Skip main navigation

These are a few of my favourite visualizations

There are so many different kinds of visualization! In this video, Lovisa and Jeremy review some more exotic diagrammatic data representations.
6.2
JEREMY: Hi, Lovisa. I thought we could talk about some of our favourite data visualisations today. This is a word cloud here. And this is a representation of a piece of text, a corpus of text. And the size of each word indicates how frequently that word appears in the text. So this is actually a story. I quite like this story. I want you to guess what the story is by looking at the text. What do you think? Have a look at the words and see if you know what the story is. So what are the big words?
37.4
LOVISA: I do notice that there is one big word of “Alice.”
40.4
JEREMY: “Alice.” Yes. That’s good.
42.3
LOVISA: Could it be Alice in Wonderland?
44.3
JEREMY: Yeah. That’s right. It’s Alice in Wonderland. Look. Can you see there’s the Hatter. And, oh, dear, the turtle. There’s the queen. She’s the baddie. There’s the king. He gets a bit confused. But you’re right. It’s Alice in Wonderland. So I generated these from Lewis Carroll’s text from an online copy on Project Gutenberg to generate this word cloud visualisation. So I really like this. It’s a very quick overview of the story, and the main characters, and the main concepts in the story. You’ve got a visualisation here which you sent me. Can you explain this one to me, please?
77.4
LOVISA: I do. Well this one represents the poverty rates in Edinburgh.
84.8
JEREMY: Oh, yes. I can see the shape of the city there. Yes, that’s right.
88.6
LOVISA: And it is almost like a heat map except, instead of using colour to indicate magnitude, so how high the poverty rate is, we use the height.
100.7
JEREMY: Yes. Good. I can see that. So in the middle of the city, the areas are quite low. But then on the outside, they’re much higher. What are you actually measuring here? Sorry.
115
LOVISA: Poverty rates.
115.8
JEREMY: Poverty rates. OK. And then towards the edge of the city, in the suburbs, this higher– OK. So higher levels of poverty. That’s really interesting.
123.1
LOVISA: Yeah. Along the coast they begin to spike.
125.4
JEREMY: OK. Yes. Yes. Yes. What’s the big bar here?
130.1
LOVISA: The big bar is almost like an axis. It’s there for scale. So it represents 100%. And anything that is a third of it could be 33% poverty rate.
140.8
JEREMY: OK. Yeah. Yeah. That’s a really striking visualisation, isn’t it? Yeah. And it’s nice that you get the kind of geographical correspondence there. You kind of see the city. Very good. What do you call this kind of visualisation?
153.5
LOVISA: Well, I have checked the pronunciation of this. And apparently it’s a “choropleth.”
158.1
JEREMY: A choropleth.
159.2
LOVISA: I do believe it’s Greek.
160.7
JEREMY: Ooh. Very nice. And there’s one more visualisation to look at. I’ll just get it up on the screen here. Wow. So I really like this one. This visualisation here is very historical. I mean, the first one, that was literary I suppose. Alice in Wonderland. And the second one is geographical within the Edinburgh city. This is very much a historical narrative here. So what we’ve got is Napoleon’s French armies starting off over here. It was a very big army there. And as they travelled towards Moscow in 1812, the army gets smaller and smaller and smaller until eventually– how many people got in Moscow? There were 100,000 soldiers, I think.
196.5
And then they retreat and come back from Moscow in the snow. And the line gets thinner and thinner because there are fewer and fewer soldiers until eventually there were only 10,000.
205.3
LOVISA: Wow. It’s so clean and stylish. What programming library was used to generate this?
212.1
JEREMY: Well, you know what? This visualisation was drawn by hand. A French information visualisation expert from the 1800s called Charles Minard generated this drawing himself and didn’t use any software at all. You can use programs these days to draw this kind of dia– it’s called a Sankey diagram nowadays. And there are JavaScript and Python libraries to draw Sankey diagrams. But this was entirely drawn by hand.
244.6
LOVISA: Wow. That’s incredible.
246.5
JEREMY: There’s more data on it as well, actually. So you can see as they’re retreating, the army’s retreating here. You’ve got the temperature that was measured. And you see it starts off at zero degrees, freezing point there in Moscow. And it goes down as low as minus 30 here. So they were certainly in extreme conditions as they were retreating through Russia. OK. Great. So quick question. Now that we’ve looked at visualisations, what do you think really is the essence of a good data visualisation?
278.8
LOVISA: Well, I think as much as we like things that are clean and clear, what also matters is that it’s visually appealing. So it’s pretty.
288.2
JEREMY: It looks beautiful. Yeah. I think that’s true of my Alice in Wonderland story there. I think that looks really nice. And it’s bright and colourful. I don’t think the colour actually means anything. It’s just the size that means something in terms of the word frequency. But, yes. It does look very visually appealing. Good. What else makes an effective data visualisation?
309.3
LOVISA: Choosing how to represent your quantities very carefully. So what dimensions are you going to use to represent magnitude, poverty rate. You use height, you use colour. Think carefully about it.
322.8
JEREMY: OK, great. Anything else? If you were going to choose an appropriate data visualisation technique?
329.4
LOVISA: I think ideally a visualisation is almost like a narrative in itself. It’s a full story that you can read. Self-contained.
337.8
JEREMY: And that’s definitely true for this visualisation of the progress of the French armies in Russia here, isn’t it? You can see the story just from the graphic. Yeah. Great.

A wide variety of different data visualizations is possible. In this video we explore three more exotic graphical data presentations.

Jeremy shows a word cloud, which is a graphical summary of a corpus of text. The size of a word indicates its relative frequency in the text – the most commonly occuring word should be the largest. Often stop words like ‘the’ or ‘and’ are removed before a word cloud is generated.

Lovisa describes a cloropleth, which indicates quantitative data for different geographical areas – in this case different districts of Edinburgh City region. This is a nicely intuitive visualization, but requires some quite intricate design.

Finally, Jeremy presents a Sankey diagram which indicates the size of population subsets over time as the population splinters into smaller groups. The original Sankey diagram charted Minard’s visualization of Napolean’s military campaign in Russia.

There are online tools or Python libraries available to generate custom diagrams in these formats. Check out sankeymatic.com or wordclouds.com, for instance.

What is your favourite visualization technique? Please share your ideas in the discussion section.

This article is from the free online

Getting Started with Teaching Data Science in Schools

Created by
FutureLearn - Learning For Life

Our purpose is to transform access to education.

We offer a diverse selection of courses from leading universities and cultural institutions from around the world. These are delivered one step at a time, and are accessible on mobile, tablet and desktop, so you can fit learning around your life.

We believe learning should be an enjoyable, social experience, so our courses offer the opportunity to discuss what you’re learning with others as you go, helping you make fresh discoveries and form new ideas.
You can unlock new opportunities with unlimited access to hundreds of online short courses for a year by subscribing to our Unlimited package. Build your knowledge with top universities and organisations.

Learn more about how FutureLearn is transforming access to education