Unique Risk and Protective Factors for Partner Aggression in a Large Scale Air Force Survey

Abstract

The objective of this study is to examine risk factors of physical aggression against a partner in a large representative Active Duty Air Force sample. A stratified sample of 128,950 United States Active Duty members were invited to participate in an Air Force-wide anonymous online survey across 82 bases. The final sample (N = 52,780) was weighted to be representative of the United States Air Force. Backward stepwise regression analyses were conducted to identify unique predictors of partner physical aggression perpetration within and across different ecological levels (individual, family, organization, and community levels). Relationship satisfaction, alcohol problems, financial stress, and number of years in the military were identified as unique predictors of men's and women's perpetration of violence against their partner across ecological levels. Parental status, support from neighbors, personal coping, and support from formal agencies also uniquely predicted men's but not women's perpetration of violence across ecological levels. This study identified specific risk factors of partner violence that may be targeted by prevention and intervention efforts aimed at different levels of impact (e.g., family interventions, community-wide programs).

An ecological model of intimate partner violence perpetration at different levels of severity.

“…Twenty-one items assessed partici pants' sense of shared mission, teamwork, unity, and connected ness in the community (e.g., "Civilian spouses/active duty mem bers find it easy to make connections with other families"). Low community cohesion is a predictor of neighborhood disorder and a risk factor for family violence (e.g., Bowen, Mancini, Martin, Ware, & Nelson, 2003;Slep et al, 2010). Items are rated on a 6-point scale from strongly disagree to strongly agree.…”

“…The following four-step model building approach was used with the development sample: (a) previous theoretical and empirical mod els were reviewed (e.g., O'Leary, Slep, & O'Leary, 2007); (b) corre lation matrices of study variables were examined; (c) exploratory factor analyses were conducted on the independent variables to iden tify indicators of hypothesized latent constructs; and (d) backward stepwise regression analyses were conducted to determine which study variables predict the most variance (and are most likely proxi mal correlates) in IPVPerp and CS-lPVPerp (see Slep, Foran, Heyman, & Snarr, 2010 for the results). Based on the above-listed steps, an initial model was tested in the development sample.…”

Emotional processing in Colombian ex-combatants and its relationship with empathy and executive functions

“…The Colombian conflict has several characteristics that make it a special scenario to test human behaviour. Combatants have ideological preconceptions bent by the political influence of the group they belong to (McCarroll et al, 2000;Slep, Foran, Heyman, & Snarr, 2010). This produces a lack of empathy for the enemy.…”

Ajuste Diádico, Victimización y Perpetración en Mujeres y Hombres Mexicanos

“…Varios estudios han relacionado el desajuste diádico con la violencia de pareja, observándose un tamaño de efecto grande (Cáceres & Cáceres, 2006;Slep, Foran, Heyman & Snarr, 2010), mayor sobre victimización que sobre perpetración (Capaldi, Knoble, Shortt & Kim, 2012). En la medida en que hay menos ajuste diádico, aparece más violencia, especialmente en situaciones de ruptura o separación (Scorsolini-Comin & dos Santos, 2012).…”

“…De acuerdo a los resultados de los estudios empíricos revisados, se formularon como hipótesis para los modelos de relación y los aspectos diferenciales entre ambos sexos que: (a) el desajuste diádico tiene un efecto grande sobre la violencia de pareja (Cáceres & Cáceres, 2006;Slep et al, 2010), mayor sobre victimización que sobre perpetración en ambos sexos; (b) la asociación o efecto del desajuste diádico sobre victimización es más alto en mujeres que en hombres y, sobre perpetración, más alto en hombres que en mujeres (Capaldi et al, 2012); (c) el número de hijos y años de relación tienen una asociación positiva y pequeña con violencia de pareja y mayor con victimización en mujeres y con perpetración en hombres (Capaldi et al, 2012;Stith et al, 2004); (d) el número de hijos y años de relación tienen una asociación inversa con el ajuste diádico y este último media totalmente el efecto de ambas variables sociodemográficas sobre la violencia de pareja (Acevedo & Aron, 2009;Cano-Prous et al, 2014) y (e) la violencia es reactiva en ambos sexos (Moral de la Rubia & López Rosales, 2012, 2014 o solo en hombres, mientras que la victimización y perpetración son independientes en mujeres, como se ha hallado previamente en población mexicana con datos de autorreporte (Oxtoby, 2012). Al realizarse el estudio en población general, se esperaba que el modelo de violencia cruzada y el modelo de violencia proactiva tuvieran mal ajuste a los datos, al ser estos modelos más adecuados en ámbitos clínicos de personas asistidas o en terapia por violencia de pareja (Johnson, 2008).…”