The Centers for Disease Control (CDC), in the National College Health Risk Behavior Survey, indicated that college students engage frequently in risky behaviors (Centers for Disease Control and Prevention, 1997). According to the CDC's survey, almost one third of college students were current cigarette smokers. In addition, one third of college students reported episodic heavy drinking of alcohol, and 27.4% reported drinking and driving during the 30 days prior to completing the survey. Students aged 18 to 24 years were more likely than students older than 25 to report riding with a driver who had been drinking. Regarding sexual risk-taking, only 29.6% of college students had used a condom at last sexual intercourse during the three months preceding the survey. With so many college students taking risks involving so many negative consequences, it would be beneficial to have one general model or framework from which to understand college student risk-taking behavior.
One theory indicates that individual traits, such as self-esteem (Kaplan, 1980), social skills (Botvin, 1986), impulse control (McCord, 1990), sensation-seeking (Kohler, 1996), and locus of control (Crisp & Barber, 1995), may explain risk-taking. Caspi and colleagues (1997) used a multidimensional trait model of personality in a longitudinal study to examine the characteristics of youth who participated in health-risk behaviors. They found that those who scored lower at age 18 on the traits of traditionalism, harm avoidance, social closeness, and control, and higher on aggression, were more likely to engage in a health-risk behavior at age 21.
Problem-behavior theory (Jessor & Jessor, 1977) has been offered as a way to explain risk-taking in adolescents and young adults. Problembehavior theory leeks at risk-taking from a developmental and personenvironment interaction perspective. Jessor and Jessor's framework entails the relationship between three components--personality, perceived environment, and behavior--each composed of many social-psychological variables. For example, the personality component consists of variables such as alienation, self-esteem, internal-external locus of control, moral attitude, and importance placed on achievement; the perceived environment component consists of variables such as parental controls, friend controls, and parents' vs. friends' influence; and the behavior component consists of variables such as problem drinking, marijuana use, cigarette smoking, and general deviant behavior (Jessor, Donovan, & Costa, 1991). In a longitudinal study, Jessor, Donovan, and Costa found that measures of problem-behavior proneness in adolescence were significantly related to young adult problem behavior. Personality and perceived environment variables consistently predicted problem behaviors such as drinking, drug use, and cigarette smoking in young adulthood. The findings were similar for the behavior component.
College student risk-taking could also be explained from the decision-making perspective. Furby and Beyth-Marom (1992) have proposed that adolescents may not be capable of competent decision-making. Adolescents may not sufficiently consider the possible consequences of their actions, and they may have a perception of invulnerability to consequences. Additionally, adolescents may consider different options than adults, as well as identify different consequences, place a different value on consequences, and assess the likelihood of consequences differently. Furthermore, adolescents may rely on peers (e.g., for information) when considering risky behavior.
The majority of studies on adolescent risk-taking have looked at those under 18 years old, yet there are numerous opportunities for older adolescents to participate in risky behavior, especially college students (Furby & Beyth-Marom, 1992). Further, most studies have addressed older adolescent risk-taking from only one theoretical perspective (e. …