هوش هیجانی و رفاه ذهنی در حال بازبینی

This study aims at extending previous research on the predictive validity of “maximum performance” measures of emotional intelligence (EI) in relation to cognitive and affective facets of well-being, by way of a prospective research design. Participants were 202 Israeli adolescents who were administered the following three predictor measures at time one: (a) the MSCEIT, an ability-based measure of EI, (b) the OCEANIC, a measure of the “Big-Five” factors of personality, and (c) the Vocabulary subtest of the WISC-R, as measure of verbal ability. At time two, data were gathered on cognitive and affective facets of well-being, as criterion measures. Overall, the MSCEIT failed to demonstrate predictive validity against affective criterion. The null outcomes were discussed and explicated.

A growing body of empirical evidence suggests that emotional intelligence (EI) correlates robustly with a variety of outcomes that signal social-emotional success, including more frequent positive affect, higher self esteem, greater life satisfaction, social engagement, and well-being (see Zeidner, Matthews, & Roberts, 2009 for a recent review). A number of related mechanisms have been hypothesized to account for the nexus of relationships between EI and adaptive outcomes, such as subjective well-being (SWB). First, under the assumption that high EI individuals are more aware of their emotions and also better able to regulate them, they should experience lower levels of distress and stress-related emotions, and concomitantly, experience higher levels of well-being (Salovey, Bedell, Detweiler, & Mayer, 1999). Second, given the working assumption that high EI individuals have an advantage in terms of greater social competence, richer social networks, and more effective coping strategies, this should serve to enhance their sense of SWB (Salovey et al., 2000 and Salovey et al., 1999). Third, because emotions provide information about one’s relationship to the environment and others, interpreting and responding to that information can direct action and thought in ways that enhance or maintain well-being (Lazarus, 1991 and Parrott, 2002). Finally, EI has been found to be associated with a lower propensity to experience negative emotions and a higher propensity to experience positive emotions, thus contributing to a richer sense of SWB (Mikolajczak, Nelis, Hansenne, & Quoidbach, 2008). For these reasons, EI has been commonly hypothesized to predict one’s subjective sense of well-being and positive mental health. A basic difficulty in researching EI and its correlates is that different psychologists have disparate visions of how to best conceptualize and assess EI, and more broadly, what a science of EI should look like. Two complementary conceptualizations of EI – i.e., as ability and as personality trait – exist side by side in the literature. Proponents of ability models of EI favor defining the construct as an intelligence, resembling other standard intelligences (e.g., verbal, numerical, figural), with abilities best measured through objective tests akin to IQ tests. Proponents of trait models, by contrast, view EI as personality dispositions, and aim at organizing the key affect-related aspect of personality under a single conceptual framework (Petrides, Perez-Gonzalez, & Furnham, 2007). Trait models of EI offer a broader conception of emotional intelligence, incorporating both personal competencies and qualities that assist the person in using EI in real-life. Researchers in the trait tradition have typically used self-report measures to assess EI. As suggested by Petrides and Furnham (2001), EI tests of ability capture maximal performance, whereas tests of EI as personality traits; capture typical performance. There is accumulating evidence in support of the incremental validity of trait EI measures, controlling for personality factors, in predicting adaptive outcomes (Bastian et al., 2005, Day et al., 2005, Gallagher et al., 2008, Palmer et al., 2002, Petrides et al., 2007 and Saklofske et al., 2007). A recent study by Petrides et al. (2007) reported that EI, as assessed by the EQi, was significantly predictive of life satisfaction, when statistically controlling for variance in the “Big-Five”. In direct contradistinction, ability-based (maximum performance) measures of EI have not fared very well in predicting affective indicators of personal adaptation or well-being. In fact, the vast majority of previous studies yielded only weak or non-significant associations between the Mayer–Salovey–Caruso Emotional Intelligence Test (MSCEIT; Mayer, Caruso, & Salovey, 2002) – or other ability-based maximum performance measures of EI – and indicators of personal adaptation such as SWB, coping styles, perceived stress, life satisfaction, satisfaction with social relationships, and depression (e.g., Bastian et al., 2005, Brackett and Mayer, 2003, Brackett et al., 2006, Freudenthaler et al., 2008, Gohm et al., 2005, Lopes et al., 2003 and Rode et al., 2007). Hence one might expect, at best, to find low correlations between ability-based measures of EI and indicators of well-being. Mikolajczak and coworkers (Mikolajczak et al., 2008) have recently argued that maximum ability measures, such as the MSCEIT, may capture what individuals are capable of doing in an emotion-laden situation, i.e., what they could do, whereas trait measures of EI capture how much of this emotional management potential or declarative knowledge about emotions is actually translated into practice in context (see also Petrides & Furnham, 2000). Accordingly, the low or non-significant associations of declarative knowledge maximum performance measures, such as the MSCEIT, with measures of well-being might indicate that individuals may fail to optimally translate their emotional knowledge into practice. This study aims at extending previous research on the predictive validity of “maximum performance” measures of EI in relation to cognitive and affective facets of well-being, by way of a prospective research design. We aim at improving on prior studies in two major ways: (a) by employing a prospective design to assess the predictive validity of EI against adjustment criteria; and (b) by employing statistical controls for ability and personality confounds in the design. Most studies attempting to validate the predictive role of EI in SWB have used cross-sectional designs (e.g., Gallagher et al., 2008, Lopes et al., 2003, Mavroveli et al., 2007, Palmer et al., 2002, Petrides et al., 2007 and Rossen and Kranzler, 2008), with the predictor (EI) gathered at the same time as the criterion measure. This method clearly precludes any tenable generalizations about the predictive validity of EI, which presumes some time lag between the predictor and criterion measures. A second major shortcoming is that many prior studies, with few exceptions (e.g., Rossen & Kranzler, 2008), have failed to statistically control for both personality and intelligence as potential confounding variables. As in other domains, controlling for personality and intelligence tends to reduce the predictive validity of the EI scales and tests, in some cases to zero (see Day, 2004). In fact, correlations between ability-based measures of EI (MEIS and MSCEIT) and social wellbeing sometimes disappear when personality and intelligence are controlled (Brackett & Mayer, 2003). In addition, prior studies have been based on the person’s self-reports of both EI and adjustment measure, with significant observed relationships possibly accounted for, in part, by common method variance, i.e., self-report in both measures. Thus, it can not be excluded that the obtained overlap between EI measures and SWB is due, in some part, to the use of cross-sectional designs, failing to control for personality and ability, or common method variance (in the case of self-report measures). It was hypothesized that a prospective study using the MSCEIT to predict adaptive outcomes (i.e., SWB), controlling for personality and ability, would show non-significant and at best very weak predictive validity coefficients when compared to the significant validity coefficients often reported by prior studies using cross-sectional designs.