ABSTRACT
Resumen Para evaluar la experiencia, expresión y control de la ira, Spielberger construyó el Inventario State-Trait Anger Expression Inventory. El presente trabajo se propuso evaluar la validez y confiabilidad de las puntuaciones de la versión española del STAXI-2 para población general y hospitalaria de República Dominicana. Con este fin se analizó su estructura factorial, se evaluó la consistencia interna del instrumento, y se obtuvieron evidencias de validez mediante el contraste de grupo (población general vs. hospitalaria). Para realizar el estudio se trabajó con muestra de total de 1034 participantes, de población general (N = 792) y hospitalaria (N = 242) de la ciudad de Santiago de los Caballeros, República Dominicana. El análisis factorial confirmatorio permitió verificar el ajuste de los modelos originales para todas las escalas de ira (i.e. modelo de tres factores para la escala de ira-estado; modelo de dos factores para la escala de ira-rasgo; y modelo de cuatro factores para la escala de control y expresión de la ira). Todas las dimensiones descritas alcanzaron índices de fiabilidad aceptables y óptimos. Finalmente, las escalas consideradas discriminaron adecuadamente entre población general y hospitalaria, especialmente cuando se consideró la muestra psiquiátrica. En conjunto los resultados permiten concluir que las puntuaciones de la versión española del STAXI-2 son válidas y confiables para medir el estado-rasgo, y control-expresión de la ira en República Dominicana.
Abstract To measure the experience, expression and control of anger, Spielberger built the Inventory State-Trait Anger Expression Inventory. This paper under took to evaluate the validity and reliability of the scores of the Spanish version of the STAXI-2 for the general and hospital population of the DR. To this end, the factorial structure was analyzed, the internal consistency of the instrument was evaluated, and evidence of validity was obtained by contrasting groups (general vs. hospital population). To conduct the study, we worked with a sample of 1034 participants from both the general population (N = 792) and hospital population (N = 242) from the city of Santiago de los Caballeros, Dominican Republic. The confirmatory factor analysis allowed to verify the adjustment of the original models for all anger scales (i.e. three-factor model for the anger-state scale, two-factor model for the anger-trait scale, and four-factor model for the scale of control and expression of anger). All the dimensions reached acceptable or optimal reliability indexes. Finally, the scales adequately discriminated between the general and hospital population, especially when the psychiatric sample was considered. Overall, the results allow us to conclude that the scores of the Spanish version of the STAXI-2 are valid and reliable to measure the state trait, and control-expression of anger in the DR.
ABSTRACT
Esta investigación busca determinar las propiedades psicométricas de la versión mexicana del Inventario de Expresión de la ira estado-rasgo (STAXI-2; Miguel-Tobal, Casado, Can-Vindel y Spielberger, 2001). En él participaron 865 personas, entre los 12 y 60 años de edad de las cuales 54.9% eran hombres y 45.1%, mujeres., Entre ellos había estudiantes (46%), deportistas (42.4%) y población general (11.6%). Se presentó una similitud en la estructura factorial de las escalas estado, rasgo y expresión / control de la ira de la versión española. Sin embargo, se agruparon en ocho factores y no en las nueve subescalas que conformaban el instrumento original. Las estudiantes bachilleres expresaron un mayor nivel de ira que las universitarias, y los deportistas expresaron mayor ira que la población general y que los estudiantes. Los índices de consistencia interna para las escalas de ira-estado, ira-rasgo y expresión-control fueron aceptables (α = .887, α = .860, y α = .733, respectivamente), así como los porcentajes de varianza explicada (49.5%, 59.3% y 49.4%, respectivamente). Se obtuvieron correlaciones significativas entre los factores, excepto con el 'control de la ira' y la 'expresión de la ira' que no se relacionaron entre sí. El estudio permite concluir que el inventario es una herramienta verbal válida y confiable para la medición de la ira en población mexicana.
The aim of the study was to determine the psychometric properties of the Mexican version of the State-Trait Anger Expression Inventory (STAXI-2; Miguel-Tobal, Casado, Can-Vindel y Spielberger, 2001). Participants were 865 people, with ages between 12 and 60 years old; 54.9% were men and 45.1%, women. Among them there were students (46.0%), athletes (42.4%) and general population (11.6%). Results showed a similar factorial structure of the state, trait and anger expression / control scales of the Spanish version STAXI-2. However, these were grouped in eight factors rather than the nine subscales that constituted the original instrument. Senior high school students expressed higher levels of anger than college students and athletes expressed more anger than the general population and the students. Internal consistency indexes were acceptable for anger-state, anger-trait and expression-control (α = 0.887, α = 0.860, and α = 0.733, respectively), as well as the percentages for the explained variance (49.51%, 59.31% y 49.44%, respectively). Significant correlations were obtained among the factors, except for the anger-control and anger-expression subscales, that were unrelated to each other. The study concludes that the STAXI-2 inventory is a valid and reliable verbal tool for anger assessment in Mexican population.
Esta pesquisa visa determinar as propriedades psicométricas da versão mexicana do Inventário de Expressão de Ira Estado-Traço (STAXI-2; Miguel-Tobal, Casado, Can-Vindel y Spielberger, 2001). Estiveram presentes 865 pessoas, entre 12 e 60 anos, das quais 54,9% eram do sexo masculino e 45,1% mulheres. Havia estudantes (46%), esportistas (42,4%) e população corrente (11,6%). Houve uma semelhança na estrutura fatorial das escalas estado, traço e expressão/controle da ira da versão em espanhol. No entanto, foram agrupados em oito fatores, e não nas nove subescalas que compõem o instrumento original. As estudantes do ensino médio expressaram maior nível de ira que as universitárias, e os esportistas expressaram mais ira do que a população geral e os estudantes. Os índices de consistência interna das escalas de ira de ira-estado, ira-traço e expressão-controle foram aceitáveis (α = 0,887, α = 0,860 e α = 0,733, respectivamente), e as porcentagens de variância explicada (49,5%, 59,3% e 49,4%, respectivamente). Obtiveram-se correlações estatisticamente significativas entre os fatores, exceto em "controle de ira" e "expressão de ira", que não estão relacionados entre si. O estudo conclui que o inventário verbal é um instrumento válido e confiável para medir a ira na população mexicana.