462: quantum of sollazzo

#462: quantum of sollazzo – 5 April 2022

The data newsletter by @puntofisso.

Hello, regular readers and welcome new ones :) This is Quantum of Sollazzo, the newsletter about all things data. I am Giuseppe Sollazzo, or @puntofisso. I’ve been sending this newsletter since 2012 to be a summary of all the articles with or about data that captured my attention over the previous week. The newsletter is and will always (well, for as long as I can keep going!) be free, but you’re welcome to become a friend via the links below.

I was on the Geomob podcast again, this time alongside the Countryle developers to talk about my Wardle game – basically a brilliant episode covering Wordle’s geo-clones. You can listen to it on the Geomob website, and on all your favourite aggregators.

Every week I include a six-question interview with an inspiring data person. This week, I speak with Francesco Piccinelli Casagrande, Content Manager at Evocon, whose work was featured on WIRED and POLITICO.

‘till next week,
Giuseppe @puntofisso

Six questions to...

Francesco Piccinelli Casagrande

Francesco is Content Manager at Evocon.

What is your daily data work like and what tools do you use?

There is no standard workflow. I usually ask myself a basic question (e.g., Do European Cohesion Funds work in Italy?), then I try to answer it using data. I try to collect as much data as I can, possibly via API. Then, I perform statistical analyses with R, visualize, and try to write a story that explains my findings. When I write for Wired Italy, I visualise data with Tableau. Otherwise, I do ggplot2 charts.

Tell me about a data project that you're proud of...

I am incredibly proud of this story I ran about Måneskin. First, I tried to understand what is special about them. Secondly, I asked myself if they belong more to the Italian or the Anglophone musical tradition. To answer my questions, I downloaded data from Spotify’s API (thanks to the developers of SpofityR – the R package that allows you to run Spotify data on R) and visualised the results on Tableau. It is a bit complicated as a story, but I like to think that readers are smarter than most journalists believe.

...and a data project that someone else did and you're jealous of.

I choose The Economist's analysis of more than 200,000 songs to see if English is declining in pop music. It is super interesting because it uses clustering techniques, advanced statistics (principal component analysis in mainstream media? Crazy stuff), and clever visualization solutions. Allowing people to listen to music in a chart defined a new standard for interactive data visualizations in digital newspapers. We can all learn a lot from that article.

If I say "dataset", you think of...

I think of a piece of marble you must carve to get a story. The lines of code I use are the chisel. The story is there, you just have to take the unneeded material away. Things become interesting when you realize that the piece of marble you have is not enough, but I guess this part is for another interview :D

Give someone new to data a tip or lesson you wish you'd learned earlier.

Always save your code and store it on GitHub, even if you think you are stuck. Maybe, in a few months, you want to go back to your idea, but you don’t remember what you did. No matter what, storing your code allows you to keep track of your work and formalise the lessons you learn over time. If anything, it will make you more productive: having scripts beforehand will prevent you from reinventing the wheel every time you open R-Studio.

Data is or data are...

Data is. Sometimes I think of data as a sort of philosophy I try to adapt to everything I do. Working with data makes you imagine things from an opposite point of view to demonstrate your theories (yes, I am talking about the null hypothesis). It doesn’t matter if you are into microeconomics or social media analytics: the principles of data at large are roughly the same. That’s also why I think of data as a unitarian philosophy that should permeate every aspect of the public sphere -- if not our day-to-day lives.

Topical

2022 Oscar Outfits as Public Health Graphs

Absolutely hilarious.

The State of the Nation: a 50-state Covid-19 Survey

This is an interesting project, through which a collective of researchers is seeking “to identify links between social behaviors and virus transmission, as well as and the impact of messaging and regulation on individual and community outcomes during this crisis.“

The U.S. has gone xxx days without a temperature record…

Great work at The Pudding, with this map that shows temperature records in a slightly different way. ““Daily high” records show that never-before-seen warm days are occurring year-round, not just in the heat of the summer months.”‘

The pandemic has been deadlier in red states

Axios analysis of CDC data shows that “COVID is killing more people per 100,000 in red states than in blue states, a reflection of GOP resistance to vaccines”.

Russian consumers are already feeling the cost of war

“Inflation is reaching heights not seen since the financial crisis of 1998”, shows The Economist.

Tools & Tutorials

Word2Vec Explained

[]”Explaining the Intuition of Word2Vec & Implementing it in Python”](https://towardsdatascience.com/word2vec-explained-49c52b4ccb71).
You probably know that Word2Vec is a little obsession of mine (some of its tools are used in my Parli-N-Grams tool), and this tutorial explains it well.

Frustration: One Year With R

I would described this Github repository by Reece Goding as an “anti-tutorial”, because it goes through the issues found in using R for a whole year. Or, as the repository’s description goes, this is “an extremely long review of R”.

A detailed guide to colors in data vis style guides

“How to create an organizational color palette“. Datawrapper’s Lisa Charlotte Muth continues her series of tutorials on good style in dataviz. In this article, she focusses her attention on some famous brands and their styling choices as a way into good palettes.

Data thinking

Stop aggregating away the signal in your data

“By aggregating our data in an effort to simplify it, we lose the signal and the context we need to make sense of what we’re seeing.“
A very good article, not least because of the beautiful example visualizations, created using Observable plot.

Dataviz, Data Analysis, & Interactive

Lightyears, mustard and diamonds

A dataviz of “the fantastical streetnames of Gothenburg”. (via Duncan Geere)

Car per domestic property by LSOA

Quite a FOI snatch by Owen Boswarva, who convinced the DfT to release how many cars are owned per property by Census geography rather than the (much less useful, I think) postcode aggregation used in the official release.
As last week’s “Six Questions” interviewee Alasdair Rae notes, there are some brilliant finds in the data, such as the “property” with 40,000 cars.

Mapped: Global Happiness Levels in 2022

“This map pulls data from the World Happiness Report to uncover the average happiness scores of 146 countries. It shows average scores from 2019 to 2021, and highlights which countries are the happiest—or unhappiest—and why.“

London’s Housing and Climate Crises Are on a Collision Course

“At least 19 of the 28 opportunity areas contain zones that have been flagged as vulnerable to flooding in a map compiled by London’s City Hall and Bloomberg Associates, the consulting arm of Bloomberg Philanthropies. The map takes into account rainfall patterns along with 16 other metrics that affect residents’ ability to deal with flooding, including socioeconomic status and proficiency in understanding English-language warnings.“

Statistical methods for healthcare regulation: rating, screening and surveillance

“We pay particular attention to the problem of simultaneously monitoring over 200000 indicators for excess mortality, both with respect to the statistical issues surrounding massive multiplicity, and the organizational aspects of dealing with such a complex but high profile process.“
Ok, it’s an academic paper, but if you’ve ever read the story of Harold Shipman’s murders, you’ll be excited to know that its lead author is Sir David Spiegelhalter.

quantum of sollazzo is supported by ProofRed’s excellent proofreading. If you need high-quality copy editing or proofreading, head to http://proofred.co.uk. Oh, they also make really good explainer videos.

Sponsors* casperdcl and iterative.ai Jeff Wilson Fay Simcock Naomi Penfold Steve Parks

[*] this is for all $5+/months Github sponsors. If you are one of those and don’t appear here, please e-mail me