DS 1001 Active Learning Modules

Preface

This site contains three active learning modules for an introductory undergraduate course on programming for data science.

These modules have been developed as part of the Data Science Active Learning Lab (DSALL) project, sponsored by the President and Provost’s Fund for Institutionally Related Research at UVA. The project is headed by Professor Brian Wright, in collaboration with Professor Pete Alonzi, of the School of Data Science at UVA.

Motivation and Design

DSALL is motivated by the observation that active learning approaches (AL) have not been widely applied and tested in data science courses, even though a body of significant research demonstrates the benefits of AL and many instructors of computer science have adopted AL successfully.

A key element of AL is the design, development, and testing of authentic and transparent assessments.

Authentic assignments ask students to either inhabit a role seen in the discipline or do the work of the discipline.

Transparent assignments explicity state their purpose, tasks, and criteria. Research has shown that transparent assignments improve a wide range of student success metrics, including a sense of belonging which can be a significant factor in the retention of underrepresented student populations, particularly in STEM.

A major obstacle to creating authentic and transparent active learning assessments is the time and effort required to develop, implement, and test them. DSALL will provide the resources and structure necessary for collaborative course development. The Lab will enable faculty and graduate students to work together to develop and empirically test active learning activities in their data science courses. This will result in the identification and validation of best practices along with practical tools that can be shared broadly and openly across the field of Data Science.

The Lab will reduce the time burden on faculty to create effective assessments. More important, it will establish a trusted research framework and tools that have been empirically tested for effectiveness. Such a resource is currently unavailable in data science pedagogy. We also have a unique opportunity, given the newness of the field, to develop cultural norms around teaching that are known to benefit underrepresented groups and first-generation students. This funding will go a long way to advance that effort by establishing a pedological research lab specifically focused on student success in a historically fast-growing field.

Project Goals

In pursuit of the mission to provide resources to enable faculty to create effective active learning assessments for data science, the project has the following goals.

• Be a preeminent resource for empirically tested best practices in data science education.

• Establish a testing and validation framework for data science oriented active learning tools.

• Create and socialize teaching norms in data science that are proven to benefit underrepresented groups and first-generation students.

• Develop and test active learning assessments (labs) specifically for courses in the proposed data science undergraduate major. (Projected courses include the Foundation of Data Science, Foundations of Machine Learning, Computation Probability, Data Science Ethics and Policy, and Data Science Systems.)

• Disseminate open instructional content in accordance with the School of Data Science open access policy and UVA’s Open Scholarship agreement.