In our sample of adjudicated adolescents, we found initially weak to moderate bivariate relationships between future orientation and impulsive sensation seeking and risk variables in the direction that would be expected. Less positive future orientation and higher impulsive sensation seeking were related to higher levels of risk behavior and less perceived risk, which was directly in line with our first hypothesis. These relationships weakened considerably, however, when age and gender were controlled in the regression analyses. In the final analyses, future orientation was more strongly related to our outcome behaviors than was impulsive sensation seeking. Less positive future orientation was significantly related to having sex while using alcohol, more alcohol problems, higher perceived risk associated with deviant behaviors, more marijuana use, more hard-drug use, and more quantity and frequency of alcohol use. This lends additional empirical support to previous studies indicating a negative relationship between future orientation and risk behaviors. Joining with previous research on impulsivity, we found that higher impulsive sensation seeking was related to more alcohol problems, alcohol use, cigarette smoking, and less condom use. Much of the prior work on the relationship of impulsivity and future orientation to risk behavior has been conducted with low-risk populations (e.g., college students, non-adjudicated adolescents). Our results show that findings obtained with predominantly low-risk populations appear to generalize to young people who engage in substantially more risk behavior.
Though we thought it likely that these variables might interact, only one of the eight interactions tested approached significance (that for alcohol problems). We found, as predicted, that those adolescents lowest in future orientation and highest in impulsive sensation seeking had the highest levels of risk behavior (see ). It would be desirable to test these relationships again with a larger sample to replicate these findings and have more power to detect other interactive effects that might exist. In sum, the variables of age, gender, impulsive sensation seeking, and future orientation account for modest proportions of the variance in risk behavior. The strongest prediction occurred for alcohol use and alcohol problems, where 10% and 16% of the variance was accounted for, respectively.
As in previous studies (e.g., Bryan, Aiken, & West, 2002; Loeber, 1998), age was a significant predictor for four outcome variables (see ), exhibiting a positive relationship to sex with alcohol, alcohol problems, alcohol use, and cigarette smoking. This indicates that as age increases, participants are more likely to use alcohol during sexual intercourse, have more alcohol-related problems, use alcohol more frequently, and smoke cigarettes. To some degree, this is simply a pragmatic resultthose who have been alive longer have simply had more time to engage in problem behaviors. The important point this finding makes is that age is a crucial variable to take into consideration when conducting research on risk behavior with adolescents at different stages of development. Had we not controlled for age in these analyses, we would have concluded that there were much stronger relationships between our personality variables and risk variables. Gender was related to condom use such that males were actually behaving safer (i.e., they were using condoms more) than were females. This gender difference is consistent with previous research among adjudicated involved adolescents (e.g., Bryan et al., 2002) and may reflect either the use of other birth control by women or a lack of control over condom-use behavior with their partners. The only other gender difference was for hard-drug use, where females were more likely to report hard-drug use than were males.
One limitation of this study is that the probation department requires adolescents to undergo drug testing via urinalysis on a regular basis while they are involved with the criminal justice system. Rates of recent drug and alcohol use reported by participants are probably lower than when they are not on probation. But the argument could also be made that the rates of alcohol use, marijuana use, and hard-drug use were surprisingly high given the consequences of testing positive for these substances. To fail a regular screening is considered violation of probation. A second limitation of the study was the location of data collection. Having participants fill out questionnaires at probation offices and self-report illegal behaviors may have influenced how they answered questions via social desirability. Finally, the data are cross-sectional, so we are unable to draw any conclusions about causality with regard to these data. Though it is reasonable to conclude that dispositions involving impulsivity and future orientation are likely to be fairly stable, and to exert influence on levels of both perceptions of risk and on actual risk behavior, our design does not provide any certainty in this regard.
Our results indicate small but reliable relationships between future orientation and risk behaviors. We also show some relationships of impulsive sensation seeking to alcohol use and problems, cigarette smoking, and condom use. There is also some evidence to suggest an interactive effect such that higher levels of impulsivity combined with more negative future orientation results in higher levels of risk behavior. As impulsivity is conceptualized as a relatively immutable personality factor, it appears that it might be useful to target future orientationwhich may be more malleablein interventions with young people. An alternative strategy is to gear intervention content toward present-oriented consequences. For particularly high-risk, present-oriented youth, it may make more of an impact to discuss the immediate consequences of behavior (e.g., girls will not want to kiss you if you smoke) rather than future consequences (e.g., smoking causes lung cancer). Present-oriented messages are becoming the norm in media-based interventions targeted to young people (e.g., Worden, Flynn, Solomon, & Secker-Walker, 1996
), and this focus may be even more important for impulsive and sensation-seeking high-risk youth.