The purpose of the impact evaluation was to gauge the success of the victim's rights clinics in attaining each of the following goals:
- Aid in enforcing rights for individual victims and getting them help for crime-related needs, thereby increasing satisfaction of victims with the justice process;
- Change attitudes of criminal justice officials towards victims' rights and increase their knowledge about rights;
- Change the legal landscape: establish victim standing and develop positive case law;
- Increase compliance of criminal justice officials with victims' rights; and
- Sustain the clinic through developing alternative sources of funding
Researchers conducted surveys with prosecutors, judges, victim advocates, and defense attorneys to determine whether they had changed their attitudes toward victims' rights since the local clinic opened its doors. Surveys were fielded in South Carolina, Maryland, and Utah (Criminal Justice Official data, n=552) where clinic evaluations were conducted. An additional survey was fielded in Colorado (Colorado data, n=583) where the victim rights clinic had not yet started to accept cases. The surveys were completed using web-based technology using the survey development tool QuestionPro, working through the state associations for each of the four groups of criminal justice officials. Victim rights clinic directors introduced researchers to the heads of the state associations of prosecutors, judges, prosecutor victim advocates, public defenders, and the private bar. Working through the state associates, researchers sent the survey via e-mail to the membership roster of each group. The invitations were followed up by two e-mail reminders sent at approximately two week intervals. The first mailing discussed the survey purpose and invited participants to complete the survey on line. Included in the initial letter was the survey URL, and a unique ID. Each participant was allowed to complete only one survey but they were able to change responses at any time as long as the survey remained open. In Colorado, a second wave of the survey was fielded approximately one year after the first wave using the same procedures.
To determine the effect that clinics had on observance of victims' rights, researchers collected three samples of cases from prosecutors (NCVLI Case File Data, n=757) in the jurisdiction or jurisdictions in which each local clinic had done the most work: (a) all clinic cases closed since the start of each local clinic; (b) cases closed during the most recent 12-month period which did not involve representation by a clinic attorney; and (c) cases closed in the year prior to the start of each local clinic. Data abstracted from the cases sampled from electronic databases and paper records maintained by prosecutor and victim advocate offices included the charge, case disposition, contacts between prosecutor staff and victims and information about the observance of victim rights.
To assess the impact of the clinics on victims' satisfaction with the criminal justice process and its compliance with their rights, researchers conducted telephone interviews (Victim Survey Data, n=125) with two samples of victims in each evaluation site, one drawn from the sample of cases at prosecutor offices and one drawn from the crime victim legal clinics. To preserve victims' confidentiality, the initial contact with victims from cases in the prosecutor case file samples was made by staff of the victims rights clinics (for clinic cases) or prosecutors' staff (for non-clinic cases). Names and contact information for victims who agreed to participate were passed to research staff to conduct interviews by phone. Victims who participated in the survey were compensated between 25 and 50 dollars.
National Crime Victim Law Institute clinics included in the impact evaluation were purposely chosen to ensure more mature clinics were included and to ensure that there would be a large enough volume of cases to include in the sample. Researchers chose Utah, South Carolina, and Maryland. Together these sites had opened about 300 cases since the clinics' inceptions within either two or three counties in their states. In addition to these sites, when conducting surveys of criminal justice officials, researchers added Colorado. The clinic in Colorado was just getting started, giving researchers the opportunity to gather baseline survey data on attitudes of criminal justice officials at the beginning of the evaluation period and to compare those results to a second round of survey results obtained a year after the clinic started accepting clients. In order to recruit survey respondents (Criminal Justice Official Data, n=552 and Colorado Data, n=583) researchers asked the victim rights clinic directors at each site to introduce them to the heads of the state associations of prosecutors, judges, prosecutor victim advocates, public defenders, and the private bar. Working through the state associates, the surveys were sent via e-mail to the membership roster of each group.
Sampling of case files varied at each location (NCVLI Case File Data, n=757), though common rules were applied to the process. Once the clinic sample was drawn in each state, researchers drew the baseline samples from prosecutors files in a manner which would ensure that it would as similar as possible to the clinic sample in terms of county of origin and type of crime. The number of baseline cases sampled was keyed to the number of clinic cases in that state. Where possible, researchers sampled cases with each crime category randomly from computer databases. When sites lacked the capability to sample from computer files, researchers developed schemes for obtaining a representative sample from manual files. The process for drawing the concurrent sample of non-clinic case from prosecutor files was the same. For both the clinic and non-clinic baseline sample, researchers oversampled by fifty percent. The final sample of 757 cases included 174 clinic cases, 282 pre-clinic cases and 297 post-clinic cases not represented by a victim attorney.
Victim surveys (n=125) were conducted with two samples of victims in each evaluation site, one drawn from the sample of cases at prosecutor offices and one drawn from the crime victim legal clinics. Names and contact information for victims who agreed to participate were passed to research staff to conduct interviews by phone.
All criminal justice officials, criminal court cases, and crime victims in jurisdictions with victims rights clinics in the United States between 2009 and 2010.
administrative records data
The Criminal Just Official Data (n=552, 65 variables) and Colorado Data (n=583, 38 variables) includes variables on respondent's occupation, types of crime typically prosecuted or defended or which type of court they preside in, opinions on victims' rights, legal standing, legal representation for victims in court, and legal remedies for victims' denied their rights. Other variables focus on how the respondent view other criminal justice officials' application and response to victims' rights. The NCVLI Case File Data (n=757, 24 variables) includes variables on state, county, charge type, court disposition, minimum sentence, whether victims were informed of their rights, court hearings, compensation, restitution, privacy, speedy trial, and victim services. Other variables include whether restitution was ordered, the number of victims, and if referrals to victims services were provided. The Victim Survey Data (n=125, 169 variables) includes variables on who informed the victim of their rights, who notified the victim of developments in the ongoing legal proceedings, who helped protect the victim's privacy, who helped the victim prepare a witness statement, and who referred the victim to services, and what type of services the victims were provided. Other variables ask about the victims' perception of fairness in the criminal justice process and whether or not their rights had been respected.
For the Colorado baseline survey, 1,918 invitations were issued with 378 surveys completed for a response rate near twenty percent. For the follow up Colorado survey, a total of 63 surveys were completed. In Maryland, a total 96 surveys were completed. In South Carolina, a total of 141 surveys were completed. In Utah, a total of 219 surveys were completed. A total of 125 victims completed interviews.
Several Likert-type scales