An examination of definition, causes, variables, and strategies of implementation suggests the kinds of data necessary for measuring school effectiveness. By focusing on students who first enter school below grade level standards and by understanding that student learning is embedded in a system of key elements or variables that, over time, are interrelated in a circular pattern of causality, longitudinal research provides such data on effective schooling. Essential structural differences between effective and ineffective schools are highlighted by causal loop diagrams illustrating both key variables and the concept of circular causality. The variables, which identify areas requiring further empirical research, are teacher effectiveness and expectations, time allocated to instruction, engaged learning time, and student motivation, behavior, and achievement. The variables identify points of intervention in the system where the principal's leadership role is critical in achieving an effective school. After the problem has been defined, causes understood, and points of intervention identified, a theory of implementation is required. The authors thus emphasize the need for research on the organizational dynamics of implementing school effectiveness programs. (PB)