The University of Georgia will carry out a two-year Synthesis Project that aims to provide a comprehensive review of the research and practices for modeling-based instruction (MBI) in K-12 science education. The project will synthesize existing literature on MBI in K-12 science education over the last three decades. It will rigorously code and examine the literature to conceptualize the landscape of the theoretical frameworks of MBI approaches, identify the effective design features of modeling-based learning environments with an emphasis on technology-enhanced ones, and identify the most effective MBI practices that are associated with successful student learning through a meta-analysis.
The project will build a systematic and analytic framework to conceptualize MBI, recommend best design strategies of technology-based modeling environments, evaluate MBI teacher professional development strategies associated with improved student learning, and propose appropriate assessment strategies created to evaluate and inform MBI. In addition to the comprehensive analysis of the theory and design of MBI, a meta-analysis will study the four components of student learning: theory, design, implementation, and assessment. Based on qualified quantitative studies, an examination of the four components will be made to evaluate how different empirical studies have established their effectiveness, examine the correlations among key components, and chart the impact of associated factors on student learning.
This Synthesis Project, Achievements and Challenges of Modeling-based Instruction (ACMBI) in Science Education, provided a comprehensive review of the research and practices of modeling-based instruction (MBI) in science education. MBI is an innovative way for science teaching and learning that encourages students to use, create, share, and evaluate models to represent and explain scientific processes and phenomena. ACMBI has synthesized existing literature on MBI in K-12 science education over the last three decades. Through the project, we rigorously coded and examined the literature to conceptualize the landscape of the theoretical frameworks of MBI approaches, identified the positive design features of modeling-based learning environments with an emphasis on technology-enhanced ones, and identified the most effective MBI practices that are associated with successful student science learning through a meta-analysis. We also pointed out weaknesses in current MBI practices and research. MBI fits the dual need of improving the current situation of U.S. science education as well as nurturing the next generation of competitive science professionals. To help students better comprehend big ideas in science, MBI activities engage students in building scientific models via modern computer information technology as well as hands-on materials, employ multiple ways of representations to accommodate diverse learners, and facilitate a collaborative community as students build models together, communicate their models to others, and evaluate alternative models. In spite of the accumulation of research on MBI, the effective application and the scaling-up of MBI are thwarted by fundamentally different theoretical frameworks, difficulties in applying modern technologies, and a lack of evidence of how various MBI factors work together to contribute to student learning. Through synthesis and meta-analysis, ACMBI builds a systematic and analytic framework to conceptualize the assessment of MBI, recommends best design strategies of technology-enhanced modeling environments, and identifies the MBI elements that result in positive student learning in science. Some significant findings are highlighted here. Overall, our metanalaysis indicates that most empirical MBI studies indicated a positive effect size on the impact of using MBI on students’ outcomes (the effect sizes ranged from -0.94 to 6.01). The average effect size for the qualified studies was 1.33, which indicates an overall large effect of MBI on student learning. The variables that had the most positive contribution in terms effect size included Inquiry Level, Theoretical Description, and Embedded Scaffolding. Furthermore, technologies were widely used in MBI interventions, and it has an overall positive effect on science learning performance of K-12 students. Student outcomes in technology-supported MBI programs were reported higher than those in MBI without technology support. More than half of these technology-enhanced MBI environments are highly interactive. However, two other important design features of technology used in MBI, Embedded Collaboration and Embedded Scaffolding, were less discussed in MBI literature. We also revealed that although relatively a small number of studies assessed student learning in terms of modeling in the MBI literature, a diverse set of strategies were used in these studies to examine student modeling ability. We also noticed that within the quantitative studies, a common problem is the lack of reporting sufficient statistics for a metaanalysis. We believe our project creates a synergy for the science education community to better understand and to effectively implement and assess modeling-based science teaching and learning. We have created a database that compiled all the relevant MBI literature from 1980 to 2010 and disseminated it through a website that allows public access. Our work can benefit education researchers (e.g., new analytic frameworks and MBI database), practicing teachers (best strategies to implement MBI with or without technology), and policy makers (what is needed in terms assessment of modeling, a core practice proposed by the new framework for K12 science education).
