Archival Version

This study was based on both primary and secondary data obtained from crime victims and institutional sources. The research involved the analysis of multiple datasets related to victims of bias- and non-bias-motivated assaults, and was intended to yield a spectrum of information on the victims' psychological and behavioral experiences and the patterns of prior criminal behavior by alleged offenders and victims. Data were collected on all 560 cases from the Boston Police Department's Community Disorders Unit from 1992 to 1997 that involved a victim of a bias-motivated aggravated assault. In addition, data were collected on a 1-percent stratified random sample of victims of non-bias assaults within the city of Boston from 1993 to 1997, resulting in another 544 cases. For each of the cases, up to 49 items of information were collected from each police incident report. These included the victim's date of birth, sex, and race, the name of the known offenders, and the date, time, location, and nature of the incident. Additionally, the researchers attempted to contact each victim in the sample to participate in a survey about their victimization experiences. Initial contact with each victim was made through an introductory letter sent to the home address recorded on the police report at the time of the incident. Approximately 50 percent of the introductory letters sent to the victim groups were returned as undeliverable by the United States Postal Service. The researchers then used an online computer service called Autotrak to locate probable current addresses for the undeliverable sample. A total of 441 (79 percent) of the surveys from the sample of 560 bias crime victims and 418 (77 percent) of the surveys from the sample of 544 non-bias-motivated assault victims appeared to be successfully delivered using these methods. Due to a low response rate, a random sample of 100 nonrespondents from each victim group was recontacted and offered a $15 incentive to complete the survey. These overall efforts yielded a final total of 91 completed surveys (21 percent) from the bias victim sample and 45 completed surveys (11 percent) from the non-bias victim sample. The victim questionnaires included open-ended, matrix, and contingency questions in five general categories: (1) incident information, (2) police response, (3) prosecutor response, (4) personal impact of the crime, and (5) respondent personal characteristics. Victims whose names appeared to be of Latino or Vietnamese origin were sent versions of the questionnaire in their native language as well as in English. Additional information was also obtained from criminal history records for identified offenders and victims in each case.