Abstract
This study investigated the relationship among emotional intelligence, gratitude, and subjective well-being in a sample of university students. A total of 365 undergraduates completed the emotional intelligence scale, the gratitude questionnaire, and the subjective well-being measures. The results of the structural equation model showed that emotional intelligence is positively associated with gratitude and subjective well-being, that gratitude is positively associated with subjective well-being, and that gratitude partially mediates the positive relationship between emotional intelligence and subjective well-being. Bootstrap test results also revealed that emotional intelligence has a significant indirect effect on subjective well-being through gratitude.
Introduction
Subjective well-being (SWB) refers to the overall evaluation of the quality of life that an individual experiences according to his own standard (He et al., 2014). SWB reflects social function and adaptability, and it serves as a comprehensive indicator of one’s quality of life (Diener et al., 2013). Diener (2000) stated that SWB consists of two components: cognition and affect. Cognition refers to satisfaction with life, and affect refers to positive and negative affects. High levels of SWB indicate high satisfaction with life, high positive affect, and low negative affect (Yamaguchi et al., 2016; Zhang et al., 2014). SWB is closely related to depression, suicide, violent behavior, and substance abuse; thus, improving SWB can effectively promote the mental health of an individual (Muffels and Headey, 2013; Zhang et al., 2013). Researchers have extensively studied the factors that influence SWB. The objective factors of an individual, such as economic income, daily life experiences, and cultural atmosphere, can significantly affect SWB (Armitage, 2016; Parker et al., 2015). However, internal factors, such as self-efficacy, coping style, and personal trait, significantly affect SWB (Wang et al., 2014; Xiao et al., 2014). Studies documented that personal traits, such introversion, extroversion, and nervousness, can significantly predict SWB; an individual displaying more extroversion and less introversion tends to have high SWB (Pu et al., 2017).
Emotional intelligence (EI) is the ability to perceive, evaluate, and express emotion; the ability to use emotion to promote the way of thinking; and the ability to understand, analyze, and express emotion (Ouyang et al., 2015). Goleman (1996) speculated that EI is a decisive factor of success and a synthesis of the ability for self-recognition, emotional management, self-motivation, transference, and interpersonal skills. The relationship between EI and SWB is currently being investigated. For example, individuals showing high EI can recognize and express their emotions; these individuals have a positive self-identity and can fulfill their potential and live a happy life (Bar-On, 2005). Overall life satisfaction is influenced by how an individual recognizes and deals with emotional information regardless of whether they can quickly recognize their emotional response, their response to emotional traits, and their ability to regulate the involved mood (Larsen, 2000). According to Gallagher and Vella-Brodrick (2008), individuals with high EI can adopt a positive coping style when facing pressure, thereby improving their SWB.
Gratitude is an emotion in which individuals feel grateful for the favor or help that they receive and try to return (Ng and Wong, 2013; Proyer et al., 2013). Gratitude includes trait gratitude (ongoing emotion) and state gratitude (spontaneous emotion). Individuals with high levels of trait gratitude tend to experience and express gratitude more frequently and with greater intensity (Gallagher and Vella-Brodrick, 2008; Watkins et al., 2015). Gratitude is a life orientation in which an individual focuses on and appreciates the positive side of life (Tsang et al., 2014). Studies showed that gratitude can significantly predict SWB (Chaves et al., 2016; Jackowska et al., 2016; Kong et al., 2015). Hill and Allemand (2011) found that compared with individuals displaying low levels of gratitude, those who show a high level of gratitude show less negative affect and pessimism and more positive affect and optimism, and they experience higher life satisfaction. Watkins et al. (2015) drew a similar conclusion: grateful individuals relatively enjoy active affects, such as satisfaction, happiness, and hope, and they experience less jealousy, depression, and other negative emotions. Moreover, these researchers pointed out that gratitude and SWB can reinforce each other in a virtuous circle (Watkins et al., 2015). These results were also found in other adult and youth populations (Adler and Fagley, 2005; Froh et al., 2009).
Although few studies have investigated the relationship between EI and gratitude, certain reports have showed that they are positively correlated. For example, Rey and Extremera (2014) found a positive correlation between EI and gratitude. Regression analysis indicated that EI can significantly predict gratitude. EI is the ability to manage emotional and feeling information. Gratitude is an important life variable; it is a feeling and an emotion that encompasses state gratitude and trait gratitude, as well as the attitudes toward life. As an emotion or feeling, gratitude may become the object of EI regulation and management. Gratitude is influenced by EI or personality trait and predicts the category of EI. This circumstance suggests that gratitude can be predicted by EI. Moreover, a logical relation exists between EI and gratitude. Based on the above findings, our first hypothesis is that EI significantly influences the positive prediction on gratitude. Studies show that both EI and gratitude significantly influence SWB; however, the trilateral relationship among SWB, gratitude, and EI remains unknown. Individuals with high EI tend to experience and express gratitude and may further achieve high SWB. Thus, our second hypothesis is that gratitude mediates the effect of EI on SWB.
Methods
Participants and procedure
Participants were recruited through random-digit selecting students ID from two general universities in Beijing. In total, 365 undergraduates participated in this study, which consisted of 179 women and 186 men with a mean age of 20.17 years (standard deviation (SD) = 1.47) ranging from 18 to 23 years, all unmarried. In total, 87 of them were freshmen, 96 were sophomore, 92 were junior, and 90 were senior. A total of 365 questionnaires were distributed, and all of them were collected and valid. The purpose and the significance of this study were informed and all participants signed written informed consent before completing the questionnaires. Participants received 10 RMB for compensation. The research described in this article meets the ethical guidelines of the Beijing Jiaotong University and was approved by its ethics committee. Moreover, the research was conducted in adherence to the legal requirements of the People’s Republic of China.
Instruments
EI scale
The Wong & Law Emotional Intelligence Scale was used to assess EI (Wong and Law, 2002). It consists of 16 brief statements. The scale consists of four dimensions as follows: self-emotion appraisals (SEAs), others’ emotion appraisals (OEAs), regulation of emotion (ROE), and use of emotion (UOE). SEA relates to the individual’s ability to understand their deep emotions and be able to express these emotions naturally. OEA relates to people’s ability to perceive and understand the emotions of those people around them. ROE relates to the ability of people to regulate their emotions, which will enable a more rapid recovery from psychological distress. UOE (or Emotional Facilitation of Thought) relates to the ability of individuals to make use of their emotions by directing them toward constructive activities and personal performance. Some sample items are as follows: “I have a good understanding of my own emotions” (SEA), “I have good understanding of the emotions of people around me” (OEA), “I am quite capable of controlling my own emotions” (ROE), and “I would always encourage myself to try my best” (UOE). Items are rated on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). Higher scores reflect higher levels of EI. The scale was written in Chinese and has good reliability and validity (Hu et al., 2016; Kong and Zhao, 2013). In this study, Cronbach’s alpha coefficients for the four subscales were SEA: 0.75; OEA: 0.70; ROE: 0.73; and UOE: 0.88.
Gratitude Questionnaire-6
Gratitude Questionnaire-6 (GQ-6) is a six-item self-report measure assessing the frequency and intensity of grateful experiences (McCullough, 2002). Items consist of statements such as “I have so much in life to be thankful for” (Item 1) or “I am grateful to a wide variety of people” (Item 4) and are responded on a 7-point Likert scale from “1” (strongly disagree) to “7” (strongly agree). GQ-6 items load onto a single factor which is distinct from constructs such as happiness, well-being, optimism, and vitality (McCullough, 2002). GQ-6 was translated into Chinese by Chan (2010) and has good reliability and validity. In this study, Cronbach’s alpha coefficient for GQ-6 was 0.79.
SWB measures
The SWB Scale includes three subscales that measure life satisfaction, positive affect, and negative affect. Life satisfaction scale consists of five items that ask participants to respond whether they agree or not with a given description, using a 7-point Likert scale ranging from “1” (strongly disagree) to “7” (strongly agree). Examples of items include the following: “In most ways, my life is close to my ideal” and “I am satisfied with my life.” Scores are the sum of items with reverse coding of relevant items. Positive and negative affect scales are made up of six and eight words, respectively. Each scale describes one kind of positive or negative emotion (e.g. angry, shameful, and proud). Participants are then asked to respond on how often they experienced the emotional state on a 7-point rating scale ranging from “1” (not at all) to “7” (all the time) (Diener, 2000). SWB measures were translated into Chinese by Knight et al. (2009) and have good reliability and validity. Cronbach’s alpha coefficients of the life satisfaction scale and the positive and negative affect scales in this study are 0.79, 0.87, and 0.77, respectively.
Analysis strategy
To analyze the mediation effects, the two-step procedure suggested by Anderson and Gerbing (1988) was adopted: first, the measurement model was tested to evaluate whether the latent variables can be represented by relevant indicators; if the hypothesis of measurement model meets the requirements of fit indices, second, the structural model was constructed adopting the maximum likelihood estimation in AMOS 17.0 program. The following four indices were used to evaluate the goodness of fit of the model: (a) Chi-square statistic/degree of freedom (χ2/df), (b) the standardized root mean square residual (SRMR), (c) the root mean square error of approximation (RMSEA), and (d) the Comparative Fit Index (CFI) (Mowlaie et al., 2016; Yu et al., 2016). In this study, a model was considered to have a good fit if all the path coefficients were significant at the level of 0.05, SRMR was below 0.08, RMSEA was below 0.08, and CFI was 0.95 or more.
Results
Bivariate analyses
Mean values, SDs, and inter-correlations for each variable are shown in Table 1. All dimensions of EI were significantly positively correlated with gratitude. In addition, all dimensions of EI and gratitude correlated with life satisfaction and positive affect positively and correlated with negative affect negatively.
Inter-correlations between the variables of interest.
SEA: self-emotion appraisal; OEA: others’ emotion appraisal; ROE: regulation of emotion; UOE: use of emotion; SD: standard deviation.
p < 0.05; **p < 0.01.
Measurement model
For the purpose of evaluating whether the measurement model fits the data adequately or not, confirmatory factor analysis (CFA) was conducted. Two latent constructs (i.e. EI and SWB) and seven observed variables (all of the items for the latent variables) were included in the measurement model. Results showed the measurement model fit to the data well, with χ2/df = 2.10; RMSEA = 0.055; SRMR = 0.050; and CFI = 0.979. More importantly, the factor loadings of all the indicators on the latent variables were significant (p < 0.01). The above results showed that the measurement model was well represented by relevant indicators.
Structural model
The structural model was tested in subsequent analyses using the maximum likelihood estimation in the sample of undergraduates. The direct path coefficient from the predictor (EI) to the criterion (SWB) in the absence of mediator was significant (β = −0.32; p < 0.01). Then, a completely mediated model with the mediator (gratitude) revealed a not very good fit to the data (χ2/df = 2.25; RMSEA = 0.061; SRMR = 0.082; and CFI = 0.966), since RMSEA was above 0.08. Finally, the partially mediated model, which included the mediator and direct path, was tested. The results indicated that the model exhibited a good fit for the data (χ2/df = 1.73; RMSEA = 0.045; SRMR = 0.046; and CFI = 0.982), and the direct path from EI to SWB was significant (β = 0.25; p < 0.01). Additionally, all of the factor loadings for the indicators on the latent variables were significant (p < 0.01). Therefore, the partially mediated model was selected as the best representation of the data (Table 2). All of the structural paths for the final model are presented in Figure 1.
Fit indices of candidate SEM.
SEM: structural equation modeling; CFI: comparative fit index; SRMR: standardized root mean square residual; RMSEA: root mean square error of approximation; df: degree of freedom.

Mediational effect of gratitude on the relationship between emotional intelligence and subjective well-being.
Mediating effect testing
The bootstrap estimation procedure in AMOS 17.0 was used to test the significance of the mediating effect of gratitude. The basic principle for the bootstrapping approach is that the standard error estimates and confidence intervals (CIs), which are calculated based on the assumption of a normal distribution, will usually be imprecise because the indirect effect estimates generally do not follow a normal distribution (MacKinnon et al., 2004). A bootstrap sample of 1200 tested the mediating effect. The 95 percent CIs of the direct and indirect effects are shown in Table 3; all the intervals did not overlap with zero. Thus, gratitude partially mediated the effect of EI and SWB.
Direct and indirect effects and 95 percent confidence intervals (CIs) for the final model.
EI: emotional intelligence; SWB: subjective well-being.
Discussion
Structural equation modeling (SEM) analysis was used to examine the prediction effect of EI on SWB. Results show that EI significantly predicts SWB. Individuals with high EI experience more happiness than those with low EI. This result is consistent with previous research (Bar-On, 2005; Sánchez-Álvarez et al., 2015). Ciarrochi et al. (2000) found that even though variations in intelligence and personality were controlled, EI and life satisfaction remained positively correlated, demonstrating that EI alone can explain the variations in life satisfaction. Petrides and Furnham (2000) showed that the population variance will exceed 50 percent when EI is used to explain SWB. Individuals with high EI believe that they can control and regulate their emotions to improve their SWB (Ruiz-Aranda et al., 2014). These individuals can relatively cope with and make full use of emotional information (Koydemir et al., 2013). Simply put, if individuals can process an emotional event or emotional information, then they can perceive more positive feelings at the psychological level, thereby improving their SWB and making them experience more happiness. The SWB of individuals can be enhanced if they can initiate the regulation of their emotion when they are pessimistic, if they can think rationally over their behavior when they are ecstatic, if they can reevaluate themselves when they feel satisfied with life, if they can regulate themselves and consider others’ feelings whenever frustrated with their interpersonal relationship, and if they can maintain a good state in a good interpersonal relationship (Malouff et al., 2014). These psychological operations are significant to the life of individuals. If these analyses are reversely deduced, they may show that individuals with low EI cannot efficiently process their emotions. This problem can further lower the intensity, frequency, and duration of one’s positive experiences (Koydemir et al., 2013). Furthermore, individuals may lower the overall evaluation of the quality of life, which can cause problems in their interpersonal adaption and mental health (Martos et al., 2016).
A significant correlation exists between EI and gratitude, and the former can significantly predict the latter. These results coincide with a few studies (Rey and Extremera, 2014). Gratitude is an “empathic emotion” rooted in the ability for transference (Emmons and McCullough, 2003), which is the basic component of EI (Schutte et al., 2001). Salovey and Mayer (1990) defined EI as one’s ability to exactly and effectively process emotional information and perceive, use, understand, and manage emotion as its core; moreover, EI as an emotional skill developed through learning and experience. According to McCullough et al. (2008), gratitude is an emotion or feeling. Therefore, the relationship between EI and gratitude can be understood easily. Petrides and Furnham (2000) asserted that EI can significantly predict emotion-related indicators. Goleman (1996) reported that EI can predict more than 80 percent of the variations in life-related variables. As a virtue of human nature, gratitude is an emotion or feeling and an important life variable. Theoretically, gratitude is influenced by EI, which also affects the former.
This research documented that gratitude can partially mediate the influence of EI on SWB and that EI can positively predict gratitude. The higher the EI, the easier an individual can become grateful. Individuals who display a high level of gratitude tend to adopt positive coping styles and enjoy prosocial behavior that promote good interpersonal relationships (Wood et al., 2007), thereby pushing forward the inner objective of struggle and minimizing materialism objective (Froh et al., 2011; Lambert et al., 2009). In this case, the SWB of an individual improves. Gratitude can also cushion the negative effects of environmental risk on SWB and provides resilient protection to an individual (Lanham et al., 2012). Gratitude itself is a positive emotional experience and a component of SWB (Emmons and Stern, 2013). Thus, the significant mediating effect of EI on SWB can be easily understood. Moreover, EI directly affects SWB because the latter can regulate and promote positive emotional experiences and allow people to actively experience life (Higgs and Dulewicz, 2014).
In sum, this research found that EI significantly predicts gratitude, and gratitude partially mediates the effect of EI on SWB. Despite its contributions, this research has certain limitations. For example, since this was a cross-sectional study, a causal inference could not be made. Thus, a follow-up longitudinal study is encouraged to explore the long-term mechanism of EI, gratitude, and SWB. This research found that the mediating effect exerted by gratitude on EI and SWB is not exceedingly high, suggesting that other important mediators possibly exists, and these mediators may display interaction effects. These factors are worthy of in-depth investigation. This study only enrolled undergraduates as participants. Thus, the results should not be generalized with other populations without further studies.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
