The Division of Corrections Failed in its Attempt to Measure Recidivism After a 1994 Performance Audit

Performance Evaluation and Research Division
Building 1, Room W-314
State Capitol Complex

(304) 347-4890

June 1999


The West Virginia Department of Corrections (DOC) was established in 1909. The Executive Reorganization Act of 1989 transferred the Department of Corrections to the Department of Public Safety and re-designated the former department as the Division of Corrections. The state constitution charges the Division of Corrections with the responsibility of incarcerating adult offenders sentenced under state law to terms of confinement in excess of one year. The Division's mission is: provide a safe, secure and humane correctional system for the public, staff and offenders.

The Commissioner of the Division of Corrections is appointed by the Governor, with the advice and consent of the Senate. The Commissioner governs eight correctional institutions, three work release centers, 14 parole offices, and a central office.

DOC Budget for Fiscal Year 1999

For fiscal year 1999, the Division's budget for it's central office is $569, 359. This includes: $334,984 for Personal Services; $8,260 for Annual Increments; and $115,111 for Employee Benefits. A further $111,004 is unclassified. The total budgeted in fiscal year 1999 for correctional units is $67,253,017. This includes: $13,297,271 for personal services; $191,988 for annual increments; $5,168,839 for employee benefits; $3,916,250 for payment to counties and/or regional jails; $8,021,199 for the St. Marys Correctional Center; $2,498,732 for the Denmar Facility; $15,712,403 for the Mt. Olive Correctional Complex; $5,342,505 for the Northern Correctional Facility; $5,118,439 for inmate medical expenses; $1,200,00 for the Ohio County Jail; and an additional $6,785,391 for unclassified expenses.

History of the DOC Computer System

DOC began implementing a centralized inmate tracking system in 1991. DOC decided to upgrade its Prime computer platform in 1991, for the tracking system. The Prime platform was originally purchased in 1985. DOC uses the inmate tracking system to collect and maintain the following inmate information:

  1. inmate/personal and booking information
  2. personal property inventory
  3. offense information
  4. classification
  5. program participation
  6. sentencing and parole eligibility information
  7. medical and dietary information
  8. movement history
  9. visitation information
  10. victim information
  11. furloughs and special events
  12. work record/work release information

In 1991, the Division of Corrections was ordered by the Secretary of Military Affairs and Public Safety to cooperate with the Regional Jail Authority (RJA) in developing a unified inmate tracking system for the Department. However, during the three year period between 1991 and 1994, DOC and RJA had continuous disagreements as to the necessary data elements needed in the system and whether the system should be centered on DOC's Prime mini computer. The Secretary subsequently informed the Legislative Auditor that:

...problem after problem surfaced in an effort to integrate both agencies into a single system. As a result, both agencies were unable to get their systems operating. Further, due to the different operations required by each, I believed it was in the best interest of both agencies to pursue separate tracks.

The Department of Military Affairs and Public Safety is currently reviewing proposals for a information management system that will be used by the Division of Corrections, the Regional Jail Authority and the Division of Juvenile Services to track inmate information. When the system is implemented, West Virginia will have the ability to track and share inmate information.

West Virginia Supreme Court of Appeals Ruling

State correctional capacity has not kept pace with increases in the average daily population of inmates committed to DOC facilities. In July 1998, the West Virginia Supreme Court of Appeals issued an order for the Division of Corrections to develop a plan to immediately transfer into correctional facilities one half of the 843 state-sentenced inmates housed in jails and awaiting correctional beds. The court order indicated that by incarceration in regional and county jails, state responsibly inmates are denied access to rehabilitative programs to which they are entitled. The court further noted that:

When states or other jurisdictions develop written plans for managing inmate population problems in the 1990s, they first spend a great deal of time and energy making certain that they understand the they are attempting to solve. Most of the steps necessary to develop a legitimate inmate population management plan have not yet occurred.

A good master plan includes three basic elements: 1) a detailed understanding of the nature and extent of the current and projected demand for inmate services; 2) a detailed understanding of the nature and capacity of existing resources to meet that demand, and; 3. Evaluation of how best to use, and augment, existing resources to meet the needs implied by the projected future demand.

The DOC obtained technical assistance from the National Institute of Corrections in preparing a comprehensive master plan. The assistance was funded by the NICs Prisons Division. The direct on-site technical assistance and the subsequent report were intended to help the DOC enhance its effectiveness. On September 17, 1998, the Division of Corrections submitted to the WV Supreme Court of Appeals its plan to transfer state sentenced inmates from regional and county jails. By January 11,1999, the Regional Jail Authority reported only 404 state-sentenced inmates in the RJA's six facilities. According to the Commissioner, the Division of Corrections made headway on relieving overcrowding after the Supreme Court ruling. However, the state prison system is again becoming saturated. The DOC takes in an average of 150 prisoners a month and releases only 90.
Because of the dynamics of the Regional Jail System and the Division of Corrections' penal institutions, it is critical for the DOC to have an effective and useful information management system. The DOC's ability to measure the impact of its rehabilitation programs on its repeat offenders is crucial for effective inter-agency planning.

Issue Area 1:The Division of Corrections Failed in its Attempt to Measure Recidivism After a 1994 Performance Audit.

In 1994, a legislative review identified that the Division of Corrections and the Regional Jail Authority did not track recidivism rates. Recidivism is the primary measure of rehabilitation program effectiveness. The DOCs inability to track recidivism is a direct result of its antiquated Information Management System. The West Virginia Supreme Court of Appeals mandated that the Division of Corrections establish adequate rehabilitation programs at its institutions. In order to assess the adequacy or effectiveness of these programs, recidivism must be monitored. In 1996, the Director of the West Virginia Statistical Analysis Center at Marshall University worked with the Division of Corrections to develop a research protocol and methodology for determining recidivism rates (Appendix A). The director of the Statistical Analysis Center stated to the Legislative Auditor that:

The WVDOC has no means of measuring recidivism rates to effectively evaluate their existing rehabilitation programs. This is due to: The WVDOC does not possess a standard protocol for the collection of pertinent data useful for calculating recidivism rates. The division's current tracking system does not include all the necessary data elements to calculate these rates. The WVDOC has not yet defined what inmates they are able to use to calculate these rates.

The Statistical Analysis Center developed a survey (Available in Appendix b) to be completed on each individual inmate at each correctional facility in order to collect necessary data. Then the Center submitted a proposal to the Division of Corrections regarding the recidivism study. The Division of Corrections informed the Center that money was not available in their budget to cover the $25,000 cost of the study (Appendix C ). The DOC then began a manual review of documents in order to develop some recidivism figures. The results of this study are compared to other states' recidivist rates in Table 1. (Standards for calculating recidivism vary by state, as seen in Table 1.)

Table 1 - Current Recidivism Studies
FloridaRecidivism is defined as a return to state custody or supervision for a new crime occurring within 24 months of the offender's date of release from prison. 18.8%
IllinoisRecidivism is defined as a return to [state] prison within three years after release for a new crime or a violation of release agreement.39.2%
West VirginiaRecidivism is defined as a return to state custody, after having been released, for a new felony conviction.4.1%
KentuckyRecidivism is defined as a return to state custody within two years, after a new felony conviction or a parole violation.33.1%
Source: 1999 PERD survey and analysis of other states' recidivism studies.
The Division of Corrections found a 4.1% recidivist rate for CY 1994 releases and a 1.3% recidivist rate for CY 1995 releases as of June 30, 1996. These findings are significantly lower than the national average of 34.2%. The Division's own recommendation was to re-publish a study in 1997 that included 1994, 1995 and 1996. The DOC failed to follow its own recommendation and the study was never completed. The 1998 Corrections Yearbook publishes individual state's self-reported recidivism rates. (A complete list is available in Appendix D) West Virginia is one of 9 states which failed to report recidivism rates. . The federal recidivism rate is reported as 40.8% . Clearly, the West Virginia Division of Corrections data system is not capable of measuring recidivism, as stated in an intra-agency memo in March 1999:

The Division of Corrections does not routinely track recidivist rates, as our automated system was not set up to track the recidivist rate. We are in the process of purchasing a new automated system wherein we have stipulated that the system must be capable of tracking recidivists. Currently, we are only able to determine the recidivist rate by manual computation. In 1996, we conducted a limited recidivism study of inmates released form our custody in 1994 and 1995...The Commissioner at that time did not feel our recidivist rate warranted additional time and effort on another recidivist study.

The Division of Corrections is not able to accurately report expenditures on rehabilitation because it is not a specific line item in the Budget. Expenditures reported to the Legislative Auditor's Office include, but are not limited to: personal services and other related operating expenses; supplies; and contractual services. The DOC reports that the majority of expenditures are for personal services and benefits for Unit Managers, Case Managers, Correctional Counselors, and Corrections Program Specialists. The DOC reports expenditures of:

Approximately $350,000 a year from Operating Expenses for drug testing, counseling supplies, books, training and contract services, etc.

The DOC dedicates personnel to the administration of rehabilitation programs, however, corrections employees perform other functions as well. At this time, PERD is unable to determine the cost of rehabilitation services, because DOC does not track the time correctional employees devote to these responsibilities. The total amounts budgeted for rehabilitation programs, the impact of which is unknown (particularly as it relates to recidivism) is available in Table 2.

Table 2 - DOC Investment in Rehabilitation
Fiscal YearEmployeesCost
1996124$3.8 million
1997152$4.5 million
1998203$6.2 million
1999215$6.9 million
Source: 1999 PERD Analysis of DOC rehabilitative costs.

Theoretically, rehabilitation infers that if services are provided, repeat offenders will decrease. The public is implicitly promised that rehabilitative programs do something to address the crime problem. Without accurate data, the Division of Corrections is unable to determine whether its rehabilitative programs produce positive outcomes in West Virginia. The Division of Corrections reports that their 'rehabilitative' measures include: Testing and assessing inmate programming and treatment needs; Development of an Individual Program Plan for each Inmate; Unit Team establishes goals along with inmate, including institutional adjustment, behavioral and educational issues, placement in appropriate counseling programs, work, victim restitution and medical treatment; and, Work with inmates in establishing aftercare plans.

Another effect of the Division's lack of accurate recidivism studies is that they are unable to effectively plan ahead for future construction needs. The DOC needs to avoid building more prisons than necessary, as well as another overcrowding crises. Without adequate research and planning, the DOC cannot adequately or efficiently prepare for the future.

In January 1999, the Department of Military Affairs and Public safety issued a RFP for a Jail and Prisoner Information and Facility Management system. This system is intended to allow the Division of Corrections, the Regional Jail Authority and the Juvenile Services Authority to access at least five years of active offender records, with off-line storage of an additional ten years of records. The importance of shared information within the Department of Military Affairs and Public Safety cannot be understated. The Division of Corrections, the Regional Jail Authority and Juvenile Services Authority serve a common and overlapping clientele. In 1994, the Legislative Auditor found that the lack of shared information results in inefficient use of staff time and raises the potential for data entry errors. An information system which allows the state's criminal justice system to access mutually needed information, will enable the DOC and the RJA to better classify and make decisions regarding the offender population of the RFP states:

The system must record details of an offender's return to custody following a parole, probation or home confinement violation. states:

Users must be able to inquire about the entire history of an offender's encounters with the agency.

The DOC reported to the Legislative Auditor's Office that:

We envision that each inmate within the DOC's system will have a unique number assigned with each separate period of imprisonment in DOC identified by a sub number. This will automatically tell DOC how many times the inmate has been in the DOC system. Upon admission to DOC, a search will be conducted of the database by name (full, partial, or soundex spelling) or number (offender #, DL #, etc) to determine prior record existence. If a prior record exists, it is activated with the assignment of a sub-number and existing data is updated. The prior record will remain stored with no change. The reason for each imprisonment term will be reviewed to determine if the inmate is a Recidivist.
According to the DOC's Director of Administration, "The RFQ for the Electronic Management Information System does not specifically refer to a recidivism report per se. It does include fields of information that will be used as a data base for such a report which our staff will be able to effect through the use of crystal adhoc reporting once the parameters of all fields necessary are established."

Once the management information system is operational, the DOC will still lack adequately trained staff to perform recidivism studies. There are three possible options: 1) The DOC could create an internal unit with the sole responsibility of research and planning; 2) The DOC could contract with one of the state Universities' Statistical Analysis Centers; or, 3) The West Virginia Division of Criminal Justice Services (WVDCJS) could be expanded to include these responsibilities.

Recommendation 1:

The DOC should continue to implement the new management information system as a top priority by assuring that a new system will produce the data which is necessary to measure recidivism.

Recommendation 2:

The DOC should develop a methodology with the assistance of a consultant to measure recidivism so that once the new management information system is operational, another recidivism study may be immediately undertaken and reported to the Committee.

Recommendation 3:

The DOC should regularly update the Joint Committee on Government Operations on the cost and progress of the new management information system.