Characterizing Science Talk in Early Elementary Classrooms: The Development of an Observational Tool

• In Event: 47.032 - Science Disciplinary Talk in K–2 Classrooms: Developing Supports for Teachers

Abstract

Purpose:
In order to enact the three-dimensional learning opportunities called for by the Framework for K-12 Science Education (NRC, 2012), it is important for early elementary teachers to support productive sense-making talk that engages students with science practices and focuses on their ideas. Supporting this type of ambitious science discourse in classrooms is difficult (Tekkumru Kisa & Stein, 2015), especially for early elementary teachers, many of whom lack strong science backgrounds (Davis, Petish, Smithey, 2006). Our project believes that we must provide teachers with clear evidence of what this type of talk looks like and how to achieve it. In this study, we present the findings from the iterative development and gathering of validity evidence (Kane, 2006; Mislevy, 2007) for an observational tool to support early-elementary teachers in improving the quality of classroom science discourse.

Methods & Data
In order to develop and gather validity evidence for the observation tool, we followed the iterative design, data collection, and refinement process in the Standards for Educational and Psychological Testing (AERA, APA, & NCME, 2014) and draw on Hill, Charalambous, and Kraft’s (2012) work on an interpretive argument approach for validating observation protocols for classroom teaching practice. After conducting a systematic literature review, we developed a first draft of the tool and received feedback from an expert panel of science and literacy education researchers. In this study, we report on analysis of video data from six teachers, three of whom used the [Name] curriculum developed for this research project and three of whom used other science curricula. We double coded all video data and qualitatively analyzed the findings (Miles, Huberman, & Saldaña, 2014) to identify overlaps, omissions, and themes across how science talk is enacted in the classroom.

Results
Preliminary findings suggest that there are four main dimensions that that can be used to characterize and measure science talk that occurs in early elementary classrooms: talk that supports:
1. equitable student participation and engagement in science
2. science language development and use
3. deepening of understanding within a science activity
4. synthesis of learning across science activities
These dimensions were first developed by synthesizing findings from research studies (e.g., Monteira, & Jiménez-Aleixandre, 2016; Samarapungavan, Patrick, & Mantzicopoulos, 2011; Varelas, & Pappas, 2006). However, literature tended to describe only high-quality talk or very low-quality talk. Through video analysis, we identified the range of ways science talk was enacted in various activities (e.g., reading informational texts, engaging in explorations) and the levels with which the talk supported sense-making. Table 1 lists sample teacher practices that indicate a high level for each dimension. The final paper will provide examples from the videos and will discuss how the iterative design process accounted for Hill and colleagues’ (2012) validation considerations.

Significance
While there are non-subject specific classroom observation protocols (e.g., CLASS, Hamre et al., 2013), few characterize the science-specific talk that can happen in early-elementary classrooms. Using mixed-methods to create an observation tool helps explicate and illustrate these abstract concepts for teachers so they have observable, measurable ways to reflect on their teaching.

Authors

• Tanya S. Wright, Michigan State University

• Amelia Wenk Gotwals, Michigan State University

• Lisa Domke, Michigan State University