An official website of the United States government
Abstract The ever increasing popularity of machine learning methods in virtually all areas of science, engineering and beyond is poised to put established statistical modeling approaches into question. Environmental statistics is no exception, as popular constructs such as neural networks and decision trees are now routinely used to provide forecasts of physical processes ranging from air pollution to meteorology. This presents both challenges and opportunities to the statistical community, which could contribute to the machine learning literature with a model‐based approach with formal uncertainty quantification. Should, however, classical statistical methodologies be discarded altogether in environmental statistics, and should our contribution be focused on formalizing machine learning constructs? This work aims at providing some answers to this thought‐provoking question with two time series case studies where selected models from both the statistical and machine learning literature are compared in terms of forecasting skills, uncertainty quantification and computational time. Relative merits of both class of approaches are discussed, and broad open questions are formulated as a baseline for a discussion on the topic.
more »
« less
New viruses are inevitable; pandemics are optional—Lessons for and from statistics
Abstract We explore ways in which statistics can be used to understand disease spread and support decision‐making by governments. “Past performance does not guarantee future results”—we hope. We discuss and show examples from the National Science Foundation (NSF)‐funded COVID‐Inspired Data Science Education through Epidemiology (CIDSEE) project. Throughout, the emphasis is on the relationships between evidence, modeling and theorizing, and appropriate action. Statistics should be an essential element in all these aspects. We point to some “big statistical ideas” that underpin the whole process of modeling, which can be illustrated vividly in the context of pandemics. We argue that statistics education should emphasize the application of statistics in practical situations, and that many curricula do not equip students to use their understandings of statistics outside the classroom. We offer a framework for curriculum analysis and point to some rich teaching resources.
more »
« less
- Award ID(s):
- 2313212
- PAR ID:
- 10524345
- Publisher / Repository:
- Wiley-Blackwell
- Date Published:
- Journal Name:
- Teaching Statistics
- Volume:
- 46
- Issue:
- 3
- ISSN:
- 0141-982X
- Format(s):
- Medium: X Size: p. 132-140
- Size(s):
- p. 132-140
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Due to the applied nature of statistics and data science, many educators in these fields recognize the need to teach their students how to be effective interdisciplinary collaborators. Some prior research considers different approaches to teaching interdisciplinary collaboration skills. However, missing from this literature are the connections between teaching collaboration and education theory. Thus, there is a lack of understanding about why the various pedagogical approaches may be effective. In this descriptive study, we describe an approach to teaching interdisciplinary collaboration using a Community of Practice (CoP) and highlight connections between potentially reproducible elements of this approach and education theory that explains why this approach may be effective from the perspectives of both education and collaboration theory. Our results show that students and content-area experts recognize this approach to teaching statistical and data science collaboration to be effective. By grounding our methods for teaching statistics and data science collaboration skills in education theory, we focus attention on which aspects can be replicated in other contexts, why they work well, and how they can be improved. We recommend instructors intentionally create a CoP within their courses, encourage peer mentorship, and emphasize a growth mindset.more » « less
-
Abstract Science educators are integrating more and more computational thinking (CT) activities into their curricula. Proponents of CT offer two motivations: familiarizing students with a realistic depiction of the computational nature of modern scientific practices and encouraging more students from underrepresented backgrounds to pursue careers in science, technology, engineering, and mathematics. However, some studies show that increasing exposure to computing may not necessarily translate to the hypothesized gains in participation by female students and students of color. Therefore, paying close attention to students' engagement in computationally intense science activities is important to finding more impactful ways to promote equitable science education. In this paper, we present an in‐depth analysis of the interactions among a small, racially diverse group of high school students during a chemistry unit with tightly integrated CT activities. We find a salient interaction between the students' engagement with the CT activities and their social identification with publicly recognizable categories such as “enjoys coding” or “finds computing boring.” We show that CT activities in science education can lead to numerous rich interactions that could, if leveraged correctly, allow educators to facilitate more inclusive science classrooms. However, we also show that such opportunities would be missed unless teachers are attentive to them. We discuss the implications of our findings on future work to integrate CT across science curricula and teacher education.more » « less
-
IntroductionThis study was undertaken to explore the potential of developing a working theory of improvement for creating a more equitable system of science education at the level of a US state. We ask: How can tools from a long-term research-practice partnership support a state team in initiating improvement research toward promoting a more equitable system of science education? MethodsThis design study took place in winter 2024 in a single state. External partners supported leaders of a single state in the US Northeast to support a process of articulating aims, specifying primary and secondary drivers, and identifying change strategies to promote a more equitable system of science education in the state, grounded in the vision ofA Framework for K-12 Science Education(National Research Council, 2012). In this paper, we rely on descriptive analyses of joint meetings and a focus group with state leaders describe the tools supporting the process of development, the team’s use of the tools to generate an early draft of the Driver Diagram, and issues surfaced while developing it with a team of interest holders in the state. ResultsTwo meanings of equity emerged as significant within the series of meetings: that of the importance of universal access to professional learning and the importance of students having opportunities to experience culturally relevant instruction. The issues surfaced highlighted the need for infrastructures for professional learning to reach a diverse group of interest holders in science, including teachers, school leaders, and district leaders across the state. They also saw curriculum materials that connect to students’ everyday lives and community priorities as key drivers for equitable change in the system, around which professional learning activities should be organized. The team also surfaced several policy changes needed to implement change strategies, only some of which team members felt they had some authority. DiscussionWhere past researchers have observed that equity can disappear as a focus during implementation of reforms, this study found that developing an aim statement and driver diagram helped energize and refocus a team’s implementation efforts geared toward a vision for science teaching and learning that is focused on ensuring all students can engage in meaningful science learning that is culturally and locally relevant to them.more » « less
-
Abstract Machine learning (ML) has become commonplace in educational research and science education research, especially to support assessment efforts. Such applications of machine learning have shown their promise in replicating and scaling human‐driven codes of students' work. Despite this promise, we and other scholars argue that machine learning has not yet achieved its transformational potential. We argue that this is because our field is currently lacking frameworks for supporting creative, principled, and critical endeavors to use machine learning in science education research. To offer considerations for science education researchers' use of ML, we present a framework, Distributing Epistemic Functions and Tasks (DEFT), that highlights the functions and tasks that pertain to generating knowledge that can be carried out by either trained researchers or machine learning algorithms. Such considerations are critical decisions that should occur alongside those about, for instance, the type of data or algorithm used. We apply this framework to two cases, one that exemplifies the cutting‐edge use of machine learning in science education research and another that offers a wholly different means of using machine learning and human‐driven inquiry together. We conclude with strategies for researchers to adopt machine learning and call for the field to rethink how we prepare science education researchers in an era of great advances in computational power and access to machine learning methods.more » « less
