high-risk adolescent males worked, the more likely they were to engage in delinquent behavior. Hours of work had no direct effect on likelihood of delinquent involvement among females or low-risk males. The researchers analyzed the relationship between work intensity and delinquency using the 1988 National Survey of Families and Households, a cross-sectional, nationally representative sample of 13,079 individuals within 9,643 households. The respondents in this study were the parents of the adolescents. 3

Adolescents who work more than 20 hours per week have been found to be prone to using cigarettes, alcohol, and illegal drugs (marijuana, cocaine) (Bachman and Schulenberg, 1993; Greenberger and Steinberg, 1986; Mihalic and Elliott, 1997; Mortimer et al., 1996; Resnick et al., 1997; Schulenberg and Bachman, 1993; Steinberg and Dornbusch, 1991). Students in St. Paul, Minnesota, who worked more than 20 hours per week engaged in more alcohol use than their classmates each year during high school (Mortimer et al., 1996). In fact, the link between intensive work and substance use is one of the strongest findings in this area, manifest even when the data are subjected to extensive statistical controls for background variables and pre-existing differences in substance use between the groups. For example, Mortimer and colleagues (1996) found that hours worked during high school were associated with alcohol use, an association that held after statistical control for frequency of past alcohol use, sex, parental socioeconomic status, race, family composition, and nativity. Mihalic and Elliott (1997), in studying the short-term effects of work hours on the use of alcohol, marijuana, and other drugs among 11-to 17-year-olds from 1976 through 1980, statistically controlled for sex, age, parental socioeconomic status, place of residence (urban/suburban, rural), ethnicity, and prior drug and alcohol use (at first interview): They found that employment had an effect on marijuana and alcohol use above and beyond any preemployment differences.

Greenberger and Steinberg (1986) and Bachman and Schulenberg


In this study (Wright et al., 1997), the researchers developed a measure that categorized youths by risk for delinquency, based on research findings in criminology. Risk factors assessed in this study were parental criminality, parental role rejection, family mobility, household size, self-control, parent-child conflict, family income, a nonintact marriage, and low school commitment. Each factor was dichotomized at the median to show presence or absence of risk. Youths with four or more risk factors present were considered high risk.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement