident to think or express ideas or opinions”) might have been perceived as “showing-off.” Indeed, it is important to take a closer look to the cultural interpretation of each well-translated survey item prior to actual data collection (Heggestad et al., 2019).

6 Conclusion

Our study investigated two affective cross-level mechanisms on employee job flourishing and performance. By drawing on the tenets of EASI theory, we elucidate the underlying multilevel affective mechanisms of perceived leader emotional intelligence on both outcomes via individual follower emotional intelligence. Leader emotional intelligence, manifested through emotional expressions and communicated via emotional reactions and inferences, is being sensed or noted by employees which in turn can significantly enhance followers' emotional intelligence. This significant effect is found moderated by positive team emotional climate which is likely to facilitate the transfer and adoption of the learned leader emotional intelligence to followers' emotional intelligence, and, in turn, positively influences two desirable individual work outcomes. To assure high job performance and flourishing of non-managerial employees, organizations must seek to reinforce any extant positive emotional work climates in the absence of which organizations run the risk of low levels of job flourishing and performance. Given this conclusion, we recommend organizations to focus on enhancing the emotional intelligence level of their managers/leaders.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by M. de Visser, Board of Ethics Commission, University of Twente. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

RA: Conceptualization, Methodology, Project administration, Data curation, Investigation, Funding acquisition, Formal analysis, Writing— original draft, Writing—review & editing. LC: Conceptualization, Methodology, Supervision, Writing—review & editing. CW: Conceptualization, Methodology, Supervision, Writing—review & editing.

Funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This research was funded by the Indonesia Endowment Fund for Education (Lembaga Pengelola Dana Pendidikan-LPDP).

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/forgp.2024.1283067/full#supplementary-material

References

Aguinis, H. (2013). Performance Management, 3rd Edn. London: Pearson.

Google Scholar

Aiken, L. S., and West, S. G. (1991). Multiple Regression: Testing and Interpreting Interactions. London: Sage.

Google Scholar

Ashkanasy, N. M. (2003). Emotions in organizations: a multi-level perspective. Res. Multi-Level Issues 2, 9–54. doi: 10.1016/S1475-9144(03)02002-2