LRPP EXHIBIT IV: PERFORMANCE MEASURE VALIDITY AND RELIABILITY
Department: Corrections
Program: Community Corrections
Service: Community Corrections
Measure: Status of offenders two years after the period of supervision was imposed: number absconded
Action (check one):
Requesting Revision to Approved Measure
Change in Date Sources or Measurement Methodologies
Requesting New Measure
Backup for Performance Measure
Data Sources and Methodology:
Data Source: The Offender Based Information System (OBIS), the Department’s database into which both field and institutional staff enter information on offenders and inmates.
Methodology: Community Corrections field staff enters gain, loss to absconding, and termination data on the PP02 screen. Then extract files pull that data directly off OBIS and these datasets are then converted to SAS datasets. SAS programs are then written to run against the datasets to determine to outcomes of offenders.
Determine the Florida case origin offenders who were admitted to supervision during the target year. Eliminate offenders whose final outcome code as of two years after their admission date had been court terminated or died or whose sentences were vacated. Examine the remaining offenders to determine their status as of 2 years after the date of admission to supervision, dividing them into those who absconded without a return from absconding; those who were revoked; those who had an early/normal termination; and those who were still under supervision and not absconders. Report the number absconded without a return from absconding.
Validity:
Internal validity speaks to the certainty with which the results of this measure can be accepted. Staff has used the PP02 screen for many years, with the listings continuously being verified and checked and accepted with a high level of certainty. External validity speaks to the results being generalizable. The results are based on the entire fiscal year cohort population rather than a sample. Court terminations and deaths are excluded since it is not appropriate to consider either of these outcomes as successful or failure outcomes.
The purpose of the Department of Corrections community supervision program is to carry out the orders of the court. Supervising offenders in the community requires an officer to notify the courts if the offender is behaving inappropriately. An absconding event means that an offender has fled supervision, his/her whereabouts are unknown, and the court has issued a warrant for a violation of supervision. This is an appropriate measure of one aspect of offender failures under community supervision, and the appropriate Departmental response to protect public safety.
The OBIS data constitute an appropriate measure of the outcome of offenders under supervision by the Department.
Reliability:
Since all supervision movement data are used, rather than a sample, the measure is by definition very reliable. The data reported are consistent from one measurement to the next and have been shown to be consistent, complete, and correct.
Office of Policy and Budget – July, 2006