8 Percent Problem Study Findings

Exploratory Research Findings and Implications for Problem Solutions

Executive Summary

Note: This Executive Summary was prepared in March 1994 by Gwen A. Kurz and Louis E. Moore of the Orange County Probation Department to facilitate the distribution of the results of the Probation Department's studies on chronic juvenile offender recidivism to criminal justice professionals and other interested parties. Notations were added by Ms. Kurz in March 1999 to reflect changes resulting from the continuing research effort.

In the 1980s, Orange County, California, experienced a rise in juvenile crime along with a rise in population, increased urbanization, and other changes. Yet the resources available to county and city governments did not keep pace, and sometimes shrank.

As a result, the Orange County Probation Department began focusing its efforts on the most serious offenders, with little left to devote to early intervention. But given limited resources, was this the best approach?

As part of Strategic Planning efforts for the 1990s, the department's management directed its in-house research staff to undertake studies to answer the question, "How well is the Probation Department doing with the youthful offenders of today?" As a by-product of these studies, a group of chronic juvenile re-offenders was identified as the "8% problem."

Forlorn in Juvenile Hall

This report summarizes the significant findings of those exploratory studies, conducted in Orange County between 1987 and 1993.

The Probation Department research staff had previously conducted research on risk factors with juvenile offenders and was knowledgeable of other studies in the field. This experience and knowledge provided direction to the formal study effort, which ultimately involved three phases of exploration.

In Study Phase I, two sets of data were examined, each comprising more than 3,000 juvenile offenders who entered Orange County's juvenile justice system for the first time during the first six months of 1985 and 1987, respectively. Each cohort of minors was tracked for three years to determine the overall volume of offenses committed and to examine differences between those minors who commit just one offense versus those who become low rate or chronic re-offenders.

During Study Phase II, a sample of the 1987 study cohort was drawn to further examine the differences between three subgroups:

Non-recidivists: Those minors with one referral to the Probation Department for a criminal offense during the three-year study period.

Low-rate recidivists: Those minors with two or three criminal justice referrals during the three-year study period.

Chronic recidivists: Those minors with four or more referrals during the three-year study period.

For the 1987 study subsamples, researchers gathered additional profile data and extended the tracking period for subsequent offenses to a total of six years.

Study Phase II resulted in a recommended target population for the development and testing of early intervention strategies to reduce chronic juvenile recidivism in Orange County. It also provided data indicative of the costs associated with the "8% problem" group.

In Study Phase III, the specific factors which were found to best predict chronic juvenile offending during Study Phase II were tested with a second, much larger data set. This led to specific recommendations for a pilot intervention project and follow-on study effort.

Overall, these study results are hopeful, concluding that through improved information-sharing and risk assessment techniques, a larger proportion of high-risk minors can be turned around before they become part of the "8% problem." There is also ample evidence that even a small reduction in Orange County's rate of chronic juvenile recidivism can pay major dividends to individual families and the safety of our communities for years to come.

The following provides a summary of the major findings of each study phase and the study conclusions. Also included is a brief description of the "8% problem" solution - the pilot intervention project that is being designed at the Orange County Probation Department.

Study Phase I: "8% Problem" Identification

The tracking of two cohorts of more than 3,000 first-time juvenile offenders revealed that, in the vast majority of cases, the juvenile justice system in Orange County was successful in deterring repeat offenses. Some aspect of each minor's contact with police, probation or the courts apparently had a positive influence on their lives.

On the other end of the spectrum, a small, troublesome group of frequent re-offenders was identified.

More specifically, the study showed that:

  1. At least two-thirds of the minors in both studies (71% in the second study) did not have a new probation referral during the initial three-year study period. Referrals to the Probation Department consist of an application for a petition to be filed in Juvenile Court, alleging a criminal offense.
  2. Some offenders (21% in the second study) went on to commit one or two additional offenses during the study period.
  3. A small percentage of minors (10% in the first study and 8% in the second) committed at least three additional offenses during the study period. These youths accounted for more than half of the repeat offenses committed by each study group.

After the second recidivism analysis was completed on the 1987 cohort, the group of minors with four or more applications for petition during the three-year tracking period began being called the "8% problem" (see Table 1).

Study Phases II and III: "8% Problem" Definition

The next two study phases were aimed at better understanding the characteristics and profile of the "8%" repeat offenders and the costs associated with their handling.

The study effort focused exclusively on the 1987 study group, reexamining the full cohort of 3,164 minors and more in-depth analysis of representative subsamples.

A major conclusion from Study Phase II was that a highly significant proportion of the chronic juvenile offenders in Orange County could be accurately identified and targeted for early intervention at the time of their first-ever system referral. This was done by combining an Age factor (15 or younger at the initial contact) with the presence of a Multi-Problem factor (see Table 2).

Below are the key findings from Study Phases II and III:

  1. A majority of the chronic recidivist (8%) group was age 15 or younger at the time of their initial case disposition (57% compared with only 23% and 31% of the non- and low-rate recidivist groups, respectively).
  2. Nearly half of the minors who became recidivists were made wards of the Court at their initial system referral versus only 22% of non-recidivists.
  3. The chronic recidivism rate for first-time wards age 15 or less (32%) was four times as great as that of first-time wards age 16 or older (8%). These findings did not vary based on gender, ethnicity, or referral offense.
  4. The chronic recidivist group was found to have significantly more problem areas in their lives, such as drug abuse, dysfunctional families, or failure in school, based on an initial evaluation of six problem variables. These chronic juvenile offenders averaged 3.25 problems each, compared to 1.74 for the low-rate recidivist group and 1.06 for the majority of youths who committed only a single offense. These problem areas were later refined and grouped into the four composite problem factors listed in Table 2.
  5. Utilizing the 1987 sub-samples, minors age 15 or less and minors declared wards of the Juvenile Court after their initial offense were also found to have a higher average number of problem factors (see Table 2) than those who were age 16 or older or whose initial cases were dismissed or handled with informal probation. >
  6. Based on a six-year follow-up of the1987study sub-samples, chronic juvenile offenders averaged nearly 20 months of incarceration, costing Orange County taxpayers $44,000 apiece in custody costs alone. Because at least 500 new "8% problem" cases are added to Orange County's criminal justice system annually, each new group could potentially cost taxpayers $22 million to incarcerate.

During Study Phases II and Ill, the researchers also conducted a variety of tests to see how well various factors worked as predictors of youths who would become serious, chronic juvenile offenders.

The previously referenced Multi-Problem profile and Age factors were tested as predictors of chronic recidivism with the study sub-samples from the 1987 cohort. In 70% of the cases, these factors accurately predicted whether a youth would become a chronic juvenile offender. (This test produced 19% false positives and 11% false negatives.) With youths ages 15 and younger, the degree of accuracy rose to 77%, and with older minors it fell to 64%.

In Study Phase Ill, a similar test was conducted with 905 first-time wards of the court -- the recommended target population and the more serious of the first-time offenders. In 66% of the cases, the recommended factors correctly identified youths as chronic, low rate or non-recidivists. (This test produced 28% false positives and only 6% false negatives.) By correcting problems with variable definitions for the first-time ward data set, the number of false positives can be significantly reduced.

Based on the study results, the authors recommend targeting younger minors with multiple problem profiles as defined in Table 2 for the design of new program strategies aimed at reducing chronic recidivism. Two notes of caution should be considered:

  • There is as yet no proof that the recommended strategies (see Study Conclusions) will work better than those currently employed. Therefore, a pilot program is recommended, with a formal program evaluation component.
  • The initial target population should consist of young, first-time wards of the Court. The Probation Department already has a mandate to take appropriate action to prevent further criminal activity with this population.

Table 1: Orange County Juvenile Justice System Recidivism Analyses

1985 Cohort Study Results

No. of Referrals per minor during 3-year Tracking Period No. of Minors In each Category Percent of Total Minors No. of Referrals For Each Category No. of Subsequent Referrals Percent of Subsequent Referrals
1 2,190 66% 2,190 0 0%
2 541 16% 1,082 541 22%
3 248 8% 744 496 20%
4-14 325 10% 1,771 1,446 58%
Total 3,304 100% 5,787 2,483 100%

1987 Cohort Study Results

No. of Referrals per minor during 3-year Tracking Period No. of Minors In each Category Percent of Total Minors No. of Referrals For Each Category No. of Subsequent Referrals Percent of Subsequent Referrals
1 2,234 71% 2,234 0 0%
2 472 15% 944 472 24%
3 205 6% 615 410 21%
4-14 253 8% 1,339 1,086 55%
Total 3,164 100% 5,132 1,968 100%

Table 2: Recommended Composite Problem Factors

As defined in the 8% study, the "multi-problem factor" constitutes two or more of the problem factors listed below. For each factor, a "yes" on any one sub-measure constitutes a "problem" in that area. 1999 Note: For the first-time ward population addressed by the 8% Solution, three or more problem factors are required.

1. School Behavior/Performance Factor

This problem factor consists of three individual measures:

  • Attendance Problems (Truancy or a pattern of "skipping" school in certain classes or at certain times of day).
  • Behavior Problems (Recent suspensions or expulsion).
  • Poor Grades (Failing one or more classes). 1999 Note: Factor now defined as failing two or more classes.

2. Family Problem Factor

Four individual measures were aggregated to create this factor, each addressing a different dimension.

  • Poor Parental Supervision and Control (Parents do not know where the minor goes, what he or she does, or with whom, and have little or no influence in such matters.)
  • Significant Family Problems (Illness, substance abuse, recent trauma, major financial problems, marital/family discord or other significant stressors.) 1999 Note: This factor used at Intake only to indicate need for better assessment of family needs or problems subsequent to Intake or court disposition.
  • Criminal Family Members Exerting a Negative Influence on the Minor
  • Documented Child Abuse or Neglect (Dependent child status or recent petitions filed on the minor's behalf) 1999 Note: This factor now includes family violence.

3. Substance Abuse Factor

This includes the use of alcohol or drugs by minors in any way but experimentation.

4. Delinquency Factor

Three measures were included. Each appears associated with a somewhat different criminal pathway, in terms of early onset.

  • A Stealing Pattern of Behavior
  • A Runaway Pattern of Behavior
  • Gang Member or Associate

Study Conclusions

Based on the findings of the entire exploratory research effort, the authors have concluded that:

  1. The number of chronic juvenile recidivists in Orange County can be reduced through a coordinated program of aggressive early intervention and treatment of young, high-risk juvenile offenders and their families.
  2. A significant proportion of chronic juvenile offenders can be accurately targeted for early intervention the very first time they are referred for juvenile justice system handling. The problems in their lives (from Table 2) are evident before they are influenced by the juvenile justice system or involved in further crimes.
  3. Significant risk factors are often overlooked at key points in the processing of youth through Orange County's juvenile justice system due to a lack of critical information. Information-sharing among youth-serving agencies and improved risk assessment techniques hold significant potential for increasing overall system effectiveness.
  4. Cooperative, concerted efforts to empower and build the families of high-risk youth can pay major dividends for years to come. More than half of the families of high-risk youth studied for this report had significant problems impeding their ability to provide adequate supervision, structure, or support to their children.
  5. Even a modest reduction in recidivism rates for the "8% problem" group identified in this study effort could result in major, long-term savings for Orange County's criminal justice system.

Toward the Development of "8% Problem" Solutions

In the spring of 1993, the Orange County Probation Department was awarded a National Institute of Corrections (NIC) Program Development grant which provides technical assistance from NIC and Temple University staff to design an "8%" intervention program. For the past seven months, a multi-agency group has been meeting to plan the pilot project.

The recommended case identification procedures and assessment tools are currently undergoing field tests. The formal pilot project and research component are expected to be implemented in July 1994.

Key program components will include:

  1. Providing adequate levels of supervision, structure, and support to minors and their families throughout the intervention process.
  2. Promoting accountability by the minors for their actions and developing increased sensitivity to the impact of their actions on others.
  3. Developing strategies that produce educational success, in part by assisting families to ensure that their minors attend school regularly.
  4. Promoting pro-social values, behavior and relationships.
  5. Developing individualized intervention strategies that are close to home and have strong follow-up beyond the "crisis" stage.
  6. Strongly promoting teamwork among the family, professional staff, and community volunteers.

1999 Note: Due to Orange County's declaration of bankruptcy in December 1994, the field tests were continued into 1996. Based on the field test results, the validity of the theoretical model was confirmed. However, a number of process issues and critical program resources were identified as important for sustaining short-term positive program outcomes, i.e., for the first 6-12 months of wardship, in the longer term, e.g. to curtail serious, chronic juvenile offending and prevent the development of adult criminal careers.

Through a combination of local, state and federal funds, the proposed 8% Problem Solution was implemented with the desired formal experimental research component in June 1997. This demonstration program/research project will continue through June 2001.