Data Curious 12.02.2019 - Global warming charts, dress codes, and a look at PyViz
12.02.2019
How can we be smarter about how we tell stories with data? It's a question I think about almost daily.
Scrolling through Twitter, I see countless examples of people using data to find and tell stories. The difference between the average ones and the great ones are often in the details.
This edition of Data Curious contains a few examples of exceptional data storytelling. It also includes links to the data behind these stories, so you can reinvent and discover a story for yourself. Happy digging!
Read_
Reproducible R charts, global warming charts, and data viz with audio
How do newsrooms create reproducible workflows for producing charts and graphcs?
Read this behind the scenes article on how the BBC Visual and Data Journalism team works with graphics in R—from raw data all the way to web-ready graphics.
Read on →
How should we interpret charts about global warming?
Line charts can be deceptive depending on the scale they use. This Datawrapper blog takes a look at how to accurately interpret the "hottest year on record" charts that have been flying around recently.
How can I use audio to tell a story with data viz?
The Flourish team just introduced a new feature to allow audio playback alongside their data-slides. It allows you to record an explanation of your data for people to listen to as the animation plays. Interesting to see where this one goes, as it seems to be the first intentional feature that allows for an audio walkthrough.
Explore_
People having kids, music tastemakers, and dress codes
How many kids do people have? When do they have them?
Animated steppy charts, ftw. From FlowingData.
Why do radio stations play hits way after they spike on Spotify charts?
The DataFace team dive into song chart data to explain why streaming services have replaced radio DJs as the new "tastemakers" of music.
Groovy →
How are schools enforcing dress codes in the US? And what is it doing to how we perceive students?
The Pudding opened up 2019 with a huge investigative story: "The sexualized messages dress codes are sending to students." Using original data they collected and coded, it breaks down how schools are sending mixed messages about students' bodies (especially with girls).
Analyse_
The Next Bechdel test and school dress codes
Are films becoming more diverse?
FiveThirtyEight recently published a story analyzing films that pass the Bechdel test. The test asks two questions of a film: Does it have at least two named female characters? And do those characters have at least one conversation that is not about a man? Most films fail this test. So the FiveThirtyEight team created a new set of tests, with new criteria, to judge the top 50 movies of 2016. And you can analyse their data below.
The New Bechdel data →
What dress codes are still enforced for girls and boys? Are they held to the same standard?
No. Remember that Pudding story I just told you about? They released all the data behind the piece on Github for you to download at will. It's a gold mine.
Learn_
Jupyter Notebook power-ups, quick data merging, and publishing charts with Flask
How do I choose a visualization library when working with data in Python?
Matplotlib. Seaborn. Bokeh. Altair. How to choose? Checkout PyViz. As a central visualization ecosystem, it looks like a really promising development in interactive data analysis. PyViz lets you create interactive charts and dashboards with a few lines of code for quicker EDA.
Which library should I use for NLP?
Stanford released a new library called NLP. It allows you to "convert a string containing human language text into lists of sentences and words, to generate base forms of those words, their parts of speech and morphological features, and to give a syntactic structure dependency parse".
Spread the love.
Share this edition of Data Curious.
Thanks for reading. More to come next week.
Ben_ say hi: twitter | medium | github