We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article


Elliott DS, Huizinga D, Morse B. J. Interpers. Violence 1986; 1(4): 472-514.


(Copyright © 1986, SAGE Publishing)






VioLit summary:

The intent of this article by Elliot et al. was to analyze the patterns and frequency rates among juvenile violent offenders.

The authors employed a quasi-experimental design by administering a longitudinal survey, the National Youth Survey (NYS), to 1,725 juveniles in the United States who were selected through a multistage, cluster sampling procedure. The juveniles, aged 11-17 during the first wave, were asked to report forms of delinquency including alcohol use, drug use, and various forms of violence. The survey was first administered in 1976, with five subsequent waves as of 1983. Data in this article was limited to the first 5 waves (1976-1980). Within these first 5 waves the attrition rate was 13%. Respondents were interviewed between January and March of each year and asked to report their behavior from the preceding year. Information was gathered in confidential, face-to-face interviews where the juveniles were asked to report various problem behaviors. The Self Reported Delinquency (SRD) scale included 47 offenses which were reflective of almost all offenses categorized by the Uniform Crime Reports (UCR); nearly all items in the NYS were violations of criminal statutes. Three sets of delinquency scales were established (level of offense homogeneity, seriousness and relative rate of occurrence), 7 offense-specific scales were used to analyze the data (felony assault, minor assault, robbery, felony theft, minor theft, vandalism and illicit drug use), 1 offense-category scale was identified (public disorder) and 2 summary scales were identified (index offenses and general delinquency). Individuals who reported 3 or more of the offenses in the serious violent scale during a given year were considered seriously violent; those who reported 3 or more delinquent offenses including at least 3 violent offenses, but no more than 2 serious violent offenses were considered minor violent offenders; those who reported three or more index offenses but no more than 2 serious violent offenses were considered serious nonviolent offenders; those who reported 3 or more delinquent offenses, but no more than 2 index offenses, and no involvement in either minor or serious violent offenses were considered nonviolent minor offenders; finally, those reporting fewer than 3 delinquent offenses of any kind and no involvement in any index offenses were considered nonoffenders.

Findings were categorized according to prevalence rates (the proportion of serious violent offenders), survival rates (the proportion, at each age, who remained free of the serious violent offender classification), and hazard rates (those who became classified as violent offenders after a year or more of being in another classification). 1) Prevalence rates: approximately 5% of each age between 12 and 17 were classified as serious violent offenders with a dramatic decline between ages 17 and 21; male prevalence rates of serious violence during adolescence were higher than female prevalence rates (7%-8% compared with 2%-3%). 2) Survival rates: between ages 12 and 21 the survival rates decreased from 93% to 77%, which suggested that the major period for initial involvement was in the mid to late teens. 3) Hazard rates: the greatest difference between male and female hazard rates occurred at age 18, with an overall classification of high risk between ages 16 and 18.
Of the 7% who were classified as serious violent offenders in 1976, males outweighed females (3:1), blacks outweighed whites (3:1), lower class outweighed upper class (6:1), and urban outweighed rural (2:1). Serious violent offenders were more likely to believe that their peers and family viewed them as emotionally disturbed and were more likely to have come from broken homes. Neighborhood crime, environmental problems and social isolation were not differentiated among violent offenders, other offenders and nonoffenders; data from the 4 waves between 1977 and 1980 revealed similar trends. School, marital, and employment status were examined in wave 5 (1980) and these indicated that the serious violent offender was more likely to be a non-high school graduate, unmarried, and employed full-time. The most common serious violent offender appeared to be a white male, lower class, urban dweller, who lived with both biological parents and attended high school or college. There was little support for the common view that black offenders had higher rates of violent offenses than whites. Serious violent offenders committed an average of 8 violent offenses per year, in addition to other forms of delinquency and seemed heavily involved in all types of offenses, including drug sales and drug use, with an annual rate of 132 delinquent offenses (compared to 54 for the serious nonviolent offenders). Career continuity over the 5 years had a mean length of approximately 1 and a half years, although nearly 4% of the total had a career length of 5 years, with 10% of the serious violent offenders having a career length of 5 years. The length of careers did not vary significantly by gender, race, class, age or residence. The probability of continuing a violent career increased with time for both genders. Use of marijuana also appeared to be more closely associated with violent careers than with other categories of offenses, and the use tended to increase during their careers as violent offenders. This pattern, however, was not observed for alcohol or hard drug use, only marijuana. The two general patterns in the violent career were an escalation in the frequency of offending and a lack of specialization in violence among those classified as seriously violent, even as time progressed. Eighty four percent of the most serious and frequent offenders had no criminal record.

Because the analysis of the self-reported data differed from the data available through arrest reports, the authors argued that further research needs to be conducted on the difference between conviction rates and self-reported rates and on the difference between prevalence rates and individual offending rates.

(CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado)

KW - Violence Incidence and Prevalence
KW - Juvenile Offender
KW - Juvenile Violence
KW - Self Report Studies
KW - Offender Characteristics
KW - Life Course
KW - Statistical Data
KW - Crime Rates
KW - Violence Rates
KW - Early Adolescence
KW - Late Adolescence
KW - Child Offender
KW - Child Violence
KW - Late Childhood


All SafetyLit records are available for automatic download to Zotero & Mendeley