March 31 – April 3, 2019
The National Association for Research in Science Teaching (NARST) is a worldwide organization of professionals committed to the improvement of science teaching and learning through research. Since its inception in 1928, NARST has promoted research in science education and the communication of knowledge generated by the research. The ultimate goal of NARST is to help all learners achieve science literacy. The theme of the 2019 annual international conference is “Creating and Sustaining Collective Activism through Science Education Research.”
Monday, April 1
Using Technology to Promote Students’ Modeling Practice and Complex Systems Thinking
9:30 – 11:00 AM, Maryland E
Since the release of A Framework for K-12 Science Education and the Next Generation Science Standards there has been a greater emphasis on modeling and complex systems thinking in science classrooms. Technology-based tools provide opportunities to meaningfully engage students in modeling complex systems to support their science learning. In this related paper-set, we present findings from four different studies of student learning, which focus on the effect of integrating one or more modeling tools in science curricula. We discuss the advantages and affordances of using technology to promote student modeling practice and content learning, the challenges and difficulties students experience in learning with tech-based modeling environments, and the type of scaffolds needed to support student engagement in using such tools. Altogether, these findings provide important insights to be considered by science education researchers, educational environment designers, and policy-makers that are interested in the learning outcomes of students engaged in modeling and systems thinking facilitated through technology.
Introduction: Using Technology to Promote Students’ Modeling Practice and Complex Systems Thinking
Dan Damelin, Joe Krajcik
Logical Discrepancies in Semi-Quantitative System Models: Visual Cues to Causal Modeling Issues vs. Accurate Modeling of Alternative Concepts
A. Lynn Stephens, Steven Roderick
Indirect relationships or relationship arrows indicating causality opposite to what we expect can lead to logical contradictions in complex causal models. A question arises as to whether these are always problematic. An understanding of cause and effect relationships is central to systems modeling and to explaining phenomena in science (Jonassen & Ionas 2008; Koslowski & Masnick 2002). However, Authors (1998) found that while some students were able to engage in deep and full causal reasoning and systems thinking when using dynamic modeling tools, other students faced challenges such as distinguishing between causal and correlational relationships and providing shallow explanations when reasoning about the systems represented in their models. We ask how different issues with causal reasoning manifest in student models and whether these suggest corresponding different instructional implications.
Agent-Based and Systems Dynamics Modeling of Complex System Behaviors
In science class, students are expected to learn scientific phenomena associated with complex systems such as epidemics and global warming (Yoon, 2017). One of the most difficult concepts students need to learn about complex systems is system emergence, e.g. “how parts of a system and their relationships give rise to the collective behaviors of the system, and how the system interrelates with its environment” (Bar-Yam, 2002, p. 2). To teach system emergence, agent-based simulations and systems dynamics models are often used. This paper addresses how high school students constructed these two types of models to address epidemics. Two research questions were asked: (1) what differences and similarities existed between agent-based modeling and the systems dynamics modeling when students developed models to explain how epidemics get started, progress, and are controlled, and (2) how high school students distinguished affordances of the two modeling types based on their model building experiences.
Embedding Computational Thinking into a Middle School Science Meteorology Curriculum
Carolyn Staudt, Nanette Dietrich (Millersville University of Pennsylvania), Meridith Brouzas (Argonne National Laboratory)
4:15 – 5:15 PM, Maryland C-D; Maryland and Baltimore Foyer
All students need to understand the role of computation and computational thinking within disciplinary problem solving. Opportunities to learn and apply computational thinking are absent from most students’ experiences. Yet there is no need for these opportunities to be inaccessible. With proper tools and approaches, compelling student experiences within science class can be imbued with fundamental computational thinking skills. This project designed, developed, and enacted an innovative, technology-rich curriculum for middle school students that addresses critical NGSS-related science standards, engages students in an intriguing, ongoing inquiry-related investigation, and supports the development of key computational thinking practices. This study suggests that with purposeful integration of the curricular design elements, embedding computational thinking practices into the science classroom is possible.
Tuesday, April 2
Integrating Science and Engineering with a Focus on Evidence of Student Learning
Selcen Guzey (Purdue University), Senay Purzer (Purdue University)
Kerrie Douglas (Purdue University), Jim Pellegrino, (LSRI, University of Illinois at Chicago), Corey Schimpf, Kristen Wendell (Tufts University), Jessica Watkins (Vanderbilt University)
8:00 – 9:30 AM, Baltimore A
The integration of engineering and science is gaining an interest worldwide, as part of the NGSS in the U.S. and under the label of STEM education more globally. These learning environments involve real-world problems and engage students in learning new concepts, learning to apply of core ideas to tangible situations, and learning practices that reflect the work of scientists and engineers. This means learning is happening in multiple dimensions. In this panel, we bring together experts who examine questions around collecting and noticing evidence of student learning through research in diverse areas such as assessment, teacher education, and on student learning. Together, these scholars will discuss, “What learning is happening when science and engineering are integrated?,” “How do we collect evidence of such learning?,” and “How do teachers notice and respond to such learning?”
Uncertainty Manifested within Science and Computational Thinking Practices
2:30 – 4:00 PM, Homeland
That uncertainty is involved in science enterprise is one of the essential epistemological beliefs students need to develop about the nature of science. The purpose of this symposium is to delineate and demonstrate how uncertainty permeates through science and computational thinking practices with which students engage in science class and how it can serve as an opportunity to enrich students’ overall inquiry experience. This symposium features four projects that pedagogically incorporate uncertainty as part of articulating limitations involved in using model-based theories to explain real-world data, carry out systematic experimentation, recognize and treat conceptual and data-embedded errors, and interpret computationally generated data to resolve instability in inquiry-based activity.
Uncertainty in Using Model-Based Claims to Explain Real-World Evidence
Data play an important role in scientific argumentation. Data invoke two types of reasoning. In the use mode, Staley (2015) mentioned that one interprets data to arrive at substantive conclusions based on one’s theoretical and empirical assumptions. In the critical mode, he pointed out that one “turns a critical eye towards the assumptions employed in one’s inferences from data must be concerned with the state of one’s own knowledge and the constraints it sets upon the ways in which her inferences might go wrong” (p. 41). As such, uncertainty arises when one engages in the critical model with data because “well developed scientific theories predict and explain facts about phenomena but not facts about the data which are the raw material of evidence for these theories. Many different factors play a role in the production of any bit of data, and the characteristics of such factors are heavily dependent on the peculiarities of, for example, the particular experimental design, detection devices, or data-gathering procedures which the investigator applies. Data are idiosyncratic to particular experimental contexts, and typically cannot occur outside these contexts” (Boumans & Hon, p. 2). This paper addresses whether and how students evaluate theoretical and empirical assumptions when they use model-based claims to explain real-world evidence.
Entropy as Proximal Measure of Systematicity in Experimentation
In order to draw valid conclusions from experimentation, students need to systematically explore a set of variables associated with the experimental setup. In particular, when students need to discover the relationship between an independent variable and a dependent variable, the control of variable strategy (CVS) is essential. During physical or virtual experimentation, students’ execution of CVS can be demonstrated by changing one variable at a time while controlling all other variables in order to attribute any observed effect to the single changing variable. While CVS has been a coveted student learning outcome from experimentation, how to quantify the degree to which students utilize CVS in their experimental procedure is neither clearly articulated nor measured. In this paper, we invoke the concept of sample entropy utilized in natural and medical sciences to create a proximal measure of how closely students execute CVS during the entire run of their experimentation and discuss how this measure can be traced to reduction in student uncertainty about their conclusions based on data from that experimentation. In this study, we developed an indicator that can represent the systematicity of students’ experimentation. The sample entropy is a practical way to measure the disorderly nature of a time-dependent process of parameter changes committed by students.
Student Agency in Discovering Scientific Practices: Leveraging Uncertainty
Concern has been expressed that students are rarely positioned with epistemic agency and that a complacent approach to the NGSS scientific and engineering practices could be to teach them directly, as an authority-described version of what scientists do (Miller et al., 2018). When engaging with the practice Analyzing and Interpreting Data, we have observed students respond to data they have generated merely as a product to be turned in, without regard to its meaning. Masnick et al. (2007) found that even in the absence of domain knowledge, students from third grade to college were aware that there is variation in data and that this was something for them to consider in their reasoning. According to Ben-Zvi (2006) and Paparistodemou and Meletiou-Mavrotheris (2008), when using a statistical visualization tool, students from third grade up were observed using an informal process to reason about signal and noise and other types of variability in data, without their having been taught the mathematics of statistical inference. Manz (2015) argues that building material resistance and uncertainty into learning environments can establish a need for scientific practices and lead to practices and purposes being emergent for students. We describe a lesson in which this appeared to happen and investigate what led students to discover the need for a scientific practice in three of the classrooms taught by the teacher but not in the fourth.
What Do Students Do About Uncertainty? Resolving Instability in Inquiry Activity
Contending with uncertainty is an important component of science learning (Metz, 2004). We consider uncertainty as an interactional rather than purely conceptual phenomenon, to examine the role uncertainty plays in students’ scientific inquiry. We present a case from a classroom study in which we introduced computational tools for data acquisition and control, of how uncertainty seeded by one anomalous dataset propagated throughout the classroom, and how three groups took up three different types of investigations as responses to that uncertainty. We argue that uncertainty can play an important role in science learning by opening space for students to exercise conceptual agency. If promoted and sustained by a teacher, uncertainty can open pathways for students to exercise conceptual agency in inquiry. This research will contribute to science teaching and learning by examining the role of uncertainty in promoting student agency in science, with the role of the teacher both as sustaining instability in inquiry and positioning students as capable of managing or responding to it.
Kindergarteners’ Use of Particle Models of Matter to Explain Material Phenomena
Lynn A. Bryan (Purdue University), Carolyn Staudt , Ala Samarapungavan (Purdue University)
4:15 – 5:45 PM, Gibson
Recent data indicate that large numbers of students leave school without having acquired basic science proficiencies. One major challenge to improving science achievement is the paucity of science instruction in the early grades. Recent research indicates that the percentage of time spent on elementary school science instruction—estimated at between 10-13% in the years ranging from 1987-1994—has dropped dramatically since then. Studies indicate that this drop is associated with an increased emphasis on and allocation of instructional time to teaching reading and mathematics at the expense of other content. Our own prior research in kindergarten classrooms indicates that teachers in the primary grades often feel that young children lack the cognitive resources do real science. A key challenge for science education is to support the implementation of reform-oriented science instruction in the early grades and to document how and what children learn from such instruction. Therefore, the overarching goals of our research are: (a) to work collaboratively with primary grades teachers to develop and implement physical science instruction through a discourse-rich, model-based inquiry approach, and (b) to examine and document children’s developing sense of physical mechanism as they build models to explain physical phenomena and learn to apply a core set of physical mechanisms to a variety of diverse phenomena. As part of this research agenda, we examined kindergarten students’ use of particle models to explain the properties and behavior of matter in solid, liquid, and gas states, and phase changes such as melting and freezing.
Wednesday, April 3
Using High School Students’ Initial Perceptions of Evolution Across Biological Levels to Inform Curriculum Development
8:30 – 10:00 AM, Maryland F
Students are often taught evolution isolated from genetic and cellular mechanisms. In reality, a complete understanding of evolution requires knowledge spanning multiple biological levels. To address this issue, the ConnectedBio project has developed a set of technology-enhanced units for high school biology designed for NGSS. The goal of each unit is to have students explore biological phenomena from the molecular level to the population level and make connections across levels by developing and revising models. Our lessons use evolutionary phenomena from the previously developed Evo-Ed Cases as the basis for each unit. To inform unit development, we developed survey questions asking students to explain different hierarchical levels for each phenomena. The first case addresses the phenomenon of different coat colors in deer mice. In this study, we report our analysis of written survey responses and follow-up interviews from students using the three dimensions to explain phenomena. Our preliminary results suggest that 9th grade General Biology students include more alternative explanations to case phenomenon, but 11th/12th grade AP Biology students use more evolution key concepts and terms. Both levels appear to require more scaffolds to make connections among levels in their explanations.
Teachers and Engaging Environmental Education
10:30 AM – 12:00 PM, Fells Point
The study investigates whether curriculum that is purposefully designed to provide students with meaningful opportunities to connect conceptual understanding of watersheds to real-world decision-making using authentic data and tools leads to an increase in students’ content knowledge and environmental interest and action. The Teaching Environmental Sustainability: Model My Watershed (MMW) curriculum and toolset situates student learning in the exploration and evaluation of the conditions of their local watershed using outdoor exploration, field experimentation with probeware, and a scientifically valid watershed modeling application. The curriculum was implemented by 38 teachers in 8 states with data collected from 1,533 students. The instruments used in this study include a 15-question content knowledge test developed by Stroud Water Research Center, portions of the NOAA B-WET Secondary Science Self Report, and a semi-structured Critical Incident Technique interview protocol to assess students’ engagement and action. The study indicates that a place-based watershed modeling curriculum is an effective tool for increasing students’ understanding of watersheds, encouraging personal environmental action and serving as a critical incident for watershed engagement.