Peers have long been recognized as a key influence on adolescents' choices with respect to substance use. Some studies that look at peer influence in relationship to family influence are reviewed here.

Barnes and Farrell (1992), studying parental support as predictors of adolescent drinking, also found support for peer orientation as a significant predictor of drinking behavior.

Bahr et al. (1995) examined social development and social control theories that suggest children with strong bonds to social groups with antidrug norms will be less likely to use drugs. They hypothesized that family bonds and family substance use are exogenous variables that influence choice of friends, whereas educational commitment and number of friends who use alcohol and other drugs are intervening variables with more direct influences on adolescent alcohol use. Studying a large sample of 7th- through 12th- graders in Utah schools, the authors found that the influence of family bonds on risk of alcohol use is primarily indirect, through educational commitment and choice of peers. The indirect effects of family bonding on substance use were moderate. Findings support an integrated social learning/social control model of adolescent substance use. Family bonding appears to be an important social control mechanism that could decrease the risk of adolescent alcohol use. Findings about relative strength of risk factors studied should be useful for prevention program planning.

Conger and Rueter (1996) showed that factors within the family can promote deviant peer relations. Adolescents can acquire alcohol-using friends through three distinct avenues. The behavior of other family members, including parents and siblings, is expected to increase directly the risk of choosing friends who drink. The adolescent's own drinking behavior plays a major role in determining choice of friends. Also, the role of siblings and their drinking behavior is important. The authors employed several measures in their analysis: parents' history of substance use problems, alcohol use of the target adolescent and his or her siblings, harsh and inconsistent parenting of the youth, and friends' alcohol use. Findings supported the hypothesis that frequent and problematic drinking by siblings would exacerbate the target adolescent's tendency to drink. Also, early drinking appeared to exacerbate other problems. The "flocking phenomenon," i.e., the tendency of teens who drink or use drugs to acquire friends with similar habits, was noted. Target adolescent drinking in 7th grade appears to generate more harsh and inconsistent behavior by parents. Parental history of substance abuse problems is indirectly directed to adolescent risk for later alcohol use, and a stronger effect for mothers' drinking versus fathers' drinking was noted.

The MORI Survey, carried out in 1996 by the Imperial Cancer Research Fund in Great Britain, showed that getting teens to quit smoking can be effectively done by their sweetheart, more so than by a television personality or pop star. "Love is really the key to getting youngsters to quit," says Professor Gordon McVie, director of the Cancer Research Campaign. Next on the list of influencers are parents. Celebrities influence teens in a more subtle way: by their behavior, not their preaching. Featuring celebrities who had successfully quit smoking would send a strong message to teens. Teens also report that they would be encouraged to quit if it became more difficult to buy cigarettes, or if smoking were banned in more public places, or if the price of cigarettes were raised.

Kafka and London (1991) explored the link between communication in relationships and adolescent substance use, specifically the openness in communication between youth and their parents and closest friends. Predictive value of perceived pressure from friends, friends' substance use, and parents' substance use was also analyzed. The investigators found that openness of communication is negatively associated with substance abuse in the case of parent/child communication, but not in the case of peer communication. Perceived pressure from friends was not correlated with substance use. In fact, high school students do not perceive much pressure from friends about their behavior. A possible interpretation is that overt peer pressure is less common than previously believed. Teens did not report pressure from friends to engage in negative behaviors. Perhaps the internal concerns of adolescents are at the root of their choices regarding substance use: wanting to be accepted, to belong, to be noticed.

Keefe (1994) noted that many prevention programs focus on teaching refusal skills, assuming that peers exert direct pressure on each other to influence behavior. As explained in the previously cited study, recent research has shown that adolescents rarely use explicit pressure, yet conformity to group norms does occur when one values being a member of that group. This study examined perceived social pressure among adolescents, and explored age difference in perceived social pressure. Parental and peer pressure were examined in a study of 386 7th-, 9th-, and 11th-graders. Students were asked to compare "costs" and "benefits" of using alcohol; significant differences by age were apparent, with older kids finding more benefits. Findings showed normative pressure against alcohol use from both peers and parents. The authors suggest that programs that focus on coping with negative peer pressure may not be necessary, but that peer norms and programs encouraging peer support for abstinence could be used as effective deterrents. Since older adolescents perceived more benefits to alcohol use, this suggests that programs focusing on negative consequences mightbe ineffective, and a better choice would be programs focusing on alternative ways to have a good time.

Ried (1989) investigated differential social control theory, which proposes that delinquent behavior becomes more probable as an adolescent's bond to society weakens. The bond, which includes attachment to parents and close supervision, communication, and affectionate ties with parents, is inversely related to delinquency. Another aspect of the bond is attachment to peers; delinquents do not form attachments to conventional peers nor to one another. The third aspect of the bond relates to school: School is not seen as by delinquent adolescents as a source of conventional norms and influences. Ried discusses differential association-reinforcement theory: the adolescent symbolically interacts with many primary groups (parents, peers, school). These groups have verbalized normative expectations for adolescents' behavior. The adolescents are rewarded for the desired response, or punished for failing to make the response, depending on the group's expectations.

This study of 5th- through 8th-grade students showed that none of the above three attachment variables had significant, direct effects on drug use. Instead, the drug use of this sample of students was shown to be directly affected by 1) their perception of whether their friends think they should use drugs or not; 2) their peers' own drug use; and 3) their attitude towards drugs.

Schneider and Perney (1993), surveying adolescents' major concerns and perceived resources, collected information on 42 concerns (e.g., divorce, war, grades, salary, peer drug and alcohol use , family finances, appearance, sexual behavior, etc.). Drug use was in the top 11 percent. Concerns were grouped in various contexts: family, peers, school, etc. Peers were viewed by adolescents as their major source of support, followed by parents, counselors, and self. The peer context included two concerns. Peer drug and alcohol use was seen as less of a concern to Hispanics and Asians, more for African Americans, and most of all to whites. The other major concern, sexual behavior, was significantly more important to rural than to suburban students.

Lifrak et al. (1996) studied the relationship between perceived self-competence and social support, gender, and substance abuse in middle schoolers. In girls, social support was unrelated to substance abuse except for support from classmates, which was associated with more cigarette and marijuana use. In girls with low scholastic competence, more support from peers was consistently associated with more substance abuse.

Ianotti et al. (1996) examined the relationship between adolescent substance abuse and the adolescents' perceptions of their friends' substance use. The study involved 4th- and 5th-grade students who were surveyed and tracked for 4 years. The sample included nearly 2000 students, most of whom were African American. Self-reported substance use of friends and classmates was also assessed. Perceived substance use of friends had stronger association with prior substance use than friends' self-reported substance use. Perceived family use and classmates' self-reported use were also contributors. Perceived friends' use is more likelyto be a product of an adolescent's previous substance use than a precursor of subsequent substance use.

Parenting IS Prevention Training of Trainers Workshop, 1998	SAMHSA

Office of National Drug Control Policy