The Effect of Emotional Intelligence and Leadership Styles on Job Satisfaction

Transformational leadership, defined by Yukl (1989) as "the process of influencing significant and major changes in the attitudes and behaviors of the members of the organization and creating loyalty to the mission and goals of the organization," focuses on the leader–follower relationship that benefits both whole organization and individuals working in the organization (Kent and Chelladurai, 2001). In today's sense, the foundations of the dimensions of transformational leadership were introduced by Bass and Avolio (1990). These dimensions listed as "idealized influence, inspirational motivation, intellectual stimulation, and individual support" are explained below (as cited by Bass and Avolio, 1990; Deluga and Souza, 1991; Karip, 1998).

In the dimension of idealized influence, feelings toward the leader are expressed. Level of appreciation and respect, admiration, and trust of the followers toward the leader determine the idealized influence (Kelsay, 2010). One of the dimensions of transformational leadership is being able to motivate followers by inspiring them. In this dimension, credibility of the person who is embraced as a leader is not questioned (Yukl, 1989) and he/she is respected. When intellectual stimulation is in question, followers are encouraged to question, criticize, or generate new ideas about the way things are done or the procedures of doing work (Karip, 1998). In addition, leaders help their followers to be more creative as well. Considering the individual support dimension, it is known that the leader pays special attention to the subordinates, treats each employee individually and diligently, while, at the same time, coaching and giving advice to them (Bass, et al., 2003).

Transactional leadership was verbalized by Burns in 1978, and then it was re-evaluated by Bass in 1985, and the behavior patterns of the leaders were examined accordingly (Bass, et al., 2003; Gedik, 2020). In the behavior between leader and his/her followers, the leader fulfills the basic needs of his/her followers, their security requirements, and their needs for social dignity, such as having a reputation. Followers, on the other hand, are extremely sensitive to the demands and expectations of the leader (Deluga and Souza, 1991).

In the literature, the dimensions of transactional leadership are defined as "contingent reward, management-by-exception (active and passive), and laissezfaire leadership" (Karip, 1998). In the dimension of contingent reward, needs of subordinates and resources of the organization are determined; also, everyone knows from the very beginning what are the conditions for granting rewards (such as which goals are achieved or which cases are experienced) (Deluga and Souza, 1991; Bass, et al., 2003). The leader keeps his/her followers under observation in the dimension of active management-by-exception and endeavors to discipline employees with a punitive method when mistakes are made, and deviations from goals or objectives occur (Bass, et al., 2003). In the dimension of passive management-by-exception, the leader either intervenes after the problems arise or does not intervene at all. According to laissez-faire leadership style, when followers ask questions or expect solutions to problems, they are not answered, and the method of procrastination is resorted (Rowold and Schlotz, 2009).

In the literature, it is stated that transactional leadership positively affects job satisfaction (Bryman, 1992; Duckett and Macfarlane, 2003; Walumbwa, et al., 2005; Bushra, et al., 2011; Belias and Kousteliosi, 2014; Tanrıverdi and Paşaoğlu, 2014; Boamah, et al., 2018). In addition, many studies that address two leadership forms together claim that both transformational leadership and transactional leadership positively affect job satisfaction (Fuller, et al., 1999; Duckett and Macfarlane, 2003; Walumbwa, et al., 2005; Bushra, et al., 2011; Belias and Kousteliosi, 2014; Tanrıverdi and Paşaoğlu, 2014; Boamah, et al., 2018;). Thus, we formulated the following hypotheses:

H₁: Transactional leadership positively affects job satisfaction.

H₂: Transformational leadership positively affects job satisfaction.

The concept of job satisfaction is characterized as “the satisfaction or dissatisfaction of employees with the work they do, and it is indicated that if the employees are satisfied with their jobs, their job satisfaction is high, otherwise it is low” (Özkalp, et al., 2013). Job satisfaction is an attitude rather than a behavior and interacts with performance factors (Robbins and Judge, 2012).

When the literature is examined, it is understood that factors leading to job satisfaction are generally grouped under the main headings as individual and organizational factors. Organizational factors that are effective in job satisfaction are stated as wage, the job itself, promotion opportunities, working conditions, counseling, colleagues, perception of justice, perceived organizational support, availability of social support, stress level of the job, presence of work–family balance, and management style (Özkalp, et al., 2013). In addition to organizational factors, there are also individual factors such as personality, age, gender, education level, marital status, time spent in the profession, title, suitability of the individual for the job, ability, and intelligence level of the individual (Eğinli, 2009; Kantar, 2010).

The concept of emotional intelligence first emerged with Thorndike's studies on social intelligence in 1920. In his article for Harper's Magazine, Thorndike argued that "social intelligence," which is an aspect of the concept of emotional intelligence, is individually a part of cognitive intelligence (Goleman, 2019). In the literature, the concept of emotional intelligence is defined as “the ability of individual to follow emotional mobility of both oneself and others, to distinguish and identify these emotions, to reflect and guide these data acquired to their behaviors by managing their emotions” (Mayer and Salovey, 1993). Emotional intelligence is the order of non-cognitive capacities, competencies, and skills that affect the ability to cope effectively with environmental pressures and demands (Bar-On, 2006).

According to the literature, there is a positive and significant relationship between emotional intelligence and job satisfaction (Wong and Law, 2002; Law, et al., 2004; Sy, et al., 2006; Joung and Kim, 2006; Wong, et al., 2010; Jordan and Troth, 2011; Ealias and George, 2012; Liu and Liu, 2013; Sudak and Zehir, 2013; Tokmak, et al., 2013; Saeed, et al., 2014; Ngirande and Timothy, 2014; Meisler, 2014; Miao, et al., 2016;). Thus, the following hypotheses was put forward:

H₃: Emotional intelligence positively affects job satisfaction.

H₄: Emotional intelligence undertakes an intermediary role in the transactional leadership and job satisfaction relation.

H₅: Emotional intelligence undertakes an intermediary role in the transformational leadership and job satisfaction relation.

The purpose of the study is to determine the impact of transformational and transactional leadership forms applied by senior managers to employees working in different positions in Turkey on their job satisfaction behaviors. Moreover, the study also aims to determine the impact of employees' perceptions of emotional intelligence on job satisfaction behaviors. In addition, the intermediary role of emotional intelligence in the impact of transformational and transactional leadership forms on job satisfaction was examined as well.

In line with this objective, a questionnaire study was conducted on 542 employees working in different positions in the private and public sectors.

Based on the findings obtained from the literature review and the theory, five hypotheses were formulated for the study.

The research model is shown in Figure 1.

Figure 1.

Conceptual model

(Source: Authors’ own research)

The study population consists of private sector and public sector employees aged between 18 and 65 with different demographic characteristics in business life. The convenience sampling method was implemented in the study. The research was carried out through Google forms, and the questionnaire was distributed electronically via social media, e-mail, and messaging applications of smartphones. The support of social circles was requested for the questionnaire and they were also asked to forward it to their contacts. The forms were answered and sent back electronically, and 542 questionnaires were retrieved. It is considered that the size of the research population is sufficient, since the size of research population should be at least 384 subjects, which can provide a population size of 100 million with a confidence interval of 0.95 according to the literature (Yazıcıoğlu and Erdoğan, 2004).

As part of the study, it was aimed to measure the perceptions of public and private sector employees, aged between 18 and 65 with all kinds of demographic characteristics, on emotional intelligence, transformational leadership, transactional leadership, and job satisfaction. The transformational–transactional leadership style scale implemented in the study was created by Bass and Avolio (1995) and was also applied by Uçar (2018) in his doctoral dissertation study. The emotional intelligence scale developed by Wong and Law (2002) was implemented. The scale developed by Wong and Law (2002) was employed by Atilla (2012) in his doctoral thesis and by Özdemir (2016) in his master's thesis.

The scale developed by Keser and Bilir (2019) was used to measure job satisfaction. The scale developed by Keser and Bilir (2019) was employed by Yıldız (2021) in his master's thesis. The scale is in a 5-point Likert system and consists of a total of five statements, and there are also two reverse-coded statements in the scale.

For analysis, SmartPLS 3.3.3. packaged software was utilized together with the method of Partial Least Squares Structural Equation Modeling (PLS-SEM). To examine the validity and reliability of the study measurement model, consistency, reliability, convergent validity, and discriminant validity were assessed. In this context, Cronbach Alpha (α), Composite Reliability (CR), and Average Variance Extracted (AVE) values were used. Table 2 shows the measurement model results.

Internal consistency reliability was preserved in the study (Cronbach alpha coefficients were between 0.853 and 0.969, and CR coefficients were between 0.887 and 0.972).

Considering the values shown in the table, it is understood that loadings of factor are between 0.645 and 0.921. According to Hair, et al. (2016) factor loadings ≥ 0.708. Authors recommend that statements with factor loads between 0.40 and 0.70 be excluded from the model in case of an increase in AVE or CR value. AVE and CR values calculated in the study were above the threshold value. For this reason, questions measuring emotional intelligence statements numbered 9, 10, and 11, with factor loadings below 0.708, were not removed from the scale. Convergent validity was achieved in the study (factor loadings of the structures are between 0.645 and 0.921, and the AVE values are between 0.529 and 0.781). In the analysis of divergent validity, the Fornell and Larcker (1981) criteria and the Heterotrait-Monotrait Ratio (HTMT) criteria proposed by Henseler, et al. (2015) were applied. Tables 3 and 4 show that divergence validity was achieved.

The values of HTMT are shown in Table 4.

Figure 2 shows the results of the structural equation model.

Figure 2:

Structural equation model

(Source: Authors’ own research)

(Variables used in the research: BD: individual support, DK: using emotions, EU: intellectual stimulation, IEA: idealized influence-attributed, IED: idealized influence-behavior, İT: job satisfaction, IVM: inspiring motivation, KDD: evaluating your own emotions, KÖ: contingent reward)

With respect to the research pattern, the PLS algorithm was run to calculate linearity (VIF), path coefficients R², and effect size (f²). Blindfolding analysis was also run to calculate the prediction power (Q²). Table 5 shows the research pattern coefficients.

When evaluating R² coefficients, conditions of the research discipline should not be ignored. When the marketing discipline is considered, 75% and above R² coefficients are evaluated as strong, while 50% and above R² coefficients can be interpreted as strong in the organizational behavior discipline. However, in some scientific studies, it would be sufficient to calculate the R² value as 10% (Yıldız, 2020). When we look at the R² values in Table 5, it is found that job satisfaction is defined by 31%. In the light of this information, it is observed that job satisfaction within the scope of working discipline has moderate explanatory power for the model, while emotional intelligence is among the desired values, yet it does not have explanatory power.

In the study, the effect size (f2) coefficient was calculated for all independent variables. When we look at the explanatory rate of the dependent variables, it is understood that f2 coefficient indicates the share of the independent variables (Yıldız, 2020). When the impact size (f2) coefficients are examined, it is observed that transactional leadership has a low impact size on job satisfaction, while emotional intelligence has a low impact size on job satisfaction. However, it is not possible to mention that transactional leadership has an effect on emotional intelligence and transformational leadership has an effect on job satisfaction and emotional intelligence. In Table 5, it is determined that the Q2 value is greater than zero. Therefore, the research pattern has predictive power.

According to Table 5, VIF coefficients are below 5. In this context, there is no problem in ensuring linearity between the variables. Standard errors within the scope of the research are estimated using the bootstrap method. In this study, 5000 samples were recreated with the bootstrap method and the t values were calculated. Table 6 shows the research design (hypothesis tests).

To explore the mediation effect, at the first stage, the intermediary variable was withdrawn from the research pattern and the significance of path coefficients was tested. It was determined that transactional leadership affected job satisfaction (β = 0.295, p < 0.01). In the light of these findings, hypothesis 1 of the research was backed up. In addition, it was found that transformational leadership also affected job satisfaction (β = 0.220, p < 0.01). In the light of these findings, hypothesis 2 of the research was backed up.

At the second stage, the intermediary variable was incorporated in the model and path coefficients’ significance was tested. It was determined that emotional intelligence affected job satisfaction (β = 0.253, p < 0.01). Thus, hypothesis 3 of the research was also supported.

When the model is examined, the intermediary effect cannot be discussed because the independent variables have no significant effect on the dependent variable, the independent variable has no significant effect on the mediator variable, and the mediator variable has no significant effect on job satisfaction. In the light of this information, hypotheses 4 and 5 of the research were not accepted.