March 15-18, 2020
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 2020 annual international conference is “School, Community, Citizenship: Science Education Across Places and Contexts.”
Sunday, March 15
Pre-Conference Workshop: Next Generation Labs for Next Generation Science Standards: Mobile Sensing as an Example
8:00 – 11:45 AM, Salon A, Lower Level
During this workshop, participants will interact with the mobile sensor technology that offers great potential in supporting out-of-school citizen science programs. Science lab will be redefined as a concept that sustains science practices without costly equipment, substantial science knowledge, confined space, or profound experience. The next generation science lab speaks directly to the NARST theme “Science Education Across Places and Contexts” in three aspects: 1) Fostering science inquiry without context or resource constraint; 2) Facilitating crowdsourcing and collaboration during science practice; 3) Providing a venue to engage in daily-life science investigation. We will introduce the development of mobile sensors, engage the participants in hands-on experiments, and discuss how the technology enabled diversity in science practices. We will share the potential data analytics tools that could be used to understand student’s science practices by taking advantage of backstage sensor data. Tips and resources for preparing successful grant proposals, which supported the research at a non-profit organization to thrive, will also be discussed.
Automated Assessment of Argumentation in School Science: Developments and Challenges
Automated Real-Time Argument-Text and Model-Interaction Feedback to Support Secondary School Students’ Revision of Scientific Arguments
2:40 – 4:10 PM
Most environmental systems encountered in science class are not amenable for direct manipulation for students to study real-world implications. In such cases, simulations provide ample opportunities to explore and investigate aspects of otherwise complex systems. In this study, students engaged in three tasks where they used different simulations to investigate groundwater flow, aquifer types, and fresh water management and wrote scientific arguments about sustainability embedded in these topics. To scaffold students, we developed in real time argument feedback based on automated scoring of students’ text responses to structured argument prompts as well as simulation feedback based on automated detection of students’ interactions with simulations. Two delayed cohorts of students were studied: one cohort receiving Argument plus Simulation (AS) feedback and the other cohort receiving Argument Only (AO) feedback. We compared how AS and AO students interacted with simulations and wrote scientific arguments before and after receiving their respective feedback. Results indicate that (1) similar percentages of AS and AO students revised arguments (49% each) and both AS and AO students significantly improved scientific arguments as expected from the use of argument feedback. Only 5% of AO students reran simulations while 36% of AS students did. AS students who reran simulations significantly improved their simulation interactions and improved their scientific arguments to a greater extent than AS students who did not rerun simulations. Based on these findings, we discuss pedagogical and research implications related to the development of intelligent feedback along with current research and design limitations.
Technology Tools to Support Scientific Thinking
Impacts of Sequential Experience with Agent-Based Modeling and System Dynamics Modeling on Students’ Ability to Link Across Levels in Reasoning About Complex Phenomena
2:40 – 4:10 PM
Real-world systems dynamics problems involve multiple complex systems and the interactions between them. Preparing learners to reason about such systems requires developing their ability to navigate and understand systems at multiple levels. The goal of this study was to characterize the development of student understanding of the links between agent-level and system-level entities, properties, and events in complex systems, as a result of creating and analyzing models in an agent-based modeling environment followed by equivalent activities in a system dynamics modeling environment. In a week-long unit on modeling evolution, students were guided to create and analyze computational models for three experiments of bacterial growth. Students completed a Modeling Evolution questionnaire at the beginning of the activities and again at the end. The authors developed a coding scheme that first identifies each expression as describing agent-level or system-level entities, properties, and events; then examines the causal links between expressions and labels the links as cross-level links, same-level links, or other types of links. The authors are currently analyzing the data using the coding scheme described above. Findings will reveal the impacts of sequential experience with agent-based modeling and system dynamics modeling on students’ ability to link across levels in reasoning about complex phenomena.
Monday, March 16
Characterizing Computational Thinking in the Context of Technology-Enhanced Multilevel System Modeling
10:15 – 11:45 AM
The science education community recognizes the importance of systems and systems models and computational thinking (CT) to help the next generation of learners engage in STEM-related fields and lead a more informed civic life. The goal of our research effort is to explore how students’ CT develops over time through investigations of phenomena and system models of increasing complexity. We examined the progress of ~100 students’ computational artifacts (system models) associated with their explanations of those models in 10th grade science classes over a two-week instructional unit. The students constructed and iteratively revised their models while they learned new concepts using a system modeling tool that provides various ways to support systems modeling. The four papers in this related paper set work together to show 1) the CT framework in the context of system modeling practices, 2) How prior content knowledge and CT influence constructing an appropriate dynamic model, 3) how students’ understanding and CT evolve as they engage in system modeling practices, and 4) system models as a window into student conceptual thinking. The mixed-method qualitative and quantitative data results show that there is a need to scaffold system modeling throughout the unit in order to promote students’ use of computational thinking skills and practices.
Tuesday, March 17
Technology-Enhanced Framing of Data to Facilitate Classroom Enactment of Science Practices
1:45 – 3:15 PM
In science, data play a central role in testing knowledge-based hypotheses, exploring new phenomena, and developing explanations of phenomena under investigation. While the Next Generation Science Standards (NGSS) strongly support the integration of disciplinary core ideas with science practices, how to design and enact data-rich classroom activities remains a challenge. The purpose of this session is to delineate and demonstrate how various types of technologies can provide data-intensive contexts to support a range of science practices such as using models, planning and carrying out investigations, analyzing and interpreting data, communicating information, and using computational thinking. This session consists of four papers that feature student activities designed for high school and college students in physical and biological sciences. Technologies addressed in this session include data visualizations, simulation models, an experimentation entropy tracking engine, an interactive data collection and analysis platform, an online collaborative simulation environment, and interactive sensor data ow controls. This session will inspire ways in which technologies can frame data exploration, collection, use, and analysis appropriately for target students in disciplinary contexts and across science practices.
Wednesday, March 18
Motivating Youth Engagement
8:30 – 10:00 AM
Youth-oriented community and citizen science (YCCS) projects are a promising setting to explore how youth interact with data, with implications for science learning and data science education (DSE). Community and citizen science (CCS) involves members of the public working with scientists to generate new scientific knowledge. In environmental YCCS projects, youth participants work in their classrooms or with local groups to produce data that are nested within larger, sometimes national or international, datasets. These data help professional scientists, agencies, and community-based organizations answer questions, make management decisions, inform policy, or advocate for community needs. Investigating how youth engage and act with CCS data offers insights into how science learning can connect youth across multiple communities, and raises questions for what DSE can and should be.
Digital Tools: Research and Demonstration Showcase
Examining High School Students’ Scientific Practices during an Augmented Thermal Perception Lab
10:30 AM – 12:00 PM
Students encounter difficulties in learning science concepts that involve ‘hidden’ processes and phenomena that cannot be observed by the naked eye. We adopted augmented vision technology to investigate behavior related to scientific practices during the ‘latent heat’ lab. We chose this topic because it presented a perplexing scenario, where heat seems to come out of thin air to warm up the paper placed on a cup of tap water. Ninth-grade students (N = 112) from Northeastern US participated in the study. Students used thermal imaging application developed by the team to “see” thermal energy and to carry out investigation in the prediction-observation-explanation cycle. Their behavior was screencast and recorded as log data by means of an infrared camera and the cellphone sensor, respectively. Semi-supervised algorithms were adopted to classify student’s scientific practices based on four indexes related to the augmented thermal perception lab. Students were categorized into three groups—planner, stabilizer, and roamer. We demonstrated representative behavior and common mistakes found during active experimentation. Our ultimate goal is to scaffold the scientific practices by providing just-in-time instruction or cue on the App. The App and the automatic profiling technique have great potential. Educational implications and future studies are discussed.