Portfolio of Allen Downey at Olin College
Allen Downey
January 2021
Overview of my professional activities in 2020.
I am continuing work on an undergraduate curriculum in Data Science for use at Olin and other schools.
In Spring 2020, I taught Data Science at Olin in parallel with the first offering of DS10 at Harvard. Both classes were based on a new online curriculum I developed in collaboration with faculty at Harvard, called “Elements of Data Science”.
The curriculum includes two case studies on:
In Fall 2020, I taught Data Science at Ashesi University in Ghana, as part of their Archer Cornfield Fellowship.
My online course, “Exploratory Data Analysis in Python”, is ongoing. At this point, more people have taken this course online than I have taught, in person, in my academic career.
From July 2019 to July 2020 I was a member of DataCamp’s Instructor Advisory Board.
In 2020 I published several blog articles related to Data Science, including:
In November I led a half-day workshop, “Exploratory Data Analysis with Pandas and Matplotlib” for PyData Global 2020.
As part of my work on Bayesian Statistics for undergraduates:
I worked on a major revision of Think Bayes. The second edition will be published by O’Reilly Media in 2021.
I started work on Bite Size Bayes, an introduction to Bayesian statistics designed to be incorporated into existing statistics classes, which I hope will make it easier to adopt than a completely new class. The Python version is nearly done; I plan to translate it into R.
As a classroom activity, I designed a dice game called The Shakes that demonstrates Thompson sampling, a strategy for Bayesian medical testing.
In October I led a half-day workshop, “Bayesian Statistics Made Simple” for the Open Data Science Conference (ODSC West) 2020, and presented “The Bayesian Zig Zag” for the first annual PyMC Conference.
I collaborated with The Carpentries to develop a new curriculum on computational tools for large astronomical datasets (supported by funding from the American Institute of Physics and the American Astronomical Society). We taught a pilot workshop based on this curriculum in November and debuted in January 2021 at the annual meeting of the American Astronomical Society.
I have published the astronomy curriculum as an online book, Astronomical Data in Python.
I am continuing work on
Physical Modeling in MATLAB, to be published by No Starch Press in 2021, and
In March I revised Think DSP and released version 1.1.
In September I presented a talk for the first DSP Online Conference: “In Search of The Fourth Wave”.
All of my books are available under free licenses that allow readers to copy, modify, and redistribute them. These licenses make possible many translations, adaptations, and derivative works, and lead to collaborations with co-authors all over the world.
In 2020, new translations of my books include:
Think Java in Polish (Helion).
Think Python in Greek (Klidarithmos Publications) and Spanish
Think Julia in Korean (Hanbit Media), simplified Chinese (China Electric Power Press), traditional Chinese (GoTop Information Inc.), Spanish, Brazilian Portuguese, and Persian.
Think DSP in Italian
Think Complexity in simplified Chinese (China Machine Press).
Because my books are freely available, they are used in classes all over the world. Since my last report, I have heard from people using them at Allegheny College, A.L. Brown High School, The Athenian School, Carleton College, The Citadel, Dallas College, De Anza College, Faith Academy, Harper College, Longwood University, Merced College, Miami Dade College, Minnesota State University, Miss Hall’s School, Missouri Western State University, Pierce College, Prince George’s Community College, Santa Monica College, Simon Fraser University, Temple University, Trident University, UCLA, UCSC, University of the People, University of Wisconsin, CETYS University (Mexico), Chang Gung University (Taiwan), Istanbul Data Science Academy (Turkey), National University of Engineering (Peru), and the University of Applied Sciences Upper Austria.
In all of my projects, I “release early and release often”, getting feedback from readers and inviting collaboration. Following the principles of Open Science, I publish data and code along with research results, allowing others to reproduce and modify my analysis. And I don’t publish in venues that do not make papers and supporting materials freely available.