Resumen El constructo de desvinculación moral tiene implicaciones muy importantes en el estudio de diversos fenómenos psicosociales. Ante la ausencia del análisis de las propiedades psicométricas de la escala de desvinculación moral de Bandura en población mexicana, el objetivo de esta investigación fue la obtención de evidencias de validez factorial y confiabilidad de dicha escala. Como hipótesis central de esta investigación, se consideró que la escala podría mantener las dimensiones estructurales reportadas en la versión original, así como la fusión de los ocho mecanismos propuestos por Bandura en un factor de segundo orden. La muestra estuvo conformada por 480 estudiantes universitarios de tres instituciones públicas de nivel superior. Los resultados mostraron que el modelo con una estructura de primer orden en su versión adaptada al español, incluyendo los 32 reactivos propuestos originalmente, presentó características psicométricas adecuadas; sin embargo, el modelo de segundo orden mostró una mejor variabilidad observada en las puntuaciones de 27 reactivos. Además, todos los reactivos tuvieron un valor adecuado con respecto a los índices de bondad de ajuste, así como una suficiente variabilidad explicada tanto en las mujeres como en los hombres. El análisis de fiabilidad de esta escala indica que posee una elevada consistencia interna (α = .89). Con base en estas evidencias, se concluyó que esta escala es un instrumento válido y fiable para evaluar la desvinculación moral en la población general.

Abstract The construct of moral disengagement has relevant consequences in the study of several psychosocial phenomena. Due to the lack of the psychometric characteristics of the Bandura’s Moral Disengagement Scale among Mexican populations, the purpose of this research was to examine the scale’s factorial validity and reliability. As a core hypothesis, it was considered that this scale could keep the structural dimensions reported in the original version as well as the combination of the eight mechanisms proposed by Bandura in a second order factor. The sample included 480 undergraduate students from three public universities. Results showed that the model with a first order factor structure in its version adapted to Spanish, including the original 32 items, revealed acceptable psychometric characteristics; however, the second order model showed a better variability in the scores of the 27 items. All items had an appropriate value regarding the goodness-of-fit indices, as well as an appropriate explained variability both in women and in men. The reliability of this scale revealed an internal consistency (α = .89). In conclusion, this scale is a valid and reliable instrument to measure moral disengagement in the general population.

Resumen La violencia en el noviazgo es un fenómeno multifactorial que requiere de estudios que profundicen respecto a los efectos que tienen las creencias culturales en el comportamiento tanto del agresor como de la víctima. Teniendo esto en cuenta, el objetivo de la presente investigación fue determinar la relación entre las creencias acerca de la violencia y la prevalencia de la violencia en el noviazgo. Para ello, se contó con una muestra de 420 estudiantes de dos universidades públicas mexicanas que respondieron el Inventario de creencias acerca de la violencia hacia la esposa, el Inventario de conflictos en las relaciones de noviazgo, y un cuestionario de información sociodemográfica. Los datos recolectados fueron examinados por medio de un análisis de correlación canónica, y los resultados mostraron que el modelo en general fue estadísticamente significativo (Wilks X = .654, F(20, 677.54) = 4.626,p < .05); que el tamaño del efecto del modelo general fue de .346, lo que indica que este explicó el 34.6 % de la varianza compartida por los dos conjuntos de variables; y que, específicamente en la primera función, el coeficiente de mayor magnitud fue el de la variable de justificación de la violencia (r2 s = 76.2; h 2 = 90.0), seguido por la del apoyo que se le puede brindar a la víctima (r2 s = 57.1; h2 = 94.5).

Abstract Dating violence is a multifactorial phenomenon that requires in-depth studies regarding the effects that cultural beliefs have on the behavior of both the aggressor and the victim. With this in mind, the objective of this research was to determine the relationship between beliefs about violence and the prevalence of dating violence. To this end, a sample of 420 students from two Mexican public universities answered the Inventory of Beliefs about Wife Violence, the Inventory of Conflicts in Dating Relationships, and a sociodemographic information questionnaire. The data collected were examined through a canonical correlation analysis, and the results showed that the overall model was statistically significant (Wilks X = .654, F (20, 677.54) = 4.626, p < .05); that the effect size of the overall model was .346, indicating that it explained 34. 6 % of the variance shared by the two sets of variables; and that, specifically in the first function, the coefficient of greatest magnitude was that of the variable of justification of the violence (r2 s = 76.2; h2 = 90.0), followed by that of the support that can be given to the victim (r2 s = 57.1; h2 = 94.5).

Abstract The effects of family of origin violence and intimate partner violence have been extensively documented; however, very few studies have examined the interaction with emotion regulation strategies. Thus, the objective of this research was to analyze whether different types of emotion regulation strategies, both adaptive and maladaptive, mediate the relationship between family of origin violence and intimate partner violence in the Mexican population. A total of 838 participants (45.9% men and 54.1% women) responded to instruments addressing family of origin violence, emotion regulation strategies, and intimate partner violence. The results revealed that both structural models were significant. For women, the model showed an adequate fit X 2 (11, N = 838) = 22.75, p = .288, GFI = .95, AGFI = .91, NFI = .98, CFI = .97, RMSEA = .05. Likewise, we found similar indexes for men X2 (11, N = 838) = 28.20, p = .348, GFI = .97, AGFI = .93, NFI = .97, CFI = .95, RMSEA = .04. Specifically, the direct effects of adaptive strategies on intimate partner violence were statistically significant. Meanwhile, the direct effects of family of origin violence on maladaptive emotion regulation strategies were significant, as were the direct effects of maladaptive strategies on intimate partner violence. In turn, the indirect effects of family of-origin violence were significantly related to intimate partner violence via maladaptive emotion regulation strategies. In addition, the results clearly showed that men reported higher levels of aggression against women. Finally, regarding the selection of emotion regulation strategies, while women employed more adaptive emotion regulation, men showed a more definite tendency to use maladaptive emotion regulation.