Select Page

Making Maps Make Sense For The Reader

This week I had the utter privilege of speaking at Tiny Tableau Talks, an event run to provide public speaking opportunities for inexperienced speakers who happen to be Tableau enthusiasts. Being a Tableauphile with very few speaking events to my name, this was a fantastic opportunity for me, and Emma was very kind indeed to provide me with five minutes on a topic of my choice, and a forgiving audience of 50 or so.

Here I’ve reproduced my talk in an extended, written form, and also provided links to all the resources and references along the way.

Tableau is amazing at presenting data in maps. Particularly with the advent of spatial file integration since Desktop version 10.2, all it requires is a single click on ‘Geometry’ and you’re away. The problem is, however, that I feel Tableau has allowed us to become lazy cartographers – me included.

I want to use this opportunity to demonstrate, with good reason, that we should not take our audience’s mapreading skills for granted. But also that there is a relatively simple solution that goes a long way to bridging any gaps that is also very easy to apply using Tableau.

I, for one, love maps. I feel I have a good sense of direction and spent far too many hours driving up and down the motorways every other Saturday during my adolescence finding football grounds. I believe that I have a good sense of where places are and, as it turns out, I’m not alone. The Ordnance Survey conducted a survey around this time last year to promote its National Map Reading Week, and gathered information on how people rated their own map reading skills. They spoke with 951 men and 1066 women.

Only one in seven (14%) answered “no” to the question “Would you say you typically have a good sense of direction?” and only one in nine (12%) felt that their geographical knowledge of the UK was “poor”. More than half said “good” or “excellent”. That’s pretty good going, and suggests that people know their way around a map. We could play a game of “Where is Place X” with a blank map like this one and be confident that people would get pretty close.

Well, that’s just what the Ordnance Survey did. Their pool of 2,017 people were asked to locate a series of places on the map, ranging from pretty obvious to quite tricky – from London, Birmingham and Edinburgh down to Swindon and Aberystwyth. Asked to locate Manchester on that map, here’s the range of places that people clicked on:

Just 22% picked the correct square. Not even two-thirds picked a square in the 3×3 box surrounding the correct square. Outside of those based in the North West of England, fewer than a quarter of those from any other part of the country were able to correctly locate Manchester. Let’s look at another example, this time Aberystwyth. Here’s what the panel thought:

Yup, it’s just generically ‘somewhere in Wales’, it seems. Still, at least they knew where Wales was. You can check out the answers to this question for all sixteen locations at this website (click the ‘images’ dropdown menu in the top left to switch between cities).

Clearly from this evidence, we might need to recalibrate our expectations as to the ability of our audience to locate a place on a map. But that poses a problem for me, since earlier this year I created a viz mapping petition signature prevalence around the UK.

Further investigation into the OS survey results reveals that those aged 55 or over are significantly better at identifying locations in the UK, and that as you’d expect those living in the same region as the town or city they’re trying to locate are much better at it. So if my target audience is of working age and location-agnostic, I need to rethink how I present this to them. I chewed on this for a long time, complaining to colleagues about these figures but without any bright ideas as to what I or anyone should be doing about it. And then gradually the penny began to drop, and once I saw the solution I started to see it everywhere…

On the plane:

On the telly:

In the Wall Street Journal:

The key is in a smattering of visual cues, or signposts, provided on the map that allow the viewer to find their way around. They don’t need to have every town marked, but even a few locations pinpointed for them provide points of reference that they can use to find their way around. Compare the BBC weather map above with that from only a decade or so ago – here’s Ian McAskill in his heyday:

image from Daily Mail website

The difference is immediately apparent. In the older version of the weather map a relatively easy mistake could have significant consequences. Mistakenly placing yourself on the wrong side of Newcastle could be the difference between being on the receiving end of a blizzard or not. With the new style, you can locate a place that you’re familiar with and establish where your place of interest is relative to that signpost.

And now that we have a simple but effective tip, how can we apply it using Tableau? Well, by default Tableau won’t help you, but with just three clicks you can help yourself. When in a map view, a menu option ‘Map‘ is shown. Click here and find the option ‘Map Layers‘.

This provides a range of options, many of which you’ve probably seen before. However, tucked away at the end of the list of layers is ‘Place Names‘, an choice which is unticked at default. When the map has been sufficiently zoomed out, this option cannot be toggled, but if you move closer in this should appear:

For the avoidance of doubt, this video walks through the process in a matter of seconds, and demonstrates the impact on the map shown. This is the default map for the ‘Regional’ workbook which is shipped with Tableau.

About The Author

Mark Edwards

A statistician at heart, Mark’s approach is always numbers-led. Already visualising data in other side-projects, Mark was introduced to the world of Tableau in 2016, when he and Pablo started working together in financial services. A keen participant in social Tableau challenges, Mark is building his skills and appreciation of powerful visuals, discovering interesting and untapped data sets.

Leave a Reply