eah, it’s that time of the year again, when Tableau hosts another round of the #IronViz competition. This time the topic is “Safari”: Animals & Plants.
After playing around with a couple of datasets like the IUCN list of threatened species, and the London Fire Brigade animal rescue records, I decided to have a close look to some of the most interesting animals in the world: Penguins!
Here below is my final submission, built around a map of the Antarctic Penguins colonies which uses a polar projection in Tableau. If you want to know I achieved it, with a mix of alteryx and mapbox magic, this post is for you.
Since 2010, the unstable situation in the middle east generated a massive inflow of migrants in all European countries. Through different routes, people illegally cross borders until they ask asylum and apply for refugee status in a chosen country. However, too many people don’t make the final destination, as they disappear – or die – in the mediterranean sea, or on their way to a better life.
I gathered data on refugee status applications in European countries, illegal crossing of European borders, and people who disappeared or died trying to reach Europe.
The final result aims to illustrate the hope-guided journey of people who risk everything for the simple hope of a better and safer life for themselves and their families.
No person is illegal.
I really appreciate the power of constructive criticism in the #DataViz space. Given the importance of the audience in the data visualization process, feedback from peers is a gift to ask for and to encourage.
When criticizing a visualization, there are both technical and ethical rules to be followed, and I think most of them can be found in two main sources:
1. The #MakeoverMonday project, that every week picks a data visualization to improve, asking participants to stick to the original data, and constrain their time to a hour (to seek simplicity);
2. A framework recently published by Stephen Few, where he purposed a structured way to assess the quality of a data visualization.
Some week ago, my good friend, colleague and great data wrangler Ben Moss published a new viz about flight delays.
As an Italian expat in London, I can easily qualify as an interested audience, and I definitely had questions to ask the dashboard.
This peculiar cocktail of interest, good data, and friendship made me willing to apply to Ben’s dashboard the rules of both the MakeoverMonday project and the framework outlined by Stephen Few.
Very often we use Tableau Public to showcase interesting visualizations, that we would never bring to our workplace.
Sankey charts, radial charts, etc. are all fun to experiment and great learning experiences, but unlikely to be the visualization of our choice in a business environment.
At the same time, we can’t really have those as main examples of BI dashboards, can we?
#VizBI wants to be a series of Tableau visualizations, based on dummy data and curated by Ravi Mistry (@Scribblr_42) and myself.
The idea is to model real-life business questions and get to actionable insights as quickly and accurately as possible.
The aim is to explore and share practices of Visual Business Intelligence, across different real-life-like scenarios.
Feedback will of course be encouraged; it is essentially the soul of the whole project: Finding different options and approaches to the same questions. Or even find different/more interesting questions.
Watch this space!
This week I spent my free time stretching and refreshing my (high school) knowledge of plotting exponential functions. Here below the end result: A visualization in Tableau of all the missions to Mars and the Moon so far, represented by curved lines and sorted by recency.
Last week, Buzzfeed released the output of a research they have carried on all the connections that Donald Trump holds with people and organizations.
I decided I wanted to visualize this dataset on a network chart, showing both the direct and indirect business connections of Trump. See below the end result (click for the interactive version).
Yeah, the title is provocative.
Of course, I have nothing against storytelling with data: it’s a great way to prove points and impact opinions. Experts in the data space proved how storytelling can be crucial when working with data: check out the great work by Cole Nussbaumer, or have a look at most of the visualizations published by Andy Kriebel.
However, as storytelling is usually widely celebrated, here I want to put a spotlight on the other face of the #DataViz coin: Exploration.
Explorative data visualizations are interfaces that allow users to ask questions to a dataset, and find their answers. They embody the concept of “Guided Analytics”, and are usually found in the form of dashboards of interactive and inter-related charts and controls.
As we strive for a more transparent, informed and data-driven society, I genuinely think exploratory data visualizations can be as powerful as storytelling.
In fact, because they don’t aim to change opinions, they empower people with visual instruments to check what’s going on themselves. Thanks to Data Explorers, people can make sense out of data when they are complex and – yes – big.
On Tuesday, the world wake up with a new POTUS, and the media were filled with data visualizations of the result. An electoral dataviz that never gets old is the “choropleth map”, like the one below:
However, despite popularity and ease of interpretation, this map can be told inaccurate for a very simple reason: it suggests to the eye a disproportioned result, based on State’s areas, not actual votes.
At the beginning of this year, I decided to get involved in the Tableau Foundation Service Corps: a group of people volunteering for charities and no profit organizations using Tableau. At the time, I was really excited to join this project, as to me it represented a great opportunity to put my skills at the service of good causes.
And this has really been the case. Since then, I have been involved in a project that let me feel really helpful, and for which I am so grateful to who entrusted me with its implementation: the Humanitarian Response Map Review.
The output of the project is the dashboard below: basically a visual way for decision makers on emergency fields to quickly filter the multitude of maps created by various organizations and find the ones that suit their needs.
Siccome questa settimana, per lavoro, mi e’ servito imparare come si produce un Marimekko Chart in Tableau 10 (istruzioni in fondo), ho deciso di applicare subito quello che ho imparato a un dataset interessante sul livello di soddisfazione dei cittadini europei riguardo i loro mezzi pubblici.
In questo modo ho anche prodotto il mio primo contributo al progetto social di data visualization #MakeoverMonday, coordinato dal mio coach e mentor Andy Kriebel e da Andy Cotgreave.
Premessa obbligatoria: Il Marimekko non e’ una Best Practice di Data Visualization, e si basa sull’assunto erroneo che piu’ informazioni si codificano in un grafico, piu’ questo sia informativo. Al contrario, un singolo grafico dovrebbe codificare un solo messaggio in modo tale che questo sia desumibile a una prima occhiata.
Nel grafico qui sotto la larghezza delle barre e’ proporzionale alla popolazione della citta’, mentre le barre stesse sono colorate in base alla percentuale di risposte per livello di soddisfazione. I colori delle barre sono facilmente identificabili anche da persone daltoniche, al contrario del piu’ classico – ed escludente – verde/rosso.
Indovinate a quale paese appartengono le barre piu’ gialle?
Clicca per esplorare la dashboard