From Deficits in Emotional Intelligence to Eating Disorder Symptoms: A Sequential Path Analysis Approach Through Self-Esteem and Anxiety

Abstract

Past studies have reported emotional intelligence (EI) as a relevant factor in development and maintenance of eating disorders (ED), as well as in increasing self-esteem and reducing anxiety. Similarly, research has showed that anxiety and self-esteem are positively and negatively associated to ED criteria, respectively. However, no prior studies have yet tested the multiple intervening roles of both self-esteem and anxiety as potential mediators of the association between EI and ED symptomatology. The present study aims to bridge these gaps by testing a sequential path model. Specifically, we examine the potential sequential mediation effects of self-esteem-anxiety on the link between EI and ED. A sample composed of 516 Spanish undergraduate students and community adults completed measures of EI, self-esteem, anxiety, and ED symptomatology. The results show that high levels of EI were positively associated with self-esteem and negatively associated with anxiety and ED symptoms. Anxiety was positively associated to ED symptoms, while self-esteem levels were negatively linked to ED symptoms. Moreover, path analyses showed that self-esteem and anxiety fully mediated the relationship between EI and ED symptoms in sequence. These findings suggest that EI plays a key role in reducing symptomatology of ED through increased self-esteem and reduced anxiety symptoms, providing novel evidence regarding psychological mechanisms through which EI contributes to a reduction of ED symptomatology. Implications for assessing and improving these psychological resources in ED preventive programs are discussed.

Keywords: anxiety; eating disorders; emotional intelligence; path analysis; self-esteem.

Figure 1

Multiple mediation model for the effect of EI and ED symptoms via self-esteem and anxiety controlling for age and sex as covariates. Total effect (c-path) is given in parentheses; standardized coefficients; *p < 0.05; ***p < 0.001.