The little of visualisation design: Part 12

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns the use of interactive visual guides to help readers perceive values in a stacked area chart, specifically the classic ‘How Different Groups Spend Their Day‘ produced by the New York Times all the way back in 2009.

LoVD-Part12a

Wherever the user’s cursor is positioned on the chart (it sensibly snaps to only 10 minute intervals), a vertical guide is illustrated to help assist the eye towards the true height of each layer at that position on the chart. It also thickens the stroke width of the line for the specific layer they are focusing on reading.

LoVD-Part12b

This type of annotated chart assistance is helpful because judging the plotted value on an area chart at any given point is open to be misreading – as explained by this nice post.

The little of visualisation design: Part 11

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns colour choices for divergent scales, ie. when you are showing quantitative scales in two directions either side of a pivot/breakpoint (often zero or an average). In this analysis by the Washington Post, we see a choropleth map showing the gender gap in adult employment rates for children of low-income families, based on the county where they grew up.

LoVD-Part11a

The little issue here concerns the respective shade of purple/green used for the extremes – the highest value colour bands – which, inevitably, tend to represent the most interesting insights of such analysis. Whilst the underlying choice of a purple > green colour scheme makes loads of sense (and importantly will be colour-blindness friendly) the darkest shades of the purple and green are very similar upon scanning the map.

LoVD-Part11c

LoVD-Part11b

Sure, you *can* distinguish between these shades but it is perhaps more effort for the eye to perform this separation, quickly, than it needs to be. Maybe making the second darkest shade of purple/green representative of the highest value banding would make it a little easier.

The little of visualisation design: Part 10

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns the enhanced role of annotated captions. In the well-known work ‘Gun Crimes‘ by Periscopic, some of the main findings of analysis are provided in captions located beneath the main chart (this exists in both years of analysis, I’m focusing here on the 2010 view).

LoVD-Part10a

The smart feature here is that when you click on the respective captions, this provides a shortcut for the user by automatically applying the necessary criteria in the main chart above to formulate the associated view of the data that supports the finding described.

LoVD-Part10b

The little of visualisation design: Part 9

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns the juxtaposition of photo-imagery within a chart display. In this analysis by Vox 2,000 voters were surveyed to find out which of the 2016 candidates they were able to correctly recognise. The charts produced show the breakdown of the results and (logically) include a small image showing the faces of each candidate for reference.

LoVD-Part9a

One of the subtle design choices I like here is the deliberate cropping of the least recognised candidates’ images. Due to the smaller ‘correctly recognised’ values, the available space to accommodate the images is already diminished*. Rather than see this as a reason to relocate or perhaps proportionally shrink the images, notice how the faces of Kasich and O’Malley partially disappear below the axis, editorially reinforcing the findings of the analysis.

LoVD-Part9b

(* Incidentally, I probably wouldn’t normally repeat the value label word ‘Correct’ on each bar – once should be enough – and so removing this would create more space in practice).

The little of visualisation design: Part 8

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns creative thinking about the most meaningful orientation of a chart, in this case a scatterplot produced by FiveThirtyEight that shows analysis about the average favourability and unfavourability scores of a range of political names across the first few months of 2015.

LoVD-Part8a

Rather than leave the chart in its typical y-axis up/x-axis across layout, it is rotated clockwise by 45° to draw focus on the relative mapping of the plotted records according to four meaningful quadrant regions within the chart, indicating the general popularity and profile of each politician. In particular, and substantiated by colour, there is vertical significance in being more popular (above the line) and less popular (below the line).

Although some readers may find it a little more challenging, as a consequence of this rotation, trying to read off the coordinate values compared to the more standard approach (simulated below), the presentational emphasis on the meaning of the position rather than the position itself represents an astute editorial choice.

LoVD-Part8b

(See also this chart by Adam Pearce)

The little of visualisation design: Part 7

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns a clever (but quite simple) way of helping to guide a viewer’s eye towards changes on a map between two points in time. This project from the Guardian plots the results of the UK Election in 2010 compared to the projected results for 2015 using adjacent cartograms.

LoVD-Part7a

Only the constituencies that are forecasted to switch to a different political party are emphasised in colour – the exceptions – but it is the inclusion of the arrow and the shape outlining, as you hover over each area, to draw focus towards the related pairs that I find works ever so well.

LoVD-Part7b

Think of this as the equivalent of using your arm/hand/fingers to draw attention to a key feature in a display if you were presenting a visualisation in person.

The little of visualisation design: Part 6

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns the use of colour and specifically the restrictions caused by the universal application of ‘corporate’ colour palettes. There are benefits from applying consistent colours to facilitate brand recognition but sometimes this can cause unnecessary obstruction.

LoVD-Part6a

In the Gallup chart above, showing trends for how people identify with the US political parties, it would seem to be more logical to use the established associations of Republican = red, Democrat = blue and Independents = grey.

LoVD-Part6b

Whilst I appreciate it only requires the reader to learn 3 new colour associations, by not utilising the classic colour associations it does undermine the ‘available immediacy’ that this chart should have been able to offer.

The little of visualisation design: Part 5

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns a nice way, proposed by Tim Brock, of handling some of the concerns that people raise about the potential misleading effect of using of a non-zero value-axis origin in line charts. (You don’t need to start line chart value axes at zero but I’m not going to get into that here and now).

LoVD-Part5a

In an excellent article Tim discusses various approaches that help avoid giving the reader the sense that the ‘bottom’ of a chart should be read as a zero baseline. One solution that catches my eye is the use of a fading effect at the lower end of the chart. By decreasing the opacity for the colouring of the axis line, any tick marks and the value label, this makes visually clearer that the chart’s view is only presenting the observed range of values.

LoVD-Part5b

The little of visualisation design: Part 4

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of this next design concerns a really neat feature demonstrated in a project I reckon 99.9% of visualisation people are well familiar with: Gapminder. Specifically this is a new version of the classic tool, described as being pre-alpha (not sure really what that means).

LoVD-Part4a

The feature I want to point out here is the ‘DATA DOUBTS’ link positioned just below the chart. Data is rarely perfect. The journey it goes through from measurement, processing, statistical treatment and finally on to visualisation will often introduce a need for assumptions, application of counting rules, small inaccuracies, rounding errors etc. ‘Good enough’ is usually a necessary attitude to take otherwise we’d be frozen by the reluctance to publish any information.

LoVD-Part4b

What I love about this feature is that it acknowledges doubts about the data in a very open way: it is acknowledged front and centre, not scuttling around in the shadowy outposts of the site. The trustworthiness of a visualisation has to be of fundamental importance and so this kind of feature is so refreshing to see. Clicking on the link brings up a dialogue box with a brief comment about the reliability of the data, details about some of the necessary adjustments and assumptions and a link to read more in a blog post. Really very sensible and helpful for a reader to get this kind of contextual guidance so transparently before one launches in to forming meaning from the display.

LoVD-Part4c

The little of visualisation design: Part 3

This is part of a series of posts about the ‘little of visualisation design’, respecting the small decisions that make a big difference towards the good and bad of this discipline. In each post I’m going to focus on just one small matter – a singular good or bad design choice – as demonstrated by a sample project. Each project may have many effective and ineffective aspects, but I’m just commenting on one.


The ‘little’ of design today concerns a matter I’ve discussed before relating to the integration of graphic devices into written sentences, as demonstrated in this graphic about the quality of signings being made by football clubs in China signalling the possible emergence of a new power in the game.

LoVD-Part3a

Specifically, in this project, you will see how the ‘key’ explainers that would normally be segregated by a legend adjacent to the chart are instead incorporated into a written guide explaining what the various marks and attributes used in the chart represent. It might look simple but I’m sure it would have been quite a fiddly task making the sizing and alignment of the small graphics align seamlessly with the text and row size.

LoVD-Part3b

(Via a tweet from John Burn-Murdoch who created this graphic and authored the article)