|
The youth in these urban samples experienced a substantial number of family transitions during adolescence. In Rochester, about two-thirds of the sample (64.5 percent) experienced at least one change in family structure over the 4-year period and about 45 percent experienced two or more transitions (see figure 1). Almost half of the Denver youth (49 percent) had one or more family changes and 29 percent had two or more (see figure 2). Family instability is less pronounced in Pittsburgh; about 30 percent of the boys experienced one or more family transitions (see figure 3).
The number of family transitions had a clear and statistically significant effect on the prevalence of delinquency and drug use for the Rochester youth (see figure 4). About two-thirds (64.1 percent) of those who experienced no changes in family structure reported delinquency; this rate increased steadily as the number of transitions increased, reaching a peak of 90 percent for youth who experienced five or more transitions. A stronger pattern was seen for drug use—about 28 percent of adolescents with no change in family structure reported using drugs, but that rate increased to almost 60 percent for those who experienced five or more transitions.
In Denver (see figure 5), the prevalence of delinquency increased significantly with an increase in family transitions, from 61 percent for youth with no transitions to a peak of 85 percent for youth with three or more transitions. About one-third (31 percent) of adolescents with no family transitions used drugs, and 58 percent of those with three or more transitions used drugs.
In Pittsburgh (see figure 6), the relationships between family transitions and both delinquency and drug use were not statistically significant, but the trend is the same as that observed in Denver and Rochester. While 64 percent of juveniles who experienced no transitions reported delinquency, 80 percent of those who experienced three or more transitions reported delinquency. About 27 percent of youth with no family transitions reported drug use, and more than one-third (37.5 percent) of those with three or more transitions used drugs.
A more refined analysis (results not shown) using the Rochester data (Smith, Rivera, and Thornberry, 1997) examined these issues in greater depth to ensure that the basic results reported in this Bulletin from data that compare only two variables—family transitions and delinquent behavior—are still accurate when other factors are taken into account. First, in addition to measuring the prevalence of delinquency and drug use (the data reported in figure 4), the frequency of involvement in delinquency and drug use was examined. Second, the number of family transitions that occurred near the beginning of the study was compared with delinquency and drug use later in the study. This ensures that the predictor variable—experiencing family transitions—actually occurs before the outcome variable—delinquency or drug use. Finally, the effects of gender, family poverty, family structure at the beginning of the study, parental supervision, and earlier delinquency or drug use were controlled when the relationships between family transitions and delinquency and drug use were reexamined. This helps ensure that the relationships reported earlier (see figure 4) were not due to these other variables. (See Smith, Rivera, and Thornberry, 1997, for these results.) In all of these comparisons, the results were the same as those reported here: a greater number of family transitions was significantly related to a higher rate of delinquency and drug use.
|
