Engaging Students With Uncertainty Through Repeated Measure of Biological Qualities

• In Event: Exploring Data Science Across the Curriculum and Across Grade Levels

Abstract

Scientific practice creates models to represent and study the world, and these models are how scientists specify the “conditions of seeing” for others critiquing scientific claims (Bazerman, 1998; Lehrer & Shauble, 2010). These conditions often take the form of a data model, where sampling and measurement procedures convert qualities of the natural world into quantities for analysis. Science education, then, should not only support students to develop understandings of what scientists know, but also the data modeling practices scientists deploy to generate, critique and revise knowledge (AAAS, 2011; NRC, 2012; Duschl, Schweingruber, & Shouse, 2007).

In this presentation we describe our efforts to design learning environments that engage undergraduate students with opportunities to construct variable data within the context of measuring sea sponge abundance. Our approach to instruction was motivated by a body of research in statistics education that has shown the importance of a deep understanding of variability and distribution in reasoning with data (Franklin et al., 2007; Konold & Pollatsek, 2002), and of engaging students in the construction of variable data as they model the natural world (e.g. Lehrer & Romberg, 1996; Lehrer, Kim, & Schauble, 2007). Students in these classes independently measured the volume of identical 3D printed sponge fragments from a digital scan of a sea sponge specimen. Students then aggregated their independent measures to consider the distributional shape formed from their measurements and discussed the aspects of the distribution that informed our guiding question about sponge abundance. Students also created mathematical procedures to quantify the distributional characteristics they described in class.

We carried out this project using design-based research methodology (Cobb, Confrey, diSessa, Lehrer, & Schauble, 2003) in order to create detailed accounts of how students reasoned about measurement, distribution, statistics, and the biology of sea sponges. We video-recorded 4 class meetings in two sections of a non-major biology course at a large public university in the southeastern United States. We also collected written artifacts from 218 students enrolled in these two classes. We open coded the video records and artifacts by cataloging the different ideas students made use of in the tasks and then created categories of similar ideas across the students’ comments during class and the artifacts.

We found that students imagined many different measurement strategies to account for sponge abundance, but also came to value shared measurement procedures as critical to biological research. Their measurements of the identical sponge fragments created opportunities for students to come to see the center of the distribution as a signal about true measure, and the variability as an indication of error. Students developed strategies for quantifying distributional characteristics, such as center and variability, that have strong correspondence to disciplinary conventions of mean and standard deviation. In our presentation, we will emphasize the opportunities students’ ideas provide for building correspondence between their thinking and disciplinary norms within biology research. This work contributes to a growing body of knowledge about how to support students in undergraduate biology courses to authentically engage in data modeling of biological phenomena.

Authors

• Ryan Seth Jones, Middle Tennessee State University

• Anna Strimaitis Grinath, Idaho State University

• Zhigang Jia, Middle Tennessee State University

• Angela Google, University of South Alabama