The Aggression Questionnaire: A Validation Study in Student Samples

The goal of this study was to examine the psychometric properties of the Aggression Questionnaire (AQ) in Spain. The AQ is a 29-item instrument designed to measure the different dimensions of the hostility/anger/aggression construct. It consists of 4 subscales that assess: (a) anger, (b) hostility, (c) verbal aggression, and (d) physical aggression. In Study 1, reliability, construct validity, and convergent validity were evaluated in a group of 384 male and female university students. Test-retest reliability was evaluated using a group of 154 male and female university students. The results of the factor analysis were similar to the scale structure claimed for this instrument. The subscales also showed internal consistency and stability over time. The AQ and its subscales were also compared with the scales and subscales of the Spielberger State-Trait Anger Expression Inventory (STAXI), the Cook-Medley Hostility Scale (Ho), the Buss-Durkee Hostility Inventory (BDHI), and the Jenkins Activity Survey-Form H (JASE-H). The results show that the AQ evaluates some aspects of anger, such as Anger-Trait and Anger-Out, rather than other elements, such as Anger-In or Anger-State. In Study 2, two new male groups were used to evaluate the criterion validity of the AQ: 57 prison inmates and 93 university students, finding that this instrument discriminated between the scores obtained by common offenders and university students. El objetivo de este estudio fue examinar las propiedades psicométricas del Cuestionario de Agresión (AQ) en España. El AQ es un instrumento de 29 ítems que mide las distintas dimensiones del constructo hostilidad/ira/agresión. Consta de 4 subescalas que evalúan: (a) ira, (b) hostilidad, (c) agresión verbal y (d) agresión física. En el Estudio 1 se evaluó la fiabilidad, la validez de constructo y la validez convergente en un grupo de 384 estudiantes universitarios de ambos sexos. También se midió la fiabilidad test-retest, usando un grupo de 154 estudiantes universitarios de ambos sexos. Los resultados del análisis factorial revelaron una estructura similar a la encontrada con muestras de habla inglesa. Las subescalas mostraron una adecuada consistencia interna y fiabilidad test-retest. Igualmente, se calcularon las correlaciones existentes entre el AQ y otros instrumentos de medida del constructo, tales como el Inventario de Expresión y Estado-Rasgo de Ira de Spielberger (STAXI), la Escala de Hostilidad de Cook y Medley (Ho), el Inventario de Ho