Longitudinal studies into science learning - methodological issues

The Gold Standard for education research promotes randomized controlled trials (RCTs) that can produce generalizable knowledge claims across similar problems and situations. Unfortunately, the Gold Standard does not fully recognize the need for developmental research to better understand the problem space, formulate theory and approaches to teaching and learning, and formulate and pursue associated research questions. This developmental research has been a precursor to the development of interventions together with the necessary instrumentation and technologies required to fully investigate these through the more formal evaluative processes imagined by the Gold Standard. This chapter focuses on longitudinal studies that cover a continuum from such developmental research to research that uses control-experimental features to evaluate interventions. These studies attend to a set of issues dealing with  developmental progressions and learning trajectories that require  investigation over an extended period of time. It will be argued that
these longitudinal studies of a variety of methodological types represent quality research in that rigorous design and implementation produce  evidence-based claims. The chapter examines the nature of the relationship between evidence and claims in these studies, to show the possibility of building in control features every bit as strong as those in classic Gold Standard designs. Further, it will be argued that, given the complexity of learning pathways, a simplistic interpretation of RCTs conducted over the shorter term can be misleading in terms of both internal and external validity claims.