Search for: All records

We develop, advance, and promote a previously existing framework called the Qualitative-Quantitative-Qualitative workflow (Q1Q2Q3, pronounced “Q-Q-Q”) to systematically guide the content of interdisciplinary collaborations and improve the teaching of statistics and data science. The Q1Q2Q3 workflow is designed to help statisticians and data scientists develop skills and techniques for collaboration to work with domain experts across academic fields, industry sectors, and organizations. The Q1Q2Q3 workflow explicitly emphasizes the importance of the qualitative context of a project, as well as the qualitative interpretation of quantitative findings. We explain Q1Q2Q3 and provide guidance for implementing each stage of the workflow. We describe how we teach Q1Q2Q3 within a statistics and data science collaboration course and present data evaluating its effectiveness. We also describe how Q1Q2Q3 can be useful for educators teaching introductory, projects-based, and technical statistics and data science courses. We believe that the Q1Q2Q3 workflow is an easy-to-implement technique that is beneficial and necessary for statistics and data science education and practice. It can be used to weave ethics into each stage of practice so that statisticians and data scientists can successfully transform evidence into action for the benefit of society. 

Despite growing calls to develop data science students’ ethical awareness and expand human-centered approaches to data science education, introductory courses in the field remain largely technical. A new interdisciplinary data science program aims to merge STEM and humanities perspectives starting at the very beginning of the data science curriculum. Existing literature suggests that humanities integration can make STEM courses more appealing to a wider range of students, including women and students of color, and enhance student learning of essential concepts and foundational reasoning skills, such as those collectively known as data acumen. Cultivating students’ data acumen requires a more inclusive vision of how the knowledge and insights generated through computational methods and statistical analysis relates to other ways of knowing.