|
Abstract
|
VioLit summary:
OBJECTIVE:
This aim of this study by Osgood et al. was to compare U.S. national estimates of age trends and time trends in self-reported illegal behavior with corresponding figures for officially recorded arrests.
METHODOLOGY:
The authors conducted a quasi-experimental study. Data was obtained from the Monitoring the Future study. This study collected information from a nationally representative sample of U.S. high school seniors. The sample was selected yearly by a three-step national probability sample. The questionnaire was administered to approximately 130 high schools each year, and resulted in 15,000 to 19,000 respondents.
Specific to the authors' study, a random 20% of the questionnaires included the items analyzed for this article. The time trends analysis was based on high school seniors' responses (1976-1986). The age trends analysis was based on the follow-up data collected from a subsample of each senior class. Half the subsample completed the questionnaire on odd years, the other half on even years: ages 17, 18, 20, 22, and 17, 19, 21, 23. Serious drug users were purposely over sampled in the follow-up study. The authors had an 80% response rate for senior year subjects and about a 75% response rate for all follow-up years. The instrument asked participants to complete a self-report of frequencies (0, 1, 2, 3, 4, 5 or more) for different types of illegal activities. The authors looked at rate of offenses per individual and rates of individuals who engage in each act.
Arrest rates were estimated from calculations that looked at both age-specific arrest totals from the Uniform Crime Reports (1976-1986) and corresponding national population estimates from the Bureau of the Census' Current Population Reports (1982 and 1985). The authors used descriptive statistics, Chi-square tests and linear and quadratic regression equations to analyze the data.
FINDINGS/DISCUSSION:
The authors reported the following age trends findings. Overall, criminal behavior declined throughout the age span of 17 to 23. At age 17, 74.3% of males and 50.8% of females reported at least one offense; the percentages dropped to 29.7% and 24.3%, respectively, at age 23. Overall, arrest rates peaked at ages 16 to 17. However, arrest rates for assault peaked at age 21. Only two kinds of offenses showed meaningful time trends. Rates of assaults increased, and rates of theft decreased over time. Regression analysis revealed significant increases in rates of assault: the linear coefficient for time was significant, but the quadratic was not. Regression analysis revealed that the significant linear decrease in self-reported theft was totally attributed to shoplifting. Arson was the least frequent offense.
AUTHORS' RECOMMENDATIONS:
The authors recommended that policy-makers take age into account when assessing the need for different types of crime programs. The authors also suggested that the public would gain a more accurate picture of crime by knowledge of crime rate trends across several years rather than reports of differences from one year to the next.
EVALUATION:
In general, this study provides some important, albeit dated, findings on crime age and time trends. It should be noted that an important group of teenagers were excluded - those who drop out before their senior year. Also, self-reports present the problem of inaccuracy: the validity of respondents' answers was not assessed. In addition, the measure for frequency of criminal activities did not distinguish between those who committed a particular crime 5 times, and those who committed it many more times. The generalizability of this study is limited by the above short-comings. In conclusion, this study offers some insights into crime patterns from 1976 to 1986.
(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 - Senior High School Student
KW - Late Adolescence
KW - Early Adolescence
KW - Arrest Rates
KW - Age Factors
KW - Crime Rates
KW - Self Report Statistics vs. Arrest Statistics
KW - Self Report Studies
KW - 1970s
KW - 1980s
KW - Juvenile Crime
KW - Juvenile Offender
KW - Juvenile Violence
KW - Adult Crime
KW - Adult Offender
KW - Adult Violence
KW - Young Adult
KW - Violence Rates |