Faculty Bibliography

Historically, the political context of partner physical aggression policy and research has focused on protection of physically victimized women and mandated interventions for male batterers. This emphasis is understandable when one considers the injuries and deaths of women by men. However, physical aggression against partners among teens is a very different phenomenon than battering. Intimate partner violence (IPV) in the form of physical aggression, the focus of this review, often starts in junior high school, and approximately 35% of male and female senior high school students report engaging in IPV. The specific trajectory of IPV varies by sample, but IPV appears to decrease in the late teens or early 20s. IPV is generally reported by both males and females, and not attributable to self-defense. IPV is significantly stable in couples who remain together, but stability appears lower if partners change. Given the importance of physical aggression by both males and females, prevention and early intervention programs need to address relationship factors, and targeted prevention and early intervention would be prudent with young high-risk couples. Decades of intervention programs for batterers have not proven very successful, and IPV appears easier to prevent than treat. Thus, emphasis on prevention of IPV seems both timely and promising. This review is intended for diverse audiences including educational administrators, policy makers, and researchers. It reviews issues such as who and when to target for IPV prevention programs, and it summarizes data relevant to these issues.

Objectives: Using the actor-partner interdependence model (Kenny, 1996), the current study is the first to examine: (1) the relation among 4 forms of family-of-origin aggression (FOA), namely, father-to-mother, mother-to-father, father-to-child, and mother-to-child aggression, and subsequent experience with physical intimate partner violence (IPV) at the couple level; and (2) the gender-specific intergenerational transmission hypothesis. Method: A representative sample of 453 married or cohabiting heterosexual couples from the U.S. northeast completed self-report measures of IPV and FOA as part of a larger study on family and relationship violence. Results: Although both individuals' (respondent effects) and partners' (partner effects) FOA histories generally predicted physical IPV victimization and perpetration, dual-FOA couples were not at increased risk for TV. Respondents' interparental and partners' parent-to-child aggression experiences were most predictive of IPV. Gender-specific transmission of aggression across generations was only partially supported. Last, mother-to-child aggression was a significant predictor in 3 of the 4 models. Conclusions: Findings support the intergenerational transmission of aggression (Widom, 1989) and social learning/cognitive (Bandura, 1977, 1997) theories, and suggest that both partners' IPV and FOA (which often includes multitype maltreatment) experiences should be assessed and considered when developing prevention and treatment programs. Violence prevention parent training programs are also discussed.

OBJECTIVE: The U.S. Air Force recently implemented system-wide changes that both (a) clarified the criteria used to determine when family maltreatment has occurred and (b) made the process by which these decisions are made more consistent. The current study examined the effects of these changes on family maltreatment recidivism. METHOD: Official records were obtained from the Air Force Family Advocacy Program. All cases decided during the last year of the old system and the first year of the new system at each base (total N = 14,298) were examined. For each incident, it was determined (a) whether the incident met criteria and (b) whether the same offender committed family maltreatment again within 1 year of the initial incident. RESULTS: Overall substantiation rates were significantly lower (p = .003) under the new system (47%) than the old (56%). After the change, significant interaction effects were obtained for both alleged (b = -.51, p = .004) and substantiated (b = -.55, p = .015) reoffense, in that 1-year reoffense rates decreased significantly among initially substantiated cases but remained unchanged among initially unsubstantiated cases. Indeed, rates of substantiated reoffense by substantiated offenders were cut in half (from 14% to 7%). CONCLUSIONS: Reductions in overall substantiation rates were most likely due to the use of more stringent criteria. The results of the recidivism analyses suggest that clear criteria and consistent decision processes can have secondary preventive effects on family maltreatment in their own right, possibly due to increases in informal community sanctions.

The authors sought to provide an initial evaluation of the hypothesis that corporal punishment is less strongly associated with parental emotion and impulsivity among African American ("Black") in contrast to European American ("White") parents. White-Latino and Black-Latino differences in corporal punishment, emotion, and impulsivity were explored, given the lack of existing theory predicting group differences. Couples with 3- to 7-year-old children were recruited via random digit dialing, and the parents completed questionnaires and an analog parent-child conflict task in the laboratory. Group differences were tested pooling mothers and fathers via dyadic data analyses. Black parents (N = 57) had more positive attitudes toward and used more corporal punishment than White parents (N = 730). Latino American parents' (N = 78) views and use of corporal punishment were similar to those of White parents. By and large, associations of corporal punishment with parents' impulsivity and emotion did not significantly vary by race/ethnicity. The present findings, although preliminary, do not support the emotion-impulsivity hypothesis of racial differences in the use of corporal punishment suggested by K. Deater-Deckard, K. A. Dodge, J. E. Bates, and G. S. Pettit (1996).

OBJECTIVE: Research on and intervention for child emotional abuse and emotional aggression toward children have been severely hampered because there have been no agreed-upon, clinically usable definitions. METHODS: We have (a) proposed and field-tested a set of criteria to operationally define child emotional abuse for clinical settings and (b) used these criteria to design a parent-report measure of parental emotional aggression and child emotional abuse that could be used in research. In this paper, we review the development and field trials of these criteria for making substantiation decisions. RESULTS: Agreement between master reviewers and field decisions was extremely high in a 5-site development trial (96% agreement, kappa=.89) and a 41-site dissemination trial (90% agreement, kappa=.73). We compare these criteria to other research criteria in the literature. We then present data collected using a self-report measure designed to parallel these criteria from an anonymous online survey of US Air Force personnel and their spouses. The final sample (N=52,780) was weighted to be representative of the United States civilian population. The prevalence of parents' emotionally aggressive acts was much higher than the prevalence of emotional abuse (acts plus impact), but rates of parents' acts of emotional aggression were lower than those typically reported in the literature. Additional analyses tested for differential effects due to gender of perpetrator (i.e., mothers or fathers), age of victim, and clustering within families. These factors did not drive rates of aggression or abuse. CONCLUSIONS: In sum, the criteria developed and proposed appear to support reliable clinical decision making regarding child emotional abuse and can be translated to research survey tools that better capture the continuum of parents' emotional aggression and child emotional abuse than the measures that are currently available, advancing the state of the science with respect to child emotional abuse.

OBJECTIVE: Suicidal members of the U.S. military often fail to disclose their suicidal urges and behaviors. Military suicide prevention efforts may therefore be enhanced if they also target less stigmatized psychosocial factors that may decrease risk of suicidality. In keeping with Bronfenbrenner's (1977, 1994) model, this study simultaneously examined 4 ecological levels (i.e., individual, family, workplace, and community) of factors variously associated with increased or decreased risk for suicidal ideation. METHOD: Active-duty U.S. Air Force members (N = 52,780; 79.3% male; 79.2% non-Hispanic White; mean age = 31.78 years, SD = 7.38) completed the 2006 Community Assessment survey (a biennial, anonymous survey conducted at 82 U.S. Air Force bases worldwide), including the Centers for Disease Control and Prevention's (2008) 5-item measure of past-year suicidality along with scales assessing an array of potential predictors. RESULTS: The 1-year rate of suicidal ideation, defined as (a) more than rarely thinking about suicide or (b) ever seriously considering suicide, was approximately 4%. In multivariate models, for men and women, individual- (depressive symptoms and alcohol problems), family- (relationship satisfaction and intimate partner victimization), workplace- (hours worked), and community-level (social support) variables were retained in the final model. However, some sex differences in retained predictors were noted (e.g., men: dissatisfaction with the U.S. Air Force way of life; women: workplace relationship satisfaction and financial stressors). CONCLUSIONS: Addressing depressive symptoms and alcohol use, facilitating healthy relationship functioning, and increasing job satisfaction and social support may aid military suicide prevention efforts. These findings illustrate the importance of attending to multiple levels of potential influence when designing integrated suicide prevention and intervention programs.

OBJECTIVE: Hazardous drinking is a serious societal concern in military populations. Efforts to reduce hazardous drinking among military personnel have been limited in effectiveness. There is a need for a deeper understanding of how community-based prevention models apply to hazardous drinking in the military. Community-wide prevention efforts may be most effective in targeting community functioning (e.g., support from formal agencies, community cohesion) that impacts hazardous drinking via other proximal risk factors. The goal of the current study is to inform community-wide prevention efforts by testing a model of community functioning and mediating risk factors of hazardous drinking among active duty U.S. Air Force personnel. METHOD: A large, representative survey sample of U.S. Air Force active duty members (N = 52,780) was collected at 82 bases worldwide. Hazardous drinking was assessed with the widely used Alcohol Use Disorders Identification Test (Saunders, Aasland, Babor, de la Fuente, & Grant, 1993). A variety of individual, family, and community measures were also assessed. Structural equation modeling was used to test a hypothesized model of community functioning, mediating risk factors and hazardous drinking. RESULTS: Depressive symptoms, perceived financial stress, and satisfaction with the U.S. Air Force were identified as significant mediators of the link between community functioning and hazardous drinking for men and women. Relationship satisfaction was also identified as a mediator for men. CONCLUSIONS: These results provide a framework for further community prevention research and suggest that prevention efforts geared at increasing aspects of community functioning (e.g., the U.S. Air Force Community Capacity model) may indirectly lead to reductions in hazardous drinking through other proximal risk factors.

OBJECTIVE: Intimate partner violence (IPV) is a serious health concern, but little is known about prevalence of IPV in the armed forces, as military members cope with the pressures of long-standing operations. Furthermore, previous prevalence studies have been plagued by definitional issues; most studies have focused on acts of aggression without consideration of impact (clinically significant [CS] IPV). This is the first large-scale study to examine prevalences of IPV, CS-IPV, and clinically significant emotional abuse (CS-EA) for men and women. METHOD: A United States Air Force-wide anonymous survey was administered across 82 bases in 2006 (N = 42,744) to assess IPV, CS-IPV, and CS-EA. RESULTS: The adjusted prevalence of CS-IPV perpetration was 4.66% for men and 3.54% for women. Prevalences of IPV perpetration were 12.90% for men and 15.14% for women. CS-EA victimization was 6.00% for men and 8.50% for women. Sociodemographic differences in risk for violence were found for gender, race/ethnicity, pay grade, religious faith, marital status, and career type even after controlling for other demographic variables. CONCLUSIONS: Partner maltreatment is widespread in military (and civilian) samples. Men were more likely to perpetrate CS-IPV, whereas women were more likely to perpetrate IPV. Specific demographic risk factors were identified for different types of partner maltreatment (e.g., lower rank predicted higher risk for both perpetration and victimization across men and women). Other sociodemographic differences varied across severity (IPV vs. CS-IPV) and across gender.