Análisis de las propiedades psicométricas del cuestionario: rasgo inteligencia emocional versión corta en estudiantes mexicanos
No Thumbnail Available
Date
Advisors
Evaluators
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Industrial de Santander
Abstract
Description
Introducción: La inteligencia emocional ha demostrado tener relaciones con el desempeño académico y el laboral, sin embargo, las propiedades psicométricas de escalas utilizadas para su medición han sido escasamente analizadas en culturas latinoamericanas, como es el caso del Cuestionario Rasgo Inteligencia Emocional Versión Corta (TEIQue-SF por sus siglas en inglés). Objetivos: Analizar la confiabilidad de sus factores y facetas, examinar su estructura factorial y determinar su validez concurrente. Metodología: Estudio observacional analítico realizado con base en una muestra guiada por participantes compuesta de estudiantes de una universidad mexicana (N=155). La confiabilidad se determinó con alfa de Cronbach y coeficiente omega, la estructura factorial y su validez mediante análisis factoriales exploratorios y confirmatorios, y la validez concurrente teniendo como criterio las escalas afecto positivo y afecto negativo del PANAS. Resultados: Se obtuvieron en general índices de confiabilidad reducidos para los factores de la escala original (alfa de Cronbach entre .32 a .82; coeficiente omega entre .16 a .83), y su estructura original no pudo comprobarse en el análisis factorial confirmatorio (X2=768.47, gl=390 y p=.0000, CFI=.65, RMSEA=.008, 90% IC=.071-.087). Los análisis factoriales exploratorios tampoco sustentaron la validez de constructo, al obtenerse una estructura en que se mezclaron entre sí los ítems de distintos factores y facetas. La validez concurrente tanto de la escala original como de la obtenida con los nuevos factores pareció insuficiente al obtener algunas correlaciones significativas débiles tanto con el afecto negativo como con el afecto positivo (r entre -.04 y .21). Conclusiones: La escala no mostró adecuadas propiedades psicométricas en la muestra participante, y son necesarios más estudios que sustenten estos hallazgos, recalcando que es importante publicar resultados negativos en estos casos.
Introduction: Emotional intelligence has been shown to have relationships with academic performance and work, however, the psychometric properties of scales used for measurement have been poorly analyzed in Latin American cultures, such as Trait Emotional Intelligence Questionnaire Short Form (TEIQue-SF for its acronym in English). Objectives: To analyze the reliability of its factors and facets, examine its factorial structure and determine concurrent validity. Methodology: Analytical observational study based on a sample guided by participants composed of students from a Mexican university (N = 155). Reliability was determined with Cronbach’s alpha and omega coefficient, the factor structure and validity through exploratory and confirmatory factor analyzes, and concurrent validity having criterion positive affect and negative affect scales PANAS. Results: Were obtained generally reduced reliability indices for the factors of the original scale (Cronbach’s alpha between .32 to 8.2; omega coefficient between .16 to .83), and its original structure could not be verified in the confirmatory factor analysis (X2=768.47, df=390 and p=.0000, CFI=.65, RMSEA=.008, 90% CI=.071-.087). Exploratory factor analyzes also did not sustain construct validity, by obtaining a structure in which mixed together items of different factors and facets. The concurrent validity of both the original scale and obtained with the new factors seemed weak to get some significant correlations with both negative affect as positive affect (r between -.04 and .21). Conclusions: The scale did not show adequate psychometric properties in the participant sample, and more studies are needed to substantiate these findings, stressing that it is important to publish negative results in these cases.
Introduction: Emotional intelligence has been shown to have relationships with academic performance and work, however, the psychometric properties of scales used for measurement have been poorly analyzed in Latin American cultures, such as Trait Emotional Intelligence Questionnaire Short Form (TEIQue-SF for its acronym in English). Objectives: To analyze the reliability of its factors and facets, examine its factorial structure and determine concurrent validity. Methodology: Analytical observational study based on a sample guided by participants composed of students from a Mexican university (N = 155). Reliability was determined with Cronbach’s alpha and omega coefficient, the factor structure and validity through exploratory and confirmatory factor analyzes, and concurrent validity having criterion positive affect and negative affect scales PANAS. Results: Were obtained generally reduced reliability indices for the factors of the original scale (Cronbach’s alpha between .32 to 8.2; omega coefficient between .16 to .83), and its original structure could not be verified in the confirmatory factor analysis (X2=768.47, df=390 and p=.0000, CFI=.65, RMSEA=.008, 90% CI=.071-.087). Exploratory factor analyzes also did not sustain construct validity, by obtaining a structure in which mixed together items of different factors and facets. The concurrent validity of both the original scale and obtained with the new factors seemed weak to get some significant correlations with both negative affect as positive affect (r between -.04 and .21). Conclusions: The scale did not show adequate psychometric properties in the participant sample, and more studies are needed to substantiate these findings, stressing that it is important to publish negative results in these cases.
Keywords
Emotional intelligence, validity of test, factorial analysis, students, affects, Inteligencia emocional, validez de las pruebas, análisis factorial, estudiantes, afectos