Teaching and Evaluating Data Communication At Scale

Effective statistical communication is an essential skill for the modern data scientist. In this talk, I will share three effective activities that have been used and adapted for course levels ranging from early undergraduate through capstone through Master’s level in preparation of portfolio exams. We will cover assignments based in statistical communication, covering formal peer review, reproducibility, and presentation that, while not directly covering statistical theory or methodology, require students to think critically about such material, apply it to their own work, and become better statisticians as a result. I hope that through the course of this talk, attendees are inspired to use such activities in their own pedagogical practice.

This session explores strategies for fostering transfer of written communication skills and integrative thinking across institutional and disciplinary boundaries. In “How Does Expansive Framing Promote Transfer?” Randi A. Engle and her co-authors discuss two types of instructional framing, or ways of setting up learning expectations: expansive framing and bounded framing. Whereas expansive framing encourages students to see multiple applications for learning, bounded framing constrains learning to a single task, test, or class. In contrast to narrow and siloed approaches to instruction, K16 collaborations support the development of disciplinary and transdisciplinary literacies high in transfer value. These include strategies for critical reasoning, rhetorical writing, and independent problem solving. Session participants examine resources and activities developed through two grant-funded K16 collaborations, using examples from math, social science, science, and humanities courses.

It has become increasingly  clear why the Humanities  faculty need to collaborate with STEM faculty.  But is the opposite the case?  What do Humanities faculty have to offer STEM instructors?  The answers are multiple but I propose to discuss one topic: writing.  In a world where communication skills are more and more valued, STEM students need to learn how to write, to communicate clearly and articulately about their projects and problems .  They must have the written skill set to explain their data, to tell a story about their data, to persuade others that their discoveries are consequential (not to mention excelling at grant writing).  It is not sufficient, for example, to solve a mathematical problem.  You need to explain the process  by which you came to the solution. Numbers speak but they do not speak for themselves. In this lecture, I propose to explore three interelated themes:  1) A Brief History  of Writing Pedagogy in the Sciences; 2) Creating a Syllabus for Teaching STEM Writing;  3)  Best Practices In Writing Pedagogy and Practical Advice on Assignments, Assessments, and Research.

With the use of statistics in a wide variety of disciplines, it is important that students learn how to effectively communicate statistical results. In this talk, I will share how communication is implemented as a learning objective in an undergraduate regression analysis course. I will present in-class activities and assignments designed to help students think critically about how they interpret and communicate statistical results, and I’ll discuss how students are assessed. I’ll conclude with the challenges of assessing communication in a large statistics course and share lessons learned.

Writing-to-Learn (WTL) is a practice that involves writing assignments specifically formulated to foster conceptual understanding and disciplinary thinking. When properly implemented, WTL has been shown to not only foster conceptual learning, but also support development of scientific reasoning, promote metacognition, and elicit student thinking. While the positive impact of WTL is well-established, systematic barriers inhibit its implementation at a large scale. The most effective WTL assignments are meaning-making tasks with clear writing expectations that include a metacognitive and interactive writing process. At the University of Michigan, we designed and executed a program called MWrite that makes implementing WTL accessible to instructors teaching large-enrollment courses. We provide instructors with support to design effective WTL assignments and further support their implementation of the assignments through specialized undergraduate teaching assistants known as writing fellows and an interactive, automated peer review process for students. For the past 7 years, we supported WTL in 100+ sections of 15+ courses, impacting over 23,000 students submitting over 53,000 assignments and producing a database of over 800,000 peer review comments. The research team has studied the impact of the program in several courses and demonstrated that the assignments support the intended learning goals and that the design elements of the assignments function as intended (e.g., students can provide constructive, content focused feedback to their peers). Furthermore we have analyzed students’ writing and peer review using machine learning and explored the utility of machine learning for providing feedback to both students and instructors. We will discuss how MWrite is structured and the associated research aimed at understanding whether and how the MWrite design is effective and ways in which it could be improved.

Data scientists and statisticians have essentially identical skill sets. Yet over the next decade, the U.S. Bureau of Labor Statistics predicts there will be nearly six new data science jobs for every new opening in statistics. In this talk, I will first argue that the preference for data scientists is due to a difference in mindsets. Data science emphasizes collaboration and communication, which appeal to employers despite being historically undervalued by statisticians. I will then argue that this mindset can be taught at scale through data science laboratories (like those in the physical sciences) in which students address important campus problems by collecting data and communicating the results to campus experts who can evaluate the effectiveness of the solution. I will conclude with three examples from my own teaching: working with the campus transportation office to estimate the number of traffic violations, working with the campus diversity office to identify barriers to inclusion, and working with the campus sustainability office to monitor the local ecosystem.