Although the relationship between emotional intelligence and flourishing has not yet been extensively researched, numerous studies have shown a strong association between trait emotional intelligence and various well-being outcomes. For example, literature suggests that emotionally intelligent people report better psychological adjustment (Augusto-Landa et al., 2011; Sarrionandia et al., 2018), prosocial behaviour (Wang et al., 2021), life satisfaction (Kong et al., 2019; Ruvalcaba-Romero et al., 2017), and hedonic and eudaemonic well-being (Di Fabio & Kenny, 2016). In addition, systematic reviews and meta-analyses show that trait emotional intelligence is associated with good health, including lower levels of depression, anxiety (Gómez-Baya & Mendoza, 2018), perceived stress (Newton et al., 2016), and burnout (Fiorilli et al., 2019; Jacobs & Wollny, 2022; Martínez-Monteagudo et al., 2019; Newton et al., 2016). Emotionally intelligent people also possess the social skills needed for establishing healthier and more successful interpersonal relationships with others, which is a prerequisite for well-being (Malinauskas & Malinauskiene, 2020).

Recent research on the association between emotional intelligence and flourishing suggests that emotionally intelligent people report higher levels of well-being and psychological functioning (i.e. flourishing) (Callea et al., 2019; Chamizo-Nieto et al., 2021; Di Fabio & Kenny, 2019). Findings suggest that trait emotional intelligence contributes positively to flourishing in terms of self-esteem, presence of purpose, optimism, and social and psychological prosperity in relationships.

From the discussion above, it is evident that trait emotional intelligence plays a crucial role in positive psychology, which emphasises human performance, adjustment, positive affect, and character strength. However, empirical research on the association between trait emotional intelligence and flourishing is scant. This study is aimed at addressing this research gap. Additionally, researchers have recommended that future research on trait emotional intelligence focus on potential mediators between emotional intelligence and individual well-being (Di Fabio & Kenny, 2016; Zeidner et al., 2012). The purpose of this study is to contribute to this need in research by examining the relationship between trait emotional intelligence and flourishing through positive coping behaviour.

An important factor to consider in the new normal because of its influence on individuals’ well-being is their coping behaviour. A good state of well-being, according to Ukeh and Hassan (2018), can provide immunity against stressors and reduce the utilisation of maladaptive coping resources. Effective coping behaviour is, therefore, required to adapt to ongoing digital transformation and change (Ghislieri et al., 2018).

Coping or coping behaviour signifies an individual’s conscious response aimed at diminishing a physical, emotional and/or psychological stressor (Snyder, 1999; Weiten et al., 2018). While general coping behaviour is regarded as reactive responses to a threat or loss of resources (Niessen et al., 2018), positive coping entails proactive behaviours and compelling beliefs that are associated with psychosocial resources for self-improvement (Marx, 2017). Positive coping requires psychological resources or capabilities that allow one to mitigate stressors and to deal with the conditions associated with stressors in a meaningful way (Coetzee et al., 2017; Wu et al., 2020). Individuals who adopt positive coping behaviours deal with stressors pre-emptively by (1) striving to obtain and maintain resources; (2) responding proactively when warning signs are evident; and (3) positioning themselves in circumstances that fit their resources (Niessen et al., 2018). Such individuals do not see risks, demands, and challenges as potential threats or potential causes of harm or loss. Instead, they perceive demanding situations or challenges as opportunities for finding purpose in life and creating meaning.

Problem-solving coping styles are additional manifestations of positive coping behaviours. Problem-focused coping is linked to a tendency to use cognitive reinterpretations or to solve problems in a structured manner, according to Endler and Parker (1990), who distinguished coping as a strategy and a style. Individuals who adopt this strategy experience a sense of control over the situation (Dijkstra & Homan, 2016) and greater health and well-being (Kim et al., 2022; Marx, 2017; Rudman, 2021; Wu et al., 2020). Negative coping behaviour, on the other hand, is aimed at reducing emotional distress, but the stressor is not dealt with. Consequently, negative coping behaviour is associated with pessimism, distorted thinking patterns, negative judgement, and unfair self-evaluation (Dijkstra & Homan, 2016).

The conservation of resources theory (COR) (Hobfoll et al., 2018) provides a strong theoretical basis for considering positive coping behaviour as a mediating factor in the association between emotional intelligence (a dispositional factor) and flourishing (a behavioural outcome at work). Resource loss is the main principle of the COR theory, meaning that resources lost outweigh the benefits gained (Hobfoll et al., 2018). Individuals will invest in resources to safeguard themselves from resource loss, recover from losses, and acquire resources (Hobfoll et al., 2018). As the COR theory argues that resources determine how individuals cope with stressful conditions, one can assume that individuals may utilise positive coping behaviour to prevent future resource losses. Positive coping behaviour (as mentioned above) relates to the utilisation of personal psychosocial resources (such as emotional intelligence) to contend with a stressor in a meaningful way. This approach lessens individuals’ appraisal of stressors and puts them in a resource gain cycle, which, in turn, influences their well-being and work behaviour. However, individuals do not respond to the same stressor in the same way. Even though emotional intelligence is perceived as an individual resource characteristic variable, emotional intelligence reflects individual differences in coping resources (Li et al., 2020). Emotional intelligence has four ability domains, and each domain may have a different effect on how an individual copes with stressors.

Research further supports coping behaviour as a mediator in the relationship between emotional intelligence and flourishing in two ways or conditions: (1) support for emotional intelligence as a predictor of coping behaviour (i.e. the independent variable–mediator relationship) and (2) support for coping behaviour as a predictor of flourishing at work (i.e. the mediator–outcome relationship).

In relation to the first condition, a significant amount of research has confirmed a direct relationship between emotional intelligence and positive coping behaviour. For example, trait emotional intelligence has been found to be positively associated with positive (or adaptive) coping (Asturias, 2017; Di Fabio & Kenny, 2016; Lu et al., 2022; Marinaki et al., 2017; Parker et al., 2021) and negatively related to maladaptive coping mechanisms (Espinosa et al., 2019), including social withdrawal, behavioural disengagement, anger, and self-blame (Antoniou & Drosos, 2017; Sanchez-Ruiz et al., 2021). More specifically, in a review of existing literature pertaining to the dimensions of trait emotional intelligence, Jung and Yoon (2016) reported a significant positive association between the dimensions of trait emotional intelligence – self-emotion appraisal (SEA), use of emotion (UOE), regulation of emotion (ROE), and others’ emotion appraisal (OEA) – and the cognitive-appraisal coping style. Prentice et al. (2020), however, found that individuals who rated high on SEA, OEA, and UOE mostly adopted both task- and emotion-oriented coping mechanisms. Furthermore, in the case of avoidance coping, only OEA explained a significant variance because the appraiser mimics others’ attitudes and behaviours. One could, thus, conclude that emotional intelligence is a coherent attribute of an individual that supports and contributes to adaptive coping.

Concerning the second condition, an equally large body of research has confirmed that positive coping is related to flourishing. According to the COR theory, people must invest in resources to be able to deal with stressors better so that desirable goals are achieved in the long run. Within the confines of this theory, it is maintained that a variety of responses resulting from a positive state of mental health and emotional, psychological and social well-being constitute positive coping behaviour (Coetzee et al., 2017). Each state of optimal flourishing is, therefore, supported by several self-regulating personal resources that affect the outcome of a stressor in a meaningful way. More specifically, problem-focused – and adaptive – coping strategies (including positive reappraisal, support-seeking, and planning) lead to an increased level of flourishing (Freire et al., 2016; Rahim, 2019). In contrast, emotion-focused – and maladaptive – coping strategies influence employees’ levels of flourishing negatively (Marx, 2017; Rahim, 2019). Consequently, without emotional support and management’s understanding of personal situations, employees would be unable to cope with certain job demands or stressors and, as a result, languish. Flourishing is therefore dependent on individuals’ coping behaviour (Fischer et al., 2021; Rautenbach, 2015).

Existing literature is clear on the association between emotional intelligence and flourishing, and there is a new field of research that examines the underlying process through which emotional intelligence leads to better outcomes (Extremera et al., 2020; Udayar et al., 2019). In accordance with the evidence that emotionally intelligent people report higher levels of flourishing – that is, improved mental health and emotional, psychological, and social well-being – it is important to further explore the pathways through which emotional intelligence operates. Several mechanisms have been found to mediate the positive association between emotional intelligence and individual well-being, including social support (Fiorilli et al., 2019; Schutte & Loi, 2014), trust (Kryazh & Karazin, 2018), meaning-centred coping (Sanchez-Ruiz et al., 2021), cognitive emotion regulation strategies (Extremera et al., 2020), and perceived stress (Malinauskas & Malinauskiene, 2020). Another potential mediator might be that emotionally intelligent people experience higher levels of flourishing because they adopt different coping behaviours compared to people with low emotional intelligence. Because emotional intelligence plays a significant role in how people understand and regulate their own emotions (Feldman Barrett & Salovey, 2002), one can expect emotional intelligence to have a considerable effect on coping behaviour and, therefore, on flourishing.

Considering the theoretical framework described above, the aims of this study are (1) to explore the interrelationship between trait emotional intelligence and flourishing and (2) to investigate the possible role of positive coping behaviour in mediating this relationship within the bounds of the COR theory. Consequently, the following hypotheses were formulated:

A cross-sectional study was performed between August 2021 and March 2022. A convenience sampling technique was employed to recruit knowledge workers from different industries and provinces within the South African labour market. The inclusion criteria consisted of (1) permanently employed individuals from different demographical backgrounds and sectors within the South African labour market; (2) knowledge workers, including executives, management, and professionals with a tertiary qualification who were required to perform tasks that included higher-order thinking and problem-solving (Brinkley, 2006); and (3) employees between the ages of 18 and 65 years. A total of 13 000 potential participants were invited to participate in the study.

The demographic information obtained for this study included age, gender, occupational level, and tenure. Age was measured on an interval scale ranging from 18 to 25, 26 to 35, 36 to 45, and so forth. Gender was measured on a nominal scale and included three categories, namely, ‘Male’, ‘Female’, and ‘Prefer not to say’. Occupational level was divided into the following categories: executives, senior, middle and junior management, and professionals. These demographics were asked to ensure that the inclusion criteria are met.

Three psychometrically validated measurement instruments were utilised to collect data at a single point in time.

IBM SPSS software (version 28) and the SPSS PROCESS macro programme were used for data analysis, and statistical significance was set at p ≤ 0.05. The data set was scrutinised for unresponsive and unengaged responses. A total of 435 completed responses were included in the analysis, yielding a response rate of 3.34%. Descriptive statistics were used to examine the demographic profile of the respondents and the main study variables at the item level (Appendix 1). In terms of gender, the sample included 49.6% males, 49.6% females and 0.7% who preferred not to say. Most (42.6%) respondents were between the ages of 46 and 55. In respect of job level, 29.7% of the respondents were in executive/top management positions, 37% were in senior management positions, 26.2% were in middle and junior management positions, and 7% were employed as professionals or specialists.

The next step in the analysis was to assess the psychometric suitability of the measurement instruments for the current sample. Because of concerns about common method bias (CMB) and the cross-sectional nature of the research, a one-factor confirmatory factor analysis (CFA) (loading all the items onto one factor) was conducted and Harman’s one-factor solution was calculated. Common method bias may appear when the variables are captured by the same response method, potentially influencing the reliability and validity of the measures and resulting in a bias of the parameter estimates of the relationships between the variables (Jordan & Troth, 2020; Kock et al., 2021).

Descriptive statistical techniques were used to organise, analyse and interpret the data at the construct level. The characteristics of the data in relation to the research constructs were then explained. Pearson’s correlation coefficients were determined to evaluate the bivariate correlations between the variables. Finally, the bias-corrected percentile bootstrap method was used to investigate the mediating effect of positive coping behaviour, with the 95% lower level (LLCI) and upper level (ULCI) ranges not exceeding zero. PROCESS version 3.0 SPSS macro (Model 4) was used to analyse the mediation effect (Hayes, 2018).

The data were screened for univariate and multivariate outliers prior to the testing of the mediation model, and the means, standard deviations, skewness, and kurtosis of each item were calculated. Compared to the standard error of kurtosis (SE), the ratios of kurtosis were found to be higher than average. Even though some of the variables had limited variation, there were no serious issues that necessitated item exclusion. Annexure 1 outlines the results of the univariate analysis on the item level.

The next step was to test for CMB using Harman’s one-factor (or single-factor) test and CFA. The purpose of this test is to load all of the variables into a single exploratory factor analysis (EFA) and examine the unrotated factor solution to determine the number of factors that are required to account for the variables’ variance (Podsakoff et al., 2003). The results of these tests are summarised in Table 1.

The one-factor solution for the AES showed that the construct accounted for 28.5% of the covariance among the scale variables, which is below the recommended threshold of 50% (Podsakoff et al., 2003). On the contrary, the one-factor solution for the PCBI was above the recommended threshold (59.2%) suggesting a unidimensional measurement model. The one-factor CFA solution, however, indicated that the single factor did not fit the data well, in that absolute and relative fit indices were below the suggested 0.90 threshold (Hu & Bentler, 1999). In accordance with the recommendations made by Podsakoff et al. (2003), the one-factor results and poor model fit suggested that CMB did not pose a serious threat to the research findings.

A CFA was performed to evaluate the extent to which the original four-factor models fit the data and to determine the construct validity of the measurement scales. As shown in Table 2, the original four-factor solution of both the AES and the PCBI had less than acceptable data fit, in that the normed fit index (NFI) (AES = 0.71; PCBI = 0.79), comparative fit index (CFI) (AES = 0.77; PCBI = 0.85), and Tucker Lewis index (TLI) (AES = 0.76; PCBI = 0.84) values were lower than the suggested threshold value of 0.90 (Hu & Bentler, 1999). However, the original four-factor solution of both scales had a better model fit than the one-factor CFA solution, with improved absolute and relative fit indices. Lower AIC scores (AES = 1733.4; PCBI = 4447.9) were also reported.

Model modification of the original four-factor models was conducted to improve the fit indices. The modification indices were assessed and items with standardised residuals equal to or greater than 2.50 were removed to account for the correlations between the variables in the data set. The standardised residuals (no values > 2.50) and modification indices did not reveal any additional modifications to the model. As shown in Table 2, the best model fit was obtained by the modified four-factor model for both scales. These results show that both emotional intelligence and positive coping behaviour had good construct validity.

The psychometric properties (internal reliability and construct validity) of the measurement instruments were evaluated next. The results obtained are summarised in Table 3.

As shown in Table 3, the internal consistency and composite reliability (CR) values of the three measuring instruments were above the recommended threshold of 0.70 (DeVellis & Thorpe, 2022; Hair et al., 2019). The values were determined as follows: emotional intelligence (α = 0.80; CR = 0.89), positive coping behaviour (α = 0.90; CR = 0.94), and flourishing (α = 0.94; CR = 0.94). The convergent reliability average variance extracted (AVE) values of the variables were determined as follows: emotional intelligence (0.68), positive coping behaviour (0.79), and flourishing (0.66). These values were above the recommended 0.50 threshold prescribed by Hair et al. (2019). Lastly, the maximum reliability (MaxR(H)) values exceeded the suggested threshold of 0.70 (Hancock & Mueller, 2001). Considering the measurement model fit, and reliability and validity values, the data seemed appropriate for further analysis.

Table 4 sets out the descriptive information computed for the three measurement instruments. Firstly, the AES had an overall mean score of 4.2 (maximum 5; SD = 0.51), which indicates that the majority of respondents received high scores on this scale. The utilisation of emotions subscale received the highest scores from the participants (M = 4.33; SD = 0.71) and the lowest score on the perception of emotions subscale (M = 3.99; SD = 0.62).

Secondly, the overall mean score for the PCBI was 4.76 (maximum 7; SD = 0.96), indicating that respondents generally had favourable self-evaluations regarding their coping behaviours (Coetzee et al., 2017). The respondents further scored the highest on intentional coping (M = 4.95; SD = 1.02) and the lowest on engaging coping (M = 4.60; SD = 1.19).

Lastly, the overall mean score for the FS was 5.77 (maximum 7; SD = 1.04), which indicates that the respondents perceived themselves in positive terms representing several psychological resources and strengths.

Under the assumption that the sample is large and therefore the central limit theorem is assumed, Pearson product–moment correlations were calculated to determine the magnitude and direction of the relationship between the research variables (see Table 4). The bivariate correlations indicate that emotional intelligence (AES) is positively associated with positive coping behaviour (PCBI) (r = 0.23; small practical effect; p < 0.001) and flourishing (FS) (r = 0.34; medium practical effect; p < 0.001). Also, positive coping behaviour (PCBI) is positively related to flourishing (FS) (r = 0.45; medium practical effect; p < 0.001).

When the interrelationship between each dimension of emotional intelligence and positive coping behaviour is tested, the sub-dimensions of emotional intelligence are significantly (p < 0.05) related to the four sub-dimensions of positive coping behaviour. Despite these positive associations, the results show an insignificant association between perception of emotions and inventive coping behaviours (r = 0.08; p = 0.09).

The bivariate correlation analysis provided evidence for hypotheses 1 and 2.

The mediating effect of positive coping behaviour (a single mediator) in the relationship between emotional intelligence and flourishing was explored by means of PROCESS version 3.0 macro for SPSS (Model 4) (Hayes, 2018). A simple mediation analysis was conducted as the effect of a single mediator was tested. The results of the mediation analysis are reported in Table 5 and illustrated in Figure 1.

Firstly, the results indicated that emotional intelligence was a significant direct predictor of flourishing (path c) (β = 0.69, SE = 0.09, p < 0.001; LLCI–ULCI range did not include zero: 0.51–0.87) in the absence of the mediator.

Moreover, the results further estimated that 5% of the variance in positive coping behaviour is explained by a variation in emotional intelligence (R² = 0.05; F[1, 43] = 24.05; p < 0.001; small practical effect). The results showed that emotional intelligence was a significant predictor of positive coping behaviour (path a) (β = 0.44, SE = 0.09, p < 0.001; LLCI–ULCI range did not include zero: 0.26–0.61), and positive coping behaviour subsequently predicted higher levels of flourishing (path b) (β = 0.43; SE = 0.05; p < 0.001; LLCI–ULCI range did not include zero: 0.34–0.52). A 95% bias-corrected confidence interval based on 5000 bootstrap samples indicated that the indirect effect through positive coping behaviour (path ab) (β = 0.19; SE = 0.05; p < 0.001; LLCI = 0.09, ULCI = 0.29) was above zero, providing evidence of a mediating effect.

Lastly, higher levels of emotional intelligence were associated with an increase in flourishing when considering the indirect effect of emotional intelligence on flourishing through positive coping behaviour capabilities (path c’) (β = 0.50; SE = 0.09; p < 0.001; LLCI–ULCI range did not include zero: 0.33–0.68). The results further showed that approximately 26% of variance in flourishing was accounted for by a variation in both emotional intelligence and positive coping behaviour (R² = 0.26; F[2, 42] = 75.76; p < 0.001; large practical effect).

The construct descriptives show that the knowledge workers who participated in this study are emotionally intelligent. The participants furthermore scored the highest on the utilisation of emotions construct, which involves using emotions and moods in activities related to problem-solving. In addition, the results indicate that the participants mostly adopt intentional coping behaviours. These findings suggest that the knowledge workers who participated in this study feel confident in their efforts and capabilities to adopt positive, purposeful, and adaptive behaviours (Marx, 2017). Moreover, the findings suggest that knowledge workers flourish despite the changes and challenges brought forth by the 4IR. From these findings, the following conclusion is drawn: As knowledge workers invest resource capabilities (such as moods and emotions) into addressing problems or challenges, they will naturally flourish because they adopt intentional, positive coping behaviours. These findings are consistent with those of Rudman (2021), who found that South African legislators had relatively strong positive coping behaviour capabilities, which were supported by adequate psychological resources such as emotional intelligence. Similarly, existing research supports the notion that professionally qualified workers (such as nurses and educators) with high emotional intelligence tend to adopt positive coping styles, which are significantly associated with well-being outcomes (Extremera et al., 2020; Lu et al., 2022).

The findings of the present study further show empirical support for the theoretical relationship between trait emotional intelligence and flourishing (H₁). These findings are consistent with the results reported by Di Fabio and Kenny (2019), namely, that trait emotional intelligence predicts higher levels of flourishing because emotional intelligence is positively associated with social and psychological prosperity in respect of relationships, self-esteem, presence of purpose, and optimism. These findings are to be expected owing to the strong association between emotional intelligence and various well-being outcomes. In the context of the 4IR, these findings are even more significant because emotional intelligence is thought to be a precursor to resilience, which enables employees to effectively deal with change and uncertainty in the workplace (Mrugalska & Ahmed, 2021; Reshman & Sripirabaa, 2020).

Moreover, trait emotional intelligence significantly and positively predicted knowledge workers’ positive coping behaviour (H₂). More specifically, the results indicate that participants with high emotional intelligence are more likely to exhibit positive coping behaviours. These findings agree with those of previous studies that found significant relationships between trait emotional intelligence and adaptive, task-oriented and emotion-focused coping (Fteiha & Awwad, 2020; Prentice et al., 2020; Sanchez-Ruiz et al., 2021). There was, however, no significant relationship between perception of emotions and inventive coping behaviours, which suggests that knowledge workers’ awareness and observation of their emotions and those of others (Schutte et al., 2009) may not necessarily allow them to be innovative in dealing with stressful situations or to confidently devise a plan to deal with problems and/or stressful events (Coetzee et al., 2017). Similar findings were reported by Rudman (2021).

Regarding the mediating role of positive coping behaviour, the bootstrap mediation test showed that trait emotional intelligence predicted flourishing through positive coping behaviour capabilities (H₃). This finding implies that knowledge workers who display high levels of emotional intelligence are more inclined to flourish through positive coping capabilities. In line with the COR theory, emotionally intelligent individuals will naturally invest in personal psychosocial resources (such as extroversion, social support, and agreeableness) to lessen their appraisal of a stressor, which, in turn, influences their well-being and work behaviour. Research conducted by Extremera et al. (2020) supports this finding by explaining that the positive association between emotional intelligence and well-being might be attributed to the different coping mechanisms (such as planning and positive reappraisal) that emotionally intelligent people use to maintain subjective well-being. More importantly, individuals’ ability to take control of their emotions allows them to choose psychosocial resources that downregulate their negative emotions, allowing them to maintain positive affect, which, in turn, is associated with less psychological distress and more resilience (Sanchez-Ruiz et al., 2021). Positive coping behaviour is therefore an important mechanism for enhancing knowledge workers’ sense of well-being.

On a theoretical level, the study offers a better understanding of the pathway through which emotional intelligence leads to better well-being outcomes. This study shows that emotionally intelligent knowledge workers experience a greater sense of well-being than workers with low emotional intelligence because of their positive coping capabilities. One could further argue that knowledge workers flourish in a world of work that is characterised by change and uncertainty owing to their ability to regulate their emotions. Knowledge workers will, in an effort to attain well-being, attempt to improve their personal psychosocial resources.

On a practical level, the following suggestions are made: