This paper presents findings of a study that investigated factors that contribute to the engagement of students. Based on the theoretical model of Newmann, Wehlage, and Lamborn (1992), the study examined the effect of instructional approaches of teachers, contextual features of classrooms, and attributes of student experience on students' engagement in academic work. The sample, obtained through a nationwide selection process, included 24 public schools (8 elementary, 8 middle, and 8 high schools) deeply involved in restructuring. Data were collected from a survey of 5th-, 8th-, and 10th-grade students and their teachers from six core classrooms (3 mathematics and 3 social sciences) in each of the 24 schools. Over 3,660 students in 143 of the 149 core classrooms completed surveys. Two-way analysis of variance (ANOVA) and hierarchical linear modeling (HLM) were used to analyze the data. Findings confirmed the importance of authentic work and structures of support for students. Student self-concept had a moderate effect on engagement, and girls were more engaged than were boys. Students in mathematics classes had higher levels of engagement than those in social studies classes. Surprisingly, the use of traditional instructional methods exerted a positive influence on student engagement, while progressive methods did not. The data indicate that progressive methods must be combined with challenging content and a clear focus of inquiry. Specific restructuring content, such as authentic academic work and structures of support for learning, proved important in raising student engagement even where generalized restructuring was taking place. Two figures and six tables are included. Appendices contain statistical data and information on the construction of variables and methodology. (LMI)