Beginning with the first quantitative evaluations of the Head Start program (e.g., Westinghouse Learning Corporation, 1969), there have been lively discussions about the usefulness and limitations of quantitative evaluations of early childhood programs. Questions have emerged about the types of outcomes and magnitude of change that can be expected from a wide range of interventions and the value of such outcomes to children, parents, schools, and the society as a whole (Center for the Future of Children, 1995; Farran, 1990; Hamburg, 1994). Conflicting views and beliefs about the efficacy of early childhood interventions have made it necessary to broaden evaluations so that they provide meaningful information to the necessary audiences.

This challenge to expand evaluation studies has emerged from fundamental questions about 1) the nature of developmental change, 2) the inviolability of the traditional scientific paradigm, and 3) the dichotomy between quantitative and qualitative methodologies. Simultaneously, evaluations have benefited from a growing sophistication about how systems operate within political contexts and how programs themselves evolve over time. In this chapter, we explore each of these areas with a view toward expanding the scope of early childhood intervention program evaluations through the incorporation of multimethod approaches.

BELIEFS ABOUT DEVELOPMENTAL CHANGE

Program evaluators, similar to other researchers, are guided by their theoretical and metatheoretical perspectives and assumptions about the nature and process of developmental change (Lerner, Hauser-Cram, & Miller, 1998). Beliefs about key principles of development are likely to affect the perspective of the evaluation team and lead to decisions about the research design and methodology employed.