Assessing the Role and Impact of Public Policy on Child and Family Violence

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Public Administration


Leonard M. Lopoo


child and family violence, Gun Control Act expansion, intimate partner violence, pro-arrest policies, unemployment

Subject Categories

Public Policy


This dissertation is comprised of three essays, which study how public policy affects child and family violence. In addition to evaluating the efficacy of existing policies, this dissertation seeks to understand where there might be opportunities for interventions to reduce family violence. Specifically, the first two papers investigate the effects of existing policies - state-level "pro-arrest" policies and the federal Gun Control Act Expansion of 1996 - on domestic homicide rates. The third paper uses New York State as a case study to understand how county level unemployment rates affect county level child abuse referral rates.

In Chapter 2, I use a difference-in-differences design to estimate the effect of state-level pro-arrest policies on domestic homicides, which include intimate partner homicides, child family member homicides, and other family member homicides. I employ the FBI's Supplementary Homicide Reports along with homicide data collected from selected state law enforcement agencies to construct measures of domestic homicides. I assess the effect of pro-arrest policies targeted toward a range of domestic violence crime scenes and find mixed results. I conclude that pro-arrest policies have little effect on intimate partner homicides and may reduce homicides among male domestic child victims (sons and step-sons). The policies appear to have large effects on homicide rates among "other" family members' (parents, siblings, and other relatives), which are driven by changes in female family member homicide rates.

Chapter 3 explores the effect of the 1996 expansion of the federal Gun Control Act, which prohibits defendants convicted of a qualifying domestic violence misdemeanor from possessing or purchasing a firearm. I continue to use the augmented homicide data used in Chapter 2, and I investigate if this expansion was successful in reducing homicides among the target groups. I use variation from a legal loophole and a series of circuit court decisions to generate difference-in-differences estimates. I find that the Gun Control Act expansion led to 12 percent fewer gun related homicides among female intimate partner victims (current and former wives) and 34 percent fewer gun homicides among male domestic child victims. The law also has protective benefits for those that were not targeted by the legislation. Other family members also experience a 26 percent reduction in gun homicides. I find no evidence that reductions in gun homicides were offset by an increase in non-gun homicides.

Finally, in Chapter 4, I use child abuse referrals made to Child Protective Services in New York State from 2000 to 2010 to investigate the relationship between county level unemployment and county level child abuse referral rates. I find that a 1 percentage point increase in unemployment rates reduces the child referral rate by approximately 5.4 percent. Referral rates among young children (children under the age of 6) and older children (children ages 6 and over) respond similarly to changes in unemployment; though young children's referral rates appear to be somewhat more sensitive to changes in the unemployment rate. The relationship between unemployment rates and child abuse referrals does vary by a county's metropolitan designation. The negative relationship between unemployment and child abuse referrals is largely contained to metropolitan counties. The relationship between unemployment and child abuse referrals in non-metropolitan counties is often positive but not statistically significant.


Open Access