Increases in alcohol and marijuana use during the transition out of high school into emerging adulthood: the effects of leaving home, going to college, and high school protective factors *.

THE TRANSITION OUT OF HIGH SCHOOL is one of the most critical passages in life because of pronounced changes in social environment and role responsibilities (Newcomb and Bentler, 1987). Increases in substance use are evident during this transitional period (Arnett, 2005; Bachman et al., 1997). Arnett (2005) has argued that these increases occur because this stage of the life cycle, which he has termed "emerging adulthood," provides more freedom and less social control than the high school years. Several researchers have attributed these increases, especially in heavy drinking, specifically to the college experience (Goldman et al., 2002). White and colleagues (2005), however, found that increases in cigarette smoking, marijuana use, alcohol intoxication, and alcohol problems occurred as individuals left high school, regardless of whether they attended college (see also Bingham et al., 2005). Bachman and colleagues (1997) suggested that it was living situation (with friends and roommates vs parents or spouses) that accounted for increased substance use during this transitional period, rather than the college experience per se. As these increases are not universal, however, it is important to identify the individual and environmental factors that influence changes in substance use from late adolescence into emerging adulthood. The purpose of this article is twofold: (1) to examine the effects of leaving home and going away to college on changes in substance use approximately 6 months after leaving high school, and (2) to examine how protective factors measured in late adolescence predict change in substance use and moderate the effects of leaving home and going to college.

Transitions during emerging adulthood

Emerging adulthood is defined as the stage of the life cycle that begins following high school and ends with the adoption of adult roles (e.g., marriage, parenthood, and career); it spans roughly the ages 18-25 years (Arnett, 2000). This stage represents a time when most youths initiate new roles; many develop new friendship networks, and many separate from their families (Arnett, 2005; Schulenberg and Maggs, 2002). Arnett (2005) has argued that decreasing social control and increasing instability and stress contribute to increases in alcohol and drug use during emerging adulthood. In addition, the weakening of parental monitoring and increased importance of peer relationships can lead to increased substance use (Borsari and Carey, 2001).

Several prospective studies of college students have demonstrated that as they moved out of their parents' homes into dormitories or off-campus living situations, students' heavy drinking increased (e.g., Baer et al., 1995; Harford and Muthen, 2001). In general, these studies have focused on students at 4-year colleges. Studies of students at 2-year colleges indicate that they drink less than those at 4-year colleges (Presley et al., 2002; Sheffield et al., 2005). This research, however, has not accounted for differential living arrangements (i.e., more 2-year than 4-year college students remain living at home); neither has it examined changes in use from high school to college. Although increases in substance use have been most often identified for those young adults who go to college and move away from their parents, noncollege-bound youths who move out of their parents' homes experience many of these changes as well (Dawson et al., 2004; White et al., 2005). Bachman and colleagues (1997) found that college students (2- and 4-year combined), as compared with their noncollege peers, reported lower rates of heavy drinking while in high school; however, their use increased when they entered college, resulting in higher levels than those of their noncollege peers. College students also increased their marijuana use from high school to college, but they did not surpass their noncollege peers, who reported a higher prevalence both during high school and in emerging adulthood. Increases in alcohol and marijuana use for college students were attributed to living in dormitories or similar housing shared with other young adults (see also Crowley, 1991).

Using national data, Gfroerer and colleagues (1997) found that educational status and living situation were significant predictors of cigarette, cocaine, marijuana, and alcohol use, as well as heavy drinking. Rates of past-month marijuana use were highest among high school dropouts who lived with their parents and college students not living with parents; current and heavy drinking were highest among college students who did not live with their parents. College students who lived with their parents reported the lowest rates of marijuana use and heavy drinking. These studies highlight the importance of considering both changes in living situation (i.e., moving away from home) and school status (i.e., going to college) as potential risk factors for increases in substance use during the transition out of high school into emerging adulthood. Nevertheless, not all emerging adults increase their substance use as they move away from home or enter college. Therefore, it is important to delineate those protective factors in high school that moderate the transition to higher levels of substance use after high school.

Potential high school protective factors

Protective factors have been defined differently across studies (see Stouthamer-Loeber et al., 2002, for a detailed discussion). Some researchers have referred to variables as if they are either uniquely protective or uniquely related to risk (e.g., Rae-Grant et al., 1989). Others consider risk and protection as a continuum, with scores on one end of a variable indicating risk and scores on the other end indicating protection (Stouthamer-Loeber et al., 2004). Still others have conceptualized protective factors as processes that play a special role in the presence of risk (Hawkins et al., 1992; Rutter, 1990), reflecting interaction effects; in this case, the effect of a protective factor is greater when risk is high than when risk is low. We follow the latter two approaches and examine main effects, as well as whether protective factors moderate the effects of two potential risk factors: going to college and moving away from home.