Smallest Geographic Unit:
Date of Collection:
Unit of Observation:
All sex offenders released from prison between January 2006 and March 2009 and residing in the state of California. All parole agents from the California Department of Corrections and Rehabilitation in fall of 2008.
administrative records data,
Data Collection Notes:
These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped for NACJD release, but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.
The HRSO parolee data files can be linked on the variable "Study_ID".
The purpose of the evaluation was to determine the effectiveness of global positioning system (GPS) monitoring of high-risk sex offenders (HRSOs) who were placed on parole.
This study integrates both outcome and process evaluation components. The outcome component assessed the impact of the California Department of Corrections and Rehabilitation (CDCR) GPS supervision program by employing a nonequivalent-group quasi-experimental design with a multilevel survival model. A propensity score matching procedure was used to account for the differences between the treatment and comparison groups.
Data about high-risk sex offenders (HRSOs) were collected from five sources:
- CDCR Data Management System: Most data used for this study were derived from three databases: the Parole Law Enforcement Automated Data System (Cal-Parole), the Offender-Based Information System (OBIS), and the Revocation Scheduling Tracking System (RSTS).
- Official Arrest Records: Official records of arrests, convictions, and custody were collected for each study subject. With the help of CDCR staff, the all of these records were manually entered into a database specifically developed for this study.
- Parole Supervision Records: Each parole agent maintains a record of supervision for each parolee under supervision. The parole agent notes the date and specific type of contact. The record of supervision is stored only in hardcopy format in the parolee case file, which typically is located in the parole unit of record. Consequently, a set of site visits was conducted to obtain the record of supervision data from the parole agent case files. The data were then keyed directly into a database specifically developed for this study.
- GPS Monitoring Data: The GPS monitoring system into which HRSO parolees were enrolled was operated by two vendors: Satellite Tracking of People (STOP) LLC and Pro Tech. Data were collected directly from the vendors.
- Cost Information: The cost information elements used in the study are grouped into four broad categories: personnel, facilities, equipment and materials, and other inputs. This information was obtained through communications with CDCR staff and from a review of budget documents.
A survey instrument was developed to collect process data from CDCR parole agents. The instrument was emailed to all parole agents in the fall of 2008. The parole agents were sent numerous requests to complete the survey during the next few months. They were also asked to fill out the survey during training and other events hosted by CDCR. The survey was closed at the end of November 2008.
The initial sample included more than 18,0000 offenders (all sex offenders released from prison from August 1984 through April 2009). This sample was initially narrowed down by eliminating duplicate records (10 records) and subjects who were ineligible. Ineligible subjects included offenders who were non-high risk sex offenders (non-HRSOs) (12,293 subjects), paroled outside of the timeframe of the study (2,429 subjects), and paroled out of state, deported, or outside the purview of the California Department of Corrections and Rehabilitation (CDCR) (54 subjects). In addition, offenders who were classified as Sexually Violent Predators (SVPs) or failed to appear before the parole agent for the initial parole visit were removed from the study (33 subjects). The treatment group was then restricted to parolees whose elapsed time between release from prison and place onto GPS monitoring was fewer than seven days. Propensity scoring was used to match parolees in the treatment group with parolees in the control group. The final sample included 516 subjects equally divided between the treatment and control groups.
The parole agent survey was mailed to all parole agents in fall 2008. The survey received 747 responses from a population of roughly 1,000 parole agents. A subsequent review of the responses indicated that 120 of the 747 responses contained no data, resulting in 627 survey responses. The survey focused on GPS parole agents with HRSO caseloads. Consequently, supervisors (125 surveys), agents without GPS caseloads (315 surveys), and agents without HRSO caseloads (24 surveys) were removed from the sample. Finally, 10 duplicate surveys were removed, resulting in 153 unique surveys from GPS parole agents with HRSO caseloads.
Mode of Data Collection:
Description of Variables:
HRSO Final Sample Data (n=516) includes 149 variables on parolee demographics (age, sex, and race), total number of days on GPS, parole revocations and suspensions and violations, and offense information.
HRSO Pre-Match Sample Data (n=747) includes 72 variables with parolee demographics, information on parole revocations, suspensions, and violations, and offense information.
Offense Static Data (n=1,188) includes 7 variables on charge code, offense description, severity score per California Department of Corrections and Rehabilitation (CDCR), offense type, "registerable" sex offense, sex offense description, and estimated severity score.
Arrest Events Data (n=13,713) includes 16 variables on the total number of charges, charge code and description, charge disposition, and arrest date.
Court Events Data (n=9,084) includes 19 variables on charge code, number of charges, charge level, charge sentence code, charge length, charge disposition, and date of the court event.
GPS Events Data (n=30,136) includes 16 variables on event type, violation start time, violation stop time, duration of the violation, and notifications sent.
Violations Data (n=1,180) includes 17 variables on date of parole violation, charge code and description, sex violation, severity score, and violation type.
Custody Events Data (n=4,874) includes 13 variables on date released from prison, status of prison admission, release destination, number of days in prison, and number of days between release and study start date.
Caseload Data (n=73) includes 9 variables on parole region and district, year of parole, number of parole agents, number of parolees or cases, and number of months included in the sample.
Parole Agent Survey Data (n=153) includes 179 variables in the following area:
- Overview of the GPS system
- Reduction of caseload size
- Screening of HRSO parolees
- Enrollment and orientation
- Integration of GPS monitoring into the intensive supervision regime
- Synthesis of parole GPS and law enforcement data
- Program staffing; and
- General summary.
The response rate for the Parole Agent Survey was 67 percent, with 153 of roughly 230 GPS patrol agents with HRSO caseloads responding to the survey.
Presence of Common Scales:
Parole Agent Survey Data includes Likert-type scales.