Behind the Data

Political polls are in our faces these days whether we want them or not -- especially if you're in the US during election season. We talk with polling expert and database journalist Dhrumil Mehta of Columbia University (formerly Nate Silver's FiveThirtyEight) about how to think about polls. Are they all just noise? Do they tell the future? Something in the middle? Join us to explore where political polls come from, how to evaluate them, and how to make the most of the information they offer.

Materials referenced in the show:

• Example polling tracker from Nate Silver (Dhrumil and Andrea did not work on this version directly, but it is a "descendant" of a model they did work on.)
• Nate Silver article, "The Media Has a Probability Problem", FiveThirtyEight
• Harry Enten article, "Fake Polls are a Real Problem", FiveThirtyEight
• Harry Enten article, "Trump is Just a Normal Polling Error Behind Clinton", FiveThirtyEight
• Dhrumil's polls LLM: http://pollfinder.ai

More about Andrea here and on Instagram. More about Dhrumil here and on X.

Hosted on Acast. See acast.com/privacy for more information.

We take it as a given that Americans are politically polarized, but how do we actually know if empirically this is the case? We talk with Prof. Patrick Egan (NYU) about how we can quantify something as abstract as a political attitude, why doing so helps us understand polarization, and how all of this helps reveal opportunities where we can make progress on areas where we're most divided -- such as climate change.

Explore Pat's research and writing: https://wp.nyu.edu/egan/.

Papers and resources mentioned in the episode:

• An example of Pat's work on issue ownership is here.
• The data Pat mentioned on Americans' political attitudes since 1948 is from the American National Election Studies (ANES), which you can explore for free here.
• An example of measuring leaders' ideologies based on their roll call votes is here.
• An example of measuring ideology based on campaign contributions is here.
• Learn more about Hanna Pitkin's concept of representation in her 1972 book The Concept of Representation (helpful summary here).
• Pat's 2024 climate change paper (with Megan Mullin) is US partisan polarization on climate change: Can stalemate give way to opportunity? (appeared in PS: Political Science and Politics 57(1): pp. 30-35).
• BTW: the adage that states that headlines that pose a question tend to have the answer "no" is Betteridge's law of headlines and it's very fun.

Follow Andrea at @jonesrooy on Instagram and/or learn more at jonesrooy.com. Be sure to check out our partner show The Daily Tech News Show!

Hosted on Acast. See acast.com/privacy for more information.

ALS is a fatal motor neuron disease that has no cure and is estimated to affect hundreds of thousands of people worldwide. Why is a cure, or even a meaningful treatment, so elusive? We talk with data engineer, scientist, and rare disease advocate Danielle Boyce at ALS TDI about her work helping us all better understand this terrible illness. While the topic is grim, Danielle provides a lot of hope -- as well as inspiration to all of us to get involved in solving problems through data.

Follow Danielle and her amazing data & statistics tips on LinkedIn: https://www.linkedin.com/in/data-danielle/.

Find out more about ALS Therapy Development Institute (ALS TDI): https://www.als.net/ (and donate if you like!).

Two other resources Danielle mentioned to get involved: https://ohdsi.org/ and https://www.geoals.org/.

Follow Brooke Eby on the various social medias: @limpbroozkit.

Check out my father Robert Rooy's film about our friend John Godinet: https://www.lovingjohnmovie.com/ and watch the trailer here!

Follow me at @jonesrooy and https://www.jonesrooy.com/. Behind the Data is proud to be part of the Daily Tech News Show ecosystem. Special thanks to Tom Merritt and our producer Roger Chang!

Hosted on Acast. See acast.com/privacy for more information.

When we think of data and movies the first (and maybe only) thing that comes to mind for many of us is movie reviews. But there is so much more we can learn about movies and ourselves if we just think outside the box a little bit and apply data and scientific thinking in creative ways. We talk with Walt Hickey, author of the book You Are What You Watch: How TV and Movies Affect Everything, about all kinds of exciting applications of data to track how movies affect us physically and emotionally, the value of exporting (and importing) culture internationally, whether superpowers predict evilness, how the plots of many movies have changed over time, why we like what we like, and how movies can change our entire life trajectory. Plus, prepare to learn why we should all watch Titanic again, as well as maybe consider (gasp) ignoring reviews of movies altogether.

Relevant materials and links mentioned:

• Buy Walt's book
• Subscribe to Numlock
• Check out Sherwood News
• More about Walt
• Follow Walt on X

More on Andrea here and on Instagram.

Hosted on Acast. See acast.com/privacy for more information.

Crime data is all over the news these days and, at least in the United States, it seems like you can tell any story you want to about whether crime is going up or down and whose fault it is. How should we be thinking about this data to figure out of what's really going on? We speak with Jeff Asher, crime data analyst and co-founder of JH Datalytics, which recently launched the Real-Time Crime Index, about where crime data comes from, why everyone seems to be disagreeing about it, and how to think and talk about it more clearly.

Materials referenced in the show:

• Jeff's company: https://www.ahdatalytics.com/
• Jeff's substack: https://jasher.substack.com/
• Real-Time Crime Index: https://realtimecrimeindex.com/
• One of Jeff's articles about the crime data mess of a few years ago.

Andrea is here and on Instagram.

Hosted on Acast. See acast.com/privacy for more information.