Although most states differ in the kinds of agencies that are collecting data and the type of information being collected, there are similarities across states. This section presents a general overview of how states are collecting data and how agencies are using the information.
Domestic Violence Legislation
All states have some kind of legislation that defines domestic or intimate partner violence. This definition, however, ranges widely across states. In most states the definition includes those who have ever had some kind of romantic relationship, regardless of gender and cohabitation status. Several, however, limit the definition to couples who have lived together or share a child.
Domestic and Sexual Violence Data
There are five basic sources of data available in the states: law enforcement, victimization surveys, service providers, victim compensation offices, and health/medical agencies. Most of the statistics available on domestic and sexual violence come from law enforcement data. Since these data represent only incidents that are known to the police, it is widely viewed that many offenses are underrepresented in these data. This is especially true for acts of violence against women, whose victims may not report to the police for fear of reprisal or embarrassment. As a result, multiple sources of data may be necessary to approximate the true number of domestic and sexual violence incidents. As with most data, the quantity and quality of the information differs from state to state.
Law Enforcement Data
Law enforcement data are reported to the FBI and consist of incidents that are reported to the police. There are two national reporting systems currently in place - a summary-based and an incident-based - both of which are voluntary. The Uniform Crime Reporting summary system is made up of counts of offenses; each agency submits a tally for each offense type. Since domestic violence is a characteristic of an offense and not an offense in its own right, these are not reported in the summary system. The exception to this rule occurs for homicide incidents; each agency must submit a Supplemental Homicide Report for each homicide that provides information on the offense, victim and offender. Domestic homicides can be isolated by looking at the relationship of the victim to the offender.
The National Incident-Based Reporting System provides a more in-depth look at offenses since offense, victim, offender, property, and arrestee information is provided for each incident. Using the relationship of the victim to the offender, domestic incidents can be isolated and analyzed. Incident-based data sets provide more insight into the characteristics of offenses and the people involved. Most importantly, in the incident-based systems, the information from a single incident can be matched together; the summary data, on the other hand, only gives counts for each data category and cannot be compared across categories.
As of August 2007, 31 states are certified to report incident-based data to the Federal Bureau of Investigation (FBI), although only 10 of these have 100% of their law enforcement agencies reporting. In most states, there is a mix of reporting, with most agencies continuing to report summary data with a slow move towards incident-based reporting, mostly in medium-sized agencies. Fifteen states are either testing or developing incident-based systems, while 5 states report no plans for incident-based reporting. All state Uniform Crime Reporting (UCR) units provide law enforcement data to the public through reports, the release of data, or Web sites that allow users to manipulate the data directly to create customized tables.
States may also require law enforcement agencies to collect and report information in addition to what is required by the FBI. Several states have specialized data collection systems specifically for domestic and/or sexual violence. Such systems may be either summary or incident-based. These data are not collected by the FBI but may be available from the state UCR program, and some states produce specialized reports for the public.
The information collected by both summary- and incident-based specialized law enforcement systems often mirrors what is collected by the FBI's NIBRS, except for property information.
Random surveys of the public can also provide information, particularly on incidents that are not reported to the police or additional information that is not collected by law enforcement agencies. Research has suggested that the anonymity of speaking to a researcher on the phone rather than in person may allow victims to discuss criminal incidents that they are unwilling to report to the police, or may allow for the recording of crimes victims may have deemed too insignificant to report. In this way, incident-based survey data can be gathered to provide an alternative source of data to estimate crime rates.
The National Crime Victimization Survey is funded by the Bureau of Justice Statistics and conducted by the Census Bureau twice a year and several states have instituted statewide surveys as well. Few states, however, conduct victimization surveys on a regular basis. Like the law enforcement data, victimization survey data are usually made available to the public in the form of reports.
Service Provider Data
Agencies that provide services to victims often keep statistics; at the very least, the number of people receiving services is often required to fulfill grant obligations. All states have coalitions that exist at the state level; most have separate coalitions for domestic violence and sexual assault. The role that they take differs from state to state. Some coalitions function as data collection agencies with sophisticated reporting systems and standardized report forms. Others collect no information and simply act as lobbyists or provide technical assistance to local programs.
Service providers are often required to report statistics as stipulated by granting agencies. Violence Against Women Act (VAWA) and Victims of Crime Act (VOCA) grants have allowed for states to pass federal money to local agencies to provide services to victims of domestic or sexual abuse. Often agencies are required to report summary statistics, and, less frequently, incident-based statistics, to the granting agencies.
Service providers and coalitions may provide some statistics on their Web sites, and some produce annual reports providing summary data. Generally the data provided to funding agencies are rarely published or made available to the public.
Service providers often collect a wide variety of information on clients. Many also collect information on whether the incident was reported to police, allowing researchers to use both law enforcement and service provider data without the issue of duplicate incidents.
All states provide some kind of compensation to victims based on certain criteria. To ascertain who is eligible to receive payment, victims complete forms that collect various information about the offense, victim, and offender. Most states collect information on the relationship of the victim to the offender and some forms ask victims to indicate whether the offense was domestic in nature. While some agencies collecting this information publish annual reports on the amount of compensation paid for different offense types, this information is not usually provided to the public.
All of the states participate in the Center for Disease Control and Prevention's (CDC) Behavioral Risk Factor Surveillance System survey. While the focus of the survey is on health, the CDC introduced optional modules for use in the 2005 survey, including questions on intimate partner violence and sexual assault. In the first year, 10 states and 2 territories included the intimate partner module and 18 states and 2 territories included the sexual violence module in the survey.
||Intimate Partner Module
||Sexual Violence Module
||AZ, HI, IA, MO, NV, OH, OK, PR, RI, VT, VI, VA
||AZ, CO, CT, DE, FL, HI, ID,MS, MO, NV, OH, OK, PR, RI, SC, TN, VT, VI, VA, WI
The CDC releases the data from the survey, including the optional module data. The survey does not, however, ask if the incident was reported to the police. States may also publish reports summarizing the findings from the survey, and some data are available from the state agency Web sites.
Most states have statutes that include stalking, but stalking is not usually reported to the state UCR program as a single, identifiable offense. Instead, stalking is grouped with other offenses. Since the FBI does not currently have an offense code for stalking, most NIBRS states are not able to identify stalking incidents at the state level.