Metaepistemic Discourse in Middle School Science Inquiry Classes

• In Event: Following the Thread From Planning to High-Quality Discourse About Science

Abstract

Objectives and theoretical framework. A promising way to promote growth in science reasoning may be to promote meta-epistemic discourse (Author, 2011). By meta-epistemic discourse, we mean discourse exhibiting metacognitive awareness of epistemic phenomena--the aims and goals of scientific inquiry, the criteria used to evaluate the products of inquiry, and the reliable processes used to carry out inquiry (Author, 2014, in press). Students who develop meta-awareness of these processes may use them more regularly and more appropriately (Author, 2011). However, researchers have struggled to identify exactly what kinds of talk counts as meta-epistemic (see discussions in Barzilai & Zohar, 2014; Hofer, 2004). Formal, philosophical discussions about the nature of science or the nature of knowledge and knowing are clearly meta-epistemic. However, a heavy instructional emphasis on formal levels of meta-epistemic knowledge may have limited benefits in promoting progress in practical reasoning (Allchin, 2011; Author, 2005).

Data sources and methods. What is needed is a richer analysis of meta-epistemic discourse and meta-epistemic thinking that bridges the formal and practical levels. We have developed such an analysis through examining student and teacher discourse from daily recordings in 8 classes of a yearlong implementation of a model-based inquiry curriculum (Author, 2011, 2013, 2014). Our analysis focused on 40 class discussions and 40 group discussions during model-based inquiry activities such as developing class criteria for what counts as good models, developing and revising models based on evidence, and evaluating alternative models in the light of a body of evidence. We examined transcripts of these discussions, carefully coding for a range of indicators of different levels of meta-epistemic awareness of epistemic constructs.

Results. Like other researchers, we identified a practical level of epistemic discourse (e.g., students giving evidence to support a model of heredity) and a formal level (e.g., students stating explicitly what the characteristics of good models are). But we also found several important categories of meta-epistemic discourse that bridged these two levels. One of these was the application of explicit meta-epistemic criteria for good models and good evidence to the practical work of model revision and model choice (e.g., class-developed criteria for good models specify that good models have mechanisms, and students explicitly invoke this criterion in evaluating a model of chloroplast function). A second was explicit discussion of how epistemic criteria are to be tailored to particular situations (e.g., explaining specifically why a particular model of chloroplasts function fails to incorporate an adequate mechanism). Early analyses of two teachers’ classes indicate further that the teacher who engaged students in significantly more of these intermediate forms of meta-epistemic discourse gained more on a posttest of student reasoning than the teacher who promoted less of this bridging talk.

Significance. The implication of these results is that teachers may be able to productively advance students’ reasoning by facilitating these “intermediate” forms of discourse, and designers can develop learning environments that encourage these forms of talk. We will present a taxonomy of these and other forms of meta-epistemic discourse uncovered in our analysis.

Authors

• Brandon Mauclair-Augustin, Rutgers University

• Clark A. Chinn, Rutgers University

• Ravit Golan Duncan, Rutgers University