April 11–14, 2015
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 2015 annual international conference is “Becoming Next Generation Science Educators in an Era of Global Science Education Reform.”
Saturday, April 11
Innovative Technology in Science Inquiry: Preparing Students for STEM
2:45 PM–4:15 PM, Columbus GH
This study investigates a comprehensive approach to inquiry-based science education by engaging students in inquiry-based science projects that use open source computational models and real-time data acquisition with probeware. We present four years of research findings from our research program with more than 4000 students from upper elementary to high school. In particular, we are interested in the following research questions: (1) did a comprehensive approach to inquiry-based science education increase students’ understanding of standards-based content? (2) Did this approach increase students’ interest and attitudes towards STEM careers? We investigated students’ understanding of standards-based content using pre- and post-tests that were customized to the content in particular units. Also, the STEM interest and career data was measured using pre- and post-surveys in the beginning and end of the school year. In addition, students’ interest in a STEM Career was measured using their written responses to the question “Name a CAREER you are interested in at this time.”
Involving Teachers in Developing Assessments Aligned with NGSS using a 7-Step Process
Chanyah Dahsah (Michigan State University & Srinakharinwirot University), Jane Lee (Michigan State University), Angela DeBarger (SRI International), Daniel Damelin (The Concord Consortium), Joseph S. Krajcik (Michigan State University)
2:45 PM–4:15 PM, Grand D North
It is significant challenge to develop assessments that provide evidence of three-dimensional learning. Therefore, it is necessary to build design processes that support science teachers and educators in developing assessment tasks that align with the NGSS. In this study, we used evidence-centered design (ECD) to guide our 7-Step process for developing NGSS-aligned classroom assessments. ECD is a systematic approach to constructing assessments that focus on the evidence. It has been used in developing assessments in various fields, and has also been successfully used in developing large-scale and classroom assessments in science education. Given the teacher’s central role in implementing assessments, the purpose of this study is to examine how the 7-Step process can help teachers construct their own NGSS-aligned assessments.
Sunday, April 12
Symposium – Toward Building a Foundation for Teaching and Learning in Elementary Science: Highlighting Six NSF Projects
Nancy Romance (Florida Atlantic University), Michael R. Vitale (East Carolina University), Lynn A. Bryan (Purdue University), Ala Samarapungavan (Purdue University), Annemarie Palinscar (University of Michigan), Jonathan Osborne (Stanford University), Hilda Borko (Stanford University), Deborah L. Hanuscin (University of Missouri-Columbia)
& Carolyn Staudt (The Concord Consortium)
10:15 AM–11:45 PM, Columbus KL
Quality elementary science programs are faced with the challenge of adapting/building upon the NGSS in order to provide an essential foundation for student learning and interest in science. In meeting this challenge, researchers and practitioners alike must contribute toward the development of a knowledge base that identifies components that result in quality science teaching and student achievement. This session provides a forum through which NSF researchers addressing a variety of approaches for improving elementary science will discuss – with active audience involvement – how the ideas presented can contribute to the development of quality elementary science programs. While these projects use different tools and/or instructional models for doing so, all the projects are similar in that each addresses in some manner the following components: (a) Conceptual Models and Modeling, (b) Curricular Approaches Linking Science and Literacy, and (c) Specialized Models of Professional Development. Highlighted in the symposium will be implications for researchers and practitioners for systemic issues in elementary science education.
Embedding NGSS Science Practices in Digital Game-Based Genetics Materials: Measuring Content Knowledge, Argumentation, and Motivation
10:15 AM–11:45 PM, Columbus GH
This paper describes the design, development, and testing of Geniverse: a game-based digital learning environment in high school genetics. Geniverse supports students as they learn crosscutting concepts and engage in science practices, performing experiments that generate data sets for scientific explanation and argumentation. This deep engagement with scientific argumentation helps students understand both science content and the process through which scientists learn about the world. Geniverse provides a set of interactive, game-like challenges that increase in difficulty, scaffolding students into a more genuine experience of scientific investigation. We studied the impact of these materials with a quasi-experimental research design involving 48 teachers – 24 using Geniverse and 24 using their business-as-usual genetics materials. Impacts on students’ science content knowledge, argumentation skills, and motivation in science were compared. Results from hierarchical linear modeling analyses indicate that Geniverse led to significant gains on students’ science content knowledge and argumentation skills. However, these gains were not significantly greater that learning gains in the comparison group. Our findings are discussed in light of other studies on game-based digital materials, and in the context of methodology for examining impacts of materials that are challenging to implement with high fidelity.
Poster Session: A Design Framework on Assessing Modeling Practices
Ji Shen (University of Miami), Charles Xie (Concord Consortium) & Bahadir Namdar (University of Georgia)
3:15 PM–4:15 PM, Riverside East
Poster Session: Teaching Environmental Sustainability Using a Place-based Watershed Modeling Application
Nanette Marcum-Dietrich (Millersville), Susan Gill (Stroud Water Research Center) & Carolyn Staudt (Concord Consortium)
4:15 PM–5:15 PM, Riverside East
Monday, April 13
Sensing Science: Assessing K-2 Students Readiness for Reasoning with Kinetic Models of Heat
2:30 PM–4:00 PM, Gold Coast
We present findings from our two-year study on the preconceptions young children (4 to 8 years old) hold in regards to the nature of temperature and heat and how their use of visual representations can improve their interpretation of temperature and heat. We use the word preconceptions to indicate the naïve ideas that children hold before formal instruction. Although thermodynamics does not appear in the science curriculum until middle school in the Next Generation Science Standards, our goal is to see if much younger children could explain concepts of temperature and heat. We hypothesize that the use of visualizations of heat concepts that emerge from interactions between the particles can serve as a better “tool for thinking” to help K-2 students better understand those concepts. Our findings suggest that, when aided by real-time visualizations, young children can use the kinetic heat model to disambiguate their current theories about temperature and heat.
Tuesday, April 14
Performance Assessment of Engineering Design Using Process Analytics Based on CAD Software
10:15 AM–11:45 AM, Roosevelt
In educational settings, engineering design is a complex cognitive process in which students learn to generate, implement, and evaluate ideas for systems or processes whose form and function meet specified criteria and constraints. Because of the multifaceted, systemic nature of engineering design, student assessment necessitate tracking of a comprehensive set of performance indicators over a significant period of time. This requirement poses a bottleneck to large-scale studies. This paper introduces a new assessment technique based on 1) moving design learning to a computer-aided design (CAD) platform that supports the full cycle of design including inquiry, construction, test, evaluation, and communication; 2) logging all the student-generated process data comprising actions, artifacts, and articulations behind the scenes; and 3) computing time-varying student performance curves against the specifications with novel process analytics to identify possible cognitive roadblocks in their design paths.
Symposium: Big Data and Learning Analytics: A New Frontier in Science and Engineering Education Research
Hee-Sun Lee (University of California, Santa Cruz), Saeid Nourian (The Concord Consortium), Kyle R. Cheney (WPI), Raha Moussavi (Worcester Polytechnic Institute), Janice Gobert (Worcester Polytechnic Institute),
Charles Xie (The Concord Consortium), Gey-Hong Gweon (University of California, Santa Cruz), James Lester (North Carolina State University) & Eric N. Wiebe (North Carolina State University)
1:00 PM–2:30PM, Columbus GH
The rise of engineering education in K-12 schools calls for basic research that can advance our understanding about how students learn engineering. A major research focus is on the process of engineering design. Because of the open-ended, project-based nature of engineering, students can produce a large quantity of data and artifacts while solving a complex design challenge, making it difficult to discern their learning. Visual analytics is a technique of scientific reasoning that uses visual interactive interfaces to optimally combine the computational visualization power of the computer and the pattern recognition power of the brain. This paper will demonstrate how visual analytics can be used to study the learning dynamics of engineering design encoded in the fine-grained data logs of the supporting design software that record all of student actions, artifacts, and articulations. These raw process data are difficult to analyze because of their complex, irregular, and personalized forms. Visual learning analytics can provide powerful tools for researchers to see patterns and trends in these student data, from which cognitive and learning theories for engineering education can be tested or derived.