RESUMEN
Resumen El objetivo es conocer acerca del efecto indirecto de la inteligencia emocional en la autoeficacia académica a través del compromiso académico en una muestra de estudiantes universitarios del Ecuador. Para ello se realizó un análisis de tipo descriptivo correlacional, de determinación, de mediación parcial a través de SEM y de corte transversal. El instrumental empleado corresponde a la Brief Emotional Intelligence Inventory for Senior Citizens, la Utrecht Work Engagement Scale y la Escala de Autoeficacia Percibida Específica de Situaciones Académicas. La muestra estuvo conformada por 500 estudiantes (57.5 % mujeres y 42.4 % hombres), entre 18 y 45 años (M = 24.5 años; SD = 6.9), procedentes de dos universidades de Ambato y Quito en Ecuador y de 11 carreras de pregrado. Entre los resultados se encontró que los niveles de las variables en análisis son normales entre los estudiantes, aunque un 30 % se encuentran en zona de riesgo por baja autoeficacia académica. Se confirma la relación entre inteligencia emocional (r = .522; p < .01) y el compromiso académico (r = .326; p < .01) en la autoeficacia académica. Estas variables forman un modelo explicativo del 30.6 % del cambio de la varianza de la autoeficacia académica. Además, la inteligencia emocional ejerce un efecto indirecto sobre la autoeficacia académica por medio del compromiso académico (ab = .031; p < .01). Como conclusión se establece que la inteligencia emocional y el compromiso académico son predictores de la autoeficacia académica. Además, que el compromiso académico es una variable de mediación parcial entre la inteligencia emocional y la autoeficacia académica en universitarios del Ecuador. La inteligencia emocional tiene como propiedad que ejerce un efecto directo en la autoeficacia académica e indirecto-parcial por medio del compromiso.
Abstract The intrapersonal and self-influence elements of individuals are necessary to know the impact they maintain in the educational context, because they are strong internal motivators that direct the behaviour of the individual to carry out various activities, as well as their achievement and success. There is evidence of the capacities that students have to monitor, control and regulate their own behaviours in the learning process. Although historically academic performance has been associated with intelligence in general, elements such as self-monitoring, self-control, self-regulation, self-motivation and others are key pieces to prevent school failure and maintain an adequate school performance. Within educational research, phenomena such as emotional intelligence, engagement and academic self-efficacy stand out due to their impact on academic performance. There are several studies that analyse the existing dynamics of these three variables, although separately, so a multifactorial study could expand the current conception of these internal elements and their effect on education. This investigation's objectives are to identify the current state of emotional intelligence, engagement and academic self-efficacy in a sample of university students from Ecuador, as well as the relationship between these variables and the indirect effect of emotional intelligence on academic self-efficacy through academic engagement. Thus, this study is a descriptive correlational, determination and partial mediation analysis, carried out through SEM using the Brief Emotional Intelligence Inventory for Senior Citizens, the Utrecht Work Engagement Scale and the Scale of Perceived Self-efficacy Specific to Academic Situations. The sample was made up of 500 students (57.5 % women and 42.4 % men), between 18 and 45 years old (M = 24.5 years; SD = 6.9), from two universities in Ambato and Quito in Ecuador and 11 undergraduate majors. The participants were selected through an intentional non-probabilistic sampling with inclusion criteria, which were: (a) be legally enrolled in one of the universities in this study; (b) voluntary participation in the study; (c) signed participation consent; and (c) be of legal age. In general, the levels of emotional intelligence, engagement and self-efficacy are normal among students; however, in academic engagement, 30 % are in a risk zone. The relationship between emotional intelligence (r = 0.522; p < .01) and academic engagement (r = 0.326; p < .01) in academic self-efficacy is confirmed. Together, these variables form an explanatory model that can predict changes in variance by 30.6 %. The theoretical model proposed by means of structural linear regression has an adequate adjustment indicator χ. = 252.9; p < .001; df = 116; χ./df = 2.1; CFI = .96; TLI = .95; RMSEA = .049 [.040 - .057]; SRMR = .042. In addition, emotional intelligence has an indirect effect on academic self-efficacy through academic engagement (ab = .0031; p < .01). Among the conclusions of the study, it is mentioned that emotional intelligence and academic engagement are predictors of academic self-efficacy. Therefore, their interest and management in the educational context is relevant. Moreover, the academic engagement is a partial mediation variable between emotional intelligence and academic self-efficacy in university students in Ecuador. Emotional intelligence has the ability to exert not only direct effects, but also partial indirect effects through a third variable. One of the implications of the study is that the use of more contemporary methods such as SEM allows a broader vision of the modelling of academic behaviour, which is an important contribution because it significantly reduces measurement bias. In addition, these results contribute to the future development of psychoeducational intervention processes that contribute to the improvement of the academic performance of students and to the teaching-learning processes in general. Finally, it is important to replicate similar studies in order to generalize these findings.