ABSTRACT
Psychologists and educators have long emphasized the development of evidentiary reasoning as a key component of epistemic competence in the context of learning science. Despite this historical concern with evidentiary reasoning, there is still much that we do not know about its development in science learners, especially in the context of more advanced science learning. Decades of research show that students do not fully understand the nature, quality, and scope of the evidence on which advanced science concepts are founded. In this paper, I argue that the pervasive difficulties experienced by many science learners arise in part because notions of evidence embodied in current science curricula are often divorced from deep disciplinary knowledge and practice. Knowing what counts as evidence with regard to some set of knowledge claims, to what extent, and why, requires the integration of disciplinary knowledge about theoretical principles and inquiry practices. It requires knowledge of relevant variables and plausible mechanisms; how variables are typically operationalized in investigative designs; norms and standards for the precision and accuracy of instrumentation, experimental procedures, and measurement; bandwidths and density of sampling; models for aggregating and analyzing data; and conventions for communicating results. Drawing from examples in biology, I propose that, in order to support sophisticated epistemic reasoning in the teaching and learning of science, educators must unpack the notion of evidence itself and reconnect it to its disciplinary contexts.
