Allen Downey

Professor of Computer Science

Needham, MA

Allen Downey literally wrote the book, actually books, on data science (Think Stats, Think Bayes, Think Python) that are now must reads in the data science community. That's why we're thrilled to have him as part of our community!

Tell us about what you do professionally

I'm a professor at Olin College, which is an undergraduate engineering college in Needham, MA, founded in 1999 to help fix engineering education.  I teach software engineering, data science, and computational modeling.

How did you hear about DataKind?

I was thinking about starting something similar to DataKind in Boston so I looked around to see if anyone was already doing it and I found DataKind.

Tell us about your work with DataKind.

So far I have worked on the early stages of a few projects, talking with potential collaborators, reviewing data, and helping define projects for DataDives and DataCorps.

What inspires you to donate your skills to give back?

The fundamental appeal for me is DataKind's mission to use data to do good. The organizations DataKind works with are doing really exciting, inspiring work -- it's great to have a chance to help.

What is one of the most surprising things you’ve learned or seen in working with data?

I'm not sure how surprising it is, but the effect I found that got the most attention is the relationship between religious affiliation and Internet use.  It seems likely that Internet use caused a substantial fraction of the decrease in religious affiliation in the U.S. in the last 20 years.

What’s the most interesting data project you’ve seen recently?

PewResearch published a report recently about marriage rates in the U.S. and they found a record high fraction of people in the U.S. have never married. I've been looking at related data from the National Survey of Family Growth. I think there are some societal shifts going on here that we haven't understood yet.  I am very interested in exploring more.

What advice do you have for someone just getting started with data science?

Be a voracious learner: data science projects require people who can bring together skills and knowledge from many fields.  You can't be an expert in everything, so you have to learn fast, know something about everything, and work well with people who have complementary skills.

What blogs or articles do you love reading to stay up to date on all the data news?

I'm a regular on the Reddit statistics forum. I post my work there to get comments, and I try to answer questions. I get a lot of project ideas from messages that appear there. I read a lot of blogs occasionally, but almost none of them regularly.  Except, of course, my own: Probably Overthinking It.

When you’re not busy using data science to change the world, what do you like to do in your free time?

Along with teaching, my biggest project is a book series that presents topics in science and engineering using a computational, rather than mathematical, approach.  The one I am working on now is Think DSP, which is about Digital Signal Processing using Python. And my other big project is helping to take care of two daughters, who are 11 and 8 years old, and two cats, who are 4 and 18.