Examining Recidivism of Chronic Young Offenders Into the Adult Criminal Justice System in North Carolina

This article was contributed by David E. Jones, Director, North Carolina Criminal Justice Analysis Center and Kitty Brown Herrin, PhD, Statistician, North Carolina Criminal Justice Analysis Center.

Some youth in juvenile correctional facilities are released only to be arrested later as adults; others are not. Are there differences in the youth who go on to further criminal activity from those who do not? In February 1988, staff from the North Carolina Criminal Justice Analysis Center and the Division of Youth Services began work on a cooperative effort to determine which youth in North Carolina juvenile correctional facilities were more likely to become adult offenders. This information will prove helpful in providing intervention at an earlier age to deter youth from further criminal activity.

Description of the Data.
The Department of Human Resources, Division of Youth Services (DYS), operates five residential schools (called training schools) for males and females ages 10 through 17. Youth who (1) are under the age of 16, (2) have been found guilty in the Juvenile Courts of an offense for which an adult can serve time, and (3) have no other alternative resources, are eligible for commitment to training school and can remain there until their 18th birthday after which they are no longer under the jurisdiction of the Juvenile Court or the DYS. (Youth committed to training school are referred to in this study as students.)

DYS provided the Criminal Justice Analysis Center with computerized commitment records for the years 1985 to 1987. The records were then matched with the computerized arrest/ disposition records from the Division of Criminal Information (DCI) of the State Bureau of Investigation. The DCI records covered the time period of January 1985 to June 1988. The strategy was to match individuals appearing in both data sets and develop a profile of the offenders who continued their criminal careers into the adult system. The analyses were restricted to the 1985 training school release cohort to provide a three year follow-up period. Matching the data proved problematic because the Youth Services MIS database did not contain names. Instead it used a "MISID identifier" which was composed of elements of the offender's first and last names, and birthdate. The MISID identifiers, along with sex and race, were used as the match criteria. Of the 2700 MISID identifiers on the 1985-1987 DYS tape, there were 950 matches or "hits" with DCI records.

Of the 950 "hits" returned to DYS for manual verification, only 575 (61%) were verified. Since this MISID was assigned at intake, it was possible for an alias to be used or birthdate to be reported incorrectly (e.g., a youth reported being older or younger than s/he actually was). If this occurred, the assigned MIS identifier would not match the CCH records (resulting in a "false negative", i.e., there would be no match between the DYS and the CCH databases). Thus, the number of youth in the training school database may, in fact, be higher than indicated.

One solution to solve the problem of "false negatives" was provided by the matching procedure used by DCI. This procedure also uses elements of the offender's last name and birthdate to match with the Youth Services MISID but uses no element from the first name. The DCI matching procedure resulted in additional records ("false positives") resulted. If students originally were coded by middle name, a nickname, or alias instead of first name, but the last name and the birthdate were correct, the records still matched and, because DYS manually verified matches, "false positive" records were eliminated.

During 1985, 578 juveniles were released from training schools. The number of juveniles from this 1985 release cohort that matched with the adult arrests in the DCI database was 195. Therefore, over one-third of all juveniles released from training school during 1985 were later arrested as adults for a felony or serious misdemeanor. Of this release cohort, 81% were male, 46% were white, and the average age at first commitment was 14 years 6 months. Twenty percent of these juveniles had more than one training school commitment. Seventy-eight percent of all offenses were larceny, trespassing, or assault.

Gender, Age, Race, and Prior Commitments.
Almost three times as many males as females entered the adult criminal justice system and nonwhites were about 30% more likely to be arrested than whites. Figure 1 depicts relative adult arrests by age and race. Nonwhites have higher arrest rates than whites in both age groups. Race appears to be a stronger influence than age as can be seen by the fact that the arrest rate for the 10-14 age group of nonwhites is higher than the 15-17 white rate.

Figure 2 shows that the more prior commitments to training school the more likely a youth was to be rearrested as an adult. This was the most significant finding of this study and confirms other research on this subject.

Most Recent Offenses.
Looking at the last and most serious offenses of these youth, the breakdown by offense was: 8% violent offenses, 66% property offenses, and 26% public order and drug offenses. Sixty-five percent of the violent offenses were assault; 90 percent of the property offenses were larceny; and 56 percent of the public order offenses were trespassing. In all, 78% of these offenses were either larceny, trespassing, or assault. As seen in Figure 3, juveniles in training school for violent offenses are least likely to be arrested later as adults and those arrested for property offenses most likely.

Age and Offense.
Younger training school juveniles (30.5%) were more likely to have committed public order offenses than older juveniles (22.2%) while older juveniles (71.1%) committed more property offenses than younger juveniles (61.9%). Of those younger juveniles arrested for trespassing, only 15.6% (compared to 39.3% of those 15 and over) recidivated into the adult system. Similarly, for assault, only 13.3% of those 14 and under compared to 30% of those over 15 were rearrested as adults.

Race and Offense.
Nonwhite youth (65.4%) were more likely than whites (55.6%) to be committed for a larceny offense. Whites were more likely (14.6%) than nonwhites (10.1%) to be committed for trespassing. Nonwhites were more likely than whites to be arrested as adults regardless of their juvenile offense. Nonwhites who had committed property and public order/drug offenses were those most likely to be arrested as adults. Those who had committed violent offenses were least likely to be arrested as adults.

Length of Stay.
For 1985 training school admissions, mean (average) length of stay for the most recent commitment was 277 days. For females, the average length of stay was longer than for males (289 versus 274 days). Average length of stay for nonwhites (294 days) was longer than for whites (261 days). Age was the greatest determinant of length of stay: younger students stayed longer than older students. Youth with a relatively long period of stay in training school had higher rates of arrests as adults.

Summary and Implications.
This study represents the first time in North Carolina that data from the Juvenile Justice System has been merged with adult criminal justice system data to analyze the differences between those juvenile offenders that continue their criminal careers as adult offenders and those who do not. The intent of this effort to describe the long-term, chronic offender is to provide a better understanding of those characteristics that might be addressed through prevention and early intervention programs. Identifying and targeting high-risk individuals at an early age could help in the future with diverting chronic offenders.

As mentioned in the description of the data, there are some limitations -- the major difficulty involving matching different automated records, and the likelihood that the proportions of students recidivating into the adult system are larger than indicated in this study. Despite these complications, most of the study findings were as would be predicted either from knowledge of adult recidivism patterns or from other studies of juvenile offenders. Findings not consistent with those of other studies were attributable to problems with the length of time for tracking students into the adult system.

The major question the Analysis Center attempted to answer with this study was: Which youth are most at risk to recidivate into the adult system? Who should be the top priority when endeavoring to divert youth from the adult system? The answer appears to be: males, nonwhites, students with multiple commitments (the more commitments, the greater the likelihood of entering the adult system), property offenders, and those whose length of stay for their last commitment was longer than 13 months. Examining age and race together, younger nonwhites were more likely to recidivate than whites of any age.

Older students who committed public order/drug or property offenses were more likely arrested as adults while violent offenders of any age were least likely to be arrested as adults. Nonwhite students who committed public order/drug or property offenses were more likely to be arrested as adults than white students; however, violent offenders, regardless of race were least likely to recidivate. Thus, while violent juveniles may need counseling or related services, the property and public order/drug offenders are the students that should be considered when the concern is with preventing juveniles from recidivating into the adult system.

All youth, regardless of sex, age, or race, were more likely to be committed for larceny than for any other offense. Those youth who were in training school for offenses such as drug offenses, DWI and other traffic offenses, or being drunk and disorderly, were most likely to be arrested as adults. Those in training school for trespassing or other public order offenses and violent offenses were least likely to be arrested as adults.

One policy question related to this analysis is: Should chronic offenders in the juvenile justice system be protected by confidentiality if they become offenders in the adult criminal justice system. This is indeed a difficult issue to deal with but, with the limited data analyzed in this study, it would appear that chronic juvenile offenders (especially property offenders) are quite likely to continue into the adult system. Is it fair that their juvenile records are sealed and they are treated as first-time offenders in the adult system?

The Analysis Center anticipates replicating this juvenile/adult chronic offenders study. Due to changes in the DYS computerized database to include the full name of students, it seems safe to presume that more accurate projections can be made regarding chronic offenders recidivating into the adult system.