ID potential trouble spots

Florida DOC turns to SAS® to help ensure citizen safety

As one of the pillars of the state's criminal justice system, the Florida Dept. of Corrections (DOC) embraces a mission to protect the public and rehabilitate offenders by operating more than 50 major facilities housing more than 100,000 inmates and tracking an additional 145,000 offenders on various types of community supervision.

As a result of these extensive operations, the DOC collects vast amounts of information about its prison and community-supervision populations, including data on health, education, substance abuse programs, classes attended, tests taken, transfers within the prison system, court appearances, employment records, demographic data, attempted escapes, case notes, meetings with parole/probation officers, and more.

The DOC supplements that data with even more information from other criminal justice agencies in the state to create the most comprehensive perspectives possible.

With SAS, we have the right information to deliver to our stakeholders, allocate our resources more efficiently and preserve citizen safety.

David Ensley
Research Director

Dozens of data sources integrated

According to David Ensley, DOC's Research Director, a comprehensive reporting and analytics strategy driven by SAS® has enabled the department to integrate dozens of disparate data sources from a variety of platforms.

"We have more than 100 scheduled data extracts every month," he explained. "These are simple COBOL jobs that bring the data to us in text files. Then we use SAS to import that information into our data warehouse."

That's when a team of 16 SAS experts takes over, building a broad suite of standard statistical analyses and reports that are delivered to various stakeholders and requesters on an annual, quarterly, monthly and semimonthly basis.

Lawmakers get answers right away

In addition, the team responds to a constant flow of ad hoc requests that come from numerous sources: lawsuit discovery, the Governor's Office, state legislators, auditing agencies, federal officials, reporters, the general public and others.

These impromptu requests can range from a simple list to a complex question such as, "What is the recidivism rate for this subset of inmates who are enrolled in this specific program?"

"They all want their answers 'right away,' of course," said Ensley, "so we need to be very responsive. These kinds of questions usually involve multiple SAS data sets that must be merged. Our team mostly creates these reports by writing code by hand in a simple editing window, although we're starting to use SAS to build some of the unique custom requests that involve multiple steps."

Inmates get the right interventions

One of the major challenges faced by the department is the allocation of scarce program resources to a huge inmate population in need of services.

To address this, the DOC developed a risk and needs assessment tool called the Corrections Integrated Needs Assessment System (CINAS). Its purpose is to identify inmates who are most in need of programs and ensure they receive priority ranking for program enrollment.

The DOC's research group developed a model to predict inmate recidivism based on inmate characteristics captured in DOC databases. This model identifies higher-risk inmates who would benefit the most from program intervention. Survival analysis and logistic regression techniques in SAS were used to develop these models using data on thousands of inmates and hundreds of data elements.

Conditions monitored via dashboard

The DOC has applied similar concepts and approaches in analyzing inmate populations at the prison level. The Correctional Operational Trends and Analysis System analyzes a variety of factors to actually predict which facilities are at greater risk for disruptions such as escape attempts or inmate unrest.

"We've uncovered and statistically proven the correlations between disruptions and specific inmate characteristics, gang presence and other indicators," Ensley said. "Based on our SAS analyses, our IT office turned that into a Web-based dashboard that shows the condition of each facility using a familiar stoplight interface. Then the user can drill down to see the risk factor behind that warning – such as a recent wave of new inmates belonging to gangs. With SAS, we have the right information to deliver to our stakeholders, allocate our resources more efficiently and preserve citizen safety."


The Florida Department of Corrections deals with vast amounts of data on inmates, employees, and probationers. In addition to standard reports and hundreds of weekly ad hoc queries, DOC wanted to tap its data to identify most-likely reoffenders and prevent prison disruptions.


SAS® Foundation Tools


Using SAS, Florida DOC responds to constant waves of query requests. It also built a sophisticated system to ID potential trouble spots in prison facilities and strengthen community supervision – with greater accuracy.

Les résultats présentés dans cet article sont spécifiques à des situations, problématiques métiers et données particulières, et aux environnements informatiques décrits. L'expérience de chaque client SAS est unique et dépend de variables commerciales et techniques propres, de ce fait les déclarations ci-dessus doivent être considérées dans un contexte. Les gains, résultats et performances peuvent varier selon les configurations et conditions de chaque client. SAS ne garantit ni ne déclare que chaque client obtiendra des résultats similaires. Les seules garanties relatives aux produits et services de SAS sont celles qui sont expressément stipulées dans les garanties contractuelles figurant dans l’accord écrit conclu avec SAS pour ces produits et services. Aucune information contenue dans le présent document ne peut être interprétée comme constituant une garantie supplémentaire. Les clients ont partagé leurs succès avec SAS dans le cadre d’un accord contractuel ou à la suite de la mise en œuvre réussie du progiciel SAS. Les noms de marques et de produits sont des marques déposées de leurs sociétés respectives.