Share

cover art for Behind the Data

Behind the Data

The art and science of understanding the world with data


Latest episode

  • 7. The Art and Science of Measuring Democracy

    43:31||Season 1, Ep. 7
    Is democracy in decline around the world? This is a major question on many of our minds, and luckily for us there are very smart, thoughtful people working on tracking (as well as explaining, understanding, and predicting!) exactly this. We are joined by professor Brigitte Seim, who is a project manager for the Varieties of Democracy (V-Dem) dataset, which is one of the biggest and most elaborate efforts to track democracy around the world to date. We discuss how to think about turning something as abstract and unwieldy as "democracy" into a number, how to make sense of and learn from those numbers, and about her own research on the growing importance of measuring toxic political speech online to understanding the health of democracy.This is a great episode for anyone who cares about democracy or regime type broadly, but also for anyone interested in the beautiful art and science of measurement (one of my favorite topics ever!), as well as for anyone interested in learning about a dataset that is a great place to start if you're looking for some data to practice and work with in your own curiosity-based data science journey. Enjoy!Learn more on Brigitte here and check out her prolific research here. Read about and explore the Varieties of Democracy dataset at v-dem.net.Learn more about Andrea here and on Instagram at @jonesrooy.

More episodes

View all episodes

  • 6. What Can We Actually Learn from Election Data?

    46:19||Season 1, Ep. 6
    Approximately ten billion different narratives about what happened in the recent US presidential election have emerged since said election. But how many of these narratives can be substantiated by data? How can we tell?Political scientist Seth Masket joins us to discuss how to make thoughtful inferences from this data, how and why to be skeptical of evidence-free opinions, and the importance (just like in the last episode!!) of being thoughtful about whether our narratives are based on data or we (or others who are trying to persuade us of something) are simply trying to pick data to support our narrative(s). Stick around until the end to hear how polls did and Seth's thoughts on why everything feels so politically bad these days.Follow and read more by Seth:His substack: smotus.substack.comHis podcast: powerflour.substack.comHis personal website: sethmasket.comFollow Andrea at jonesrooy.com and on Instagram at @jonesrooy.
  • 5. How Doctors Use Your Data (and How AI Can Help)

    01:09:07||Season 1, Ep. 5
    Dr. Steven Novella, clinical neurologist and host of The Skeptics' Guide to the Universe podcast joins us to share how he collects and analyzes data when he works with patients (there is so much more going on than I realized!) as well as how AI can hopefully improve the practice of medicine if we get it right. It's a longer episode than usual, but I hope you'll agree it's well worth it. Follow Steve:https://www.theskepticsguide.org/https://sciencebasedmedicine.org/https://theness.com/neurologicablog/Follow Andrea:https://www.jonesrooy.com/ and @jonesrooy on socials.
  • 4. The (Often) False Promise of Personal Fitness Data

    39:57||Season 1, Ep. 4
    Science journalist and athlete Christie Aschwanden takes us inside the not-so-scientific world of the fitness and health data that's tracked by various apps we probably own and/or are tempted to buy, and she helps us more carefully discern what's actually useful to follow and what we're better off ignoring. Plus -- an existentially (and scientifically!) important discussion about the power of uncertainty (and why it doesn't have to feel bad).Christie's Scientific American podcast is Uncertain. Her book is Good to Go: What the Athlete in All of Us Can Learn About the Strange Science of Recovery. You can read more about her and by here on her website.Follow AJR, your host, at @jonesrooy on Instagram and find out more here.
  • 3. How Movies and TV Affect Us

    39:10||Season 1, Ep. 3
    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 bookSubscribe to NumlockCheck out Sherwood NewsMore about WaltFollow Walt on  XMore on Andrea here and on Instagram.
  • 2. Why We Can't Seem to Agree on Crime Statistics

    37:29||Season 1, Ep. 2
    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.
  • 1. Political Polls

    36:31||Season 1, Ep. 1
    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", FiveThirtyEightHarry Enten article, "Fake Polls are a Real Problem", FiveThirtyEightHarry Enten article, "Trump is Just a Normal Polling Error Behind Clinton", FiveThirtyEightDhrumil's polls LLM: http://pollfinder.aiMore about Andrea here and on Instagram. More about Dhrumil here and on X.