Episode 02: How to Find & Represent Seasonality in Data - Featured Viz by Nadieh Bremer & Zan Armstrong

Nadieh Bremer. Image via  Twitter .

Nadieh Bremer. Image via Twitter.

Zan Armstrong. Image via  LinkedIn

Zan Armstrong. Image via LinkedIn

Welcome to episode 2 of Data Viz Today. How can you identify seasonality in your data? How can you best visualize it? I share Nadieh Bremer and Zan Armstrong's data viz about the seasonal trends in U.S. births to show how to find and represent seasonality in data.

Listen on Apple Podcasts, Google Play, Google PodcastsStitcher, SoundCloud & Spotify.

"Baby Spike" by Nadieh Bremer and Zan Armstrong. Image via  VisualCinnamon.com

"Baby Spike" by Nadieh Bremer and Zan Armstrong. Image via VisualCinnamon.com

  • Welcome! I'm Alli Torban.

  • 00:30 - Featured viz by Nadieh Bremer and Zan Armstrong published in the Scientific American.

  • 00:40 - Today's question: How do we find seasonality in data and how can we visualize it in a beautiful way?

  • 00:50 - Nadieh Bremer is a Freelance Data Visualization Designer from Amsterdam, and Zan Armstrong is a Data Visualization Engineer and Data Analyst in San Francisco.

  • 01:10 - What seasonal time trends are visible in data about U.S. births?

  • 01:20 - The viz began with Zan's OpenVis Conference talk called "Everything is Seasonal".

  • 02:05 - Nadieh and Zan began collaborating in person on how to best display the CDC's data on seasonal trends in U.S. births for Scientific American story.

  • 02:25 - Because the data is cyclical, they decided to start with a radial area chart.

  • 02:45 - To make the chart easier to understand, they decided to draw a circle that represents the average number of babies born over a certain period of time, and then show the data rise above and below that circle.

  • 03:25 - To smooth the lines of the data around the circle, Nadieh used a technique called the Loess curve.

  • 04:00 - People tend to like circular things.

  • 04:10 - Study shows that people prefer images with curved edges rather than sharp edges, and the images with curved edges produced increased and stronger activity in the brain.

  • 05:10 - Nadieh and Zan email back and forth to decide on final touches.

  • 05:25 - Nadieh estimated she spent about 25 hours on her parts of this viz. She used R and Adobe Illustrator.

  • 06:00 - In the article, there are three radial area charts showing trends over time at various levels of granularity.

  • 06:15 - Being able to immediately find yourself in the data makes this viz even more compelling.

  • 06:35 - My top 3 takeaways from this viz:

    • Reach out to someone and try a collaboration!

    • If you have data over time, check to see if there is seasonality in your data. You can do this for free with no coding experience with Tableau Public. Here is a short tutorial how to do it. Warning: be mindful of aggregating your data by month. More details in Zan's talk.

    • Think about using curved lines in your data viz, especially if your goal is beauty or art focused.

  • 09:00 - Thanks to Nadieh and Zan for the inspirational data viz!

  • 09:20 - Connect with Nadieh on her website or on twitter.

  • 09:30 - Connect with Zan on her website or on twitter.

  • 09:40 - Next featured viz is by Pierre Massat!

  • 10:00 - Subscribe to my weekly newsletter, which compiles the top tips from the week into a short email.

  • 10:10 - Nominate a data viz to be featured on the show.

  • 10:15 - Contact me on social media @dataviztoday!

  • 10:20 - These shows are a little longer than I initially intended. What do you think? Shoot me a message!

New episodes are released every Tuesday at 6 AM EST.