Favourite chart types

Conversations with Data: #8

Do you want to receive Conversations with Data? Subscribe

Conversations WData-header-1

Hi there! Can you believe it’s our 8th edition of Conversations with Data? To celebrate, we’re bringing you eight of your favourite chart types.

But, before we begin, it’s important to bear in mind that the best chart type the depends on your dataset.

As Birger Morgenstjerne reminded us, "charts are all about making your data/information accessible, effective and powerful to your audience. A common mistake is not paying attention to what you’re trying to communicate. Visualising data is the discipline of what you want to communicate (data/story) and how to tell it (the visualisation). Always ask, does the chart support and make your story shine through?"

If you’re in doubt about which chart to choose, Mona Chalabi, Data Editor at the Guardian US, suggests: "visualise your data in two different charts and ask a friend (or better yet, your mum) which is the least confusing to them."

Okay, now, let’s chat charts!

Charts you like

1. Timelines

"I love enriched timelines because they help to provide valuable context to the data shown. In these days of avalanches of data, providing context and connecting pieces of information can help to transform them into accessible knowledge. As with most of charts, labelling is a critical part of the process, many times neglected by chart creators.

For example, these timelines summarising the lives of two well-known artists: Michael Jackson and Paul Newman. They provide a landscape vision into their existence just after they die. They also facilitate insights into their artistic careers thanks to the connections established between professional-personal dimensions. The reader can get a sense of their richest creative periods because their achievements are shown in a temporal context." - Álvaro Valiño, Publico

Paul newman career 1340 c

Álvaro's Paul Newman timeline for Publico.

2. Grid chart

"When plotting multiple data series with very similar or very different values, a single chart may obscure more than it reveals. One alternative way to visualise multiple series that overcomes this is called the 'small multiple' technique, or 'grid chart'. This technique splits each series across individual charts in a grid and makes it easy to read and compare each series." - Mustapha Mekhatria, Highsoft

3. Pictorial small multiple

"My favourite chart type is the 'pictorial small multiple'. Each of its marks is a miniature canvas. At a glance, illustrated marks anchor the viewer to the chart's topic. In focus, each mark is an opportunity for detailed inspection, and comparison between marks. The arrangement of the marks provides an order for higher meaning to emerge. Together, many levels of understanding are possible." - RJ Andrews, author, Info We Trust: How to Entertain, Improve, and Inspire the World with Data


A pictorial small multiple devised by Bashford Dean and drawn by Stanley Rowland.

4. Stacked bar chart

"Line graphs and bar charts are really the first go-to charts I use. Almost all of the time, you can quickly get your data organised and visualised using one of these two charts, but my favourite is probably the stacked bar chart. Any time someone tries to use a pie or doughnut chart, I convert it over to a stacked bar. Horizontal or vertical, it doesn't matter. Representing a portion of a whole should be done using rectangular boxes, not radial degrees like a pie chart." - Brian Suda

5. Flow map

"One type of chart I particularly like is the flow map. The arrows easily communicate that movement is taking place in this chart. Although they’re not great for accurately displaying values, flow maps are much better at giving a generating view of the amount of movement of good or people taking place over a geographical region." - Severino Ribecca, The Data Visualisation Catalogue

6. Beeswarm chart

"Although I don't truly have a favourite chart type, I do find myself using a beeswarm technique to position my data quite occasionally. I like how this technique still allows you to show the data at a very detailed level (each datapoint separately), while simultaneously showing how all of the data is distributed, typically across one axis. And you can play with the 'datapoints' as well, you can colour them, resize them or give them some other visual mark based on a variable that shows even more insights, and therefore context, about the data.

Furthermore, instead of one datapoint, you can also apply a beeswarm to 'small multiples'. Where each 'mini chart' is positioned on the screen by using some aggregate value of the mini chart (an average for example). So I love the combination of versatility, the level of detail it can show, and the general visual appeal it gives to a data visualisation." - Nadieh Bremer


Using the technique in The Top 2000 loves the 70s & 80s.

7. d3-force

"I don't necessarily have a favourite or even a go-to chart type, but I do have a favourite D3.js module. d3-force is an implementation of the force-directed graph drawing algorithm, an algorithm for calculating the positions of nodes in a network graph. I love it for its flexibility: I can calculate node-link graphs with it, but I can also calculate beeswarm plots with it. I can cluster nodes into groups with it, and I can make the nodes fly around the screen like sprinkles. I've thought up a visual for my data and gone to d3-force to make that idea into reality. And that's why I don't have any particular chart type I love because there isn't any one chart that meets even most of my needs, but there are tools that I love because they help me translate what's in my brain to visuals on the screen." - Shirley Wu

8. Column chart

"My favourite chart type is the humble column chart, here annotated for added effect. It triggers the viewer’s most intuitive comparative understanding – comparing heights of items – and is therefore extremely simple to process cognitively. Simplicity is power, especially in communicating crucial facts such as those about our rapidly changing planet and the impact we’re having on all of its inhabitants." - Mikko Järvenpää, previously CEO, Infogram


Mikko's Column chart, illustrating the CO2 impact of protein choices.

Our next conversation

Like all journalism, charts have the potential to mislead if they aren’t checked correctly. So, for our next conversation, we thought it’d be fitting to turn our minds to the work of fact-checkers.

Ever wondered how data is used in fact-checking? We’ve got a diverse and global team of fact-checking organisations ready to answer your questions.

Happy summer!

Madolyn from the EJC Data team

subscribe figure