Predicting Staff Assault in Juvenile Correctional Facilities

ABSTRACT

This study examines the predictive utility of the community-based Positive Achievement Change Tool–Prescreen (PACT-PS) for staff assault in a sample of 787 state-committed male youth. Bivariate and multivariate analyses indicated the PACT-PS failed to predict staff assault across racial/ethnic groups. Notably, this study also found youth with serious delinquent histories and prior commitments improved the PACT-PS’s ability to predict staff assault. Limitations of this study, suggestions for future research, and practical implications are discussed.

KEYWORDS:

• Incarcerated youth
• institutional misconduct
• institutional violence
• risk assessment

Notes

1. Institutional danger is a composite dichotomous outcome measure indicating whether the youthful offenders had assaulted staff, other wards, or were found in possession of a weapon while incarcerated among state-committed youth.

2. The demographic variables were race and sex. The delinquent history variables included age at first formal referral to the juvenile justice system, age at state commitment, age at release from incarceration, length of incarceration (in days), the number of felony adjudications prior to state commitment, a weighted risk-classification score, whether the youth was on probation at state commitment, the degree of the youth’s commitment offense, whether the delinquent was a known gang member at the time of commitment, and whether the youth was previously violent toward his or her family. The risk factor variables included gang affiliation of youth’s family members; number of previous out-of-home placements; the youth’s highest grade completed; presence of divorce in the youth’s family; whether the youth lived in poverty; whether the youth was confirmed to be physically abused, sexually abused, emotionally abused, abandoned by his or her family, or physically neglected; whether the youth has specialized treatment needs (emotional disturbance, sex offender, and capital offender); whether the youth received specialized treatment for these needs while incarcerated; and whether the youth was mentally challenged, mentally ill, or demonstrated suicidal tendencies.

3. Although composite measures enhance the interpretability, statistical power, and identification of general assault dimensions, they make it difficult for administrators to conceptualize problems associated with individual measures of assault (Dozois, Dobson, Wong, Hughes, and Long, 1996).

4. The demographic variable was race. The delinquent history variables included the total number of previous delinquent adjudications, total out-of-home residential placements prior to the most recent incarceration, and gang affiliated at state commitment. The commitment offense variables included length of time incarcerated, homicide commitments, serious assaultive commitments, sexual-related commitments, and other commitments. The social history variables included mental health needs, substance dependency, sexual abuse prior to commitment, history of abuse or neglect, and perpetrated violence against a family member prior to commitment.

5. The demographic variable was race. The delinquent history variables included the age at YCS commitment, total number of previous delinquent adjudications, total out-of-home residential placements prior to the most recent commitment, homicide commitment, serious person/property commitment, sexual-related commitments, other commitments, gang affiliated at state commitment, and length of time incarcerated. The social history variables included substance abuse, suicidal, mentally challenged, mentally ill, physical abuse, sexual abuse, emotional abuse, poverty, chaotic home environment, family gang affiliation, and perpetrated violence against a family member prior to commitment.

6. The family background characteristics included poverty, family criminality, family members that were in gangs, violent toward his family member, living in a chaotic home, history of physical abuse, history of sexual abuse, history of emotional abuse. The delinquent career characteristics included substance abuse history, gang member, age at first commitment, number of previous out-of-home placements, and felony adjudications prior to commitment.

7. The phrase record of referrals is not a commonly used indicator of criminal history in validation research. The developers of the PACT use record of referrals because these records provide a more complete description of a youth’s misbehavior than official data from juvenile court records. For example, youth who receive diversion and deferred adjudications or dispositions avoid the official designation “delinquent,” because their cases are dismissed before the petition is filed. Barnoski (2004) points out that “a youth’s criminal history is an indicator of the duration and established persistence of the youth’s criminal behavior. Referrals that resulted in a conviction, diversion, deferred adjudication, or deferred disposition (regardless of whether successfully completed), rather than offenses, are the unit of criminal history being counted. This reflects the youth’s persistence to re-offend even after being processed through the juvenile justice system. For example, youth who have three separate referrals are more persistent in their behavior than youth with three offenses on a single conviction” (p. 27).