ADVICE, EXPERIMENTS AND THING'S WE'VE FOUND INTERESTING IN THE WORLD OF DATA STORYTELLING AND VISUALISATION
ADVICE, EXPERIMENTS AND THING'S WE'VE FOUND INTERESTING IN THE WORLD OF DATA STORYTELLING AND VISUALISATION
Data science roles and the importance it plays in modern businesses continues to grow. It also continues to get more technical.
There’s a lot of training courses and support networks out there to support the technical side of growth, but less support out there for the softer side of things around communication, influencing and stakeholder relationships.
As data becomes bigger and more complex, it needs to be balanced by approaches that make it more accessible and understandable. This is will allow it to achieve its full value.
At Finding Stories our aim is to help individuals and teams find that balance. Our goal is to help bright and technically minded individuals and teams become more rounded and accessible, and in doing so, have more influence and help support career growth.
We’re starting out by supporting on key skills around Data Visualisation and Data Storytelling but over time will grow into areas such as influencing, facilitation and leadership - all key skills that are needed to drive change and grow careers.
Our training courses are designed to be accessible, interesting and relevant, with the learnings easy to apply in your day to day role.
Our courses are designed for individuals and teams at the start of their data journey, or for individuals ready to push on and elevate themselves above the crowd.
We’re here to help you find your story and help you learn, develop and grow as a result.
Click here to find out more about our growing training offering.
During this time of social unrest and spotlight on equality and fairness, I was reminded of an old West Wing episode called ‘The Crackpots and These Women’. In the episode, Leo McGarry runs his ‘Big Block of Cheese’ day where he asks his senior staff to have face-to-face meetings with those people representing organisations who normally have a difficult time getting their attention.
One such group is the Organisation of Cartographers for Social Equality. This group argue that the 2D map of the world that we most commonly use, known as the Mercator projection, is unfair and biassed towards the old European powerhouses of the past and penalises the true size of ‘The Third World’. The map was designed as a navigational tool, enlarging areas at the poles to create straight lines to aid navigation. From what I understand, it was the right tool for that job, but an inaccurate and unfair one for representing the world as we know it for maps in our schools or online tools such as Google Maps.
Take a watch of this clip and the come back to me:
Now this isn’t a new topic, and is one that many people have covered in other blog posts and visualisations, but I wanted my go at presenting the story of how this map is an inaccurate portrayal of our planet and how it could be one more contributing factor towards unconscious bias against different countries and continents on a global scale.
Let's start at the beginning, where's the equator??
Before writing this post I'm embarrassed to say I can't really remember ever critically looking at this map to challenge its role as a visualisation tool. A version of it has been on our son's wall for about two years and I've been alive for 38 years so I've had my chance!
Now I teach that a good visualisation should be a fair and truthful representation of the data it is portraying, and it doesn't take much to start picking holes in this one.The first obvious thing for us to ask is 'where is the equator'. The equator divides our globe into two equal halves: The Northern Hemisphere and The Southern Hemisphere. It's therefore fair to assume that the map is also equally divided into two equal halves. But if I follow this assumption and go ahead and draw a straight line through the dead centre of the map, this is what I get:
So it turns out Spain, India, China and the majority of the United States are south of the 'equator' on this map! Not a great start then for providing an accurate portrayal of our world for our classrooms.
Moving on, so how big is Greenland really?
As mentioned in the West Wing episode, Greenland on this projection looks roughly the same size as Africa and Alaska appears three times as big as Mexico. Now, based on what was said in the West Wing episode, the location of the equator and in the knowledge that the poles are enlarged on this map to help create straight lines, I wanted to discover how out of proportion different countries were and is there indeed a further bias to the Europeans on this map.
Finding data that showed the relative size of all the different countries on this projection was actually pretty hard until I came across the Pixel Map Generator at amCharts. This excellent tool allowed me to take the map and break it up into a large number of hexagonal pixels, which I could then count up and use to create my relative sizes.
So for example, here's Greenland (lots of pixels):
And then Egypt for comparison:
In total, the land mass of the world was broken up into 15,348 pixels, with Russia, Canada, Greenland and the United States making up around 60% of all available pixels:
Table 1: The number of pixels that represent the top 10 countries by size in Mercator Projection.
So that takes care of Mercator projection, what about the real world.
Using data from the World Bank, I was able to take recent views on the land mass of each country in square kms to find the real world sizes I needed to compare against. It was at this point as well that I decided to just concentrate on those countries where I had a decent enough number of pixels to be confident(ish) of their relative size in the Mercator projection (so dropping those with a pixel size of less than 20) and I also had to drop a couple of countries which didn't have a land mass listed in the World Bank data.
Finally, using my final list of 69 countries and using the relative sizes of each in my Mercator projection data, I was able to convert my pixels from my original chart to representative square kms to match my real world data so I had a fair comparison between the two. I ended up with the following table:
Table 2: My converted land masses of the Mercator projection and relative size ranks.
From the table, it is easy to see some notable differences in scale between the projection and the real world, but how could visualisations tell a clearer story of unfairness and inequality?
Let's start by plotting the relative size ranks of each country in the Mercator Projection and data from the real world.
From our plot and by starting to annotate the outliers, we can start to see that Greenland and other Northern European countries start to stand out as being 'too big' when using the Mercator projection. And on the other side of the divide, Southern Hemisphere countries such as Niger and Kenya start to appear 'too small'.
And when we look at the ratios between what a country land mass actually is and how it is portrayed relatively speaking in the Mercator Projection, we see that countries in the continents of Africa, Asia, South America and Oceania are almost all undersized (and potentially devalued in the rest of the world's eyes.)
So what does our world look like?
From this analysis it's clear that it doesn't look a whole lot like the Mercator Projection. If we are looking to share the relative size, shape and distribution of the countries of our world, so we can talk confidently and truthfully with our children about the world in which we live, we should be looking at a map that looks something more like this:
If ever it felt like the right time to reassess and retune our world view, now would be that time.
You can find my infographic of this analysis in the subsequent post.
I was listening to a podcast the other day on my daily exercise run and it started talking about Animal Crossing (a game that launched on the Nintendo Switch a few weeks back) and it described it as possibly one of the most important games of all time. It described it as a game that, at this very moment, has the potential to save lives and help with people’s mental wellbeing.
For those not in the know, Animal Crossing is a life simulation where you move to a new island and start a new life, making friends, catching fish, planting flowers and generally enjoy being outside. Especially at the moment it is extremely refreshing and gives you a different world to live in that is pleasant and rewarding. For our family while being stuck indoors we have invested a lot of time into our island paradise and has brought us substantial joy at a time when we all need it.
So how does this link to data visualisation?
One of the in-game loops is the idea of the ‘Stalk Market’. On a Sunday morning, a small orange Boar named Daisy Mae turns up selling turnips for a set price. And then throughout the week, Timmy and Tommy Nook, who are a couple of Racoons that run the local shop, (stick with me) will then buy Turnips off you for a set price. They change their prices each day, once in the morning and once in the afternoon. And if you fail to sell your turnips by midnight on Saturday they rot and you’re left with nothing.
This loop has taught our children about the concept of “buy low and sell high”, but we’ve also taken the opportunity to learn about line charts at the same time.
As per Cole Nussbaumer Knaflic’s book “Storytelling with Data”, line graphs are most commonly used to plot continuous data. Because the points are physically connected via the line, it implies a connection between the points. Often, our continuous data is in some unit of time: days, months, quarters, or years
We decided a line chart was perfect for tracking the daily movements of Turnip Prices so we could compare how much we paid for turnips against what we could sell them for on any given day to make a profit.
We plotted the daily ‘sale’ price in blue and then the weekly ‘buy’ price with orange. After tracking prices for a few weeks the chart isn’t perfect:
1. The scale isn’t great on the y axis (more complicated but a logarithmic scale could have been better),
2. and I think there could be a better way of plotting the Daisy-Mae prices so it is easier to see what is higher or lower each day.
However, having said that, it has been a great opportunity to introduce the kids to the concept of a line chart and plotting a continuous series over time.
Now, back to panicking over the 180,000 turnips I’m currently sat on waiting to trade…
Here in the UK, we're being asked to stay two metres away from people outside our household to reduce the risk of spreading or contracting Covid-19.
But just how big is two metres?
For children (and adults) the concept of what two metres might be could be quite abstract. So when telling our children to stay two metres away from others, do they really know what we're asking them to do?
Could data visualisation help make things clearer? And if it's clearer, would it help our children and others maintain that magical two metre barrier to others. Let's find out.
1. What is a Data Visualisation?
In my intro I asked whether data visualisation could make it clearer. But in saying this you might be asking yourself, "how would a chart help our children here?" But data visualisation is more than just bar charts and scatter plots.
In Alberto Cario's book 'The Truthful Art', he describes visualisation as "any kind of visual representation of information designed to enable communication, analysis, discover, exploration etc."
So data visualisation is more than just charts. As per Alberto's definition, I wanted to find a visual representation that is more relevant and relatable to a 6 year-old. But before I share that, let me talk through an experiment we created this morning to identify if we had a problem in the first place.
2. Do we actually know how long two metres is?
Here are the results:
Sarah (38) was almost bang on but me (38) and Astrid (8) under-estimated, where as Dylan (6) and Matilda (3) over-estimated.
So as a family of 5, we are all over the place in understanding how big two metres is. So how could we fix this with Data Visualisation?
3. Introducing Non-Standard Units for Measurement.
We typically measure objects and distances using standard measurement units, such as kilograms, litres and metres. These are great for precision, but if you don't know what a metre is in the first place then it can be pretty meaningless.
So what about non-standard units then?
Non-standard units are used by children in early years of school to introduce the young to the concept of measuring without having to read any scales. Rather than reading scales, they help children understand the concept of heavier, lighter, longer, shorter etc. by comparing objects they already know and understand. So for example:
I believed we could use the same principle to better understand what is two metres, or more specifically how far away do we need to stand from others to be safe?
4. Visualising two metres.
And this is a selection of what we came up with:
(and perhaps best) 2 pairs of arm lengths.
Now you might think that 46 lego mini-figures is equally as unhelpful as saying two metres to someone that can't accurately visualise what a metre is, but kids know lego and 46 is a pretty big number so they'll likely think I have to leave a decent space to fit them all in.
But even better, we found that when we used 'big non-standard units' such as people and scooters, it was much more understandable and something they can better visualise in their heads when we're asking them to keep a safe distance.
So what we ended up with was a visualisation (or series of visualisations) that were more personal, understandable and, most importantly, allowed for a more effective means of communicating how far apart we need to stand from other people when out and about.
And if we all do that, we have the potential to keep safe and save lives.
Ergo, Data Visualisation can indeed save lives.
With the kids currently stuck at home due the restrictions around Covid-19, we've taken the opportunity as a family to start learning about a number of new topics, including Data Visualisation.
Data Visualisation has been a big part of my career over the past 15 years so it has been fun to share some of that with the children as we have explored topics such as:
1. Categorising teddies by colour and visualising first with Lego and then as a Bar Chart.
2. Using Unit Charts to visualise if we have enough Fruit and Veg in the house.
3. Using Slope Charts to understand if chocolate does indeed make you run faster.
We're trying to do two of these a week while the children are out of school and I'm sharing the examples and results on my LinkedIn Page. Click here to head over to my LinkedIn to follow along and hopefully share with your children. If you create any visualisations of your own please feel free to share with me to give me more inspiration and to share the Data Visualisation love!