ABOUT
This site is currently under development by Julia Koschinsky and Jay Cordes. We aim to bring together people, teaching materials, and other resources that use scientific reasoning as the logic for how data are analyzed. The audience we have in mind are educators engaged with a general introductory undergraduate curriculum but we try to make the site accessible to anyone interested in the topic.
There are three more specific reasons why we built this site: First, to frame the problem and highlight potential solutions. Next, to have a home for the open teaching materials we both plan to release throughout 2023. And, finally, to bring together the work of others at the intersection of scientific reasoning and data science.
We connected after Julia moderated this panel on “Teaching Data Science with Scientific Reasoning” at U.C. Berkeley’s national Data-8 workshop (June 2022) and Jay was preparing his talk on “Putting the “Science” into Data Science” at Claremont McKenna College (Nov. 2022).
Julia Koschinsky is a geospatial data analyst who has been working in academia for over 20 years. She is the Executive Director and Senior Research Associate at the University of Chicago‘s Center for Spatial Data Science (CSDS). CSDS develops the open spatial software GeoDa (over half a million downloads), and hosts spatial analytics lectures with over 800,000 views on its YouTube channel.
Julia earned a PhD in urban planning with a specialization in spatial data analysis (with Luc Anselin) from the University of Illinois at Urbana-Champaign and received Masters degrees in quantitative social sciences from SUNY-Albany and the Free University of Berlin.
Her goal is to better integrate (spatial) data science with scientific reasoning to achieve: 1) Fewer mechanical applications of methods and more engagement in the thrill of discovery; 2) greater consideration of descriptive statistics in the context of explanation; 3) fewer technical errors and biases, like confirmation bias, through more rigorous testing; and 4) more critical thinking and less social and racial biases while analyzing data.
Contact: spatial@uchicago.edu
Jay Cordes has been working as a data scientist in industry for over 20 years. He co-authored the book The 9 Pitfalls of Data Science with Pomona economist Gary Smith to help guide future data scientists away from the common pitfalls he saw in the corporate world. The book won the 2020 PROSE award in the category Popular Science and Popular Mathematics. He also co-authored The Phantom Pattern Problem, also with Gary Smith. Through his work, he hopes to improve the public’s ability to distinguish truth from nonsense.
Jay earned a degree in mathematics from Pomona College and more recently received a Master of Information and Data Science (MIDS) degree from U.C. Berkeley. He writes “electronic music with color” in his free time. His albums Excavations and Jacqueline Remixed are an ideal soundtrack for computer programming.
He believes that maintaining a skeptical mindset will keep you vigilant for the “silent evidence of failures” that distorts statistical significance. For data science to work, you need to think and work like a scientist.
Contact: jjcordes@ca.rr.com