ity of caretakers as essential to the development of trust and confidence (Mahler et al., 1975; Winnicott, 1975). Similarly, applied researchers have shown that, in all settings studied, adolescents benefit from experiencing clear rules, discipline, and consistently enforced limits on their behavior (Connell and Wellborn, 1991; Dryfoos, 1990; Jackson and Davis, 2000; Lipsitz, 1980; Roth and Brooks-Gunn, 2000). More classroom and school discipline is a major cause of poor academic achievement and psychological disengagement from school.
Structure is a critical feature of all settings. On one hand, research on families shows that firm parenting and clear behavioral expectations, when coupled with warmth and emotional support, are associated with more positive developmental outcomes than lax parenting (Steinberg et al., 1992; Steinberg, 2000). Evidence is also emerging that the best adolescent outcomes are associated with parents gradually reducing their control over their adolescents and providing them with increasing opportunities to help establish family rules and participate in family decision making. Adolescents desire both consistency and structure and increasing opportunity to manage their own behavior. They appear to do better in families that over time provide both of these experiences (see Eccles et al., 1993, for a review).
Similarly in classrooms, maintenance of discipline, control, and organization by the teacher is related to student satisfaction, growth, and achievement. Once again, however, as they grow older, students desire increasing opportunity to have input into classroom and school governance and rules. Evidence suggests that their motivation is optimized when they experience this type of change in classroom and school management over time (Epstein and McPartland, 1979; Jackson and Davis, 2000; Lipsitz et al., 1997; Maehr and Midgley, 1996; see Eccles et al., 1998, for a review).
On the other hand, research on peer relations shows that the amount of time adolescents spend with peers in unstructured activities, like driving around in cars, predicts increases in involvement in problem behaviors (Osgood et al., 1996). One important study found that participation in community programs that lacked structure predicted greater involvement in problem behaviors both in the present and 20 years later (Mahoney et al., in press).
A critical element of structure is consistent monitoring and enforcement of rules and expectations. Across settings, there is more positive development and fewer problem behaviors with consistent monitoring by parents (Grotevant, 1998; Pettit et al., 1999; Steinberg, 2000), teach-