BE IT RESOLVED BY THE OREGON JUDICIAL CONFERENCE that in the course of considering the public safety component of criminal sentencing, juvenile delinquency dispositions, and adult and juvenile probation decisions, judges should consider and invite advocates to address the likely impact of the choices available to the judge on future criminal conduct.

 BE IT FURTHER RESOLVED that judges are encouraged to seek and obtain training, education and information to assist them in evaluating the effectiveness of available sanctions, programs, and sentencing options on future criminal conduct.

No. 3    December 15, 1999
Go to issue #1  Go to issue #2  Go to issue #3 Go to issue #4  Go to issue #5  Go to issue #6

Multnomah County’s Sentencing Support Product

 Multnomah County’s “DSS-Justice” got off to a quick start thanks to over $1 million in local “bond technology” funds.  We’ve constructed a data warehouse which is now operational for many criminal justice users’ needs.  The hardest and the last of the projects is the sentencing support tool, which accesses the data warehouse to compare outcomes (based on new criminal activity) for similar groups of offenders sentenced for similar crimes, as correlated with the available sentencing options.
 We can hardly claim that this product is on judges’ benches “before the end of the 20th Century,” which has been a goal.  Maybe if we quibble about when the new millennium really begins, we might be able to argue that the deadline hasn’t really passed.  As many predicted, the process is taking longer than we expected.  At the same time, the expectation of success has never been higher, and we have achieved a working model of the functional product.  Criminal justice is in for profound improvement.
 Our programmers have written the difficult part of the code, and our technicians have solved the tedious problems of extracting data from various sources, transforming it into comparable forms, and linking it to useable query tools to serve multiple needs.  Sentencing support tools are now functional in a demonstration model that was shown to all Oregon judges who attended the association meetings at Salishan in September.  The Oregon Judicial Department’s ISD Steering Committee has seen a demonstration, and many others are planned with other major partners in Oregon’s criminal justice community.
 I’ve put screen shots of the demonstration on this website , but for convenience I will repeat the description here.  As a user, a judge (clerk, DA, or defense attorney) would begin by entering an OJIN case number (alternate methods of identification are readily available).[To access the image directly, use this button; return with the "back" arrow on your browser:]  From this information, the program “knows” the offender, the present charges, the offender’s profile and criminal history.  When the user clicks on “submit,” the program prompts the user to identify the charge for which a sentence is being considered — proposing that involved in count one of the charging instrument, but allowing the user to select any state or local crime known to Oregon law.  The program now “knows” the kind of charge for which we are sentencing; it selects proposed offender profiles, outcome measures, and sentencing experiences based on the information entered so far and the data available from the data warehouse.
 Clicking “submit” again yields a bar chart that displays a great deal of information graphically.  Each bar represents the percentage of “success” for one common sentencing element when used on similar sentencing occasions on similar offenders.All data is contained in a table visible by scrolling below the bar chart, with infrequently used sentencing elements entered in red to highlight the absence of statistical significance.
 During the course of preparation for — and during — sentencing hearings, users can modify all of the program’s assumptions: the sentencing experience (“charge”) subject to comparison (in a theft II case, the program suggests “property crime” as the relevant experience to compare, but can be narrowed to the specific statutory charge, extended to all crimes, or set almost anywhere in between); the outcome measure (“recidivism”) applied to the analysis (the program suggests convictions for charges in the same category of crime within three years, but can be changed as to time period, arrest as opposed to conviction, and to more specific or more general crime categories); and the offender profile (“offender”).
 The product’s approach to defining what makes other offenders “like” the one before the court is especially important and innovative.  Based on the offender identified in the first screen and that offender’s history and circumstances, the software selects a gender, age range, ethnicity, and criminal history “theme.”  The “theme” is composed of ratings (from none to severe) in each of five criminal categories: drugs, property, violence, traffic, and sex crimes.  You can alter the program’s approach (for example, by using sentencing guideline seriousness and history designations) or details (for example, by changing age ranges, criminal history themes, deleting or expanding ethnicity).
 In any case, whenever you make any change in the assumptions, you can simply recalculate and see if any sentencing elements seem to be more or less effective than others under the circumstances your selections have defined.
 Multnomah County’s “DSS Policy Committee” has designated a “sentencing support focus group” to monitor and develop this product.  The group consists  of Deputy District Attorney Tom Cleary,   criminal defense lawyer John Connors, “DSS Coordinator” Nancy Arnot ((503) 988-3749 ext 26117;  nancy.e.arnot @co.multnomah.or.us), and myself.  We expect the newly hired technical lead Charles Hill to be part of the focus group. (charles.s.hill@co.multnomah.or.us).  Our first addition will be a sixth “domestic violence” category for purposes of composing the criminal history theme.
 Multnomah County has taken over the lion’s share of funding for the DSS project, ensuring that it will have a budgeted support in years to come, now that the bond technology period is expiring.  The contractors who worked on the project will be replaced slowly by IT staff already hired and to be hired in the county’s ISD department specifically to continue these projects.
 Our immediate tasks are to refine the sentencing elements to analyze with the data now available to us.  OJIN fields are of limited usefulness for these purposes.  We hope to identify typical sentencing “clusters” based in part on the district attorney’s typical recommended sentences.  We also hope to use text data collected by the court coordinator for misdemeanors to link programs with their graduates.
 We also need to include more data sources in the warehouse, particularly those from the Department of Corrections and the Department of Juvenile and Adult Community Corrections, as well as those containing drug and alcohol treatment sentence information.
 There is a long way to go, but we’ve made the first flight at Kitty Hawk, and there is no turning back.  We will make every effort to bring a working version to the courtroom within the year 2000.  We welcome your input and your questions.
 OJD’S ISD
 On the state level, the Chief Justice and ISD Director Carl Ward are fully committed to developing sentencing support technology for all Oregon judges.  Although the 1999 Legislature provided no new judges, no salary increases, and no retirement improvements, it did fund the “Integrated Justice Environment initiative” at $1.8 million and reserve another $1 million which we will seek for technology purposes from the Emergency Board.  The “Judges Workbench” is an element of the planned “Integrated Justice Environment,” and will include sentencing support as a major component. We demonstrated Multnomah County’s DSS product, and our sentencing support tools, to the November 30, 1999, meeting of the Information Systems Steering Committee.  ISD is filling the newly funded FTEs with the sentencing support function as a top priority.  While full functionality will require substantial progress by the State Police in their “Public Safety Data Warehouse,” OJD’s ISD is building its own “data mart” which can and will serve sentencing support purposes as well as OJD’s other information services objectives.  Within ISD, Jeff Barlow is responsible for brining us sentencing support, (503) 986-5588; Jeff.N.Barlow@state.or.us.
 The Public Safety Data Warehouse
 As reported in the last Resolution News, the Governor’s Working Group on implementation of 1997 HB 2229 (Executive Order 97-19) recommended construction of a state-wide data warehouse, in large part to provide data resources to serve the sentencing support technology we have been promoting.  The Department State Police obtained and Governor Kitzhaber authorized the expenditure of $2.2 million in federal funds to build this “Public Safety Data Warehouse.”
 The State Police held a kickoff event on November 4, 1999, to launch this effort.  “The goal of the PSDW is to provide a central point for data from which decisions can be made regarding the criminal justice system and the effectiveness of criminal justice programs, including their ability to reduce future criminal conduct.”  In its first stages, the warehouse will gather and “normalize” data from the State Police (LEDS), the Department of Corrections, and the Oregon Judicial Department; Oregon Youth Authority data will be added as soon as it is responsible to do so.
 The PSDW will be an important source of data for sentencing support analysis and, I suspect, sentencing support will be the function that most significantly promotes public safety and most profoundly alters the role of criminal justice in the coming century.
 The Oregon Department of State Police is also hiring to fill new FTEs with personnel trained in the technologies upon which sentencing support depends: data warehouse and “on-line analytical processing” (OLAP).
 The PSDW is the responsibility of David C. Yandell, (503) 378-3054, David.C.Yandell@state.or.us.

- MM  Return to Smartsentencing home page  Go to issue #1  Go to issue #2 Go to issue #3 Go to issue #4  Go to issue #5  Go to issue #6