Does emotional intelligence predict unique variance in life satisfaction beyond IQ and personality?

Abstract

Many emotional intelligence (EI) scales have been found to overlap with personality scales and it has been argued that EI scales are measuring personality. In the present study it was hypothesised that EI would explain unique variance in life satisfaction beyond that predicted by personality, IQ, and control variables. A community sample (N = 191) aged 18–79 years (M = 35.94, SD = 14.17) was recruited. Because IQ showed no bivariate relationship with life satisfaction, IQ was not used in further analyses. After controlling for marital status and income, personality accounted for an additional 34.2% of the variance in life satisfaction, and total EI scores accounted for a further 1.3% (p < 0.05). Further analysis revealed that the additional variance was explained by the EI dimension of Emotional Management. In a competing analysis, EI explained 28.3% of the variance at step 2, and personality accounted for a further 8.8% of the variance at step 3. It was concluded that EI predicted some unique variance in life satisfaction, and that there was substantial conceptual overlap between EI and personality. However, it is argued that, rather than being redundant, emotional intelligence may offer valuable insights to current conceptions of personality.

Introduction

The concept of emotional intelligence (EI; Salovey & Mayer, 1990) has received increasing attention over the past decade. Mayer and Salovey (1997) identified four components of EI, involving the ability to perceive and express emotion, to access and use emotions to facilitate thought, to understand emotions, and to manage emotions. A growing body of research has found links between EI and a wide range of important life outcomes that are not adequately predicted by traditional measures of intelligence. However, many EI scales have been found to overlap with personality scales and some have argued, therefore, that they are measuring personality traits. This issue is complicated by the different models and measures of EI, which fall into two broad categories: mixed models and ability models. Because mixed models include a variety of traits, this may account for the overlap between personality and EI. The aim of this study was to determine whether EI predicted unique variance in life satisfaction (LS) beyond that explained by personality and IQ, using a scale based on the ability model of EI.

Emotional intelligence has been theoretically linked with life satisfaction (e.g., Bar-On, 1997, Goleman, 1996, Mayer et al., 2000a), and some researchers have empirically explored the relationship between EI and individual differences in LS, finding significant correlations ranging from r = 0.11 to r = 0.45 (Bar-On, 1997, Ciarrochi et al., 2000; Mayer et al., 2000b, Palmer et al., 2002). Ciarrochi et al. (2000) further discovered that EI accounted for variance in LS after controlling for IQ and personality variables. Conversely, scores on the Twenty-Item Toronto Alexithymia Scale (TAS-20; Bagby, Parker, & Taylor, 1994), which measures an inability to evaluate and verbally express emotions, correlated negatively with LS.

As noted however, there are a number of different theories and measures of EI to consider when evaluating research on emotional intelligence. Models of EI have been classified as ‘ability’ models and ‘mixed’ models. Ability models (e.g., Mayer & Salovey, 1997) aim to measure EI as a set of abilities that result from the adaptive interaction between emotions and cognition, whereas mixed models have been so named because they include a broader range of traits and dispositions (Mayer, Salovey, & Caruso, 2000c).

The hierarchical ability model of EI developed by Mayer and Salovey (1997), based on an extensive review of previous studies on emotion (e.g., Mayer and Salovey, 1993, Mayer and Salovey, 1995, Mayer and Geher, 1996), is divided into four levels: emotional perception, emotional facilitation, emotional understanding, and emotional management. The most basic level involves an awareness of emotion, which is developed in early childhood. The next level comprises mental processing of emotions, the ability to incorporate emotional experiences into general awareness. At the third level, the individual is able to understand and reason about emotions, including how and why they develop. The fourth level involves the most highly developed ability, to manage and regulate emotions, for instance calming feelings of anger or anxiety in oneself or others (Mayer et al., 2000c).

Mixed models of EI (e.g., Bar-On, 1997, Goleman, 1995) are so named because they generally combine mental abilities and personality characteristics such as warmth, persistence, zeal, motivation and optimism. This can pose problems for the theoretical conceptualisation of EI and for the contribution of unique psychological information that is offered by the measurement of EI (Mayer, Salovey, & Caruso, 2000d).

Largely because of the conceptual problems surrounding mixed models of EI, it has been widely argued that EI scales are measuring personality traits (e.g., Davies et al., 1998, Newsome et al., 2000, Petrides and Furnham, 2000). This is of concern, because the usefulness of a psychological construct lies in its ability to tap into variance in important life outcomes that is not explained by analogous constructs (Davies et al., 1998, McCrae, 2000). In a comparison between mixed models of EI and personality, McCrae (2000) conceptually demonstrated how each of the proposed aspects of EI on the EQ test (Goleman, 1995) and the EQ-i (Bar-On, 1997) correspond with each of the Big Five personality traits. In empirical studies, medium to large correlations have been found between personality and EI scores based on mixed models (Davies et al., 1998, Dawda and Hart, 2000, Newsome et al., 2000).

Few studies to date have investigated the relationship between personality and EI measures based on the ability model. Scores on a self-report instrument developed by Schutte et al. (1998) correlated moderately with related constructs such as Alexithymia and optimism. The 33-item scale was reported by Schutte et al. to have a significant correlation with Openness to Experience (r(22) = 0.54) and non-significant relationships with the other Big Five personality traits. Similarly, studies using the Trait Meta-Mood Scale (TMMS), which measures EI as an ability, have demonstrated that the TMMS factors are sufficiently distinct from associated constructs such as Neuroticism (Salovey, Mayer, Goldman, Turvey, & Palfai, 1995).

A study examining self- and other ratings of EI loosely based on the ability model of EI found strong relationships between the EI dimensions and the Big Five personality dimensions, particularly Extraversion and Neuroticism (Van der Zee, Thijs, & Schakel, 2002). However, in this study the EI dimensions were found to predict both academic and social success above that which was predicted by academic intelligence and personality. Preliminary investigation of a new self-report ability scale of EI (Palmer & Stough, 2001) shows low to moderate correlations with the major dimensions of the NEO-FFI (Costa & McCrae, 1992), ranging from −0.09 to −0.47.

While people have dispositional tendencies to be extraverted, conscientious and so on, the way they respond emotionally requires mental awareness, appraisal, and processing of emotions and the environment (see Mayer & Salovey, 1993). This requires a learned mental ability, or intelligence, as opposed to a tendency to behave according to inherent personality dispositions. In support of such a notion, McCrae (2000) suggested a subtle distinction between personality traits and EI, positing that optimism, for example, can result from either a cheerful personality disposition, or from a deliberate cognitive effort to focus on methods of obtaining successful outcomes and seeking out available support. The latter process requires a cooperative interaction between emotion and reason. In line with this notion, the above studies suggest that scales of EI based on the ability model may be measuring something other than personality traits.

The present study aimed to further explore the unique predictive value of EI, in relation to life satisfaction, from an ability perspective. It was hypothesised that EI would predict unique variance in LS over and above that which was predicted by personality and IQ after controlling for demographic variables.